(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Journal of agricultural research"

){T.07S 




V; 





AGRICULTURAL 
RESEARCH 



Volume XXIl 

OCTOBER I— DECEMBER 3, 192 1 






PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE 

WITH THE COOPERATION OF THE ASSOCIATION 

OF LAND-GRANT COLLEGES 



WASHINGTON, D. C. 



CONTENTS 

Page 

Occurrence of Quercetin in Emerson's Brown- Husked Type of 
Maize. Charles E. Sando and H. H. Bartleit i 

Biological Analysis of the Seed of the Georgia Velvet Bean, Stilzo- 
lobium deeringianum. BarnETT Sure and J. W. Read (15 text 
figures) 5 

Effect of Soil Temperature upon the Development of Nodules on 
the Roots of Certain Legumes. FrEd reuel, Jones and W. B. 
TiSDALE (4 text figures; 3 plates) 17 

Influence of the Period of Transplanting Western White Pine Seed- 
lings upon Their Behavior in Nursery and Plantation. E. C. 
Rogers (7 text figures) 33 

A Dryrot Canker of Sugar Beets. B. L. Richards (6 plates) .... 47 

Comparative Vigor of Fj Wheat Crosses and Their Parents. Fred 

Griffee 53 

Temperature and Humidity Studies of Some Fusaria Rots of the 
Irish Potato. R. W. Goss (2 plates) 65 

Blackleg Potato Tuber-Rot under Irrigation. M. Shapovalov and 

H. A. Edson (6 plates) 81 

Microscopic Study of Bacteria in Cheese. G. J. HucKER (i plate) . 93 

Further Studies on Relation of Sulphates to Plant Growth and 

Composition. Harry G. Mii^l,er loi 

Soybean Mosaic. Max W. Gardner and James B. Kendrick (2 

plates) HI 

Influence of the Plane of Nutrition on the Maintenance Require- 
ment of Cattle. F. B. Mumford, A. G. Hogan, and W. D. 
Salmon 115 

Turnip Mosaic. Max W. Gardner and James B. Kendrick 

(i plate) 123 

Hydrocyanic Acid in Sudan Grass. C. O. Swanson 125 

Nutrient Requirements of Growing Chicks: Nutritive Deficiencies 
of Corn. F. E. Mussehl, J. W. Calvin, D. L. HalberslEben 
and R. M. SandsuEdt (ii text figures) 139 

Aecial Stage of the Orange Leafrust of Wheat, Puccinia triticina 

Eriks. H. S. Jackson and E. B. Mains (i plate) 151 

A Transmissible Mosaic Disease of Chinese Cabbage, Mustard, 
and Turnip. E. S. SchulTz (4 plates) 173 

Flora of Corn Meal. Charles Thom and Edwin LeFevre 179 

, Hopkins Host-Selection Principle as Related to Certain Ceramby- 

u. cid Beatles. F. C. Craighead 189 

in 



IV Journal of Agricultural Research voi. xxii 

Page 

Notes on the Organic Acids of Pyrus coronaria, Rhus glabra, and 

Acer saccharum. Charles E. Sando and H. H. BartlETT. . . 221 

FertiHty in Shropshire Sheep. Elmer Roberts 231 

Relation of Soil Temperature and Other Factors to Onion Smut 

Infection. J. C. Walker and L. R. Jones (2 text figures; 3 

plates) 235 

A Physiological Study of Grapefruit Ripening and Storage. Lon 

A. Hawkins (i text figure) 263 

Absorption of Copper from the Soil by Potato Plants. F. C. 

Cook 281 

Pale Western Cutworm (Porosagrotis orthogonia Morr.) J. R. 

Parker, A. L. Strand, and H. L. Seamans (i text figure; 4 

plates) 289 

Biology of Embaphion muricatum. J. S. Wade and Adam H. 

Boving (3 text figures; 2 plates) 323 

Genetic Behavior of the Spelt Form in Crosses between Triticum 

spelta and Triticum sativum. Clyde E- LeighTy and Sarkis 

Boshnakian (3 text figm-es; i plate) 335 

Plum Blotch, a Disease of the Japanese Plum Caused by Phyllo- 

sticta congesta Heald and Wolf. John W. Roberts (2 text 

figures; i plate) 365 

A Comparison of the Pectinase Produced by Dififerent Species of 

Rhizopus. L. L. Harter and J. L. Weimer (2 text figures) .... 371 

Hemotoxins from Parasitic Worms. Benjamin Schwartz 379 

Ash Content of the Awn, Rachis, Palea, and Kernel of Barley 

during Growth and Maturation. Harry V. Harlan and 

Merritt N, Pope (5 text figures) 433 

Temperature Relations of Stone Fruit Fungi. Charles Brooks 

and J. S. CoolEy (24 text figures) 45 1 

Transportation Rots of Stone Fruits as Influenced by Orchard 

Spraying. CharlES Brooks and D. F. Fisher (6 text figures) . 467 
Storage of Coniferous Tree Seed. C. R. Tillotson (2 text figures) . 479 
Susceptibility of the Different Varieties of Sweet Potatoes to 

Decay by Rhizopus nigricans and Rhizopus tritici. L. L. 

Harter and J. L. Weimer 511 

Index 517 



Vol. XXII OCXOBER 1, 1921 No. 1 

JOURNAI. OF 

AGRICULTURAL 

RESEARCH 



CONXENXS 



PagQ 



Occurrence of Quercetin in Emerson's Brown-Husked 
Type of Maize -------- i 

CHARLES E. SANDO and H. H. BARTLETT 

(Contribution from Bureau of Plant Industry and University ot Michigan) 

Biological Analysis of tlie Seed of the Georgia Velvet Bean, 
Stilzolobium deeringianum - - - - - - 5 

BARNETT SURE and J. W. READ 

(Contribution from Arkansas Agricultural Experiment Station) 

Effect of Soil Temperature upon the Development of 
Nodules on the Roots of Certain Legumes - - - 17 

FRED REUEL JONES and W. B. TISDALE 

(Contribution from Bureau of Plant Industry and University of Wisconsin) 

Influence of the Period of Transplanting Western White 
Pine Seedlings upon Their Behavior in Nursery and 
Plantation - -- - - - - - -33 

E. C. ROGERS 

(Contribution from Forest Service) 

A Dryrot Canker of Sugar Beets ----- 47 

B. L. RICHARDS 

(Contribution from Utab Agricultural Experiment Station) 



PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



WASHINOa^ON, D. C. 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 
KARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 
J. G. LIPMAN 

Dean, Sfafe College of Agriculture, and 
Director, New Jersey Agricultural Experi- 
ment Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of Ento- 
mology and Economic Zoology, Agricuh 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS, 

Dean, School of Agriculture, and Director; 
Agricultural Experiment Station, Tb4 
Pennsylvania Staf* College 



All correspondence regarding articles from the Department of AgricultiU'e should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipnaan, New Jersey Agricultural Experiment Station, New 
Bnxnswick, N. J. 






J01M£ OF AGRICiriAL RESEARCH 

Vol. XXII Washington, D. C, October i, 1921 No. i 



OCCURRENCE OF QUERCETIN IN EMERSON'S BROWN- 
HUSKED TYPE OF MAIZE 

By Charles E. Sando, Junior Chemist, Office 0/ Physiological and Fermantation Investi- 
gations, Bureau of Plant Industry, United States Department of Agriculture, and H. H. 
BartlETT, Department of Botany, University of Michigan, Collaborator, Office of 
Physiological and Fermentation Investigations, Bureau of Plant Industry, United 
States Department of Agriculture ' 

In connection witli the genetical studies of pigmentation in maize 
which have been carried on for several years at Cornell University by 
Prof. R. A. Emerson and his students, a plan was made for the writers 
to collaborate in parallel biochemical studies in the isolation and identifi- 
cation of the pigments from material of known genetical constitution, 
the latter to be furnished as a by-product of the Cornell experiments. 

As a beginning in the work it seemed desirable to undertake a study 
of the character pair purple versus brown. These are two of the general 
plant colors recognized in the Cornell experiments, the series running as 
follows: I, purple; II, sun-red; III, dilute purple; IV, dilute sun-red; 
V, brown; and VI, green. Full details of the genetic relations of these 
types are being published by Enierson,^ to whose memoir the reader 
should refer for full details. Here it may suffice to say that purple is a 
color type which is uncommon in cultivation and infrequent in experi- 
mental cultm-es. It is distinguished from other types by the fact that 
some purple color is developed even in seedlings grown in the dark. At 
maturity nearly all parts are more or less purple, including the culm, the 
brace roots, all leaf sheaths, the husks, even the inner ones, the cob, and 
the staminate inflorescence. In intensity of coloration purple stands at 
the head of the series of color types. Material of this type was furnished 
by Prof. Emerson to Dr. John W. Calvin, of the University of Nebraska, 
before we took over the general problem, and we have therefore confined 
our attention for the present to the brown type, awaiting a report on his 
study of the purple type from Dr. Calvin. 

• In connection with our work with maize we have received numerous favors from Prof. R. A. Emerson 
and Dr. E. G. Anderson, to whom we tender our best thanks. 
^ ' Emerson, R. A. the genetic relations of generai, plant colors in maize. N. Y. Cornell Agr. 
. ^ Exp. Sta. Mem. 39, 156, p., n col. pi. 1921. 

or: _______ 

■* "Journal of Agricultural Research, Vol. XXII, No. i 

j^ Washington, D. C. Oct. i, 1921 

^^ zl. Key No. G-244 

i 

CD 



Journal of Agricultural Research 



Vol. XXII. No. I 



A few preliminary tests of the purple pigment convinced us that it was 
an anthocyanin of which the nonsugar portion was of the same general 
group as cyanidin, isolated by Willstatter ^ and his students from several 
sources, including the cornflower, Centaurea. His proof of the easy 
chemical transition from the anthocyanin to the flavonol series led us to 
look for a member of the latter series in the brovvTi maize. In accordance 
with expectations, we were able to isolate a glucosid of quercetin. This 
discovery makes it seem exceedingly likely that the anthocyanin of the 
purple type is a corresponding glucosid of cyanidin. 

The brown color type is a still more unusual one than the purple. 
It first appeared in Emerson's ^ cultures as a segregate in the second 
generation of the cross purple X green and is unkno%vn outside this 
series of cultures. Seedlings and young plants are wholly green. As the 
flowering period approaches, a brown color appears in the lower sheaths, 
and at flowering time the culm, sheaths, husks, and staminate inflores- 
cences are brown. Light is not essential to the development of the 
color. Our material of the brown type consisted of husks. 

ISOLATION AND IDENTIFICATION OF THE FREE QUERCETIN 

Ground husks were extracted in a large Soxhlet apparatus with redis- 
tilled 95 per cent alcohol for about 72 hours, and the alcohol was evapo- 
rated off in vacuo. The thin aqueous sirup was filtered from tarry 
matter and the filtrate boiled with animal charcoal. By shaking with 
ether it gave an etherial solution containing a free (nonglucosidal) fla- 
vone which proved to be quercetin. The ether was evaporated off, and 
the residue, after being dried in a desiccator, was extracted in a paper 
thimble, first with benzene, to remove tarry colloids, oils, etc., and 
finally, for a short time, with ether. The latter solvent dissolved part 
of the quercetin but left the bulk of it in the thimble. This portion was 
dried and acetylated for an hour with anhydrous sodium acetate and 
acetic anhydrid. After purification the acetyl derivative was quanti- 
tatively hydrolyzed with sulphuric acid in glacial acetic acid. The reac- 
tion mixture was diluted and the recovered quercetin was washed with 
cold water. The results are given in Table I. 

Table I. — Data on hydrolysis , by sulphuric acid in glacial acetic-acid solution, of the 
acetylated free flavone of brown maize husks 



Sample I. Samples. Samples 



Weight of acetyl quercetin (gtn.). . . . 
Weight of recovered quercetin (gm.) 
Percentage of recovered quercetin. .. 



0.2521 
.1484 
58.86 



0-3165 
.1866 
58.95 



0.4908 
.2902 
59-13 



1 Willstatter, Richard, and Everest, Arthur E. tjntersuchungen uber die anthocy.vne. i. user 
DEN FARBSTOFF DER KORNBLUME. /« Licbig's Ann. Chem., Bd. 401, Heft 2, p. 189-232, 4 fig. 1913. 
' Emerson, R. A. op. ax. 



Oct. 1, 1921 Quercetin in Emerson s Brown-Husked Type of Maize 3 

The mean of the three determinations is 58.98 per cent — in exact 
accord with theory. 

The entire yield of approximately i gm. of acetyl derivative was 
divided to make the above determinations. The quercetin obtained 
(0.6254 gin.) was again acetylated, yielding 0.8352 gm. of penta-acetyl- 
quercetin. The acetyl derivative melted at 190° to 192° C. The recov- 
ered flavone melted at about 305° to 306° with darkening. When 
mixed with quercetin from Bscholtzia (melting point approximately 
305° to 310°) the mixture melted at 306° to 307°. In other charac- 
teristics the quercetin from maize was identical with a sample obtained 
by the writers ^ from rutin, a glucosid of quercetin found in Escholtzia 
petals. 

Combustions of the free quercetin and of its acetyl derivative were 
made, with the results shown in Table II. 

Table II. — Combustions of the free quercetin of brown maize husks and of its acetyl 

derivative 



Quercetin. 



Penta-acetylquercetin. 



Sample i. Sample 2 



Weight of sample (gm.) 

Weight of carbon dioxid (gm.) 

Weight of water (gm.) 

Percentage of carbon 

Percentage of hydrogen 



0-1353 
.2951 
.0404 
59-47 
3-34 



o. 1126 
.2430 
.0403 
58-85 

4.00 



0.2026 

•4338 
.0691 

58.39 
3.82 



Theory requires: For quercetin, carbon 59.59 per cent, hydrogen 3.34 per cent; for penta-acetylquercetin, 
carbon 58.59 per cent, hydrogen 3.90 per cent. 

PREPARATION OF THE GLUCOSID 

After partition of the alcoholic extract of the brown husks between 
ether and water, the aqueous solution, containing as one of its chief con- 
stituents a quercetin glucosid, was treated with four successive portions 
of lead acetate. The first fraction of the lead precipitate was discarded. 
The second consisted largely of tarry matter and was therefore not used 
for the preparation of pure glucosid but yielded quercetin on hydrolysis 
after decomposition with hydrogen sulphid. The third and fourth frac- 
tions were combined, suspended in hot alcohol, decomposed with hydrogen 
sulphid, filtered, and evaporated to small bulk. A small quantit}' of 
impure glucosid separated out on standing, but the greater part was got 
by shaking the solution with ethyl acetate. 

The glucosid was purified only with great difficulty, by fractional solu- 
tion of the dry impure product in ethyl acetate and successive crystalli- 
zation of the purer fractions from water. The yield of pure glucosid 

' Sando, Charles E., and Bartlett, H. H. rutin, the flavone pigment of escholtzia californica 
CHAM. /« Jour. Biol. Chem., V. 41, no. 4, p. 495-501, pi. 6-7. 1920. 



Journal of Agricultural Research voi. xxii, no. i 



obtained in this manner was insufficient for a thorough investigation, 
which must be deferred until a new lot of material is extracted. It was 
nearest in color to the "lemon yellow" of Ridgway's ^ color standards 
and melted to a cherrj^-red liquid at 220° to 222° C. A\Tien hydrolyzed it 
produced quercetin and apparently only one sugar, glucose, although the 
latter point is to be more thoroughly investigated. The osazone of the 
sugar melted at 204° to 206° and was evidently glucosazone. The 
quercetin obtained by hydrolysis was identified by its general properties 
and by combustions both of the free flavonol and of the acetyl deriva- 
tive. The latter melted at 191° to 193.5° and had the properties of 
penta-acetylquercetin. A sample weighing 0.4650 gm. gave 0.2735 &^- 
of quercetin, or 58.81 per cent by quantitative hydrolysis; theory requires 
58.98 per cent. The results of combustions are given in Table III. 

Table III. — Combustions of the quercetin obtained by hydrolysis of the glucosid of brown 
maize husks and of its acetyl derivative 



Penta- 
acetyl- 
quercetin. 



Weight of sample (gm.) 

Weight of carbon dioxid (gm.). 

Weight of water (gm.) 

Percentage of carbon 

Percentage of hydrogen 



Quercetin. 





1514 




32«3 




044S 


59 


13 


3 


29 



0.1570 

.3400 

• 0537 
59.06 

3-83 



Theory requires: For quercetin, carbon 59.59 per cent, hydrogen 3 .34 per cent; for penta-acetylquercetin, 
carbon 58.59 per cent, hydrogen 3.90 per cent. 

The glucosid is not one of the well-known ones but bears considerable 
similarity to one which Heyl ^ recently isolated from the pollen of rag- 
weed, probably Ambrosia artemisiijolia L., although he gives only the 
common name. 

SUMMARY 

In accord \vith the expectation that the broAvn-husked t5'pe of maize 
would be found to contain a flavonol, we have been able to isolate from 
brown husks both free quercetin and a quercetin glucosid of which a 
further investigation will be made. 

The two compounds in question are both lemon yellow in color. If 
they account for the truly brown color of the husks of this type, it must 
be through their tinctorial quality, probably through their adsorption 
on some colloid component of the brown tissues. 

It is very probable that the quercetin glucosid is the counterpart in 
the brown type of the anthocyanin of the purple type. The pigment 
of the latter will probably be found to be allied to cyanin. 

I RiDGWAY, Robert, color standards and color nomenclature. 43 p., 53 pi. (col.) Washington, 
D. C. 1912. 

' Heyl, Frederick W. the yellow coloring substances of ragweed pollen. In Jour. Amer. 
Chem. Soc, v. 41, no. 8, p. 1285-1289. 1919. 



BIOLOGICAL ANALYSIS OF THE SEED OF THE GEORGIA 
VELVET BEAN, STIZOLOBIUM DEERINGIANUM 

By Barnett Sure; and J. W. Read, Laboratory of Agricultural Chemistry, University 

of A rkansas 

The velvet bean, Stizolohium deeringianum Bort.,is annually becoming 
more important in southern agriculture, and the acreage planted to this 
crop in the cotton belt is continually mcreasing. From 1915 to 1917 it 
is estimated that the area increased from less than 1,000,000 acres to 
more than 5,000,000 acres. The acreage in 191 7 was 1 19 per cent greater 
than in 19 1 6. It is the most vigorous growing annual legume in the 
United States ; and on account of its very rank growth and the common 
practice of cultivating it v/ith the corn crop it is chiefly used as a winter 
pasture for cattle and hogs, although much larger quantities of the 
beans are harvested from year to year and ground, either with or with- 
out the pods, for market purposes. Harvesting with the com crop for 
use as silage is also growing in favor. 

Because of the rapidly increasing interest in this crop as a feed and 
its very considerable promise in this respect, particularly to the South, 
it occurred to one of the authors^ that a biological analysis should give 
very fundamental information as to how the velvet bean might best be 
utilized for feeding purposes. Accordingly the Georgia Velvet Bean, 
commonly know as the Early Speckled, was chosen for our studies be- 
cause of its early maturity, general popularity, and adaptability to the 
more northern as well as to the other sections of the cotton area. 

The Georgia velvet bean seed has a very tough, hard hull which con- 
stitutes 12 per cent of the whole seed. In grinding the beans it was 
found impossible to grind the hulls in a satisfactory manner; conse- 
quently these were sifted out. The experiments reported in this paper 
were conducted with hulled seed. It was later found, however, that 
after the sifted hulls had been dried on a steam bath for from six to eight 
hours they could be ground ; and experiments were later inaugurated, 
introducing the hulls in the same proportions as they were found to 
exist in the seed, the results of which will be reported later, together 
with other data showing the supplementary relationships of the seed 
to the leaf and the biological value of the whole plant. The nutritive 
value of the seed and the whole plant in practical rations is also being 
studied at the present time. 

' Credit for the inauguration and outline of the velvet-been studies as approved under the Adams Fund 
is due Prof. J. W. Read. 

Journal of Agriculural Research, Vol. XXII, No.i 

Washington, D. C. Oct. i, 1921 

za Key No. Ark.-i 

(5) 



Journal of Agricultural Research voi. xxii. No. i 



The experiments reported in this paper were conducted ^vith albino 
rats, employing the standard technic adopted by the Department of 
Agricultural Chemistry of the University of Wisconsin. 

Preliminary experiments showed that young rats, 40 to 90 gm. in 
weight, will exist only from 7 to 12 days on a diet composed solely of 




Fig. I. — Gain in weight of lot 51 on ration of velvet beans, So per cent; butter fat, s per cent; No. 32 salts, 
4 per cent; and dextrin, 1 1 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted wheat 
embryo. At point x 9 per cent of the velvet beans was replaced by 9 per cent casein. 

the raw hull-less velvet bean seed and a liberal supply of distilled water. 
The food consumption could be increased 50 to 60 per cent, however, 
by feeding young animals the seed after it was autoclaved for one hour 
at 15 pounds pressure. Consequently, autoclaved velvet beans were 




Fio. 2. — G^a in weight of lot 52 on ration of velvet beans, 60 per cent; butter fat, s per cent; No. 32 
salts, 4 per cent; casein, s per cent; and dextrin, 26 per cent. The dextrin carried alcoholic extract of 
10 gm. ether-extracted wheat embrvO, 

used in all- this work. In order to determine whether the water-soluble 
vitamin was destroyed in the process of autoclaving, controls were 
run with uncooked beans. 

Even when fed upon autoclaved beans, ammals, although consuming 
considerably more of the seed, existed only for a period of 17 to 21 



Oct. 1,1921 Biological Analysis of Georgia Velvet Bean j 

days when the diet was composed solely of the seed. On diets com- 
posed of 80 per cent velvet bean with 20 per cent dextrin, and 60 per 
cent velvet bean with 40 per cent dextrin, eight animals, having an 
initial weight of 50 to 60 gm., maintained their weight for a period of eight 
weeks but made no growth. Likewise, when the velvet bean formed 
60 to 40 per cent of the ration, respectively, as a source of protein, in 
the presence of all the other dietary factors, no growth resulted, but 
all the animals maintained their body weight for a period of six weeks, 
indicating that the proteins in the seed are deficient. 

Since these experiments lasted for periods ranging from six to eight 
weeks only, no charts were prepared illustrating the points mentioned. 
All the rest of our findings are illustrated in figures i to 15. 

When velvet beans formed 80 per cent as the source of protein in the 
ration (fig. i), very little growth resulted. Altliough at point xg per 
cent of the beans was replaced by 9 per cent casein, no appreciable 
change in the character of growth ensued, lack of response to the addi- 
tion of purified casein being due, as it will be noted from the following 
graphs, to the injurious effect of this high plane of velvet bean intake. 

Velvet beans fed at a 60 per cent level as a source of protein, supple- 
mented with 5 per cent casein, produced a fair amount of growth (fig. 2). 

When, however, 40 per cent velvet beans was the source of protein 
and the ration was further fortified with 9 per cent casein, the two 
females made normal growth for a period of four months and the two 
males grew at a rate even beyond the expectation curve (fig. 3). Rat 
211 was unable to rear her young, although her litter was reduced from 
nine to four. 

Figure 4 shows that young animals are unable to make any growth 
on a ration composed of 80 per cent velvet beans as the source of salts. 

When 40 per cent velvet beans served as the source of salts, some 
little growth occurred during the first 10 weeks (fig. 5). It is evident, 
then, that at least part of the failure of lot 61 (fig. 4) must be ascribed 
to the harmful effect of the higher plane of velvet bean intake. A striking 
change in the character of growth is apparent when at point x 4 per 
cent of dextrin was replaced by 4 per cent of salt mixture No. 32.^ 

When I per cent sodium chlorid (NaCl) and 1.5 per cent calcium 
carbonate (CaCOg) replaced salt mixture No. 32 in the ration, very 
good growth was obtained for a period of three months (fig. 6) , indicating 
that the calcium, sodium, and chlorid ions furnish the necessary mineral 
supplements in the velvet bean seed. 

Figure 7 shows that when 1.5 per cent calcium carbonate alone replaces 
salt mixture No. 32 only a little growth results. 

* Steenbock, H., and Gross, E. G. pat soLUBtE vitamine. n. the fat-soi,ubi,e vitamine con- 
tent OP ROOTS together with some observations on their "WATER-SOtUBLB VITAMINE CONTENT. In 
Jour. Biol. Chem., v. 40, no. 2, p. 505. 1919. 



Journal of Agricultural Research voi. xxn. no. i 



1 



Figures i to 7 have indicated that velvet beans are detrimental to 
young experimental animals when fed at an 80 per cent level. The 
experiment on which figure 8 is based corroborates that fact. Although 




60 

Fig. 3. — Gain in weight of lot 53 on ration of velvet beans, 40 per cent; butter fat, s per cent; No. 3a 
salts, 4 per cent; casein, 9 per cent; and dextrin, 42 per cent. The dextrin carried alcoholic extract of 
10 gm. ether-extracted wheat embryo. Y indicates point at which young were littered. 

when 80 per cent velvet beans served as a source of the fat-soluble 
vitamin, two animals made a fair amount of growth for a period of four 

SO 
60 
SO 



60 



Fig. 4.— Gain in weight of lot 61 on ration of velvet beans, 80 per cent; butter fat, 5 per cent; casein, s 
per cent ;and dextrin, 10 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted wheat 
embryo. 

months; two rats died after six weeks. The failure of these two ani- 
mals can not be ascribed to the low concentration of the fat-soluble 
vitamin in the seed, since lot 70 (fig. 11) made normal growth for a period 











/ 




^ 


^^ 


^ 


















^? 


/ 




%244 


,^ 


















































-*^ 




^ 




-^ 


















A24/ 


,l>^ 




^242 














2 



Oct. 1,1921 Biological Analysis of Georgia Velvet Bean 



of over five months when only 20 per cent velvet beans served as a 
source of that vitamin. 



200 

/80 

/60 

MO 

/20 

/OO 

80 

60 

40 



Fig. 5. — Gain in weight of lot 69 oa ration of velvet beans, 40 per cent; butter fat, 5 per cent; casein, 9 
per cent; and dextrin, 46 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted wheat 
embryo. At point x 4 per cent dextrin was replaced by 4 per cent No. 32 salts. 

A considerable improvement in the character of growth is obtained 
when the plane of intake of velvet beans used to supply the fat-soluble 
vitamin is reduced from 80 to 60 per cent (fig. 9) . 















/ 




























1 


























/ 






/ 






/ 
















/ 




) 


f 




/ 


/ 














i 


/ 




/ 




i 


/ 














y 


/ 


y' 


A 






/ 














/ 


/ 


\^ 


/^ 






/ 












y^ 


.>^ 


X 






9 


-^ 


B—H- 














0274 






*Z7S 




^76 
















2 




Fig. 6. — Gain in weight of lot 75 on ration of velvet beans, 40 per cent; butter fat, s per cent; sodium 
chlorid, i per cent; clacium carbonate, 1.5 per cent; casein, 9 per cent; and dextrin, 43.5 per cent. The 
dextrin carried alcoholic extract of 10 gm. ether-extracted wheat embryo. 

When the level of velvet bean intake was reduced to 40 per cent, 
normal growth was obtained. Rat 270 failed to rear her young, although 
her litter was reduced from eight to four (fig. 10). 



lO 



Journal of Agricultural Research 



Vol. XXII, No. I 



Even when the seed was reduced to as low a plane of intake as 20 per 
cent, it served as a very efficient carrier of tlie fat-soluble vitamin. It 
is also apparent that autoclaving for one hour at 15 pounds pressure had 




Fig. 7. — Gain in weight of lot 92 on ration of velvet beans, 40 per cent; butter fat, 5 per cent; calcium 
carbonate, 1.5 per cent; casein, 9 per cent; and dextrin, 44.5 per cent. The dextrin carried alcoholic extract 
of 10 gm. ether-extracted wheat embryo. 

no deleterious effect on this vitamin. Although excellent growth was 
obtained on this ration, mother rats No. 277 and 278 failed to rear their 
young in every case even when their litters ranging from 7 to 10 were 
reduced to only 4 (fig. 11). 




^O 



Fig. 8. — Gain in weight of lot 63 on ration of velvet beans, 80 per cent; No. 32 salts, 4 per cent; casein, 
5 per cent; and dextrin, 11 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted embryo. 
D indicates point at which rat died. 

Figure 12 shows that on reducing the level of velvet bean intake to 
10 per cent as a source of fat-soluble vitamin the character of growth 
is considerably impaired. 

When 80 per cent velvet beans was used to supply the water-soluble 
vitamin very little growth resulted, one animal dying after three weeks 
on this ration (fig. 13). 



oct.i. I92I Biological Analysis of Georgia Velvet Bean 



II 




-^bl 



Fig. 9. — Gain in weight of lot 67 on ration of velvet beans, 60 per cent; No. 32 salts, 4 per cent; casein, 
5 per cent; and dextrin, 31 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted 
wheat embryo. At point x 4 per cent dextrin was replaced by 4 per cent additional casein. 



GM. 



240 



220 




Fig. 10. — Gain in weight of lot 67 on ration of velvet beans, 40 per cent; No. 32 salts, 4 per cent; casein, 
9 per cent; and dextrin, 47 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted wheat 
embryo. Y indicates point at which young were littered. 



12 



Journal of Agricultural Research voi. xxn, no. 



Sixty per cent of velvet beans used to supply the water-soluble vitamin 
allowed only a very small amount of growth (fig. 14). 



G/^. 




Fig. II. — Gain in weight of lot 70 on ration of velvet beans, 30 per cent; No. 32 salts, 4 per cent; casein, 
12 per cent; and dextrin, 64 per cent. The dextrin carried alcoholic extract of 10 gm. ether-extracted wheat 
embryo. Y indicates point at which young were littered. 

In order to determine whether the water-soluble vitamin was destroyed 
during the process of autoclaving, 40 per cent velvet beans was fed un- 
cooked. Very little growth resulted, nor was there any improvement 



SA7. 




60 



Fig. 12. — Gain in weight of lot 90 on ration of velvet beans, 10 per cent; No. 32 salts, 4 per cent; casein, 
16 per cent; and dextrin, 70 per cent. The dextrin carried alcoholic extract of 15 gm. ether-extracted wheat 
embryo. 

in the character of growth when, at point x, lo per cent dextrin was 
replaced Avith lo per cent of an alcoholic extract of ether-extracted wheat 
embryo. It will be noted that after point x this ration is identical with 



Oct. 1,1921 Biological Analysis of Georgia Velvet Bean 



13 



that given to lot 53 (fig. 3) with the exception that lot 86 received the 
beans raw while lot 53 received the beans cooked. The striking dif- 
ference in the character of growth obtained in these two experiments 
must be attributed to the fact that the velevt bean seed uncooked is 
either toxic or indigestible at a concentration as low as 40 per cent. The 

/20 
/OO 

80 

60 

^O 

Fig. 13. — Gain in weight of lot 62 on ration of velvet beans, 80 per cent; butter fat, s per cent; No. 32 
salts, 4 per cent; casein, s per cent; and dextrin, 6 per cent. D indicates point at which rat died. 

cause of the deleterious effect of the raw seed is being studied and will 
be reported later. 

DISCUSSION 

The Georgia velvet bean, Early Speckled variety, has been found to 
be injurious when fed in the raw condition at as low a level as 40 per cent 
intake. This has been evidenced from an experiment where 40 per cent 











^..^ 
























* 


/ 


























/ 








> 





t 


/^ 










^ 


/ 


( 


Si^"' 




y 


/ 




/ 












245 




^ 


i-..^ 


^ 


^ 


^ 


--^ 


/ 






































WeEKS 




Fig. 14. — Gain in weight of lot 74 on ration of velvet beans, 60 per cent; butter fat, 5 per cent; No. 32 
salts, 4 per cent; casein, 5 per cent; and dextrin, 26 per cent. At point x 4 per cent dextrin was replaced 
by 4 per cent additional casein. 

velvet beans uncooked formed the source of water-soluble vitamin 
(fig. 15). Growth was inhibited during the first six weeks of experi- 
mentation, after which time lo per cent dextrin was replaced by an 
alcoholic extract of lo gm. ether- extracted wheat embryo. This addi- 
tion of the water-soluble vitamin should have rendered the ration entirely 
satisfactory, judging by the character of growth obtained in a duplicate 
experiment where the beans were furnished cooked (fig. 3). 



H 



Journal of Agricultural Research 



Vol. XXII, No. I 



The nature of the possible toxicity of the velvet bean has been recently 
suggested by Miller * to be due to dihydroxypehnylalanine. 

Cooking the seed at 15 pounds pressure for one hour destroyed for 
the most part its harmful effects, but there was still some injury when 
fed cooked at as high a plane of intake as 80 per cent. When 80 per cent 
of the velvet bean served as a source of protein, little growth resulted, 
nor was there a response obtained after 9 per cent of the seed was replaced 
by 9 per cent of casein, although when only 40 per cent velvet bean was 
served as a source of protein, supplemented with the same amount of 
casein, excellent growth was obtained. Unpublished data in this labo- 
ratory show that the better growth on the lower level of seed intake is 
not to be attributed to the higher intake of dextrin. Additional evidence 
is apparent from the fat-soluble vitamin experiment that when cooked 

Gn 

/GO 

/20 

/OO 

So 
60 

Fig. 15. — Gain in weight of lot 86 on ration of velvet beans (uncooked) 40 per cent; butter fat, 5 per 
cent; No. 32 salts, 4 per cent; casein, 9 per cent; and dextrid, 42 per cent. At point x 10 per cent dex- 
trin was replaced by 10 per cent of an alcoholic extract of ether-extracted wheat embryo. 

velvet beans are fed at an 80 per cent level some injury is still produced. 
Reduction of the plane of intake from 80 to 40 per cent results in con- 
siderable improvement in growth. 

That the velvet bean seed is very rich in the fat-soluble vitamin is 
evident from the fact that normal growth was obtained for a period of 
over five months when only 20 per cent of the seed served as the source 
of this syndrome. Reduction of the plane of velvet-bean intake to 10 
per cent resulted in inferior growth. The fact that considerably inferior 
growth was obtained on lower levels of seed intake with larger amounts 
of casein and dextiin precludes, we believe, the possibility that our 
casein and dextrin might have furnished appreciable amounts of the 
fat-soluble vitamin at the higher levels of seed intake, where we had 
remarkable success. It is also apparent from these experiments that 
autoclaving the seed for one hour at 15 pounds pressure has no deleterious 
effect on the fat-soluble vitamin. 

































































































^ 


-s 












-^ 










X 


.^"^ 






rs< 


^ 


^ 












^^ 


A 


X 


^llr^ 


<^ 






^ 












g^ 


'-3 


0^ 




^ 




^ 


5 












2 
meAS 



1 Miller, Emerson R. DrHVDRoxYPHENYLALANiNe, a constituent of the velvet bean. 
Biol. Chem., v. 44, no. 2, p. 481-486. 1920. 



In Jour. 



Oct. 1,1921 Biological Analysis of Georgia Velvet Bean 15 

The hull-less seed contained 27.5 per cent protein and therefore furnished 
16.5 per cent protein when fed at a 60 per cent level ; however, this amount 
of protein was inadequate for growth even though all the other factors 
in the diet were rendered satisfactory by the addition of isolated purified 
food substances. 

Recently Johns and Waterman ^ have isolated two globulins and an 
albumin from the Georgia velvet bean and have reported analytical data 
on their composition, using the Van Slyke ^ method of protein analysis. 
Their results show that, with the exception of the albumin, which is low 
in histidin, the three proteins of the Georgia velvet bean are quite satis- 
factory for their diamino-acid content. However, since we have insuffi- 
cient chemical data on the amino-acid content of the Georgia velvet bean 
no correlation can be made at present between the chemical composition 
and the biological response of this seed. The nature of the amino-acid 
deficiencies is being investigated. 

The velvet-bean seed has also been found to be deficient for growth in 
the character of its salts ; however, sodium chlorid and calcium carbonate 
seemed to replace salt mixture No. 32 satisfactorily. 

The concentration of the water-soluble vitamin in the seed has been 
found to be low. Unpublished data show that the addition of the 
ground hulls in the same proportions as they occur in the whole seed 
does not improve the water-soluble vitamin content. It is not apparent 
from these experiments whether this vitamin was in any way destroyed 
during the process of autoclaving, since the seed was extremely injurious 
when fed uncooked. 

SUMMARY 

(i) The Georgia velvet bean seed, Early Speckled variety, when fed 
raw was found injurious to young rats even when constituting only 40 
per cent of the total ration. 

(2) Autoclaving the seed for one hour at 15 pounds pressure destroys 
most of this injury, so that it is possible to include 60 per cent of the 
bean in a ration. A ration composed of 80 per cent velvet bean cooked 
still shows some harmful effects. 

(3) This seed, unlike most seeds so far studied biologically, is very 
abundant in the fat-soluble vitamin. The fat-soluble vitamin as it exists 
in this seed is quite stable after the seed is autoclaved for one hour at 15 
pounds pressure. The water-soluble vitamin, however, is of low con- 
centration in the hulled seed. 

(4) Both the proteins and salts of the velvet bean have been found to 
be of deficient character for growth. 

1 Johns, Carl O., and Waterman, Henry C. some proteins from the Georgia velvet bean, stizo- 
LOBIUM deeringianum. In Jour. Biol. Chem., v. 42, no. i, p. 59-69. 

2 Van Slyke, Donald D. the analysis of proteins by determination of the chemical groups 
CHARACTERISTIC OF THE DIFFERENT AMINO-ACIDS. In Jour. Biol. Chem., V. 10, no. I, p. 15-SS, 2 fig. 1911. 



EFFECT OF SOIL TEMPERATURE UPON THE DEVELOP- 
MENT OF NODULES ON THE ROOTS OF CERTAIN 

LEGUMES 

By Fred Reuel Jones, Pathologist, Office of Cotton, Truck, and Forage Crop Disease 
Investigations, Bureau of Plant Industry, United States Department of Agriculture, 
and W. B. Tisdale, Instructor in Plant Pathology, University of Wisconsin 

INTRODUCTION 

During a search for the cause of a diseased condition of alfalfa under 
observation in 191 7 and 19 18 the senior writer was led by observations 
contributed by H. L. Westover, of the Office of Forage Crop Investiga- 
tions, to believe it likely that soil temperature, within the range which 
occurs in cultivated fields, affects the initiation and the development of 
nodules on the roots of alfalfa and perhaps all other legumes to such a 
degree that the assimilation of nitrogen by these plants is greatly modi- 
fied by this lactor during the summer. The probable importance of such 
an effect of temperature, should it be demonstrated, upon the develop- 
ment of alfalfa and other legumes and especially its possible relation to 
the disease in question seemed adequate reasons for making a beginning 
at the experimental determination of the facts. Experimental methods 
suitable for the performance of this work had already been highly devel- 
oped in the course of the study of soil-inhabiting plant parasites at the 
University of Wisconsin. Thus it came about that the collection of the 
following data was begun at Madison by the junior author in 191 7 and 
continued by both authors in 191 9 and 1920. A temporary suspension 
of the work is the immediate reason for the publication of this prelimi- 
nary report. 

In the beginning, interest was centered upon ascertaining to what 
extent soil temperature determined the number of nodules which any of 
a selected group of legumes might develop. Later the size and composi- 
tion of the nudoles appeared more significant than the number. Finally, 
it is seen that soil temperature probably affects profoundly the rate of 
nitrogen fixation within the nodules of the legumes studied and its 
assimilation by the plants. A complete demonstration of such an effect 
and a quantitative determination of its amount remains for the future. 
During the progress of the work certain striking effects of soil temperature 
upon the development of the plants quite apart from any relation to 
nodule formation have been noted. 

Journal of Agricultural Research, Vol. XXII, No. i 

Washington, D. C. Oct. i, 1921 

zo Key No. G-24S 

54817°— 21 2 

(17) 



1 8 Journal of Agricultural Research voi. xxii. No. x 

ENVIRONMENTAL FACTORS WHICH HAVE PREVIOUSLY BEEN FOUND 
TO MODIFY THE FORMATION AND DEVELOPMENT OF NODULES 

In the extensive studies which have been made of the conditions 
which may favor or hinder the development of nodules, no one, so far 
as the writers are aware, has concerned himself with the factor which 
is considered here — ^namely, the temperature of the soil. There are a 
number of factors, however, which are known to have very much influence 
upon nodule development, and two of these which may have become 
modified by our experimental methods for controlling soil tem- 
perature must be considered. These are soil moisture and the concen- 
tration of nitrates. 

With regard to soil moisture, there appears to be both observational 
and experimental data which indicate that high soil moisture tends to 
increase nodule formation. Gain * notes that peas grown in wet soil 
have far more nodules than those grown in drier soil close by. Wilson ^ 
in his experimental work reports that wet soil induces the formation of 
a greater number of nodules on soybeans. Fortunately, in experimen- 
tal work with controlled temperatures, it is comparatively easy to main- 
tain soil moisture at a predetermined point with very slight fluctuation. 
In the preliminary experiments, although no attempt was made to 
control soil moisture exactly, it is not believed to have fluctuated suffi- 
ciently to affect results appreciably. In the later work, soil moisture 
was maintained in each series at one-half the moisture-holding capacity 
of the soil used (14 per cent of the wet weight) by weighing the pots 
each day, if necessary, and restoring the water lost by evaporation and 
transpiration. It is believed that this method kept variation in soil 
moisture within such very narrow limits that this factor could not have 
produced appreciable variation in nodule formation. 

That the amount of nitrate present in the soil affects nodule develop- 
ment, completely inhibiting it when high concentrations have been 
reached, has been demonstrated by several investigators. Wilson ^ has 
added a considerable number of nitrates to soils in different amounts to 
determine the concentration at which nodule formation is inhibited by 
each of the compounds. Although complete inhibition is effected only 
at concenti"ations which are not likely to occur in normal soils, the 
marked effect of variations is so well attested that any differences 
arising unavoidably during an experimental series must be taken into 
account in the consideration of results. 

The control of the concentration of nitrates in the soil solution in a 
soil held at different temperatures offers difficulties which can be over- 

1 Gain, Edmond. influence de i.'humidite sur le developpement des NODosixfes des l^gumineu- 
SES. In Compt. Rend. Acad. Sci. [Paris], t. ii6, no. 24, p. 1394-1397. 1S93. 

2 Wilson, J. K. physiological studies of baollus radiocola of soy bean (soja max piper) and 
OF factors influencing nodule production. N. Y. Cornell Agr. Exp. Sta. Bui. 386, p. 363-413, fig. 
80-94. 1917. 



Oct. 1. 1921 Effect of Soil Temperature on Development of Nodules 19 

come only within certain limits. As was expected in advance, the rate 
of nitrification in soil differs greatly at the different temperatures, 
producing greatly different concentrations within a short time after a 
series of plants have been started. In addition, there is soon consid- 
erable difference in the size of the plants at the different temperatures 
and a consequent difference in ability to absorb nitrates. The varia- 
tions which arise from these causes can be limited somewhat by the use 
of soil low in total nitrogen, thus making impossible the accumulation 
of large amounts of nitrates in any case. Variations in the concentra- 
tion of nitrates which have been observed in the experimental work 
described here will be noted later, and their possible effect upon the 
results will be discussed. 

APPARATUS AND METHODS 

The apparatus used for the control of soil temperature in these experi- 
ments is that which has been used in the Laboratory of Plant Pathology 
at the University of Wisconsin for several years and needs no new 
description.^ In all cases plants were grown in metal cans 6 inches in 
diameter and 10 inches deep. The number of plants which could be 
grown in each can vnthout serious crowding of roots was 3 for soybeans, 
5 for peas, and 10 for clover and alfalfa. The soil used was a sandy 
loam from a pasture which had never been cultivated. To this was 
added about an equal weight of sand in order that the total nitrate 
content should be kept low and that the mechanical condition of the 
soil should permit the easy removal of the roots. The temperatures 
noted in the different series were those at which the water was main- 
tained in the tanks in which the cans were set. Fluctuations of tempera- 
ture did not often exceed 1° C. from those given in the tables, and 
were of only a few hours' duration. Although record was made twice 
daily of the actual temperatures, it is not believed that a computation 
of the mean temperature from these figures would give a figure more 
significant than tlie convenient even numbers used here. It should also 
be noted that although the surface of the soil was insulated to some 
degree from loss of heat and moisture by the use of mineral wool, never- 
theless at the higher temperatures the surface soil to the depth of about 
I inch was usually cooler by i to i}4° than the water. However, it is 
believed that the larger part of the roots and nearly all the nodules were 
sufficiently deep in the soil below this cooler layer, so that error arising 
from this source is not considerable. 

Water was supplied through a glass tube which passed to the bottom 
of the metal can where it entered an inverted unglazed flower pot 3 
inches in diameter, which acted as a reservoir. In the last series the 

> Jones, 1,. R. soil temperatures as a factor in phytopathology. In Plant World, v. 20, no. 8, 
p. 229-237, 2 fig. 1917. lyiterature cited, p. 236-237. 



20 Journal of A gricultural Research \o\. xxii. No. i 

inverted pot was placed about 3 inches below the surface instead of at 
the bottom in the hope of maintaining a more uniform and rapid dis- 
tribution of moisture. This appears to have been an unfortunate 
change in method, since at the higher temperatures roots tended to col- 
lect around these pots, where they apparently developed more ex- 
tensively and produced more nodules than they had in the previous 
series. Inasmuch as no accumulation of roots took place at lower tem- 
peratures, it is not easy to explain this fact. 

Attention should here be drawn to the fact that two distinctly dif- 
ferent methods of securing data have been used. In the preliminary ex- 
periments, the plants were first grown in the cans at ordinary greenhouse 
temperature for about two weeks before inoculation was made, with 
the suitable strain of Bacillus radicicola Beyr., by pouring a water sus- 
pension of the organism around the base of the plants, and the cans were 
placed in the tanks adjusted at the predetermined temperatures. In the 
later series the plants were grown from seed in inoculated soil held at the 
required temperatures from the beginning. Several reasons led to the 
change of method. In the first place, one could not be certain that the 
bacteria poured around the plant in the first instance would become 
rapidly distributed through the soil at all temperatures. This inequality 
in rate of distribution might affect the number of infections and hence the 
number of nodules formed. At least it might tend to limit the formation 
of nodules to the roots near the surface of the ground where temperature 
is less exactly controlled. A second objection to this method appeared 
when the marked effect of temperature upon the morphology of the roots 
themselves was observed. The number of root hairs through which 
infection has been found to take place is much greater at lower tempera- 
tures than at higher. In view of the possible effect of this difference 
it appeared preferable to grow the plants from the beginning in inoculated 
soil at the designated temperatures, even though the plants thus produced 
would necessarily vary considerably in size. Data obtained by each of 
these methods will be presented. 

MEASUREMENT OF EFFECT OF SOIL TEMPERATURE UPON NODULE 

FORMATION 

When the experiments were begim it was assumed that different tem- 
peratures, if they are at all potent, would produce such a marked effect 
upon the number of nodules that count alone would give a significant 
expression of results. This expectation was fostered by the fact that 
Wilson 1 and nearly all previous investigators have used numbers to 
express similar experimental results. It will be seen from data given 
later that this hope was early disappointed. Different temperatures 
usually seemed to affect number not nearly so much as rate of develop- 

» Wilson, J. K. op cit. 



Oct. 1, 1921 Effect of Soil Temperature on Development of Nodules 2 1 

ment and size. In fact, in some instances, volume of nodular tissue 
seemed in inverse ratio to number. With most legumes it is not easy to 
get an accurate dry-weight determination of small nodules, because these 
occur as swellings so closely attached to the root that it is hardly feasible 
to separate them from the true root tissue. The one species tried which 
gave least trouble from this source by reason of the distinct separation 
of its nodules from the root, even at early stages of development, was 
the soybean. For this reason, it alone was used in the final series recorded 
here. 

Of course it was soon realized during the progress of the work that vol- 
ume of production of nodules was only an easily observed index, significant 
chiefly in so far as it revealed important eff"ects of temperature upon the 
physiological processes which are dependent upon the nodular structures. 
It would be of greater interest, for example, to measure the amount of 
nitrogen fixed in these nodules produced at different temperatures and 
that portion which becomes available to the plant for use in its vital pro- 
cesses. The demonstration of an important limiting effect of temperature 
upon nitrogen fixation would be of no inconsiderable importance. Such 
an effect would probably be indicated by large differences in size of 
nodules, though it might occur without the appearance of such difference. 
In any case it seems easily possible to determine approximately the effi- 
ciency of nodules in the fixation of atmospheric nitrogen by growing 
parallel series of plants, inoculated and uninoculated, in the same kind 
of soil and at the same soil temperatures. If the amount of nitrogen in 
the inoculated plants (aside from that foimd in the nodules on those 
plants) is greater than in the uninoculated, the gain must be credited to 
the efficiency of the nodules. The gains thus found should be an accurate 
measure of the effect of soil temperature upon the fixation of available 
nitrogen in the legume used in the experiment, and a comparison of this 
gain with the weights of the nodules found on the inoculated plants 
should give an approximate idea of the relation existing between effici- 
ency of fixation of available nitrogen and volume of nodules. In the 
last series recorded an attempt was made to carry out this experiment 
with the soybean plant. Unfortunately some of the uninoculated plants 
in the series became inoculated during the experiment and developed a 
few nodules, thus making it necessary to discard the data so far as these 
controls are concerned. Thus an exact determination of the extent of 
the effect of soil temperature upon the fixation of nitrogen in the nodules 
of legumes remains to be made. For the present we can only ascertain 
the dry weights of the nodules themselves as they are found at the end 
of a period of time and determine the amount of nitrogen found within 
them. 



22 Journal of Agricultural Research voi. xxii. no. i 

DISCUSSION OF THE LEGUMES USED AND THEIR BEHAVIOR UNDER 
THESE EXPERIMENTAL CONDITIONS 

Four legumes were selected for tlie soil temperature series requiring 
four different strains of Bacillus radicicola for their inoculation. One of 
these, the soybean, flourishes well at high soil temperatures; one, the 
Canada field pea, requires a low soil temperature for good growth; and 
red clover and alfalfa occupy intermediate positions. 

A few of the more striking reactions of the plants in these series will be 
noted. The Canada field pea does not flourish vigorously at a soil tem- 
perature as high as 30° C. (Table I) and is intolerant of temperatures 
above this point, maintaining roots only very close to the surface of the 
soil. It is perhaps misleading to infer that the lower surface temperature 
is alone responsible for the position of the roots, since in field plots in hot, 
exposed positions the death of deeper roots and the formation of surface 
roots has been noted in hot weather. 

Perhaps the more striking effect of the series of soil temperatures upon 
the soybean plants, aside from the fact of the wide range through which 
it grows vigorously, is the effect upon the color of the foliage. After 
the plants had become 5 or 6 inches tall, in both series the leaf color was 
much darker at the two ends of the series, especially at 30° C. and above, 
than at 2 1° and 24°. This difference persisted, tending rather to increase 
as long as the plants were grown. 

One striking difference in behavior between peas and soybeans on the 
one hand and clover and alfalfa on the other was noted in this series. 
The annuals formed a rather regular series of plants as judged by appear- 
ance (PI. 1) and also by dry weights (Table I). But with alfalfa and 
red clover the seedlings at the lower temperatures, 12° and 15° C, though 
little delayed in starting, remained small Alpine plants with thick dark 
green leaves and \vith much red color in the petioles; whereas at 18° the 
plants were more nearly what may be termed "normal" plants, larger, 
with fairly long petioles containing less red color. ^ 

EXPERIMENTAL DATA 

EFFECT OF SOIL TEMPERATURE UPON THE NUMBER OF NODULES FORMED 

As previously noted, in the fiirst two preliminary trials the seeds were 
planted in soil in the metal cans and grown for about 10 days at green- 
house temperature (about 22° to 23° C.) before they were inoculated 
with the suitable strains of Bacillus radicicola and placed in the tanks 
adjusted at the temperatures designated. When it was believed that 
sufficient time had elapsed for infection at all temperatiu-es, the tops 
were cut from the plants, dried, and weighed. The roots were carefully 

1 Since this was written clover and alfalfa have been grown under similar conditions at controlled tem- 
peratures. The marked dwarfing of plants at 15° and 12° C. was found to disappear when the plants became 
older, and especially later in the spring when light intensity became greater. 



Oct. 1. 1921 Effect of Soil Temperature on Development of Nodules 23 

washed from the soil and the nodules were counted. The count obtained 
is given in Table I. In the two later series the seeds were planted in 
soil which had already been placed in the tanks adjusted to the tem- 
peratures designated. The air temperature ranged from 14° to 18°. 
The counts obtained in these series are given in Table 11. 

Table I. — Average number of nodules produced on plants 26 days old grown at a soil 
temperature of about 20° C.for 10 days, after which mociilation was made and the tem- 
perature of the soil was maintained as indicated 



Temperature. 



"C. 
10 to 12 

15 

20 

25 

30 

35 

40 



Alfalfa. 



plants. 



IS 
35 
18 
16 
5 



plants. 



Red clover. 



plants. 



17 
27 
40 
69 

75 
4 



plants. 



7 
16 

47 
4 

75 
6 



Soybeans. 



6 
plants. 



46 
61 

37 
35 



6 
plants. 



14 
25 
33 
28 



Field peas. 



plants. 



27 
31 
37 
43 
128 



plants. 



21 
24 

30 

60 

64 

3 



o The plants did not survive. 

Table II. — Average number of nodules produced on plants grown at the soil temperatures 

designated 



Temperature. 



Alfalfa. 



20 plants 63 
days old. 



Red clover. 



10 plants 63 
days old. 



Soybeans. 



6 plants 63 
days old. 



9 plants 55 
days old. a 



Field peas. 



S plants 32 
days old. 



5 plants 52 
days old. 



12 

15 

18 

21 

24 

27 

30 

33 

36 



I. 19 
1.6 
16. 7 
8.0 

3-4 
II. 6 



10.7 
3-5 



1.8 

5-0 
12.5 
24. 6 
17.4 

II- 5 
8.9 

5-3 



o. o 
6. I 

5-9 
4.0 

II- 3 
8.0 

5-8 

8.8 

13.6 



13.0 
14.7 
19.7 
18.8 
16.4 
20.0 
12.4 



3-6 

3-6 

8.0 

14. o 

25.8 

30.0 

4.0 



2.6 
9.0 

27.2 
23.8 
13.0 

58.0 



o- The larger number of nodules on the plants 55 days old as compared with those on plants 63 days old 
is believed to be due largely to the fact that this series of plants was grown in spring, when longer days 
promoted a far more vigorous growth than was produced by the other plants, which were grown in winter. 

Although it will be seen at once that the data in the two tables are not 
strictly comparable, nevertheless some temperature effects upon number 
appear. Most conspicuous of all is the greatly increased number upon 
peas near the upper thermal limit. But this increased number is accom- 
panied by a more than proportionate decease in size. No plant has been 
found to produce large nodules at 30° C. or above. Clover and alfalfa 
tend to produce their largest numbers of nodules in the middle portion 
of the range. Soybeans show no decided temperature effect at all, so 



24 



Journal of Agricultural Research voi. xxii. No. i 



far as number is concerned. However unsatisfactory these figures may 
be from several points of view,' nevertheless they establish one important 
fact beyond reasonable doubt : Modified soil temperature ^vithin the range 
which these plants can be expected to encounter in tlie field and even 
within which they can be grown \vith vigor under experimental conditions 
does not prevent the infection of roots by Bacillus radicicola and the 
formation of considerable numbers of nodules. In other words, B. radi- 
cicola, considered as a parasite, does not show the strongly marked inhi- 
bition of its ability to infect roots of plants that has been found in a 
number of fungus parasites.^ 



/Of- 



__A{oho^.€^.^ 



' ... % 






/o\ 



\ 



/2 



/■5 



/8 



30 



33 



36 



2/ 24 27 

■PiG. I. — Comparison of dry weights of tops, roots, and nodules of soybeans given in Table HI, grcwn 
during November, December, and January. 



EFFECT OF SOIL TEMPERATURE UPON THE PRODUCTION OF NODULES AS 
MEASURED BY DRY WEIGHTS 



Although it is clear that soil temperature does not, in most cases, 
greatly modify the number of nodules produced, it was obvious from the 
very beginning that the size to which they developed was markedly and 
consistently affected. Such effect is shown graphically in Plate 2, where 
nodules from an equal number of plants are shown placed in rows and in 
Plate 3, where nodules from a larger number of plants are placed in tubes 
of equal diameter. Dry weights of the nodules shown are given in Table 
III. Data are given for the soybeans only for reasons already mentioned, 
but, judging by visual evidence obtained in studying all four legumes 
used, it is believed that all behaved in essentially the same manner. 

I Johnson. James, and Hartman. R. E., influence of soii, environment on the root-rot of 
TOBACCO. /» Jour. Agr. Research. V. 17, no. 2, p. 41-86, pi. 1-8. 1919. Literature cited, p. 84-86. 

TiSDALE, W. H. RELATION OF TEMPERATtree TO THE GROWTH AND INFECTING POWER OF FUASRIUM 
UNI. In Phytopathology, v. 7, no. s, p. 356-360, i fig., pi. u. 1917. 



Oct. 1, 1921 Effect of Soil Temperature on Development of Nodules 25 



Now it would be expected, and it is clearly true, that plants grown in 
soils held at such widely different temperatures would show in the given 
time considerable difference in size and degree of maturity. The first 
question which will be asked regarding this difference in nodule develop- 
ment will be whether it does not correspond more or less approximately 
with corresponding differences in root or shoot development. Does it 
show a trend distinctly different from that of other portions of the 
plant ? 

When the figures given in this table are presented in graphs, the 
differences in trend become obvious. A comparison of the weights of 
the nodules with those of the roots (fig. i, 2) will show that in both 




/s 



/s 



2/ 24 27 30 



33 



36 <^ 



Fig. 2. — Comparison of dry weights of tops, roots, and nodules of soybeans given in Table III, grown 

during April and May. 

series the maximum development of nodules occurs at 24° C, with 
very slight development at the extremes, 15° and 36°. Root develop- 
ment, on the other hand, rises much more rapidly at the lower temper- 
atures and is maintained at the the higher temperatures, reaching a 
maximum in the second series at a point 9° higher than that of the 
nodules. Root development is far more uniform at all temperatures 
than is nodule development. 

A comparison of nodule development with shoot development (fig.i, 2) 
shows that the effect of temperature upon the development of the two 
structures is quite different. As with root development, shoot develop- 
ment is relatively more vigorous at 15° and 18° C. than is nodule develop- 
ment, which increases greatly at 21° and reaches a maximum at 24°. 
"When at 27° the v/eight of nodules is beginning to diminish, that of 
shoots maintains its level or increases. Through the higher temperatures 
weight of nodules falls off rapidly, while that of shoots remains at the 
high level. 



26 



Journal of Agricultural Research voi. xxii, No. i 



Table III. — Dry weight per plant of shoot, roots, atid nodules prodticed in 6j days in 
the first series, grown in November, December, and January, and 55 days in the second 
series, grown in April and May 



Temperature. 



°C 

12 

15 

18 

21 

24 

27 

30 

33 

36 



First series. 



Shoot. 



Gm. 

o- '^33 

. 410 

• 432 
.632 

.875 
.771 
.818 
.863 
.996 



Root. Nodules. 



Gm. 
O. 040 
. IIO 

•135 
. 140 

•131 
.108 
.096 
.116 
. 108 



Gm. 

O. 000 
. 006 
.008 
•033 
•043 
.030 
. 014 
. 012 
.005 



Second series. 



Shoot. 



Gm. 



0. 922 
1.430 

1. 710 

1. 900 

2. 620 

2.540 
2.440 
2. 130 



Root. Nodules. 



Gm. 



3. 184 
.318 
. 269 
. 296 
•342 
. 296 

•437 
. 422 



Gm. 



O. 021 
. 060 
. 108 

•145 
.094 
.089 
. 076 
. 042 



OS 






















3S^ 


5.<5.5 




L 


T"— 








' 


g 




N 




5v 








!:c. 






y 


^^ 


^ 






^ 




V 


/ 




. y 






^ 






f 




\\ 




























^ 0^ 




/ 'V/ 






V 


, 




I 




f 






\ 


"*^^^^'- 






0/ 


^ 


by 








^^ 


^. 


-„«»■— "" 












^\1^ 


.3 
















^ 



/S /3 2/ 24 27 30 33 36 

Pig. 3. — Ratios ot weight of nodules to weight (A tops and of roots. First experiment. 



In order to obtain a clearer view of the contrast between the effect of 
temperature upon weight of nodules and that of roots and shoots, the 
ratios of the weights of these parts of the plant have been determined and 
plotted (fig. 3, 4). If there is a direct relation between development of 
nodules and that of either the aerial or subterranean parts of the plants — 
if nodule development is conditioned by top or root development quite 
independently of the temperature factor which was varied in these 
experiments-^then the ratio should be approximately constant, or at 
any rate should be a straight line. A glance at the graphs shows that 
this is not the case. The ratios when plotted produce curves which are 
closely similar. No direct relation appears to exist between weight of 
nodules and that of either tops or roots under the conditions of these 
experiments. At 21° and 24° C. the weight of nodules is relatively 
larger than at temperatures above or below this region. The wide dif- 



Oct. I, I92I Effect of Soil Temperature on Development of Nodules 2 7 

ference in the ratios and the consistent similarity of the curves can 
hardly be explained otherwise than as a temperature effect upon nodule 
development which is quite different from that upon development of 
root or shoot. 

EFFECT OF SOIL TEMPERATURE UPON THE COMPOSITION OF THE 
INOCULATED PLANTS 

In order that comparisons might be made of the amount of nitrogen 
found in inoculated and uninoculated plants, total nitrogen determina- 
tions were made of shoots, roots, and nodules of the plants grown in 
each series. Since the uninoculated plants did not remain free from 
nodules, the desired comparisons can not be made. Nevertheless the 

Od 



07 



\ 
%.03 












y 


\ 












/^ 


\ 














y . 
















^ 

^ 


\\ 












:5/ 4i 


^ 


































^ 


=^cc^ 


s 






^^y^ 


cPy 








\ 


'•x.^ 




/" \i 


/ 








^ 


-, X 




n^*^ 












"-. 














1 














1 



sdQ. 






/O 



12 /S /a 2/ 24 27 30 33 3$ 

0£G7?£SS C£A^r/S7?^£>£ 

Fig. 4.— Ratios of weight of nodules to weight of tops and of roots. Second experiment. 



difference in nitrogen found were so marked in the series that the analyses 
of one series, the last that was grown, is given. 

Table IV. — Percentage of total nitrogen found in shoots, roots, and nodules of soybean 
plants grown at a series of soil temperatures in April and May 



Temperature 

°c. 

15 

18 

21 

24 

27 

30 

33 

36 



Shoots. 


Roots. 


2.42 


2.89 


2.86 


3-40 


4.27 


2. 64 


4. 22 


2. 98 


4-55 


2.66 


4.58 


2. 62 


3-98 


2. 67 


3-77 


2.85 



Nodules. 



5-95 
6.95 
6.25 

5-95 
6.25 
6.25 
6. 00 
5-70 



The greatest differences in content of nitrogen are found in the shoots, 
the roots being very uniform and the nodules hardly less so. It will be 
seen that, generally speaking, the high nitrogen content of the top is 



28 Journal of Agricultural Research voi. xxii, No. i 

correlated mth the best development of nodules, though the curve 
which would be produced by these figures when plotted in the manner of 
the preceding data would not have the same shape. The largest amount 
of notrogen is found at a higher temperature than the point at which 
the largest dry weight of nodules was found. A rather sudden increase 
in nitrogen at 21° C. as compared with 18° and a sudden fall at 33° as 
compared with 30° has characterized the series obtained thus far. The 
result of the analyses which have been made seems worth recording; 
but whether the low nitrogen content of the plants grown at both ends 
of the series is wholly due to the small nodules found on these plants, 
and whether the high nitrogen content of plants in the center of the series 
is due to large and presumably efficient nodules, likely as this connection 
appears, remains to be determined by more refined methods. 

DISCUSSION OF FACTORS THAT MAY HAVE HAD AN INFLUENCE UPON 

THE DATA GIVEN 

CONCENTRATION OF NITRATES IN THE SOII, 

In view of the fact already discussed, that large amounts of nitrates 
in the soil solution decrease nodule development and even inhibit it 
before the concentration becomes great enough to injure the plant 
directly, it is unfortunate that the soil used in these series should have 
had as much nitrate as was found (Table IV), even though the largest 
amounts are far below the inhibition point. There appears to be no 
data available in literature whereby we may know what is the maximum 
or the more usual amount of nodular structure formed on the roots of 
any of the legumes. Although the amount of nodular structure which 
peas may produce may be quite different from the amount which soy- 
beans may produce under the most favorable conditions, yet it may be 
worth while to record here that the writers have found in one instance 
a variety of wrinkled peas producing at the blossoming stage nodules 
whose dry weight was 2.2 times as great as that of the entire root system 
(average of 25 plants) ; and in individual plants the ratio of weight 
of nodules to roots was as high as 4.5 to i. However the ratio of weight 
of nodules to tops in these plants was 0.085 to i , a ratio not much different 
from that found under the best experimental conditions for soybeans 
recorded here (fig. 3, 4). 

However, the question of immediate interest here is whether or not 
the nitrate content of the soil used in tliese series was greatly changed 
at any of the temperatures at which it was held, and if there is any 
evidence that this change was of sufficient size and in the right direc- 
tion to indicate that it may have been responsible for the increased or 
decreased nodule formation at this temperature. In order to obtain 
information regarding the change which soil temperature maj'- have 
produced in the series, nitrate nitrogen determinations were made by 



Oct. 1, 1921 



Effect of Soil Temperature on Development of Nodules 29 



the colorometric method of a composite sample of the soil at the begin- 
ning of the experiment and of a sample from two cans of soil at each 
temperature when the plants were harvested. In addition, in order to 
get some clue to intervening changes, an unplanted can of soil was kept 
at each temperature from which a sample was taken at about the middle 
of the period. The results obtained are shown in Table IV. The effect of 
soil temperature does not appear to have been as definite and consistent 
upon the concentration of nitrate nitrogen as was anticipated. Until 
further data are available, it seems unwise to attempt to interpret the 
results. However, the very absence of large and consistent modifica- 
tion enables us to believe that this factor was not important in its effect 
upon nodule development. The only point at which nitrate accumula- 
tion became very large occurs in the second series where the unplanted 
soil shows at the end of 26 days a very high nitrate content at 21° 
and 24° C. If it is assumed that a similar concentration took place in 
the planted pot at an early stage in the development of the plants be- 
fore they were large enough to reduce it by absorption, it would be an- 
ticipated that a reduction in nodule production would be found here. 
In fact, however, this point of high nitrate formation is the point of 
highest nodule production, just as it is in the first series where no evi- 
dence of high nitrate content at any time was obtained. 

Table IV. — Nitrate nitrogen in the soil in which the soybeans grew and also in 

unplanted soil 







First 


series. 






Seconc 


series. 




Temperature. 


Planted soil. 


Unplanted soil. 


Planted soil. 


Unplanted soil. 




At start. 


At end. 


After 24 
days. 


At end. 


At start. 


At end. 


After 26 
days. 


After 61 
days. 


°C. 


P. p. m. 


P. p. m. 


P. p. m. 


P. p. m. 


P. p. m. 


P. p. m. 


P. p. m. 


P. p. m. 


12 


20. s 
20.5 


8.6 


16.3 
II. 4 


14.4 
8-3 










15 


3-3 


90 


20.0 


34 


34 


18 


20.5 


8.6 


16.5 


10. 6 


90 


14. 


56 


34 


21 


20.5 


8.1 


19-3 


20.5 


90 


18.0 


124 


15 


24 


20. 5 


10. 6 


II. 8 


14.4 


90 


8.5 


150 


17 


27 


20.5 


10. 6 


13-9 


9.2 


90 


14. 


88 


30 


30 


20. 5 


16.9 


13-9 


8-3 


90 


4.8 


36 


58 


33 


20.5 


3-3 


15-1 


3-6 


90 


4.8 


29 


5° 


36 


20.5 


9-7 


21.8 


3-6 


90 


30.0 


22 


30 



MOISTURK CONTENT OF THE SOU. 



In view of the effect which high moisture content of the soil is reported 
by Wilson ^ and others to have in increasing nodule production, the 
moisture content of the soil in the later series was kept uniform at all 



'Wilson, ]. K. physiologicai, studies of bacillus radiocola of soy bean(soja max piper) and 
OP FACTORS INFLUENCING NODULE PRODUCTION. N. Y. Cornell Agr. Exp. Sta. Bui. 386, p. 363-413, fig. 
80-94. 1917- 



30 Journal of Agricultural Research voi. xxii. No. i 

temperatures at one-half the moisture-holding capacity of the soil, 
previously determined to be 14 per cent of its dry weight. In order to 
get further evidence as to the extent of the effect of high moisture con- 
tent, a single can containing three plants was placed at each temperature 
in the last series with moisture content of 18 per cent of the dry weight — 
a distinctly wet soil. Accidents which befell several plants in the series 
produce irregularities in the figures which would require long explana- 
tion. Suffice it to say here that though the tops were increased in size 
there is no evidence that the nodules were increased either in number or 
size. Apparently moderate differences in moisture content of the soil 
were not large factors influencing results in the previous experimental 
work where exact control of soil moisture was not accomplished. 

HYDROGEN-ION CONCENTRATION OF THE SODv SOL,UTlON 

Inasmuch as it was considered possible that the extreme temperatures 
at which the soil was held might produce changes which would alter the 
hydrogen-ion concentration of the soil solution, and hence the formation 
and perhaps development of nodules, a determination of this environ- 
mental factor was made toward the close of the last series described. 
Samples of soil were taken from the unplanted pots at 15°, 24°, and 
36° C. A determination of the hydrogen-ion concentration of the soil 
solution of the three samples by the colorometric method gave identical 
results, the Ph value being 6.3 in all three cases. Thus no evidence 
was obtained that temperature had altered this important factor in this 

series. 

SUMMARY 

(i) Preliminary studies have been made upon the effect of soil tem- 
perature on the development of four legumes, alfalfa, red clover, field 
peas, and soybeans, with special reference to its effect upon the infec- 
tion of these plants by Bacillus radicicola and the subsequent develop- 
ment of nodules. The larger part of the data were obtained by growing 
plants in soil held at a series of temperatures 3° apart from 12° to 36° C. 
The air temperature was uniform for all plants, ranging from 14° to 20°. 

(2) As was anticipated, the four plants differed in their ability to 
tolerate soil temperatures at the ends of the series. Peas were dwarfed 
at 30° C, clover developed poorly at 36°, while alfalfa and soybeans 
still grew very well at 36°. Soybean plants grown in the soils held at 
12°, 15°, 33°, and 36*^ showed very dark green color of leaves, whereas 
those toward the center of the series became progressively lighter, those 
at 24° being lightest. 

(3) With regard to the number of nodules formed on plants grown in 
soil held at this series of temperatures, irregularities were found in each 
series; but no large consistent differences were discovered, except that 
at the extreme upper and lower limits at which a plant will survive the 



Oct. 1, 1921 Effect of Soil Temperature on Development of Nodules 31 

number is reduced, and that peas usually produced greatly increased 
numbers at 30° C. All these species form nodules in soils at any tem- 
perature at which the plant can make a growth that is at all vigorous. 

(4) While variation in number was not consistent, size measured by the 
average dry weight per plant of all those formed on a number of plants 
was found to differ greatly and consistently within the series, at least so 
far as the soybean plant was concerned. The maximum weight attained 
on the soybean plant after a period of two months was found at a soil 
temperature of 24° C. Examination of nodules on the roots of the other 
legumes indicated that their maximum production occurred at about the 
same temperature. 

(5) Weight of nodules produced by soybeans was not found to be corre- 
lated with the weight of tops or of roots through the series of tempera- 
tures. Weight of tops was almost or quite as great at 30° to 36° as at 
24° C, while weight of nodules declined rapidly at the higher tempera- 
tures. Weight of roots likewise showed no such diminution at the higher 
temperatures or even at the lower temperatures as did weight of nodules. 
With the soybean plant, and to a much less marked degree with the other 
plants, there was a correlation between weight of nodules and color of 
plant, the largest weight of nodules occurring on plants with the palest 
green color. 

(6) Generally speaking, plants with large nodules had a higher percent- 
age of total nitrogen in the tops, though this correlation is not exact. 

(7) Factors of soil environment that are regarded as having an influence 
upon nodule formation have been taken into account. Soil moisture has 
been controlled within narrow limits. Concentration of nitrates and the 
hydrogen-ion concentration of the soil solution have been recorded. It 
is not believed that variations in any of these factors are to be regarded 
as having produced the variations in nodule development recorded at the 
different temperatures in these series. 



PLATE I 

A. — Alfalfa plants grown 63 days in soil held at the temperatures indicated. 
B. — Red clover plants grown under exactly similar conditions with the alfalfa 
plants shown in A. 



Effect of Soil Tomporature on Development of Nodules 



Plate I 





Journal of Agricultural Researcli 



Vol. XXII, No. 1 



Effect of Soil Temperature on Development of Nodulos 



Plate 2 



^-n .>^ r^' •* ^ 



27"" 30° ^3° 






Journal ot Agricultural Research 



Vol. XXII, No. 1 



PLATE 2 

A. — Soybean plants grown 63 days in soil held at the temperatures indicated. 
B. — Nodules from 6 soybean plants (only 5 plants at 30° C.) grown 63 days at the 
temperattu-es indicated. One-half of the plants are shown in A. 
54817°— 21 3 



PLATE 3 

A. — Soybean plants inoculated with Bacillus radicicola contrasted with uninoculated 
plants grown 55 days in soil held approximately at the temperatures indicated. The 
pots are grouped according to temperature, with the control on the left and the inocu- 
lated pot on the right in each set. 

B. — Nodules from 9 soybean plants grown 55 days in soil held at approximately the 
temperatures indicated. 



Effect of Gnil Temperature on Development of Nodules 



Plate 3 



A 



15' 








Journal of A_;ricultural Research 



Vol. XXII, No. 1 



INFLUENCE OF THE PERIOD OF TRANSPLANTING 
WESTERN WHITE PINE SEEDLINGS UPON THEIR 
BEHAVIOR IN NURSERY AND PLANTATION 

By E. C. Rogers 

Forest Examiner, Forest Service, United States Department of Agriculture 

At forest nurseries in the northern part of the United States the work 
is customarily crowded into three or four weeks in spring immediately 
following the time when the soil can first be worked. Preferably, the 
stock is lifted, packed, shipped early, and spring sowing and transplanting 
are all usually crowded into this period. At the Savenac Nursery, Haugan, 
Mont., this spring congestion has been keenly felt; and the experiments 
outlined below have had for an object the determination of the safe 
limits of the transplanting season. The results may or may not apply 
beyond the local conditions prevailing in the region of western Montana 
and northern Idaho. 

FALL TRANSPLANTING 

Work was commenced upon this problem at the Savenac Nursery in the 
fall of 1913. By the use of the Mast trencher method, 600 i-year-old 
seedlings of western white pine {Pinus monticola Dougl.) were transplanted 
on each of the four following dates: August 15, September i, September 
15, and October 10. In May, 19 14, these plants were examined and the 
overwinter losses were recorded. Loss by frost heaving, as indicated 
in the figures, includes not only plants completely thrown out but also 
those lying prostrate on the ground, even though they were quite firmly 
attached and still alive, because in that condition they would never 
recover sufficiently to be fit to plant. A few individuals showed the 
symptoms of winterkilling. These were about evenly distributed among 
the four units, in no case amounting to i per cent of the total. Figure 
I shows the loss by frost heaving. From one- third to one-half of the 
plants were heaved out during the cold nights of late October, before 
the coming of snovv^, and during the clear weather of late March and 
early April after the snow left. The loss was greatest in the October 10 
unit. This may be explained by supposing that, because of the warmer 
soil temperature the individuals transplanted earlier had had time to 
make sufficient root growth to render them more resistant to the frost 
lifting, but that those transplanted latest were virtually heeled in. 
However, precise evidence on this point is lacking. 

As it had been suggested that possibly in the Mast V-shaped trench 
a pocket of loose soil was formed around the lower roots and that this 

Journal of Agricultural Research, Vol. XXII, No. i 

Washington, D. C. Oct. i, 1921 

z p Key No. F-6 



34 



Journal of Agricultural Research voi. xxii. No. 



predisposed the plants to heaving, several rows of i-o western white pine 
seedlings, transplanted September 25, 1913, in open plowed trenches 
were likewise examined in May, 1914. From a total of about 12,000 
trees the loss from frost heaving was 29.6 per cent and that from winter- 
killing 1.2 per cent. Here, again, nearly one- third of the plants were 
thrown out — a loss hardly 6 per cent less than that by the Mast trencher 

soo 



4S.0 






^.o 



3S.O 



30.0 



£S.O 



g 20.0 
k 

^ /s.o 



/o.o 



s.o 



o 













/ 












/ 








^ 




/ 








^^ 


, 






/-OSS OP^/ia// /jrcZ/7S/0/c7/7/s 










































^ 






.__ 













/i^ay/7 



Ja/7e2 Ji//7e/S /?uy./S S(sp^/ Se/:>A/S 



Oc/:? 



Fig. I. — I,oss by frost heaving of seedlings transplanted at different dates. 



method, which was used 10 days earlier. Apparently the method mat- 
tered little, the loss having been a necessary consequence of fall trans- 
planting. 

Lorey ^ found that fall transplants of Douglas fir and European larch 
led spring transplants in height growth in the transplant rows. More 
recently, Toumey ^ states that fall transplants lead in earliness of growth 
and in size, provided they escape winter injury. While no data with 

1 Lorey, Tuisko. mitteilungen aus dem forstgarten und KtaTURBETRiEB. n. forstgarten 
INSBESONDERE. In AUg. Forst u. Jagd. Ztg., N. F., Jahrg. 70, p. 193-197. 1894. 

' TouMEv. James W. seeding and planting: a manual for the guidance op forestry students . . . 
xxxvi, 4SS p., 140 fig. New York, 1916. 



Oct. 1, 1921 



Infltience of Period of Transplanting 



35 



respect to those matters were collected, the heavy loss through heaving 
at Savenac Nursery more than balanced any possible gains of that 
kind. Fall transplanting is clearly so.unsafe that no further local experi- 
ments with it are necessary. 

SPRING TRANSPLANTING 

NURSERY TESTS 

Experiments in the spring of 19 13 had for their object the compari- 
son of three lots of 600 i-o western white pine each, transplanted on 
May 17, June 2, and June 16. The first summer's loss from drought 
increased with the lateness of the transplanting period, as is shown also 
by figure i. The June 16 lot looked less thrifty than the others at the 
end of the season, and it was concluded that^ in case of necessity, trans- 
planting could evidently be continued as late as June 15, though it is 
not desirable. 

As a control on the tests made in the spring of 19 13 it was arranged 
to transplant 1,000 i-o western white pine at Savenac Nursery every 
10 days during the spring of 19 14. This was actually done on April 
24, May I, May 9, May 20, May 30, June 12, June 19, June 30, and July 
14. On each of these dates 100 more were removed from the seed- 
bed, of which the weights and measurements appear in Table I. 

Table I. — Weights and measurements of i-o western white pine on different dates of 
transplanting in the spring of IQ14 



Date of 




Average 
length 
of stem. 


Average 
diameter 
of stem. 


Average 


weight. 


Per- 
centage 


Per- 
centage 


Per- 
centage 


Average new spring 
growth. 


trans- 


Lot 

No. 






of plants 

with 

buds 

closed. 


of plants 
with 
buds 

swelling. 


of plants 






plant- 
ing. 


Top. 


Root. 


with 
buds 
open. 


Needles. 


Rootlets. 






Inches. 


Mm. 


Gm. 


Gjn. 








Inches. 


Inches. 


Apr. 24 


I 


1-54 


I. 17 


0. 140 


0. IIS 


100 








0. 


COS 


May I 


2 


1-52 


I 


23 


•i53 


. 122 


100 















10 


9 


3 


I. SO 


I 


13 


•143 


. 090 


90 


10 












20 


20 


4 


1.90 


I 


IQ 


.186 


. 120 





34 


66 




08 




S4 


30 


S 


1.99 


I 


17 


.125 


.oSi 


I 


6 


93 




IS 




62 


June 12 


6 


3.30 


I 


28 


.265 


.150 





I 


99 




27 


I 


07 


19 


7 


2-43 


I 


22 


.203 


• 103 








100 




44 




94 


30 


8 


2. 30 


I 


46 


•313 


•139 








100 




73 


(') 


July 14 


9 


2. 60 


1-53 


.274 


.130 








100 


1. 02 


0) 



1 Not recorded. 



In Table I stem length is the distance from the ground line to the tip 
of the terminal growing point. The stem diameter was measured at the 
ground line. The average new growth of rootlets was based upon the 
longest rootlet noted in each plant examined and not upon all new rootlets. 
This figiure is only relatively correct, because some slight root breakage 
was unavoidable in taking up the plants from the seed bed. Root growth 
data were omitted in the two latest lots, since the older portions of the 



* Unpublished progress report. 



36 



Journal of Agricultural Research voi. xxii, no. i 



new roots were assuming a brown, mature appearance, and this made it 
difficult to determine the margin of growth. 

- Figure 2 shows graphically tlie condition of the seedlings as to length 
of stem and spring growth of needles and rootlets on the different dates 
of transplanting. 




Apr £4 May I May 9 



Julijl-^ 



Maij£0 May 30 June 12 June IS June JO 
Time of 7~ran^ p/a nilng 

Fig. 2. — I^ength of stem and spring growth of needles and rootlets of seedlings transplanted at different 

dates. 



Table I brings out the following points : 

I. Root growth began prior to April 24 during the spring of 19 14, or 
over three weeks before visible stem gro^vth. Although not so indicated 
by the table, it was found that the earliest visible root elongation took 
place in the superficial soil layers, gradually progressing to deeper and 
deeper levels as the season advanced, presumably in response to changes 
in soil temperature. 



Oct. 1, 192 1 



Influence of Period of Transplanting 



37 



2. The swelling of the buds began a little before May 9, and two-thirds 
of them were fully open by May 20. 

3. Although individual variations existed — due largely, it is thought, 
to a lack of uniform density in the seed bed — stem length, stem diameter, 
and weight of top increased in general as the season advanced. 

4. The proportion of the fresh weight of the plants contained in the 
root was greatest early in spring and decreased as the growth pushed 
ahead in May and June. Figure 3 brings out this point. Owing to the 
possibility of variation in the water content of top and root, dry weights 
would be of interest, but circumstances prevented their being obtained. 

so. Or 



o 



4-50 =- 



4-0.0 



JSO 



30.0 





















\ 
















\ 






"~~\ 
















Percsnta^s offrssh \A'eigh1r of plant- 

in root at different dates of transo/anting 




__- 






































































A\/eraffe fresh we/g/it ofp/ant 










■~^~- 
























' ■ 



SO 



/\pri/2'}. Maj'l MayS May 20 May JO June/ 2 June/3 June SO Ju/y/4- 



77/77G' o/' 7rt:?/7S/>/c?/7//no 

Fig. 3.— Proportion of fresh weight of roots of seedlings transplanted at different dates. 

Transplanting was done in adjacent parallel rows, and these were 
irrigated at intervals during the summer of 19 14. As early as August i 
there had come to be a marked differentiation in size and color. The 
April 24, May i , and May 9 units showed particular vigor and had a rich 
green color. The June 30 and July 14 lots showed much the same 
development as the earliest ones but had a yellow color, which gave the 
July 14 lot an almost sickly appearance. The remaining intermediate 
lots showed a healthier color than the later ones but lacked the size and 
development of the latter. This differentiation was increasingly marked 
at the end of the growing season. The loss from drought in these lots 
during the summer was very slight, the heaviest loss, 2.4 per cent, being 
suffered by the July 14 unit. 



38 



Journal of Agricultural Research voi. xxii. No. x 



On September ii, 1914, the season's growth of stem and needles was 
obtained by measuring every tenth plant in each lot, or 100 in each unit. 
The terminal buds of 500 plants in each lot were examined as to their 
maturity on this same date. Buds having a definite form, of a deep 
brown color, and covered with a protective coat of fine hairs were classed 
as mature. Plants without a single well-defined bud and those whose 
growing point had a tender green color, without the coat of hairs, were 
considered of immature development. For purposes of comparison, data 
similar to the foregoing were obtained from 2-year-old western white 
pine plants from a representative area of seed bed. These plants were 
of the same age and seed lot as the nine transplanted units, differing 
only in that they had been allowed to remain in the seed bed. These 
data are given in Table II. 

Table II. — Growth and development of western white pine during the first season in the 

transplant bed 



Date of transplanting. 



Lot 

No. 



Average 

seasonal 

stem growth. 



Average 
seasonal 
growth of 
needles. 



Percentage of 

plants with 

mature 

buds. 



Percentage of 

plants with 

immature 

buds. 



Apr. 24 

May I 

9 

20 

30 

June 12 

19 

30 

July 14 

Not transplanted . 



Inches. 

0-757 
729 

763 

734 
825 
874 
918 
974 

951 
063 



Inches. 
0.934 
952 
897 
424 
566 

570 
611 

715 
943 
131 



80.0 
75-7 
75- 1 
69. I 
70.1 
69.6 

56.4 
80.6 
92. 6 
93-5 



20. o 
24-3 
24.9 
30-3 
29.9 

32-4 

43-6 

19.4 

7-4 

6.5 



By average growth of stem and needle is meant the growth for the 
entire season, regardless of whether that growth took place in the seed 
bed, in the transplant bed, or in both. Needle measurement was made 
in the middle of the sector of currently- grown stem. 

Table II brings out the following points : 

1. The later the transplanting after the buds open, the higher the stem 
growth for the season. It appears that height growth practically ceases 
for a time after transplanting, the plant's energies being directed toward 
getting established in its new habitat. In other words, the height 
growth is roughly proportional to the length of time the plant is left in 
the seed bed. Hence, plants that were not transplanted made a higher 
stem growth than any of the transplanted lots. 

2. The longest needle groAvth at the end of the season had been made 
by the first and last lots (Apr. 24 and July 14). The needle growth of 
the first lot had been made entirely in the transplant bed and was accom- 
panied by a deep green color, but that of the last lot had taken place in 



Oct. 1, I92I 



Influence of Period of Transplanting 



3,9 



the seed bed before transplanting and the needles of these looked much 
less vigorous. As figure 4 shows, the season's needle growth commences 
to fall with the May 9 lot, drops abruptly with the May 20 lot, then 
climbs gradually until the last lot equals the earliest. The stock that 
was not transplanted produced longer needles than any of the trans- 
planted units. 

3. The difference in dates of transplanting had a pronounced effect 
upon the maturing of the fall buds. The earliest maturing lots were 
the two that were transplanted latest (June 30 and July 14). The less 



so 

k.O 

0.0 
%^^ 

c 
2.0 

1.6 

10 

06 

























Height 


growth erf 
in 1316. pric 














/ 


planfj 


r +oJali/ 1. 












/ 




/ 




\ 










/ 




/ 




N 





\ 












/ 








\ 








Growth 


\ of naedfe 


3 during 






\ 








planti 
dif 

/ 


fran^pi^ni^d at 








^ 






s/ 








__^ 


--^^ 








^v 


^ 


: 1 











AprZ^ Mayl Maij9 May20 MayJO Junel2 JunelS JuneJO Ju/(//4 Mf 

Time of Transpthntinq planted 

Tig. 4.— Increase in height and growth of needles of seedlings transplanted at different dates. 

favorable weather conditions appear to hasten preparations for winter 
by stock transplanted in summer. 

4. Seedling stock of the same age and source (2-0) led transplants 
(i-i) in current stem growth and needle development no matter what 
the period of transplanting. The shock of the treatment, expressed 
quantitatively, resulted in a loss of 0.3 inch of stem growth and 0.2 inch 
of needle growth, even when the transplanting was done at the most 
favorable period. 

At the end of the season in which the transplanting was done it ap- 
peared that the plants lined out before the buds were open had suffered 
the least shock, and, judging from their unhealthy appearance, those 
transplanted in midsummer (July 14) seemed to have suffered most. 
No single item of weight or measurement appears to be a consistent 
indicator of the degree of severity of the shock. 



40 



Journal of Agricultural Research voi. xxii. No. 



On July I, 1916, 100 plants were washed out from each of these nine 
transplanted units by the aid of water under pressure. At this time, 
the stock could be considered to be in the i-2>^ age class. Data from 
these 900 plants are assembled in Table III. 

Table III. — Weights and measurements of 1-2% -western white pine 





Lot 

No. 


Average 

stem 

height 

growth 

(cur- 

reiit).a 


Average 

stem 
diameter. 


Average number of laterals. 






Date of 
trans- 


First order. 


Second order. 


Average 

total 

fresh 

weight 

of plant. 


Percent- 
age of 
weight 
in root. 


planting. 


2 inches 
and up. 


0.5-inch 

to 2 
inches. 


2 inches 
and up. 


0.5-inch 

to 2 
inches. 










Apr. 24 

May I 

9 
20 

30 
June 12 

19 

30 

July 14 


I 
2 

3 
4 
5 
6 

7 
8 

9 


Inches. 

2-75 
2-75 
3.66 
2.09 
3-27 
3-58 
2.8s 
2.74 
1.82 


Mm. 
6.31 
6.72 
7.14 
5-65 
6.48 
6.48 
6.07 
6.27 
4-95 


10.2 
10.3 
10.3 

9.0 
10. 5 

8.1 

9.1 
10.2 

2.6 


5-4 
6.7 
6.1 
6.1 
4-5 
5-2 
5-3 
5-1 
5-3 


3-4 
6.3 
4.1 

3-2 
2.9 
3-6 
3-7 
3-8 

2-5 


15-3 
27.4 
17.2 
13.0 
13.0 

II-3 

10.5 

12.4 

9.9 


Gm. 

3-09 
3.16 

3-78 
1.84 
2.78 
2.74 

2.54 
2.62 
1. 41 


43-4 
45-6 
36.2 

45-7 
41.4 

36.9 
39-4 
42.7 
48.2 



o By current stem growth is meant the 1916 growth prior to July i. 



73 



7.0 



6J 



GO 






§ s.o 

J.S 

ao 
2.6 





^^ 














^ 
















^ 






/ 









^ 








^ Diamett 
(me, Juli 


1 
r of sfern af ground 
/.ISI6, of stock trans- 




\ 








plan-f-e 
in the 


d at differ 
spring. of 


enr dates 




\ 



















































































Apr. 2^ Maul 



Maij9 Maij20 Mat/JO JunelS Juna/S JuneJO Julijl^ 
7~irrre of 7~ransp/aniing 

"Bic. 5. — Increase in diameter of stem of seedlings transplanted at different dates. 



Oct. 1,1921 Influence of Period of Transplanting 41 

A study of the proportion of the fresh weight of plant in the root 
system shows that the relation between this proportion and the time of 
transplanting, which was so evident two years before, had entirely dis- 
appeared. All the plants were washed off, and the surfaces were allowed 
to dry in the air; but as the evaporating power of the air varied, it was 
not possible to compare directly the average fresh weights obtained at 
tlie time of transplanting with those obtained on July i, 19 16. All 
the weights obtained on the latter date are, however, comparable with 
one another. Figures 3, 4, and 5 illustrate further the current height 
growth of stem, the stem diameters, and the total fresh weight of the 
plants. 

A striking similarity will be noted in the curves in figures 3 (weight), 
4, and 5. All agree in shovnng two minima, the first for the May 20 lot, 
and the second and still lower point for the July 14 transplanting. This 
bears out the conclusion reached in the fall of 19 14 that the shock of 
transplanting fell hardest upon the July 14 lot. But it now appears that 
the transplanting on May 20 was almost equally lasting in its unfavorable 
influence upon growth in the transplant bed. Between these two dates 
there appears to have been a period of about a month in which trans- 
planting worked less injury to the plant. 

As the roots of these plants were removed by washing, and with a 
minimum of breakage, it seemed worth while to compare the average 
numbers of laterals of the first and second orders. Figure 6 shows 
graphically the data obtained. In order to bring the curves close 
together, and thus facilitate comparison, the actual average number of 
lateral rootlets of the second order, between 0.5 and 2 inches in length, 
has in each case been divided by 2 in plotting the curves, and the average 
total number of lateral rootlets in the four classes has similarly been 
divided by 5. Because of the difficulty involved in recording them, and 
the limited time available, no records were obtained of the number of 
laterals of higher orders than the second, nor, in any case, of laterals less 
than 0.5 inch in length. The figures obtained are, however, considered 
indicative of the general nature of the root system. 

There is a close relation not only between the average number of 
laterals in each of the two length classes of the second order but also 
between these and the total number of rootlets of the recorded classes. 
On the other hand, the curves for the two classes of the first order run 
quite differently. Yet, so far as determining the total curve is concerned, 
these two classes could obviously have been neglected. There is a con- 
sistent decline in the fibrous development of the root system as the 
transplanting season advances. Transplanting in early summer, and 
even more so in late summer, holds back lateral root growth in the trans- 
plant bed, the effect being strikingly noticeable two years afterwards. 



42 



Journal of Agricultural Research voi. xxii. No. 



Although several seasons' observations had shown no noticeable loss 
through winter frost heaving of i-o western white pine transplanted in 
April and in early May, it was found in the spring of 1918 that of 2,922 
i-i white pine transplanted June 15, 19 17, in connection with another 
experiment, 1,152, or 39.4 per cent, were heaved out during the late 
fall, winter, and early spring of 19 17-18. This is nearly as heavy a loss 




Ma(/9 May20 Mat/30 Junel2 JunelS JuneJO Jultjl^ 

Time of Transplanfina 



Apr 24 May I 

Fig. 6.— Number of lateral rootlets on seedlings transplanted at different dates. 



as had previously been recorded for fall transplanting. However, an 
unprecedented snowless period in December probably contributed to 
produce this result. A greater susceptibility to frost heaving on the 
part of late spring and summer transplants is the natural result of the 
poorer root development just referred to. The plant must rebuild its 
entire root system late in the season and so has a relatively poor anchor- 
age when the frost comes. Furthermore, Cannon * has found that the 

1 Cannon, William Austin, root habits oP desert plants. 96 p., x7 fig., 23 pi. Washington, D. C„ 
igii. (Carnegie Inst. Wash. Pub. no. 131.) 



Oct. 1, 1921 Influence of Period of Transplanting 43 

formation of an abundant lateral root requires a favorable water content 
in the soil and a sufficiently high soil temperature. Although summer 
soils are warm, yet, in spite of occasional irrigation, the greatest loss of 
transplants from drought at Savenac Nursery occurs during July and 
August, indicating that there is less available soil moisture during that 
period, or at least that there is a smaller balance for growth when the 
transpiration loss of the plant has been met. 

It seems, therefore, that in the foregoing series the plant organism 
was most deeply disturbed by being transplanted in midsummer. This 
appeared to be a consequence of the high evaporation and lack of moisture 
in the soil, along with the greater topheaviness of the plant. The May 
20 transplants gave evidence of having been most severely set back, a 
result which must be attributed either to external conditions or to the 
internal state of the plant. The Savenac Nursery weather records show 
a precipitation of 2.16 inches in April, 1914, well distributed throughout 
the month, with only seven clear days. In May, previous to the 
twentieth, there fell 0.58 inch of rain, and 12 out of 19 days were cloudy 
or partly cloudy. On May 20, the soil was well stored with water and 
was favorable for the reception of plants. The maximum temperature 
on that day was 72° F., and it and the eight days following were partly 
cloudy. During the period from May 22 to 28, inclusive, 0.46 inch of 
rain fell, every day yielding at least a trace. The weather and soil 
conditions were, therefore, sufficiently favorable to convince the writer 
that the reason for the marked checking of the growth of the May 20 
lot lay in the developmental stage of the plant itself. One-year-old 
western white pine seedlings, whose buds are just opening and whose 
tiny new needle fascicles are less than i/io inch long, show a particular 
sensitiveness toward removal and replanting. 

FIEI/D TESTS 

The influence of the season of transplanting upon the behavior of the 
tree in the plantation is of special interest to the forester. One hun- 
dred of the plants from each of the nine spring lots described above 
were planted October 6, 19 15, on the Wallace experimental area, near 
Wallace, Idaho. A northwest aspect — a typical white pine planting 
site — ^was selected. One row was devoted to each lot, and the rows were 
placed adjacent to each other and parallel, extending up and down the 
slope. The place where each tree was to be planted was previously 
marked by a cedar stake 16 inches long, whose top had been dipped in 
white paint to make it conspicuous among native cover plants. Each 
stake bore its lot number in black lumber crayon. The same man 
planted all the rows, using a uniform method. 



44 



Journal of Agricultural Research voi. xxii. No. 



On September 21, 19 16, and on November 7, 19 17, these plants were 
examined, their condition was noted, and the average height growth of 
stem was recorded, this average being based upon all vigorous living 
plants. Table IV gives the principal data secured. 

Table IV. — Date of transplanting, average current height growth, and percentage of trees 
surviving at the end of the igi6 and I gi J field seasons on the Wallace area 



Date of transplanting — 1914. 



Apr. 24. 

May I . 

9- 

20. 

30 
June 12. 
19. 

30- 
July 14. 



Average of all lots. 



Lot 
No. 



Average current 
height growth. 



1916 



Inches. 

0-57 
.91 

•95 
.82 
.67 
.96 
.82 
I. 00 
■65 



Inches. 
O. 62 

83 
96 
81 

75 
95 
84 
IS 



84 



Percentage of 
trees surviving. 



Fall 1916. Fall 1917 



95-9 



90.9 

95-7 
93-6 
94.6 

92-5 
92.8 

93-7 
98.9 
96. 2 



94-3 



The fact that the unusually dry summer of 191 7 caused almost 
negligible losses makes improbable any further changes of importance in 
the survival standing of the nine lots. The percentage of living trees of 
all lots in the fall of 191 7 was above 90. There is no superiority on the 
part of the early lots, the April 24 units standing lowest. The later 
lots have, on the whole, lived best. Both the May 20 and July 14 
plants, while outclassed in the transplant bed, showed better than an 
average survival in November, 191 7. In fact, the time of transplanting 
had no apparent influence in the field. 

Figm-e 7 further illustrates the height growth of these plants. There 
is a marked similarity between the growth curves for 1916 and 191 7; 
hence each resembles the total growth curve for the entire two seasons. 
For instance, in each of the three curves the average point for the June 
19 lot falls exactly upon the horizontal average line for that curve. To 
facilitate comparisons, the height growth curve from figure 4 is plotted 
in figure 7 also. This renders it possible to compare the growiJi made in 
the plantation the first year after planting (1916) Avith that made the 
same season prior to July i by individuals left in the transplant beds 
(curves A and D, respectively). 

There are certain points of resemblance between the curves of growth 
in transplant bed and in field. The May i and June 19 lots stand upon 
or very near the horizontal average line in both. The July 14 transplants 
stand low, and the May 9 and June 12 transplants stand high in both. 
But, on the other hand, the May 20 lot, which had a low growth rate in the 



Oct. 1, 1921 



Influence of Period of Transplanting 



45 



transplant bed, lias an average rate in the field; but the May 30 and June 
20 units reverse their positions with respect to the horizontal average 
lines. The planting out of the stock has evidently caused a somewhat 




Apr so May/. Maj'9 MaySO Ma/JO June I2 June/3 JuneJO July l4 
T/mci of Transplanting 
Fig. 7. — Increase in height of seedlings transplanted at different dates. 

general rearrangement of growth rates in which the later lots tend to 
overtake the earlier ones, this rearrangement being accentuated after a 
second season in the field. The inferiority of the May 20 plants has 



46 Journal of Agricultural Research voi. xxii, no. i 

disappeared; the July 14 stock slightly surpasses the April 24 lot; and 
there is nothing to indicate that summer transplants are not fully the 
equal of spring transplants, so far as growth after planting is concerned. 

CONCLUSIONS 

Transplanting western white pine seedlings at any time in the fall is 
not a safe practice at Savenac Nursery, because the frosts of the following 
late fall and early spring heave out the young plants. 

Results in the plantation thus far indicate that, where 1-2 stock is 
grown, the transplanting season may be extended from a date as early 
in spring as the ground can be worked until early July. The shock of 
removal from the seed bed is greatest when the transplanting is done in 
midsummer, on account of the high evaporating power of the air, the 
low water content of the soil, and the greater top-heaviness of the plant 
with its considerably increased transpiring surface. A particularly 
sev^e shock was also suffered when transplanting was done at the time 
of bud opening and before the rudimentary needle fascicles had reached 
■^^g-inch in length. Stock transplanted at either of these critical periods 
lagged behind the other lots for at least two years in the transplant 
bed. However, when they were planted in the field as 1-2 stock, the 
plants survived as well as the others, with little, if any, inferiorit}'- in 
growth on account of the considerable rearrangement of growth rates 
following the planting. 

There are, nevertheless, other reasons which make transplanting at 
Savenac Nursery safer in spring than in summer. June and July trans- 
plants suffer more from drought the first season and, because of their 
poor root development at the end of the growing season, are more subject 
to frost heaving the following winter. The first disadvantage can be 
met by proper irrigation, but the second can not readily be prevented. 

The safest practice, therefore, is to confine this work as much as 
possible to April and early May. 

The foregoing conclusions apply to i -year-old seedlings which are to 
remain two years in the transplant rows, 1-2 stock being the only age 
class of white pine transplants at present grown at Savenac Nursery. 

Certain points brought out by this study may have an important 
bearing upon the season for field planting. It is probable, for instance, 
that subsequent lateral root development in the plantation may be 
decisively influenced by the time of planting. 



A DRYROT CANKER OF SUGAR BEETS 

By B. L. Richards 
Department of Plant Pathology, Utah Agricultural Experiment Station 

What appears to be an undescribed rootrot of the sugar beet was first 
called to my attention on August 5, 1920, by Mr. A. H. Bateman. Speci- 
mens of the diseased beets collected at this date at Cornish, Utah, ex- 
hibited numerous brown, circular lesions that varied from j^ inch to i 
inch in diameter (PI. 4 ; 8, D ; 9, A, B) . The outer surface of the root cov- 
ering these lesions, which in most cases remained entire, had so sunken as 
to give a definite undulating contour of alternating light and dark brown 
concentric areas or rings (Pi. 4). The removal of this outer layer of cells 
of the older lesions exposed deep cankers or pockets filled with hyaline- 
mycelium embedded in the dry remains of partially decayed host cells. 
This accompanying mycelium, when exposed to the atmosphere through 
the cracking open of the outer covering, appeared dark brown in color 
and immediately suggested the typical mycelium of the sterile, or 
"Rhizoctonia," stage of Corticium vagum B. and C. The general prev- 
alence of black sclerotial bodies on the outside of the diseased beets 
(PL 7, A), together with the microscopic examinations made at this time, 
confirmed this initial suggestion. 

An examination of the field from which these first diseased specimens 
were taken revealed the trouble to be of considerable economic impor- 
tance; at least 20 per cent of the beets in this field of 40 acres were 
diseased. The disease appeared to be confined to definite areas wherein 
every beet might be found infected. These diseased spots varied con- 
siderably in size and appeared to be widening most rapidly in the direc- 
tion parallel with rows. Three adjacent fields were found at this time 
to be seriously diseased, but none to the same degree as the field first 
visited. 

The progress of the disease in these fields appeared of such ominous 
character as to require immediate investigation. However, as the 
season was well advanced, little more than preliminary experiments 
were undertaken. The results to date, while definite, are not sufiiciently 
extensive to warrant final conclusions, and many of the important rela- 
tions of the disease remain obscm-e; nevertheless it is felt that the 
apparent economic importance of the trouble justifies a preliminary 
description at this time. 

The disease is first detected in the field by abnormal wilting of the 
leaves in the daytime with partial or complete recovery at night. Later 

Journal of Agricultural Research, Vol. XXII, No. i 

Washington, D. C. Oct. i, 1921 

zm Key No. Utah-14 

54817°— 21 4 (47) 



48 Journal of Agricultural Research voi. xxii. no. i 

the older leaves fail to recover, turn brown, and die. This dying of the 
outer or older leaves continues with the progress of the disease in the 
root until all the leaves on the affected beets may succumb. Localized 
browning frequently occurs in the blade and petiole, but to date no 
suggestion of a parasitic relation has been found. Neither the petiole 
decay reported by Duggar (2) ^ nor the "western crownrot" described by 
Edson (j) have been found associated with thedryrot canker in the field. 
A peculiar type of crownrot, however, is found late in the season, usually 
well toward harvest time (PI. 6; 8, B). A study of a number of these 
crownrot specimens indicates definitely that the causal organism enters 
the beet below the stu-face of the soil and works upward in the tissues, 
eventually destroying the crown. The fungus has not been observed 
to attack the beet above the soil line. 

It is evident that the fungus is unable to destroy the outer corky cells 
of the beet root, but gains entrance to the inner tissue at a definite 
point and works tangentially just beneath this outer layer. As the 
fungus eats its way from the point of entrance the outer tissues, due to 
killing and subsequent drying out of the cells beneath, sink in such a 
manner as to produce the circular lesion with its very definite undulating 
contour of alternating raised and sunken concentric "rings" (PI. 4). 
The lesions appear first as a small, brown, sunken spot with a minute 
perforation in the center (PI. 8, D). The first definite concentric "ring" 
which is considerably sunken below the central area and usually dark 
brown in color is noted before the lesion reaches a diameter of >^ inch. 
With continued enlargement a second and somewhat broader "ring," 
less sunken and much lighter in color, results. Similar concentric areas 
are developed alternately until the fimgus reaches its limit of lateral 
spread. Individual lesions resulting from a single point of infection 
may obtain a size of from ^ to i inch in diameter and develop as many 
as eight alternate "rings" (PI. 4). When, however, adjacent lesions 
become confluent, as they frequently do (PI. 4; 8, D; larger 
lesions result which may in severe cases cover a large part of the root 
surface. In such cases large concentric rings are produced, which become 
common to a number of centers of original infection (PI. 4; 8, D; 9, A, B). 

Another characteristic feature of the disease results in cases where 
infection occurs at or near the apex of the root. The root in such an 
event is usually severed at the point of infection and the fungus advances 
upward, producing the typical dryrot with resultant concentric rings 
which may encircle the entire root (PI. 7, B). Again, cankers may occur 
with such frequency as to girdle completely the root (PI. 8, A). 

The distinctive feature of the contour, as shown in Plates 4 and 8, D, 
is obtained usually before the fungus penetrates deeply into the tissue 
of the beet and before a serious rupture of the outer layer occurs. With 

1 Reference is made by number (italic) to "I,iterature cited," p. 52. 



oct.i.i93i A Dryrot Canker of Sugar Beets 49 

the drying out and final cracking of this outer covering the fungus, 
possibly because of a better oxygen relation, eats radially into the beet, 
producing deep cankers (PI. 5 and 6). The decaying tissues rapidly 
dry out as the fungus advances inward, leaving the cavity partially 
filled with a dry, pithy residue. Frequently the content of the canker 
appears as a definite plug, which, upon wetting, may be removed intact 
from the cavity of the canker (PI. 9, D). 

Except for slight cracking, the outer layer of dead cells remains entire 
and furnishes a definite covering until the lesion has reached approxi- 
mately its limit of tangential spread. As the cells of this outer covering 
finally dry out the central perforation enlarges and ultimately gives rise 
to a definite crack which may extend the entire diameter of the lesion 
(PI. 4; 8, D; 5). Frequently adjacent cracks become confluent, 
resulting in large characteristic fissures, which in severe cases of the 
disease may obtain from 2>^ to 3 inches in length and from % to 1% 
inches in depth (PI. 5, 6). With numerous points of attack the beet by 
harvest time is converted into a dry, brittle shell filled with a pithy mass 
of host and fungous debris (PI. 6). 

During the season careful study was made of a large number of the 
beets taken from each of the different fields in which the dryrot had 
been found. In all cases the characteristic cankers exhibited the pres- 
ence of the sterile stage of Corticium vagum. This fungus, it was found, 
may be obtained regularly in a pure form from any part of the typical 
canker, provided the outer covering of the lesion is not previously de- 
stroyed. The brown layer separating the normal from the diseased 
tissue (PI. 8, A-C) has never failed to yield the fungus free from other 
organisms, and even from the open lesions cultures have been obtained 
with remarkable ease and regularity. The degree to which other organ- 
isms are found to be excluded is phenomenal. 

To determine the etiological relation of the fungus, inoculations were 
made September 3 on partially grown beets. In the process of inocula- 
tion the soil was removed to a depth of approximately 4 inches from 2 1 
beets in each of five rows. Each of the 2 1 beets in the first row was 
punctured a number of times with a sterile needle, and the inoculum, 
consisting of the beet fungus, grown for several days on potato agar, was 
then scattered throughout the soil as the latter was replaced about the 
beet. Row 2 was inoculated exactly as row i except that in place of 
needle punctures slight incisions were made in the beet by the use of a 
sterile scalpel. The beets in rows 3 and 4 were wounded as in row i, 
and the soil was inoculated with two different "strains" of Corticium 
vagum,} Row 5 was left uninoculated, and the beets after wounding as 
in rows i and 2 were covered and grown as controls. All the wounded 

1 These "strains" were obtained from the surface of a potato tuber in 1918 and have proved virulent 
on potato stems in both sterilized and unsterilized soil. 



50 



Journal of Agricultural Research voi. xxii. No. i 



beets in the control row healed normally. Infection occurred on but 
one beet in rows 3 and 4. The other beets in these two rows healed as 
perfectly as in row 5. The results of inoculation with the sugar-beet 
strain of the fungus in rows i and 2 are given in Table I. The types 
of lesions produced as a result of atrificial inoculation are shown in 
Plate 9, A. D. 

Table I. — Number of lesions on sugar beets inoculated with the sterile stage of Corticium 

vagum 



Beet No. 


Row I, 

needle 

puncture. 


Row 2, 
incision. 


I . 


23 

7 

II 

13 
8 
8 
17 
14 
II 

17 
6 

17 
13 
17 
20 

19 
10 

15 
II 
12 
16 


8 


2 





3- • 
4. . 




\.A\^.^:^uu^^ 


3 

13 

8 


5 


6 


2 


7 


6 


8 : 


2 


Q 


6 


10 .;. 


4 


II 


4 
6 


12 


I -J 


7 
8 


14 


IC 


3 
2 


16 


17 


6 


18 


4 
4 



10 


20 


21 - - 







Total 








285 
13-5 


96 
4-5 




Average 











Instructions for the inoculating of sugar beets with the beet fungus 
without puncture or incisions were not followed. As a result the question 
as to the ability of Corticium vagum to attack the sugar beet independ- 
ently of other agents remains unsettled. It is quite conceivable that 
sugar-beet root aphis {Pemphigus betae Sloane) and other insects so 
prevalent in the soil may serve an important function in the initial en- 
trance of the fungus. Having once gained access to the lower tissue, 
however, it appears evident from the results that this particular "strain" 
of C. vagum is capable of producing the type of canker and dryrot with 
which it is so constantly associated in the field. 

The peculiar method of decay, together with the sharp line of demarca- 
tion between the diseased and the normal tissue (PI. 8, A-C; 9, C-F), pro- 
vide the most distinctive characteristics of the disease. A dark brown, 
watery layer invariably separates the dry, decayed mass occupying the 
cavity of the canker from the normal host tissue beneath. This layer 



oct.i.igzx A Dryrot Canker of Sugar Beets 51 

when examined under the microscope is found to be composed of masses 
of hyaline, vigorously growing young hyphae ramifying through and 
between the rapidly decaying host cells. It is in this advance layer 
that the major portion of the tissue destruction occurs. The brown 
layer advances uniformly inward by additions from the normal host 
tissue, while the outer surface of the layer rapidly dries out and con- 
stantly contributes its substance to the pithy mass occupying the resulting 
cavity of the canker. The thickness of the layer is dependent largely 
upon the rate at which the moisture is lost from it outer surface as the 
fungus eats its way radially into the normal tissue. No evidence of direct 
penetration of tlie normal cells by the fungus has been found. On the 
other hand, it appears that dissolving enzyms precede considerably the 
advancing mass of young hyphae (PI. 8, C). 

This method of tissue destruction resembles in a very definite way that 
described by Ramsey (5), by which Rhizoctonia solani Kiihn attacks 
and produces a definite pitting of the mature potato tuber. A similar 
process of decay is described by Atkinson (r) for the "sore shin" of 
cotton. He states that — 

the fungus {Rhizoctonia solani) never seems to penetrate far into the living tissues, but 
kills as it goes, and the tissues become brown, depressed, and present the appearance 
of a plant having a deep and ugly ulcer at the surface of the ground. 

A type of decay most accurately resembling this particular beet rot is 
described by Richards (7) for the potato stem-canker caused by Corticium 
vagum. 

The early production of definite cankers by a slow corroding of the 
normal tissue, finally resulting in a complete dryrot of the beet, suggests 
a possible name "dryrot canker" for the disease here described. 

Various American workers. {2; 4, p. 243-^4; 3) have reported rootrots 
of the sugar beet which they attribute to the work of Rhizoctonia solani 
Kiihn. It appears difficult at this time, however, to determine the possible 
relation of these to the particular type of dryrot described in this article. 
The indefiniteness of the literature on the subject in fact does not justify 
any general statement as to the possible distribution of the disease. 

During September and October of 1920 a preliminary survey ^ was 
made of the beet-growing districts in four counties of Utah — Cache, 
Davis, Utah, and Salt Lake. The disease was found in 18 fields of the 
51 visited in Cache County and in 3 fields of the 20 surveyed in Davis 
County. Very serious damage occurred in a number of these fields. 
No indication of the trouble v/as found in either Utah or Salt Lake 
Counties. 

The limited survey does not permit of an estimate of the loss to the 
total sugar-beet crop of the State; nevertheless, the general prevalence 

1 This survey was conducted in cooperation with the Office of Plant Disease Survey, United States 
Department of Agriculture. The author wishes to express his indebtedness to Dr. G. R. Lyman for this 
support. 



52 Journal of Agricultural Research voi. xxii. No. i 



of the trouble would indicate that under more favorable conditions the 
disease may become a serious factor in beet culture. It is not improbable 
that a thorough survey may discover the "dryrot canker" in every beet- 
growing district in this and surrounding States. 

Since the appearance of the author's abstract (6), Dr. George L. Peltier 
reports in a letter to the author that he noted during 1920 in Nebraska 
what appears to be the same trouble. Preserved specimens in the 
plant-disease herbarium of the Utah Agricultural College show that the 
disease was collected in Utah as early as 19 15. 

LITERATURE CITED 
(i) Atkinson, Leo. F. 

1895. DAMPING OFF. N. Y. Cornell Agr. Exp. Sta. Bui. 94, p. 231-272, fig. 
55. 6 pi. 

(2) DUGGAR, B. M. 

1899. THREE IMPORTANT DISEASES OF THE SUGAR-BEET. N. Y. Cornell AgT. 

Exp. Sta. Bui. 163, p. 335-363, fig. 49-63. Some references to the lit- 
erature of beet diseases, p. 361-363. 

(3) Edson, H. a. 

i915. seedling diseases of sugar beets and their relation to root-rot 
AND CROWN-ROT. In Jour. Agr. Research, v. 4, no. 2, p. 135-168, pi. 
16-26. Literature cited, p. 165-168. 

(4) Pammel, L. H. 

1891. FUNGUS DISEASES OF SUGAR BEET. lowa Agr. Exp. Sta. Bui. 15, p. 
234-254. 6 pi. 

(5) Ramsey, Glen B. 

1917. A FORM OF POTATO DISEASE PRODUCED BY RHIZOCTONIA. In Jour. Agr. 

Research, v. 9, no. 12, p. 421-426, pi. 27-30. 

(6) Richards, B. L. 

1921. a dry rot of the sugar-beet caused by corticium vagum. 
(Abstract.) In Phytopathology, v. 11, no. i, p. 48. 

(7)- 

1921. THE POTATO STEM CANKER. Utah Agr. Exp. Sta. Bui. 178. 



PLATE 4 

Sugar beet showing typical lesions as a result of natural field infection. Lesions 
as sho^v^ may become confluent and develop common concentric rings. Initial 
stages in fissiu-e formation are also evident. Photographed August 8, 1920. 



A Dryrot Canker of Su::ar Beets 



Plate 4 




Journal of Agricultural Research 



Vol. XXII, No. 1 



A Dryrot Canker of Surrar Beets 



Plate 5 







Journal of Agricultural Research 



Vol. XXII, No. 1 



PLATE 5 

Sugar beet showing various stages in the rupture of the outer covering of the lesion 
resulting in the formation of deep fissures. The lesions shown on this particular 
beet have not reached the size normally attained before rupture occurs. 



PLATE 6 

Late stage in the development of the disease, showing the beet as a dry shell par- 
tially filled with a pithy residue composed of mycelium and dead host tissue. The 
decay of tlie crown of this beet is a result of the fungus working upward from the point 
of infection below the surface of the soil. Remnants of the concentric rings of typical 
lesions are clearly visible. The cracking of the outer surface of the beet at this stage 
is shown to extend beyond the lesions. 



A Dryrot Canker of Sugar Beets 



Plate 6 




Journal of Agricultural Research 



Voi. XXII, No. 1 



A Dryrot Cankor of Su':ar Beets 



Plate 7 





Journal of Agricultural Research 



Vol. XXII, No. 1 



PLATE 7 

A. — Portion of a sugar beet showing the typical sclerotial masses commonly found 
adhering to the beets in the infested areas. 

B. — Sugar beet showing the result of natiu-al infection near the apex of the beet, 
at which point the root has been completely severed. The dryrot advancing upward 
from the initial point of attack has produced the typical undulating contour so char- 
acteristic of the small lateral lesions. 



PLATE 8 

Sections of diseased sugar beets, showing the abrupt drying out between the dis- 
eased and healthy tissue. The prominent "feeding surfaces" composed of recently 
killed cells and the young hypha is clearly evident. 

A. — Cross section, showing complete girdling of the beet by cankers resulting from 
separate points of infection. In such cases the continued penetration of the fungus 
may completely sever the root at the line of greatest infection. 

B. — Longtitudinal section of diseased beet, showing various stages of decay and 
the pulpy material partially filling the cankers. 

C. — Sugar-beet crown, showing the definite type of crownrot caused by the fungus 
worked upward from a point of infection below the soil surface (PI. 6). A small region 
of healthy tissue is shown to which a few sickly leaves were attached. 

D. — Section of beet stuface, showing progressive stages in the development of the 
lesions resulting from natural infection. The earliest visible stage is shown to exhibit 
a slight perforation of the outer surface at the center of the lesion. This small open- 
ing, present in all lesions, gradually enlarges with age and finally results in the large 
fissures (PI. 5). Various stages in the coalescence of lesions are especially evident. 



A Dryrot Canker of Su^ar Beets 



Plate 8 







Journal of Asricultural Research 



Vi.1. XXII, No. 1 



A Dryrot Canker of Sugar Beets 



Plate 9 








Journal of Agricultural Research 



Vol. XXII, No. 1 



PLATE 9 

A, B. — Beets showing typical lesions produced by artificial inoculation. Needle 
punctures through which the fxmgus entered permitted of rapid drying out of the 
diseased tissue and of an early ruptiure of the outer layer of cells at the margin of the 
lesions. A number of the lesions, however, show the concentric rings so character- 
istic of the disease produced by natural infection. Cross sections of these lesions 
are shown in C and D. 

C, D. — Cross sections of the lesions in A and B. The lesions in C disclose the more 
advanced stage of the disease wherein the outer layers of cells are broken down. In 
D the outer layers of the lesions are more or less entire. 

E, F. — Cross sections of cankers resulting from natural infection in the field. A 
more advanced stage is shown than in C and D; otherwise the lesion produced by 
the natural and artificial method of inoculation appeared identical. 



ADDITIONAL COPIES 

or THIS PUBLICATION MAT BE PROCUBED FROM 

THE SUPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTINO OFFICE 

■WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 

SUBSCRIPTION PRICE, $3 PER YEAR 

V 



Vol. XXII OCTOBER 8, 1921 No. 2 

JOURNAL OF 

AGRICULTURAL 

RESEARCH 



CONTENTS 



Page 



Comparative Vigor of Fi Wheat Crosses and Their 
Parents - - -- - -- - - 53 

FRED GRIFFEE 

(Contribution from Minnesota Agricultural Experiment Station) 

Temperature and Humidity Studies of Some Fusaria Rots 
of the Irish Potato - - - -- - -65 

R. W. GOSS 

( Contribution from Nebraska Agriculttual Kzperiment Station) 

Blackleg Potato Tuber-Rot under Irrigation - - - 81 
M. SHAPOVALOV and H. A. EDSON 

(Contribution from Bureau of Plant Industry) 

Microscopic Study of Bacteria in Cheese - - - - 93 

G. J. HUCKER 

(Contribution from New York Agricultural Experiment Station) 

Further Studies on Relation of Sulphates to Plant Growth 
and Composition - -- - - - - - 101 

HARRY G. MILLER 

(Contribution from Oregon Agricultural Experiment Station) 

Soybean Mosaic - -- - - - - -111 

MAX W. GARDNER and JAMES B. KENDRICK 

( Contribution from Indiana Agricultural Experiment Staition) 



PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE' 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



^W^ASHINQXON, D. C, 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCUTION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

ElARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entonwlogist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 
J. G. LIPMAN 

Dean, State College of Agriculture, and 
Director, New Jersey Agricultural Etptri- 
menl Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief. Dtvisiott of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS 

Dean, School of Agriculture, and Direetor. 
Agricultural Experiment Station. Th» 
Pennsyhania State College 



All correspondence regarding articles from the Department of Agricult^ire should be 
addressed lo Karl F. Kellennan, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultural Experiment Station, New 
Brunswick, N. J. 



JOIMALOFAGBiaiTllRALffiSEARCH 

Vol. XXII Washington, D. C, October 8, 192 1 No. 2 



COMPARATIVE VIGOR OF F^ WHEAT CROSSES AND 
THEIR PARENTS ' 

By Fred Griffee ^ 

Instructor in Plant Breeding, Division of Agronomy and Farm Management, Depart- 
ment of Agriculture, University of Minnesota 

The comparative vigor of Fj crosses and their parents is a subject of 
much interest to the plant breeder. In crops where the technic of cross- 
ing is comparatively easy, the increase in vigor obtained in the F^ cross 
often more than pays for the additional trouble of producing the hybrid 
seed. In self-fertilized crops like the small grains where considerable 
labor is involved in making artificial crosses, it is apparent that F^ crosses 
can not be used commercially as a means of increasing crop yields. The 
suggestion, however, has been made by Anderson (ly that the added 
vigor of the heterozygous condition might be utilized in small grains by 
making a large number of crosses between strains which, when crossed, 
show a considerable increase in yield. Produce from Fj and F3 progeny 
could be used for seeding the general field, and the crosses could be 
repeated each year in order to keep up the supply of seed. 

Several theories have been advanced to explain the phenomenon of 
heterosis. The discovery of genetic linkage has led to the development 
of an adequate Mendelian explanation of the vigor so often obtained in 
Fi crosses. An excellent review of the development of this theory is 
given by East and Jones (<?). The theory explains the increase in vigor 
shown in the first hybrid generation as being due to the meeting in the 
zygote of dominant or partially dominant growth factors some of which 
are contributed by each parent. Linkage is given as the reason why all 
dominant factors can not be combined in a homozygous individual. Ac- 
cording to this hypothesis the maximum number of favorable growth 
factors can be obtained only in the heterozygous condition. 

In producing new varieties by crossing, forms may be obtained in the F3 
generation which appear homozygous for botanical and agronomic char- 

• Published with the approval of the Director as Paper No. 259 of the Journal Series of the Minnesota 
Agricultural Experiment Station. 

' The writer wishes to express his appreciation to H. K. Hayes, Head of the Section of Plant Breeding, 
Division of Agronomy and Farm Management, for suggestions and criticisms during the progress of this 
study. 

• Reference is made by number (italic) to "Literature cited," p. 62-63. 

Journal of Agricultural Research, Vol. XXIL No. 2 

Washington, D. C. Oct. 8, 1921 

ir Key No. Minn.-« 

(53) 



54 Journal of Agricultural Research voi.xxn.No. a 

acters but which may be heterozygous for growth factors. There is the 
possibility that this heterozygous condition may cause the Fg or F. 
hybrid to give a high yield. After several further generations this hete- 
rozygous condition may be lost, with a consequent loss in growth stimu- 
lus. A knowledge of the amount of added vigor in the F, generation is 
of value in determining whether heterozygosis in Fg and F^ lines would 
modify their yields sufficiently to interfere seriously with a determina- 
tion of their value as improved varieties. 

With these points in view a study has been made in wheat of the imme- 
diate effect of cross-pollination on seed weight and the increased vigor 
of Fi crosses. Pure lines were used of seven varieties of Triticum vulgare 
Vill. and one variety of each T. compactum Host. (Little Club) , T. dicoccum 
Schr. (Spring Emmer) , and T. durum Desf. (Mindum). Varieties of T. 
vulgare were crossed with each other and with Little Club, Spring Emmer, 
and Mindum. Little Club was crossed also with Spring Emmer and 
Mindum. 

IMMEDIATE EFFECT OF CROSS-POLLINATION 

Because of the phenomenon of double fertilization it is possible in 
some cases to obtain an increase in weight of seed as an immediate eflfect 
of cross-pollination. The increase is due principally to an increase in 
weight of endosperm in such crops as com, where the proportion of endo- 
sperm to embryo is large. 

Collins (5) observed open-pollinated ears of Chinese maize in which the 
size of seed was increased by cross-pollination. Seeds which showed by 
their color the effect of foreign pollen averaged 0.178 gm., while white 
seeds from the same portion of the ear averaged 0.153 g^. Roberts (14) 
mentions a similar instance with Chinese maize. Collins and Kempton (6) 
compared the average seed weight of com from intravarietal and inter- 
varietal pollinations. The intervarietal crosses exceeded the intra- 
varietal in seed weight by 8.8 per cent. In a similar experiment, Wolfe 
{16) found that 23 of 31 com varietal crosses yielded more grain than in- 
travarietal pollinations. Carrier (4) obtained an increase in yield of 
grain in strains of com when grown in a mixture as compared with any 
one of the strains grown alone. 

That an increase is also obtained in the size of the embryo is clearly 
shown by Lewis and Vincent {12) in a comparison of seeds of Newtown 
apple from self- and cross-pollinations. The crossed seeds showed a 
striking increase in weight over that of the selfed seeds. As there is 
little or no endosperm in apple seeds, an increase in seed weight is due 
largely to an increased size of the cotyledons. 

Since artij5cially pollinated seeds of wheat are usually smaller than nor- 
mally pollinated seeds, spikes of each variety were emasculated in the 
same manner as for cross-pollination and then pollinated with pollen 
from plants of the same pure line. Seed from this intrapollination is 



Oct. 8, 1921 Comparative Vigor of F^ Wheat Crosses and Their Parents 55 



termed "incrossed seed" and is used as a basis of comparison in deter- 
mining the immediate effect of cross-pollination. The average weight of a 
normally pollinated seed for all varieties used was 26.65 ±0.22 mgm./ 
and the average for an incrossed seeds was 18.13 ±0.24 nigm. 

A comparison is shown in Table I of the hybrid seed and the incrossed 
seed where the average dates of pollination are the same or approxi- 
mately so. 



Table I.- 



-Weight of seed of the immediate crosses compared with weight of seed of the 
incrossed parents 





Seed parent. 


Cross. 


Difference 


Name of cross. 


Num- 
ber of 
seeds. 


Average weight 
of seeds. 


Num- 
ber of 
seeds. 


Average weight 
of seeds. 


between cross 

and female 

parent. 


Marquis X Velvet ChafFa 

Marquis X Penny 


38 
38 
49 
39 
44 
104 

44 


Mgm. 

12. 6±o. 5 
12. 6± . 5 
17. 2± .8 
10. i± .5 
26. 4± .8 
19. 9± . 6 
26. 4± .8 


48 
24 
26 

50 
24 

23 

15 


Mgm. 

15. 6±o.5 
20. 2 ±1.0 
23- 5± -7 
9-4± .3 
27- I ±1.3 
15. 9± .6 
25. o±i. 2 


Mgm. 
+3. o±o. 7 
+ 7-6±l.2 
+6.3±i.o 
-0. 7± .6 


Haynes Bluestem X Marquis . . 
Little Club X Marquis 


Emmer X Velvet ChafE 

Velvet ChafE X Mindum 

Emmer X Little Club 


+0. 7±i. 5 
-4. o± .8 
— I. 4±i. 4 







a In the discussion of crosses the seed parent is given first. 

The varietal crosses in every case showed an increased seed weight as 
compared with the female parent. The largest increase in seed weight 
was 7.6 ±1.2 mgm., which was obtained from the cross Marquis X Penny. 
This hybrid gave on the average over 50 per cent heavier seeds than 
incrossed Marquis. Of the species crosses none gave a significant increase 
in seed weight. Velvet Chaff crossed with Mindum produced seeds which 
on the average were 4.0^0.8 mgm. lighter than the seeds of incrossed 
Velvet Chaff. 

F, GENERATION CROSSES COMPARED WITH THEIR PARENTS 

Some of the earliest hybridization work affords good examples of the 
vigor of Fi crosses. For an excellent review of this subject the reader is 
referred to the publication of East and Jones {8) . 

In the present experiment the Fi generations and their parents were 
grown in the greenhouse under controlled conditions. Care was taken 
to plant seeds at a uniform depth, and when the seedlings were about 4 
inches tall they were transplanted to 7-inch pots, two seedlings to a pot 
and only like seedlings together. Unfortunately an epidemic of stem-rust 

1 The probable error of an average of averages was calculated according to the formula: 

in which n is the number of individuals in a generation, e the probable error, and A" the total niunbcr of 
individuals (is). 



56 



Journal of Agricultural Research 



Vol. XXn.No. a. 



started about heading time, and some plants were rusted badly 
Measurements of height were taken on those plants which were not 
attacked previous to heading, and yield data were taken only on plants 
uninjured by rust. 

Both incrossed and normally pollinated seeds of the parental varieties 
were planted. Amy and Garber (2) have shown that in some cases there 
is a positive correlation between weight of seed planted and the vigor of 
resultant plants. In order to determine whether the size of seed planted 
was of importance in an analysis of individual plant yields in the present 
experiment, correlation coefficients were calculated for the weight of seed 
planted as subject and length of culm and yield of grain per plant as 
relative. (Table II.) 

Table II.— Correlation coefficients for weight of seed planted and the vigor of resultant 

plants 



Variety. 



Marquis 

Velvet Chaff. ... 

Barletta 

Penny 

H. B. S. 1-16-12 

Bobs 

Little Club 

Emmer 

Mindum 



CoefBcient of correlation with weight of seed planted as subject. 



Length of tallest 
culm (relative). 



— O. 084 ±0. 100 

+ .o69± .065 

- . I44± • 085 

- . i97± .075 

— . 242± . 071 

— . ii6± . 079 

- . ii8± .082 
+ • 024± . 086 

— . 121 ± . 076 



Total culm length 
(relative). 



+0. i36±o. 098 

+ • 078± . 065 

.205± .083 

— . I29± .077 
+ . I2i± . 071 
+ .i69± .078 
+ . i85± .080 

— .oo7± .085 
+ .38i± .066 



Yield of grain in 

grams per plant 

(relative). 



+0. 047±0. 173 
+ .o87± . 108 



+ . OIO± . 112 

- .I94± .095 
+ . 205± . 084 
+ .032± . 099 

— . 425 ± . 080 
+ .i36± .095 



The only significant correlation was obtained with the Mindum variety. 
A correlation coefficient of +0.381 ±0.066 was obtained for weight of 
seed planted and total culm length. In the light of these facts it was 
considered legitimate to use the plants from normal and incrossed seed 
as a single parent population. 

The Fi crosses and their parents were compared for height of tallest 
culm and for total culm length. (Table III.) 

Six of the 1 1 Fi varietal crosses showed an increase in length of tallest 
culm as compared with the parental average, and 5 showed a decrease. 
The two Fi crosses between Mindum and vulgare varieties were con- 
siderably taller than either parent. Similar results were obtained from 
crosses between Emmer and the same vulgare varieties. On the othef 
hand, the Fi crosses of Little Club with either Emmer or Mindum did 
not show a significant difference in average height of tallest culm when 
compared with the average of the parents. 

In Table IV the crosses and their parents are compared for total culm 
length. 



Oct 8, 192 1 Comparative Vigor of F^ Wheat Crosses and Their Parents 57 



TablB III. — Height of tallest cuUn of Fi wheat crosses compared with parental average 



Name of one parent. 



Num- 
ber of 
individ- 
uals. 



Height. 



Name of other 
parent. 



Num- 
ber of 
individ- 
uals. 



Height. 



Aver- 
age 
height 
of par- 
ents. 



Fi cross. 



Niun- 
berof 
individ- 
uals. 



Height 



Marquis 

Velvet Chaff 

Penny 

Haynes Bluestem. . 

Marquis 

Velvet Chaff 

Average. . . . 

Emmer 

Mindum 

Emmer 

Average. . . . 



Inches. 

5 
5 
5 
5 
7 
7 
7 
4 



73 



45 



62 

77 
77 
77 
62 
62 



70 



50.8 



Velvet chaff. .. . 

Barletta 

Penny 

Bobs 

Barletta 

Penny 

JBobs 

Marquis 

JLittleClub 

iLittleClub 

Marquis 

jVelvet Chaff . . . 

Marquis 



Inches. 



105 
60 

74 
70 
60 

74 
70 
70 
46 
66 
66 



69 



43 



66 
66 
46 
105 
105 
46 



Inches. 
45- 6 
47. 6 
41.9 

43-3 
46.7 
41. o 
42.4 

38.7 
48.6 
47.8 
46.9 



64 
38 
45 
65 
38 

49 

108 

60 

24 
62 

52 



44.6 



55 



Inches. 

42.3 
50.8 

41. I 
47-9 
50-9 

42. 6 
40. 7 

50.8 
50-5 
43-3 

45- o 



47.2 



49. 2 



72 



46.8 



14 
17 
28 
18 



51.0 
48.5 
53-2 
54-5 
55-5 
55-6 



15 



53-1 



TablU IV. — Total culm length of Fj wheat crosses compared with parental average 



Name of one parent. 



Num- 
ber of 
individ- 
uals. 



Height. 



Name of other 
parent. 



Num- 
ber of 
individ- 
uals. 



Height. 



Aver- 
age 
height 
of pat- 
ents. 



Fi cross. 



Num- 
ber of 
individ 
uals. 



Height 



Marquis 

Velvet Chaff .... 

Penny 

Hajmes Bluestem 

Average . . . 

Emmer 

Mindum 

Marquis 

Velvet Chaff 

Average. . . 



46 

46 

46 

46 

46 

loS 

105 

105 

L 105 

74 

79 



Inches. 
195 
19s 
195 
195 
19s 
149 
149 
149 
149 
90 
215 



Velvet Chaff . 

Barletta 

Penny 

Bobs 

Little Club. . 

Barletta 

Penny 

Bobs 

Little Club. .. 

Bobs 

Marquis 



73 



171 



62 

77 
77 
77 
46 

105 

74 



204 
131 
131 
131 
195 
149 

157 



^Little Club. . 

Marquis 

Velvet Chaff . 

►Emmer 



105 
60 

74 
70 
66 
60 

74 
70 
66 
70 
46 



Inches. 
149 

157 

90 

136 

149 

157 
90 
136 
149 
136 
195 



Inches. 
172 
176 

143 
166 
172 

153 
120 

143 
149 

113 
205 



64 
38 
45 
65 
62 

38 

49 
108 

52 
60 

24 



Inches. 
146 
168 
132 
211 
167 
174 
133 
153 
153 
III 
214 



69 



140 



156 



66 
66 
46 

105 
62 

62 

68 



149 
149 

19s 
149 

204 
204 

17s 



177 
140 
163 
140 
200 
177 

166 



55 



160 



14 
17 
i8 
28 

15 



169 
189 

151 
123 
210 
149 

165 



58 



Journal of Agricultural Research 



Vol. xxn. No. a 



For total culm length, 6 of the ii varietal crosses showed an increase 
over the parental average and 5 showed a decrease. The averages for 
culm length of the F^ crosses and of their parents are practically identical 
when the results of all crosses are considered together. This makes it 
doubtful whether the increases of the F^ crosses over the parental averages 
are the results of the vigor due to crossing or are due to some other 
experimental factor. 

Table V. — Average yield of grain per plant of Fi wheat crosses and their parents 





Niun- 
berof 
individ- 
uals. 


Yield. 


Name of other 
parent. 


Num- 
ber of 
individ- 
uals. 


Yield. 


Aver- 
age 
yield 
of pa- 
rents. 


Fj cross. 


Per 

centage 
of in- 


Name of one parent. 


Num- 
ber of 
individ- 
uals. 


Yield. 

Gm. 
2.7 
3-3 

2-5 

2.9 
2.8 

2-5 

2.3 
2.5 


crease 
with 
paren- 
tal av- 
erage 
as basis. 


Marquis 


1 ^5 
\ '5 

; 38 

I 38 

36 

47 

IS 
38 


Gm. 
1.9 
1.9 

1-5 
1-5 
2.4 
2.4 

1.9 
1-5 


Penny. . . . 

Bobs 

Penny. . . . 

JBobs 

Marquis. .. . 

JLittle Club. 


36 
59 

I 59 
15 

f 46 
I 46 


Gm. 
2.4 
30 
2.4 
30 

30 
1.9 

2. 2 
2. 2 


Gm. 
2. 2 

2-5 
2. 

2-3 
2.7 
2. 2 

2. I 
1.9 


18 

65 
28 

92 
23 
18 

45 
37 


33 


Velvet Chaff .... 
Penny 


32 
25 
26 

4 
14 

10 


Ha)mes B 1 u e- 

stem. 
Marquis 


Velvet Chaff. ... 


32 


Average . . 


30 


1.9 




45 


2-5 


2. 2 


41 


2.7 


23 


Little Club 

Marquis 


/ 46 
I 46 

1 ^^ 

I 38 
IS 


2. 2 
2. 2 
1.9 
1-5 
1-5 
1.9 


Emmer 

[Mindum 

[Emmer 


48 
f 49 

49 
I 49 
/ 48 
I 48 


1. I 

2. I 
2. I 
2. I 
I. I 
I. I 


1-7 
2. 2 
2 
1.8 
1-3 

1-5 


9 

I 

13 
8 

23 
18 


•3 
I. 

•3 
I. I 

•5 
.6 




Velvet Chaff 

Marquis 








Average . . 


33 


1.9 




49 


1.6 


1.8 


12 


.6 





For average yield of grain per plant, six of the eight variety crosses 
yielded more than either parent, and all variety crosses jdelded more than 
the parental average. Marquis X Bobs and Velvet Chaff X Little Club 
exceeded the parental average 32 per cent in yield of grain per plant. 

With the exception of crosses between common wheat and Little Club 
the average yield of grain per plant of the species crosses was less than 
that of the lower-yielding parent. This is due to the fact that the F^ 
plants had a high percentage of barren florets. 

STERILITY IN SPECIFIC CROSSES 

The occurrence of sterility in wheat specific crosses has been reported 
by several workers. Tschermak {13), after several years of hybridiza- 
tion work, found that hybrids of Triticum dicoccum and T. compactum or 
vulgare varieties were only partially fertile. Hybrids of T. durum with 
T. compactum or T. vulgare varieties were classed as fully fertile. Ster- 



Oct 8. tgai Comparative Vigor of Fj Wheat Crosses and Their Parents 59 



ility is mentioned by Kezer and Boyack (ii) as occurring in the F^ gen- 
eration of the cross Fultz Mediterranean by Black Winter Emmer. In 
crosses between Algerian Macaroni and Algerian bread wheats, Free- 
man (9) reports that the F^ generation developed normally but in the 
F2 generation all degrees of sterility appeared from complete sterility to 
complete fertility. 

Hayes, Parker, and Kurtzweil (10) crossed varieties of Triticum vul- 
gare with varieties of T. durum and T. dicoccum. The parental varieties 
showed an average of 4 per cent of barren florets. The F^ crosses of 
varieties of durum with varieties of vulgare and the reciprocals showed 
a barrenness of 47 per cent. The Fi crosses of T. dicoccum crossed with 
varieties of vulgare showed 26 per cent barrenness and the reciprocal 29 
per cent. The results are not in agreement with the conclusions of 
Tschermak (13). 

In the present experiment a count was made of the total number of 
outer florets per plant and the number of these which were barren. 
From these data the percentage of barren florets was computed. (Table 

Table VI. — Barrenness of outer florets in wheat varieties and Fi crosses 



Variety or cross. 



Number Percentage 

plants of barren 

consid- outer 

ered. florets. 



Marquis 

Velvet ChaJBf 

Penny 

Haynes Bluestem 

Bobs 

Little Club 

Emmer 

Mindum 

Average 

MarquisXPenny 

MarquisXBobs 

Velvet Chaff X Penny 

Velvet Chaff X Bobs 

Haynes Bluestem X Marquis 

PennyXBobs 

Marquis X Little Club 

Velvet Chaff X Little Club. . 

Average 

MarquisX Emmer 

Velvet Chaff X Emmer 

Little Club X Emmer 

Average 

MarquisX Mindum 

Velvet Chaff X Mindum .... 
Little Club X Mindum 

Average 



15 
38 
34 
47 
59 
47 
49 
44 



18 
21 
IS 

18 
17 
25 
18 

19 



42 



19 



18 

65 

27 

93 
18 

22 
57 
37 



17 
14 
15 
13 
17 
15 
17 
14 



42 


15 


18 
23 

10 


73 
67 
86 


17 


75 


13 

8 
2 


88 
67 

54 



70 



6o Journal of Agricultural Research vd. xxii. No. a 

The parental varieties showed an average of 19 per cent barren florets. 
Intercrosses of vulgare varieties and .crosses between Little Club and 
vulgare varieties showed an average of 15 per cent barren florets. The 
Fi crosses of Marquis, Velvet Chaff, and Little Club with Emmer gave 
an average of 75 per cent barrenness. These same common varieties 
and Little Club crossed with Mindum showed a barrenness of 70 per 
cent. These data confirm the results of Hayes, Parker, and Kurtzweil (10) 
and show conclusively that in some cases Fj crosses between varieties 
of T. vtdgare and T. durtim or T. dicoccum are highly self-sterile. 

DISCUSSION OF RESULTS 

It has been pointed out by East and Jones (7) that the increase in 
productivity of a cross is due to an increase in the number of growth 
factors of which the maximum number can be obtained only in a hetero- 
zygous condition. In a crop such as com, this heterozygous condition 
is kept up by cross-fertilization. Selfing corn varieties reduces the hetero- 
zygosity and consequently the vigor. In wheat the continued selfing 
natural to the crop has brought about a condition of homozygosity. 

In the present experiment all varietal crosses gave an increase in seed 
weight as an immediate effect of cross-pollination. An increase is also 
shown in the F^ crosses for average yield of grain per plant as compared 
with the parental average. The increase ranged from 4 per cent in 
Penny X Bobs to 32 per cent in Marquis X Bobs. Before attempting to 
utilize the vigor of the heterozygous condition by growing Fg and F4 
generation crosses as the commercial crop, it seems logical to combine 
in one variety the maximum number of growth factors possible. When 
the possibilities of combination have been exhausted and a variety, or 
a series of varieties, has been secured which contains this maximum 
number of growth factors, it may be desirable to follow out the sugges- 
tion of Anderson (j). This method probably could not be used to 
advantage except under intensive farming conditions. In case one 
desired to use such a method it is logical to assume that the more desirable 
crosses to make are those which show the greatest increase in yield of 
grain in the F^ generation. 

There is an indication that the increased productivity of the heterozy- 
gous condition is a factor which must be considered in comparing F3 and 
F4 lines for yielding ability. In the cross Marquis X Bobs the Fj genera- 
tion showed on an average a 32 per cent increased yield of grain per plant 
as compared with the average of the parents and a 10 per cent increase as 
compared with the higher-yielding parent. In the Fj generation of such 
a cross it is highly probable that some of the most vigorous plants will be 
those with the greatest degree of heterozygosity. These heterozygous 
individuals will produce F3 progeny the vigor of which likewise will be 
partially due to the heterozygous condition. If the F3 lines are classified 
on a basis of their yielding ability, some of these heterozygous lines will 



Oct. 8,1921 Comparative Vigor of F^ Wheat Crosses and Their Parents 6i 



be included as the best 3nelders. In subsequent generations as the lines 
become homozygous their productivity may decrease. In generations 
beyond the F5 the heterozygous condition of the population rapidly dis- 
appears. 

A method of breeding which, according to Babcock and Clausen (j), 
has been used by the Svalof Station, seems worthy of wider application. 
As self-fertilized crops approach homozygosis rapidly in generations follow- 
ing a cross, it is suggested that a cross be made between varieties selected 
because of the desirable characters which they possess. After 6 to 10 
years have elapsed, during which time progeny of the cross has been grown 
in bulk plots, selection of individual plants may be made with the assur- 
ance that a high percentage of these plants will give homozygous progeny. 
While this system requires some length of time before results are ob- 
tained, it requires a minimum of labor. 

When making crosses with the hope of increasing yield through a re- 
combination of the desirable factors of both parents, the parents will 
naturally be selected on the basis of their yielding ability. The chances 
of favorable recombinations of yield factors in generations following a 
cross will presumably be greater when dealing with a cross which shows 
maximum increased yield over the parents in the Fj generation. Genetic 
linkage, however, may make certain combinations difficult or impossible. 

The sterility of the specific crosses, with the exception of crosses of 
varieties of Triticum vulgare with Little Club, is partially or wholly 
responsible for the low grain yield of the crosses as compared with the 
parental averages. The fact that Little Club behaves in every way as a 
variety of T. vulgare agrees with the view of Tschermak (75), who believes 
that T. compactum and T. vulgare are closely related. Little Club 
crosses readily with varieties of vulgare. When Little Club or varieties 
of vulgare are crossed with Emmer or Mindum, the same high degree of 
sterility is shown. 

SUMMARY OF RESULTS 

(i) An increase in seed weight was obtained in all varietal crosses as 
an immediate effect of cross-pollination. The only significant differ- 
ence shown by the immediate hybrids of specific crosses was a decrease 
in seed weight obtained in Velvet Chaff X Mindum. 

(2) In the Fj generation some of the hybrids exceeded the parental 
average in height of tallest culm, and in total culm length others showed 
a decrease. In all varietal crosses the F^ hybrid exceeded the parental 
average in yield of grain per plant, and six out of eight crosses exceeded 
the yield of the better parent. 

(3) Crosses between Little Club and varieties of Triticum vulgare gave 
results similar to those of crosses between vulgare varieties. 

(4) The F^ generation of Emmer or Mindum crossed with varieties of 
Triticum vulgare or with Little Club showed a high degree of sterility. 



62 Journal of Agricultural Research voi. xxn.No. » 

The average percentage of barren florets of tlie parental varieties was 19. 
The average percentage of barren florets of the Fj varietal crosses, includ- 
ing crosses of Little Club with vulgare varieties, was 15. The vulgare- 
Emmer and Little Club-Emmer crosses produced 75 per cent barren 
florets, while an average of 70 per cent of barren florets was obtained 
from the durum-vulgare and durum-Little Club crosses. 

LITERATURE CITED 
(i) Anderson, T. 

I919. THE IMI'ROVEMENT OF AGRICULTURAL CROPS BY SELECTION AND HYBRID- 
IZATION. (Abstract.) In Scot. Joiir. Agr., v. 2, no. i, p. 10-20. 

(2) Arny, a. C, and Garber, R. J. 

I918. VARIATION AND CORRELATION IN WHEAT, WITH SPECIAL REFERENCE TO 

WEIGHT OF SEED PLANTED. In Jour. Agr. Research, v. 14, no. 9, 
p. 359-392, 8 fig. Literature cited, p. 391-392. 

(3) Babcock, Ernest Brown, and Clausen, Roy Elwood. 

1918. GENETICS in relation TO AGRICULTURE. XX, 675 p., 239 fig., 4 Col. 

pi. New York, London. List of literature cited, p. 622-647. 

(4) Carrier, Lyman. 

I913. THE immediate effect ON YIELD OF CROSSING STRAINS OF CORN. Va. 

Agr. Exp. Sta. Bui. 202, 11 p., 2 fig. 

(5) Collins, G. N. 

1909. A NEW TYPE OP INDIAN CORN PROM CHINA. U. S. Dept. Agr. BuT. Plant 
Indus. Bui. 161, 30 p., 2 pi. 

(6) and Kempton, J. H. 

I913. EFFECT OF CROSS-POLLINATION ON THE SIZE OF SEED IN MAIZE. In U. S. 

Dept. Agr. Bur. Plant Indus. Circ. 124, p. 9-15. 

(7) East, E. M., and Hayes, H. K. 

I912. HETEROZYGOSIS IN EVOLUTION AND IN PLANT BREEDING. U. S. Dept. 

Agr. Bur. Plant Indus. Bui. 243, 58 p., 8 pi. Literature cited, p. 

49-51- 

(8) and Jones, Donald F. 

1919. INBREEDING AND OUTBREEDING, THEIR GENETIC AND SOCIOLOGICAL 

SIGNIFICANCE. 285 p., 46 fig. (in text and on 13 pi.). Philadelphia, 
London. Literature cited, p. 266-277. 

(9) Freeman, George F. 

1919. THE HEREDITY OF QUANTITATIVE CHARACTERS IN WHEAT. In GcneticS, 

V. 4, no. I, p. 1-93, 85 tab. (i fold.) Literature cited, p. 93. 

(10) Hayes, H. K., Parker, John H., and Kurtzweil, Carl. 

1920. GENETICS OP RUST RESISTANCE IN CROSSES OP VARIETIES OP TRITICUM 

VULGARE WITH VARIETIES OF T. DURUM AND T. DICOCCUM. In Jotll. 

Agr. Research, v. 19, no. 11, p. 523-542, pi. 97-102. Literature cited, 
P- 541-542. 

(11) KezER, Alvin, and Boyack, Breeze. 

1918. MENDELIAN INHERITANCE IN WHEAT AND BARLEY CROSSES WITH PROB- 

ABLE ERROR STUDIES ON CLASS PREQUENafiS. Colo. Agr. Exp. Sta. 
Bul. 249, 139 p., 95 tab., 10 fig., 9 col. pi. 

(12) Lewis, C. I., and Vincent, C. C. 

1919. pollination of the APPLE. Oreg. Agr. Exp. Sta. Bul. 104, 40 p., 

14 pi. (in text). 



Oct. 8, 192 1 Comparative Vigor of F^ Wheat Crosses and Their Parents 63 

(13) Roberts, Elmer. 

1918. FLUCTUATIONS IN A RBCESSIVB MENDELIAN CHARACTER AND SELECTION. 
In Jour. Exp. Zool., v. 27, no. 2, p. 157-192, 3 fig., 2 pi. (in text). 
Literattire cited, p. 176-177. 

(14) Roberts, H. F. 

i912. first generation hybrids of american x chinese corn. in ann. 
Rpt. Amer. Breeders' Assoc, v. 8, p. 367-384, 5 fig. Literature cited, 
p. 384. 

(15) TsCHERMAK, Erich von. 

1914. DIE VERWERTUNG DER BASTARDIERUNG FUR PHYLOGBNETISCHE 

FRAGEN IN DER GETREiDEGRUPPE. /n Ztschr. Pflanzenziiclit., Bd. 2, 
Heft 3, p. 291-312. 

(16) Wolfe, T. K. 

1915. FURTHER EVIDENCE OF THE IMMEDIATE EFFECT OF CROSSING VARIETIES 
OF CORN ON THE SIZE OF SEED PRODUCED. In Joiu". Amer. Soc. Agron., 
V. 7, no. 6, p. 265-272. Literature cited, pp. 271-272. 



TEMPERATURE AND HUMIDITY STUDIES OF SOME 
FUSARIA ROTS OF THE IRISH POTATO ' 

By R. W. Goss 
Assistant Plant Pathologist, Nebraska Agricultural Experiment Station 

INTRODUCTION 

The ability of Fusarium oxysporum Schlect. to cause a rot of the 
potato tuber has been clearly demonstrated by a number of workers. 
The influence of temperature on this disease has been reported in a 
number o;f papers, but the experimental evidence as a whole is rather 
meager, usually only extreme temperatures being vtsed. The effect of 
moisture on the progress of the disease, except under conditions of 
extreme dryness or saturation, has received practically no attention. 
It was with the purpose of determining the relation of temperature and 
humidity to the progress of potato tuber-rots caused by Fusaria that 
the following work was undertaken. 

HISTORICAL 

The association of Fusaria with storage-rots of the Irish potato (Sola- 
num tuberosum L.) has been a matter of common observation by most 
workers in plant pathology from 1842 to date. Several species of the 
form genus Fusarium Link have been described as causes of potato 
tuber-rots, by Von Martins {12), Reinke and Berthold {18), Schacht 
(jp), Pethybridge and Bowers (14)," Longman {10), and Sherbakoff {20). 
The fact that Fusarium species could produce a rot of the tuber was 
demonstrated by Pizzigoni {15) and Wehmer {24, 25), who described 
the species they worked with as Fusarium solani (Mart). Frank (6), 
De Bary (2), and others considered that the Fusaria were unable to 
produce a rot of the tuber. In most of the earlier papers, F. solani, or 
some species thought to be a synonym of it, was given as the causal 
organism. 

Owing to the absence of clearly defined species in all the literature 
previous to Appel and Wollenweber's (i) monograph on the form genus 
Fusarium in 191 2, no attempt will be made to review in detail the earlier 
reports of potato tuber-rots caused by Fusaria. 

Fusarium oxysporum was considered by Wollenweber (27) to be a 
strictly vascular parasite producing a wilt of the potato vine but not 

• Published with the approval of the Director of the Nebraska Agricultural Experiment Station. The 
paper is based upon experimental -work undertaken at the Michigan Agricultural College in 1914-15, and 
at the University of Wisconsin in 1916-17. 

' Reference is made by number (italic) to "Literature cited," p. 77-79. 

Journal of Agricultural Research, Vol. XXII, No. 2 

Washington, D. C. Oct. 8, 1921 

zb Key No. Nebr.-3 

(65) 



66 Journal of Agricultural Research voi. xxii. No. 2 

causing a rot of the tuber. Carpenter (4) in 19 15 was the first to report 
successful infections by inoculations with pure cultures of F. oxysporum. 
He made these by dipping wounded tubers in a water suspension of 
spores, wrapping in oiled paper and keeping them at controlled tem- 
peratures ranging between 17° and 30° C. No detailed experiments 
were reported except in this saturated atmosphere. He noted, however, 
that either a dryrot or a wetrot was produced, according to the tem- 
perature and humidity used. He concluded that a constant storage 
temperature below 50° F. (10° C.) would prevent the action of F. radici- 
cola Wollenw., F. eumartii Carp., and F. oxysporum. Previous to this 
work of Carpenter's, Smith and Swingle {22), in 1910, described a bundle 
blackening and a dry endrot of the tuber as two stages of the same 
disease. They attributed this to a Fusarium for which they accepted the 
name F. oxysporum as first applied to it by Schlechtendahl (2/, p. 139). 
They noted that the disease continued in stored potatoes and that when 
potatoes were stored in warm rooms, either moist or dry, they became 
badly diseased, whereas those stored in cool places kept much better. 
They did not differentiate this species of Fusarium from others occurring 
on the potato, and no inoculation experiments were recorded. Manns 
in 191 1 (//), working with the same disease, stated that the "dormant 
internal infection" under improper storage conditions becomes so active 
as to cause a high percentage of dryrot. He noted that the disease was 
favored by high temperature and considerable moisture. At 36° to 40° 
F. (2° to 3° C.) the disease made no progress, at 45° to 55° F. (7° to 12° 
C.) it developed gradually and caused considerable rot, especially when 
accompanied by high humidity. He made no mention of pure culture 
inoculations on tubers or morphological studies. 

Jamieson and Wollenweber in 191 2 {8) described a dryrot of the 
potato tuber caused by a species of Fusarium which they named Fusa- 
rium trichothecioides Wollenw. They made inoculation experiments and 
found the most rapid penetration of the tuber to take place at 10° to 12° C 
in an atmosphere of low humidity. Rotting took place at the high 
humidities but not as rapidly. Wilcox, Link, and Poole {26) pubUshed 
on a dryrot of the potato tuber caused by a Fusarium which they called 
F. tuberivorum W. and L. but which was undoubtedly the F. trichothe- 
cioides previously described by Jamieson and Wollenweber (<?), They 
found that a temperature of 8° to 10° C. was only slightly inhibitive to the 
growth of the fungus and that when potatoes infected with the organism 
were stored at this temperature, the most rapid decay took place when 
the humidity was high. Pratt (ly), working with the same disease, 
found that temperatures ranging from 12° to 25° C. were favorable for 
the progress of the disease and that dryrot did not develop at temperatures 
below 2° C. He concluded from storage experiments that in a dry, 
well-ventilated storage house losses would be very slight at temperatures 
from 2° to 4° C. 



Oct.8.i93i Temperature and Humidity Studies of Fusaria Rots 67 

Link (9), making comparative studies of Fusarium oxysporum and 
F. trichothecioides found that both were capable of producing a rot of 
the potato tuber and that F. trichothecioides produced a typical dry- 
rot. F. oxysporum produced a softrot of the whole tuber except under 
cold, dry conditions, when a drjn-ot was produced. He ran his experi- 
ments at controlled temperatures ranging from 1° to 30° C. in an 
almost saturated atmosphere. 

Pratt {16) found that Fusarium radicicola behaved much the same as 
F. oxysporum, and he concluded from storage experiments that the 
tuber-rot caused by this organism does not make any progress in storage 
at a temperature of 48° F. (8.8° C.) or below. 

In general, then, it can be said that a high temperature favors the 
production of tuber-rots by all three of these Fusaria, although Fusarium 
trichothecioides appears to be able to produce a rot at lower temperatures 
than the other two. High humidities also appear to favor the produc- 
tion of tuber-rot. With the exception of the paper of Jamieson and 
Wollenweber {8) all the evidence points toward an increase in rotting 
with an increase in humidity. 

TEMPERATURE RELATIONS IN PURE CULTURES 

A review of the literature shows a general conformity of results re- 
garding the relation of temperature to the growth of Fusarium oxy- 
sporum. Link (9) by making dry-weight determinations of growth in 
liquid media found 30° C. to be the optimum for growth. Edson and 
Shapovalov (5), working with Petri-dish cultures, obtained the same 
optimum. They reported a maximum temperature of 37 C, where the 
spores changed to chlamydospores ; they did not observe growth at 5° C. 
Humphrey (7) gives 4° C. as the minimum temperature for certain 
strains of F. oxysporum.. 

The writer, working with three strains of Fusarium oxysporum and 
using the same methods for measuring growth, obtained somewhat 
similar results to those reported by Edson and Shapovalov (5). The 
minimum temperature for growth was 9.5° C, no growth taking place at 
the next lower temperature of 7° C. The maximum temperature was 
37.5° C, where there was a very slight growth. 

Fusarium trichothecioides is apparently unable to grow at 30° C, 
which is the optimum temperature for F. oxysporum. Link (9) found the 
greatest growth of F. trichothecioides in liquid potato extract media at 
the end of 20 days to take place at 12° C, with no growth present at 30° 
C, although the organism was capable of living in the potato tuber at 
that temperature. Edson and Shapovalov (5) obtained a much higher 
optimum for F. trichothecioides; they found the greatest growth took 
place at 25*^ C, with a sharp drop to the maximum temperature at 30° C, 
where germination of spores took place but no growth of mycelium. 



68 Journal of Agricultural Research voi. xxii.no. 2 



The writer, working with two strains of Fusarium trichoihecioides 
in Petri dishes, found 25° C. to be the optimum temperature, and with 
one strain he was able to obtain sHght growth, 7 mm. in diameter, at the 
end of one week at 30° C. At 5° C. germination took place and there 
was slight growth. 

The optimum temperature for Fusarium radicicola was 30° C, the 
same as for F. oxyspornm. The minimum was at 5° C, where a very 
slight growth was produced in 10 days. At 35° C. the growth was greater 
than with F. oxysporum, although the rate of growth was slower. Edson 
and Shapovalov (5) report a similar optimimi temperature, with germi- 
nation but no growth at 5° C. They found that at 39° C. a transforma- 
tion from normal spores to chlamydospores took place. 

In general it can be said that at 25° C. the growth for all three species 
is nearly equal, Fusarium oxysporum, and F. radicicola increasing in 
growth up to 30° C. and F. iriclwihecioides decreasing. The minimum 
temperature for F. oxysporum is higher than for the other two, and in 
general F. trichoihecioides appears to be more tolerant of the lower tem- 
peratures than the others. 

Preliminary experiments, using liquid media and determining the 
growth by dry weights, have been conducted with a number of strains of 
these three species. While on certain media the results have in general 
corroborated the foregoing cardinal points for growth, they indicated that 
these cardinal points may vary with the medium used. For instance, 
with an nutrient solution made up of ammonium nitrate (NH^NOj), 
potassium phosphate (KHoPOJ, magnesium sulphate (MgSOJ, ferric 
chlorid (FeCls), and sucrose, the results compared well with those ob- 
tained on agar in Petri dishes. With a nutrient solution made up similarly 
to the potato extract medium used by Link (9), the total growth at the 
higher temperatures was considerably less than the growth obtained in 
the first nutrient solution, while at the lower temperatures the growth 
was much greater. The optimum temperature for growth of Fusarium 
trichoihecioides in the first nutrient solution was 25° C, with no growth tak- 
ing place at 5° C. With Link's potato-extract medium the optimum lay 
between 15° and 20° C, and there was weighable growth at 5° C. These 
results would possibly account for the considerable discrepancy between 
the results obtained by Link (9) with liquid media and those obtained by 
the writer and by Edson and Shapovalov (5) with agar cultures. 

EXPERIMENTAL INFECTION OF TUBERS 

The cultures used in the following experiments, with their origin, are 
listed below. In practically all cases the various strains of the same 
species behaved alike. Several other strains of Fusarium oxysporum, iso- 
lated by the writer, were also used in the experiments in addition to the 
ones listed below. 



Oct. 8,1921 Temperature and Humidity Studies of Fusaria Rots 69 

No. I. — Fusarium oxysporum, isolated by the author from browned 
vascular bundles of potatoes and identified by H. W. Wollenweber and 
numbered at Washington as 3377. '^-''^ r. 1 ;■;■. 

No. 8. Fusarium oxysporum, obtained from C. W. Carpenter of the 
United States Department of Agriculture, No. 3395. 

No. 32. — Fusarium oxysporum, obtained from G. K. K. Link, of the 
University of Nebraska, as No. 3345a. 

No. 28. — Fusarium. irichoihecioides obtained from G. K. K. Link. 

No. 31. — Fusarium irichoihecioides, obtained from A. C. Pratt of the 
United States Department of Agriculture. 

No. 29. — Fusarium radicicola, obtained from A. C. Pratt and numbered 
716. 

METHODS 

In all inoculation experiments with tubers, potatoes which were of one 
variety, of the same age, and had been kept under the same storage con- 
ditions were carefully selected for uniformity of size, type, and freedom 
from wounds. The stem ends always were cut and examined for natural 
infection, and all tubers showing vascular discoloration were discarded. 
The tubers were always treated with formaldehyde or mercuric chlorid 
and washed in sterile distilled water. 

The inoculations were made by wounding the epidermis, usually by 
stabbing to a depth of 3 mm. with a sterile scalpel. The inoculum was 
introduced in various ways as outlined in the experiments. 

Experiment i, December, 1915- — Potato tubers of the Up-to-Date 
variety were inoculated by wounding the tubers and then dipping them 
in a water suspension of spores, wrapping in sterile waxed paper, and 
placing in moist chambers at 25° C. Controls were treated in the same 
way, being dipped in sterile water. Results were taken 18 days later. 

Set No. I. Four tubers inoculated with Fusarium. oxysporum, No. i. 
All tubers completely rotted. The two control tubers remained sound. 

Set No. 2. Four tubers inoculated with Fusarium oxysporum, isolated 
from infected tubers in storage. All tubers showed a complete wetrot; the 
tissue was soft and of a light brown color; a large cavity was present in 
each tuber containing masses of white mycelium. At the point of inocu- 
lation there was a granular mass of hyphae and starch grains separated 
from the rest of the tissue. Control tubers remained healthy. 

Set No. 3. Four tubers inoculated with Fusarium oxysporum, isolated 
from wilted potato vines. All tubers showed a dark brown dryrot pro- 
gressing only a short distance from the point of inoculation. Controls 
remained healthy. 

Reisolations were made from all the rotted tubers, and Fusarium oxy- 

sporufn was recovered in every case. No bacteria or secondary invaders 

were found in any of the tubers. These results show that F. oxysporum 

is capable of producing a rot of the tuber in a saturated atmosphere at 

54818°— 21 2 



yo Journal of Agricultural Research voi. xxn.No a 

25° C. The characteristic rot under these conditions is a soft we trot with 
no sharp line of demarkation between the healthy and diseased tissue. 
The organism appears to be unable to attack whole starch grains, which 
accumulate in a granular mass with the myceliimi, as in set 2. The 
tuber-rot under these abnormal conditions is not typical of the rots 
usually found in storage. 

Experiment 2, February 23, 1916. — Further tests were conducted at 
the same temperature but with a lower relative humidity to test the abil- 
ity of the organism to cause a rot under conditions not so adverse for the 
host as in the previous experiment. Tubers of the Up-to-Date variety 
were inoculated by wounding and then placing a little of the fungus 
mycelium and spores in the wound. The tubers were then placed in a 
sterile moist chamber but were not wrapped in paper. Controls were 
treated and wounded in the same way. The experiment was run at 25° 
C. Twelve different strains of Fusarium oxysporum were used for the 
inoculations, two tubers being used for each strain. Results were taken 
after five weeks. 

In only one case had the rot extended three-fourths of the length of 
the tuber. In all the other tubers there was only a slight rotting extend- 
ing for a short distance from the point of inoculation. The controls 
remained sound in every case. The tubers were in a saturated atmos- 
phere at the beginning of the experiment, gradually becoming drier until 
at the end the tubers were considerably dried out. Compared to the 
preceding test the amount of rotting was very slight, and its inhibition 
may be directly attributed to the dryness of the air. The slight amount 
of rot around the point of inoculation would indicate that the fungus 
progressed a short distance into the tuber at the beginning of the experi- 
ment when the humidity was high but was unable to advance further 
under the drier conditions. This would indicate that the rotting of 
tubers already started could be checked by submitting the tubers to 
lower humidities. 

Experiment 3, April, 26, 1916. — A further test on the relation of 
humidity of the atmosphere to the rot of the tuber was started. The 
inoculations were made as in the previous experiment, and the same 
variety of potatoes was used. 

Set No. I. The inoculated tubers were placed in sterile chambers, and 
moist filter paper was placed in the chambers at the start of the experi- 
ment to produce a favorable humidity for the initial penetration of the 
tuber. 

Set No. 2. The tubers were placed in moist chambers in which the 
atmosphere was kept saturated throughout the experiment. 

Both sets were kept at a temperature of 25° C. The results were taken 
after seven weeks. (Table I.) 



Oct. 8,1921 Temperature and Humidity Studies of Fusaria Rots 71 

Table I. — Comparative amount of rot produced by Fusarium spp. under different con- 
ditions of relative hum,idity 



Strain. 


Set No. I. Set No. i. 

i 




One-third rotted 

2-mm. rot 

One-third rotted 

Healthy 


Entirely rotted. 
Do. 






Half rotted. 


Control 


Healthy. 







In all cases where rotting was present the starch grains were not 
corroded. Culture No. 8 seemed to have a much slower initial growth 
than the others, thus showing a greater difference between the two sets. 
In general, it can be clearly seen that the rotting was much greater in 
set No. 2, where the atmosphere was saturated throughout the experi- 
ment. Although the organisms were capable of starting a rot under the 
moist conditions at the start of the experiment in set. No. i, they were 
later considerably checked imder the drier conditions. 

Experiment 4, March 15, 19 17. — Further infection experiments were 
started under conditions in which the relative humidity of the atmosphere 
was controlled by the use of various concentrations of sulphuric acid. 
Previous experiments conducted at the Michigan Agricultural Experi- 
ment Station in 19 15 and described under experiment 5, in which the 
relative humidities were carefully controlled, produced very good results 
with Fusarium oxysporum. 

The apparatus used in experiment 5 was not available in 191 7, so the 
relative humidities used in experiments 4 and 4A were determined from the 
tables given by Stevens (aj). One-quart Mason jars were used, in which 
were hung small wire baskets containing the tubers, the acid being placed 
in the bottom of the jar. Tubers of the Rural New Yorker variety were 
inoculated as in the preceding experiments. They were then placed in 
the baskets in the sterilized jars and were sealed with paraffin and placed 
at the desired temperatures. The experiment was run in duplicate. 
Three strains of Fusarium oxysporum, two of F. trichothecioides , and one 
of F. radicicola were used for the inoculations. The temperatures used 
were 5°, 9°, 16°, and 25° C. While these temperatures varied somewhat 
during the experiment, the extremes did not in any case overlap. The 
relative humidities obtained by using sulphm-ic acid remained fairly 
constant throughout the experiment. One hundred cc. of each of the 
acid solutions were used for each jar. At the close of the experiment the 
specific gravity of the solutions was taken, and the calculated humidity 
at this time was compared with that at the start, with the result that the 
one having 1.5 per cent relative humidity had changed to 3.6 per cent, 
the 33 per cent to 49 per cent, and the 66.5 per cent to 74 per cent. These 
variations were not considered great enough to cause conflicting results. 



72 



Journal of Agricultural Research 



Vol. XXII, No. 2 



The results were taken after seven weeks, and the penetration of the 
tubers was measured in millimeters, as shown in Table 11. The number 
of individuals was so small that slight discrepancies in the tabulated 
results are found. Fusarium trichothecioides produced a slight rot at 
lower temperatures than F. oxysporum but did not produce as extensive 
a rot at the higher temperatures. At 5° C. the only rotting found was 
with one strain of F. trichothecioides, at 100 per cent humidity. No 
rotting was found at the temperature of 9° at the lower humidities, but 
there was slight rotting at this temperature at the higher humidities, 
especially with F. trichothecioides. It is noticeable that at 9° with the 
relative humidities of 66 and 100 per cent, the amount of rotting is greater 
than at the increased temperature of 16°, with the relative humidities of 
I and 33 per cent. The same comparative results are found between 
the amount of rotting taking place under the several humidities at a 
temperature of 16° and of 25°. The results do not conform with the 
report of Jamieson and Wollenweber {8), that penetration of the tuber 
by F. trichothecioides is favored by low humidities. The work by Link (9) 
and Wilcox, Link, and Poole {26), however, would indicate that more 
rapid rotting takes place in an atmosphere of high humidity, thus agreeing 
with the results shown in this experiment. The results of F. oxysporum 
accord well with those obtained in experiment 3. 

Table II. — Extent of penetration of tubers in experiment 4 



Tem- 
pera- 


Approxi- 
mate 
relative 
hiunidity. 


Fusarium oxysporum. 


Fusarium 
trichotliecioides. 


Fusarium 
radicicola. 


ture. 


Strain i. 


Strain 8. 


Strain 32. 


Strain 28. 


Strain 31. 


Strain 39. 


°C. 


Per cent. 
















I 





o 














5...... 


33 





o 














66 





o 
















100 





o 








3 mm. 







I 





o 
















0.3 





o 














9 


66 





I imn. 








I mm. 







100 





I mm. 





I mm. 


3 mm. 


I mm. 




I 





o 














16 


33 


I mm. 


o 














66 


2 mm. 


3 mm. 





I mm. 


I mm. 







100 


10 mm. 


lo mm. 





I mm. 


T mm. 


2 imn. 




I 


5 mm. 


lo mm. 


I mm. 








I mm. 




33 


10 mm. 


15 mm. 


3 mm. 





2 mm. 


6 imn. 


25 


66 


20 mm. 


Krot. 


Krot. 


I mm. 


25 mm. 


6 mm. 




100 


Krot. 


Complete rot. 


Krot. 


I mm. 


Krot. 


Complete rot. 



Experiment 4 A, May 16, 191 7. — In order to check up the possible 
error due to differences in the age of the tubers used in the various tests, 
the following experiment was started. New tubers of the Bliss Triumph 
variety were used in comparison with tubers of the same variety that had 



Oct 8, 1921 Temperature and Humidity Studies of Fusaria Rots 73 



been kept in cold storage from the previous year. The experiment was 
conducted in the same way as experiment 4, and the same cultures of 
Fusarium oxysporum, F. trichothecioides , and F. radicicola were used for 
inoculations. Only two temperatures were used, 13.5° and 25° C, as 
well as two humidities, 33 and 100 per cent at each temperature. The 
results shown in Table III were taken after six weeks. 

As in experiment 4 the rotting was much greater at the high tempera- 
tures and the high humidities. At the lower temperature of 1 3.5 ■^ C. there 
was no distinct difference between the amount of rotting in the old and 
new tubers, due to the very slight penetration at this temperature. In 
the old tubers at 25° the infection in every case had been rapid and the 
rotting had progressed much further than in the new tubers. These 
results support the statement of Bisby (j) that old tubers are more sus- 
ceptible to rot than new tubers. 

Table III. — Extent of penetration of old and new tubers 





Ap- 




Fusarium oxysporum. 


Fusarium 




Tem- 


mate 


Tubers. 












Fusarium 
radicicola. 
Strain 29. 


ture. 


relative 

hu- 
midity. 
















Strain i. 


Strain 8. 


Strain 32. 


StiainaS. 


strain 31. 






Per 
















°C. 


cent. 




















/New . 


2 mm. 


5 mm. 


I mm. 


4 mm. 


2 mm. 


S mm. 


TO - i 


33 noid.. 


2 mim. 


2 ram. 


I mm. 


7 mm. 


10 mm. 


S mm. 


13-5 


ijNew. 


5 mm. 


5 mm. 


I mm. 


2 mm. 


2 mm. 


5 mm. 






lOld.. 


5 mm. 


4 mm. 


I mm. 


5 mm. 


6 mm. 


5 mm. 






[New. 


20 mm. 


2 mm. 


3 mm. 


4 mm. 


5 mm. 


Xrot. 




2,3 


Old.. 


Contami- 


15 mm. 


15 mm. 


Xrot. 


5 mm. 


Xrot. 






nated. 












25 


100 


[New . 


Contami- 
nated. 


Krot. 


Contami- 
nated. 


2 mm. 


2 mm. 


Krot. 






loid.. 


5 mm. 


'A rot. 


15 mm. 


10 mm. 


>^rot. 


Complete 
rot. 



Experiment 5, 19 15. — The results of earlier experiments having 
indicated that the influence of the relative humidity was nearly as great 
as that of temperature, it was decided to run a more complete test on 
the effect of the relative humidities at different temperatures. Since 
no apparatus was available by which the relative humidity and tempera- 
ture could be controlled at will, it was necessary to construct one. 

The principle employed in experiment 4 of using sulphuric-acid solu- 
tions of varying specific gravity in a closed chamber to obtain the different 
relative humidities was not used in this test. In preliminary experi- 
ments conducted in the same way as experiments 4 and 4A, the infection 
usually resulted in a softrot which gave good comparative results, but 
the type of rotting was not similar to that usually found in storage. 
Cultures from these softrots invariably yielded the Fusarium sp. used 
in the inoculation, and no bacteria were present in any case. Apparently 



I 



74 Journal of Agricultural Research voi. xxii. no. 2 

the absence of any aeration was the cause of this abnormal type of rot- 
ting, and the following method was devised to allow for aeration. 

The principle j[inally decided upon was that of passing a current of 
air, kept at a constant pressure, through sulphuric-acid towers and then 
over calcium chlorid and sodium hydrate. This gave a constant stream 
of dry, sterile air. The air was then passed over sterile water to bring 
it to a desired humidity. The amount of water necessary for a given 
humidity was determined by trials, and the air was then passed into the 
jars containing the tubers. An outlet was provided at the bottom of 
the jar. These jars were connected separately with the current of air 
and not in series. Relative humidities were obtained and used through- 
out the experiment as follows: i, 30, 70, and 100 per cent. These rela- 
tive humidities were used at three different temperatures — 9°, 12.5°, 
and 25° C. The set at 25° was placed in an incubator in the laboratory, 
the set at 12.5° was placed in a special low temperature incubator, and 
the set at 9° was placed in a well-insulated ice box. Each of these 
temperatures was maintained within a variation of 2° throughout the 
experiment. In this way four gradations of humidity at each of three 
temperatures were obtained. The method provided the tubers with suf- 
ficient aeration and secured sterile conditions throughout the experiment, 
since the jars containing the tubers were not moved or opened until the 
end of the period. 

The humidity readings were taken by the wet- and dry-bulb method, 
the thermometers being inserted into the stream of air at the entrance 
to the jar. The readings were found to vary, and at least 10 trial read- 
ings were taken for each jar after the preliminary determinations were 
made and the apparatus was set up. These readings ranged as follows : 
I to 10 per cent, 20 to 40 per cent, 60 to 80 per cent, and 90 to loo per 
cent. These were the greatest extremes found; and since a knowledge 
of the approximate relative humidity is all that is necessary in an experi- 
ment of this kind, these readings were taken to be sufficient, inasmuch 
as they showed a gradual gradation from approximate dryness to satura- 
tion. The ranges given above simply denote the possible error due to 
the method of taking the readings. The humidity necessarily remained 
constant, since the temperature, water surface, and air pressure were 
constant. It was found to be impossible to use the wet- and dry-bulb 
method to determine the relative humidity at the lower temperatures. 
The changes of temperature caused by opening the door to make the deter- 
minations were found to change the readings. Therefore the sets at 9° 
and 12.5° C. were installed temporarily at 25°, the preliminary deter- 
minations were made, and the readings were taken at that temperature 
and corrections made by the use of psychometric tables (13). 

The large battery jars were fitted with wire screen supports, and six 
tubers were used in each jar — four inoculated and two controls. The 



oct.8.i92i Temperature and Humidity Studies of Fusaria Rots 75 

control tubers were separated from the inoculated ones by a thin layer of 
cotton. The entire apparatus was disinfected with formaldehyde gas 
before the experiment was set up. 

The tubers used were of the Up-to-Date variety. They had been 
kept over winter in a cool cellar, and a few sprouts which had started 
were removed. They were inoculated by wounding the epidermis and 
placing several drops of spore suspension in the wound. They were then 
placed in the jars which were closed with cork tops and paraffined. The 
inoculations were made with Fusarium oxysporum No. 8. The jars were 
opened up and the tubers examined after five weeks. 

Set I (9° C). At 10 and 30 per cent humidity the tubers were all 
healthy. 

At 70 per cent the tubers were sound with no penetration, although 
there was a slight growth of mycelium on the surface of the tuber at the 
point of inoculation. 

At 100 per cent the condition of tubers was the same as at 70 per cent, 
except that the external growth of mycelium was greater. All the 
control tubers of this set remained healthy, and both the controls and 
inoculated tubers had sprouted. 

Set 2 (12.5° C). At 10 per cent humidity the tubers were healthy. 
There was no invasion of the tissues. 

At 30 per cent, same as above with a slight external growth of mycelium 
at the point of inoculation. 

At 70 per cent the tubers were about the same as at 30 per cent. 
(PI. 10, A.) 

At 100 per cent invasion of the tissue had taken place for about 2 mm. 
beyond the wound, causing a slight browning of the tissue. On the 
surface there was a slight brown discoloration for several millimeters 
surrounding the point of inoculation and a slight growth of aerial myce- 
lium. All the control tubers in set 2 remained healthy, and both control 
and inoculated tubers were sprouting normally. (PI. 10, B.) 

Set 3 (25° C). The control tubers remained healthy and sprouted 
at 10 and 30 per cent relative humidity, while at 70 and 100 per cent 
there was a slight disorganization of the tissue around the eyes and the 
sprouts were all dead. No actual rotting was present or any fungus 
growth. 

With infected tubers at 10 per cent humidity all inoculations were 
successful and uniform. The fungus invaded the tissue for 2 cm. around 
the wound. Immediately below the surface at the point of inoculation 
there was in every case a cavity lined with a white mycelial growth. 
The tissue surrounding the cavity was of a granular appearance. Exam- 
ined under the microscope it appeared to be made up of a tangled mass of 
mycelium and starch grains. A sof trot extended out from this area, the 
tissue being light brown in color and completely invaded by mycelium 
(PI. 10, C). 



76 Journal of Agricultural Research voi. xxii. No. 2 

At 30 per cent humidity the rotting took place in the same manner 
as at 10 per cent, except that the cavity was larger and the rot extended 
through about 50 per cent of the tuber (PI. 11, B). 

At 100 per cent humidity there was a total rot of all tubers, most of 
the surface being covered with a white mycelial growth (PI. 11, C). 

The results of this experiment show very clearly that relative humidity 
plays a very important part in determining the amount of rot produced 
by Fusarium oxysporum. The only rot appearing at the low tempera- 
ture of 12.5° C. was in an atmosphere of 100 per cent humidity. Even 
at the high temperature of 25° complete rotting did not take place 
at the lower humidities. A gradual increase in the amount of rot 
corresponding to the increase in humidity was present in every case. 
The fungus can live and sporulate at the lower temperatures and lower 
humidities used in this experiment but apparently is not capable of 
penetrating the tubers under these conditions. It can be safely con- 
cluded that F. oxysporum under good storage conditions is not capable 
of producing a tuber-rot of great importance. 

CONCLUSIONS 

(i) Fusarium oxysporum, F. trichochecioides , and F. radicicola are all 
capable of producing a rot of the potato tuber. 

(2) In pure culture the amount of growth of all three species is nearly 
equal at 25° C, Fusarium oxysporum and F. radicicola increasing in 
growth up to 30°, where they produce their maximum growth. The 
growth of F. trichochecioides decreases above 25°, until at 30° very little 
or no growth takes place. It is more tolerant of the lower temperatures 
than the other two species. 

(3) Preliminary tests with different liquid media would indicate that 
the cardinal points for growth of these Fusaria vary to some extent 
with the medium used. 

(4) Experimental infection of tubers was produced with all three 
organisms under various conditions of temperature and relative humidit}'. 

(5) Preliminary tests with Fusariivm oxysporum indicated that the 
relative hmnidity plays a very important part in determining the amount 
of rotting. 

(6) In comparative tests with new and old tubers there is a distinct 
difference in the amount of rotting under the same conditions. The 
rotting was much more rapid and progressed much further in the old 
than in the new tubers. 

(7) Comparative tests with all three species at controlled relative 
humidities from i to 100 per cent and at controlled temperatures from 
5° to 25° C. proved conclusively that — 

(a) A temperature of 25° C. is favorable for the production of a tuber 
rot by Fusarium oxysportim, F. radicicola, and F. trichothecioides. 



Oct.8.i92i Temperature and Humidity Studies of Fusaria Rots 77 



(b) Fusarium oxysporum grows more rapidly and produces a more ex- 
tensive rotting of the tuber than the other two at a temperature of 16° C. 
and above. 

(c) Fusarium trichothecioides is capable of producing a rot at much 
lower temperatures than the others, in some cases causing rotting at 
5°C. 

{d) The relative humidity plays a very important role in determining 
the progress of tuber rots and has the same influence on all three species. 
In every experiment it was noticeable that there was a gradual increase 
in the amount of rot corresponding to an increase in relative humidity. 
With a high humidity at a given temperature the rotting was always 
greater than at a temperature 5° to 10° C. higher but with a low humidity. 
The Fusaria used can all live and sporulate at the low temperature of 9°, 
and with low relative humidities, but they are not capable of producing 
a rot under these conditions. 

Inasmuch as the three species of Fusaria used in these experiments 
represent the common types causing storage-rots of potatoes, it is clear 
that considerable attention should be given to moisture as well as tem- 
peratures where incipient rot occurs in stored tubers. It is also en- 
tirely probable that a rotting of the tubers initiated at high temperatures 
and high relative humidities could be completely checked by submitting 
the tubers to lower temperatures and lower humidities. 

LITERATURE CITED 
(i) AppEt, Otto, and Wollenweber, H. W. 

I910. GRUNDLAGEN EINER MONOGRAPHIE DER GATTUNG FUSARIUM (LINK). 

Arb. K. Biol. Anst. Land. u. Forstw., Bd. 8, Heft i, 207 p., 10 fig., 3 
pi. (i col.) Verzeichnis der wichtigsten benutzten Schriften, p, 
196-198. 

(2) Bary, Anton de. 

1861. DIE gegenwartig herrschende kartopfelkrankheit, ihre ursache 
UND ihre verhutung. eine pflanzenphysiologische unter- 

SUCHUNG . . . 75 p., I pi. I^ipzig. 

(3) BisBY, G. R. 

1919. STUDIES ON FUSARIUM DISEASES OF POTATOES AND TRUCK CROPS IN MIN- 

NESOTA. TECHNICAL. Minn. Agr. Exp. Sta. Bui. 181, 58 p., 30 fig. 
Bibliography, p. 40-44. 

(4) Carpenter, C. W. 

1915. SOME POTATO TUBER-ROTS CAUSED BY SPECIES OF FUSARIUM. In Joiir. 

Agr. Research, v. 5, no. 5, p. 183-210, pi. A-B (col.), 14-19. Litera- 
tiire cited, p. 208-209. 

(5) Edson, H. a., and Shapovalov, Michael. 

1920. TEMPERATURE RELATIONS OF CERTAIN POTATO-ROT AND WILT-PRODUCING 

FUNGI. Ir„ Jour. AgT. Research, v. 18, no. 10, p. 511-524, 9 fig. 

(6) Frank, Albert Bernhard. 

1898. UNTERSUCHUNGEN USER DIE VERSCHIEDENEN ERREGER DER KARTOFFEL- 

FAULE. In Ber. Deut. Bot. Gesell., Bd. 16, Heft 8, p. 273-289. 



jS Journal of Agricultural Research voi. xxn. No. a 

(7) Humphrey, H. B. 

i914. studies on the relation of certain species of fusarium to the to- 
MATO BLIGHT OF THE PACIFIC NORTHWEST. Wash. AgT. Exp. Sta. Bul. 

115, 22 p., 5 pi. 

(8) Jamieson, C. O., and WollenwebER, H. W. 

I912. an external dry rot of potato TUBERS CAUSED BY FUSARIUM TRICHO" 

THEcioiDES, WOLLENW. In Jour. Wash. Acad. Sci., v. 2, no. 6, p. 
146-152, I fig. 

(9) Link, George K. K. 

I916. A PHYSIOLOGICAL STUDY OP TWO STRAINS OP FUSARIUM IN THEIR CAUSAL 
RELATION TO TUBER ROT AND WILT OP POTATO. In Bot. Gaz., V. 62, 

no. 3, p. 169-209, 13 fig. Literatiire cited, p. 207-208. Reprinted as 
Nebr. Agr. Exp. Sta. Research Bul. 9. 1916. 

(10) Longman, Sibyl. 

1909. THE DRY-ROT OF potatoes. In Jour. Linn. Soc. (London), Bot., v. 39, 
no. 270, p. 120-129, pi. 10. Literature referred to in the text, p. 129. 

(11) Manns, Thomas F. 

I91I. THE FUSARIUM BLIGHT (WILT) AND DRY ROT OF THE POTATO. PRELIM- 
INARY STUDIES AND FIELD EXPERIMENTS. Ohio Agr. Exp. Sta. Bul. 
229, p. 229-336, 15 pi. (in text). 

(12) Martius, Carl Friedrich Philipp von. 

1842. DIE KARTOFFEL-EPIDEMIE DER LETZTEN JAHRE ODER DIE STOCKFAULE UN 
RAUDE DER KARTOFFELN, GESCHILDERT UND IN IHREN URSACHLICHEN 

VERHALTNISSEN ERORTERT. 70 p., 3 col. pi. Miinchen. 

(13) Marvin, C. F. 

1900. PSYCHROMETRIC tables FOR OBTAINING THE VAPOR PRESSURE, RELATIVE 
HUMIDITY, AND TEMPERATURE OF THE DEW-POINT ... U. S. Dept. 

Agr. Weather Bur. Bul. 235, 84 p., 2 fig. 

(14) Pethybridge, George H., and BowERS, E. H. 

1908. DRY ROT OF THE POTATO TUBER. In Econ. Proc. Roy. Dublin Soc, v. i, 
pt. 14, p. 547-558, pi. 48- 

(15) PiZZIGONI, A. 

1896. CANCRENA SECCA ET UMiDA dellE patate. In Nuovo Gior. Bot. Ital., 
n. s. V. 3, fasc. i, p. 50-53. 

(16) Pratt, O. A. 

I916. A WESTERN FIELDROT OF THE IRISH POTATO TUBER CAUSED BY FUSA- 
RIUM RADicicoLA. In Jour. Agr. Research, v. 6, no. 9, p. 297-310, 
Pl- 34-37- 

(17) 

1916. CONTROL OF THE POWDERY DRYROT OF WESTERN POTATOES CAUSED BY 

FUSARIUM TRicHOTHEcioiDEs. In Jour. Agr. Research, v. 6, no. 21, 
p. 817-832, pi. 108. 

(18) Reinke, Johannes, and Bertholdt, Gottfried. 

1879. DIE zersetzung der kartoffel durch pilze. 100 p., 9 pi. Berlin. 
(Untersuch. Bot. Lab. Univ. Gottingen. Heft i.) 

(19) Schacht, Hermann. 

1856. bericht an das konigliche landes-oekonomie-collegium uber die 

KARTOFFELPFLANZE UND DEREN KRANKHEITEN ... 29, II p., 10 pi. 

(6, 8, 9 col.). Berlin. 

(20) Shereakofp, C. D. 

1915. fusaria op potatoes. N. Y. Cornell Agr. Exp Sta. Mem. 6, p. 
87-270, 51 fig., 7 pi. (col.). Literatiu-e cited, p. 269-270. 

(21) Schlechtendal, D. F. L. de. 

1824. FLORA BEROLINENSIS, PARS SECUNDA. CRYPTOGAMIA. Berlini. 



Oct. 8, I92I Temperature and Humidity Studies of Fusaria Rots 79 

(22) Smith, Erwin F., and Swingle, Deane B. 

1904. THE DRY ROT OF POTATOES DUE TO FUSARIUM OXYSPORUM. U. S. Dept. 

Agr. Bur. Plant Indus. Bui. 55, 64 p., 2 fig., 8 pi. Literature, p. 
61-62. 

(23) Stevens, Neil E. 

I916. A METHOD FOR STUDYING THE HUMIDITY RELATIONS OP FUNGI IN CUL- 
TURE- In Phytopathology, v. 6, no. 6, p. 428-432. Literature cited, 
P- 432- 

(24) Wehmer, C. 

1896. UEBER DIB URSACHE DER SOGNENANNTEN "TROCKENFAULE" DER KAR- 

TOFFELKNOLLEN. hi Ber. Deut. Bot. Gesell., Bd. 14, Heft 3, p. loi- 
107, 3 fig. 

(25) 

1897. UNTERSUCHUNGEN iJBER KARTOFFELKRANKHEITEN. 2. ANSTECKUNGS- 

VERSUCHE MIT FUSARIUM SOLANI (DIE FUSARIUM-FAULE.) In Centbl. 

Bakt. [etc.], Abt. 2, Bd. 3, No. 25/26, p. 727-743, pi. lo-ii (i col.). 

(26) Wilcox, E. Mead, Link, George K. K., and Poole, Venus W. 

1913. A DRY rot of THE IRISH POTATO TUBER. Nebr. Agr. Exp. Sta. Re- 
search Bui. I, 88 p., 15 fig., 28 pi. (i col.). Bibliography, p. 85-88. 

(27) WOLLENWEBER, H. W. 

1913. STUDIES ON THE FUSARIUM PROBLEM. In Phjrtopathology, V. 3, no. i, 
p. 24-50, I fig., pi. 5. Literatiu-e cited, p. 46-48. 



PLATE lo 

Tubers inoculated with Ftcsariwrn oxyspomm and kept for five weeks at the fol- 
lowing temperatures and humidities: 
A. — 12.5° C, 70 per cent relative humidity. 
B. — 12.5° C, 100 per cent relative humidity. 
C. — 25° C, 10 per cent relative humidity. 

(80) 



Temperature and Humidity Studies of Fusaria Rots 



Plate 10 





f 



B 





^ 



Journal of Agricultural Research 



Vol. XXII, No. 2 



Temperature and Humidity Studies of Fusaria Rots 





B 





Journal of Agricultural Research 



Vol. XXII, No. 2 



PLATE II 

Tubers inoculated with Fusarium oxysporum and kept for five weeks at the fol- 
lowing temperatures and humidities : 
A. — 25° C, 30 per cent relative humidity. 
B. — -25° C, 70 per cent relative humidity. 
C. — 25° C, 100 per cent relative humidity. 



BLACKLEG POTATO TUBER-ROT UNDER IRRIGATION 

By M. Shapovalov and H. A. Edson, Pathologists, Office of Cotton, Truck, and Forage 
Crop Disease Investigations, Bureau of Plant Industry, United States Department of 
Agriculture . 

OCCURRENCE AND GENERAL APPEARANCE ' '■ 

A bacterial field decay of the potato tuber, the real nature of which 
has not heretofore been adequately explained, prevails in certain irri- 
gated sections of the West. In early harvest, when the diseased tubers 
are apparently free from fungous invasion, the trouble has sometimes 
been assumed to be "sunscald"; during the winter months it has fre- 
quently been taken for a form of freezing injury. In other instances 
it has been confused with the so-called "jelly-end rot" and attributed 
either to Fusarium radicicola Wollenw. or to F. oxysporum Schlecht. It is 
probable, also, that on superficial examination some such material has 
been classed as "leak" {Pythium debar yanum Hesse), when conditions 
favored an extremely rapid progress of the decay, whether in the field 
or in transit. 

Specimens of this decay were received by the writers in 191 7 and 
19 1 8 from Idaho, Nevada, and California. In material received in 
August the decay was soft and mushy (PI. 12, A-C). The aflfected 
tissues were in part brown to black, but mostly only slightly colored or 
colorless, though with a darker margin on the border line between the 
healthy and diseased portions. Disintegration, originating at one end 
of the tuber, was advancing irregularly over the surface. In some 
areas the decay was confined to the outer layer, just beneath the epi- 
dermis, while in others the deeper tissues also were involved. As a 
rule, the disease started at the stem end, but occasionally the eye end 
became infected first (PI. 12, B). Decaying material usually possessed 
a disagreeable odor. It is this soft type of the rot which some were 
inclined to regard as sunscald injury. 

Specimens received later in the season, during the months of Novem- 
ber and December, presented an entirely different appearance. The 
affected portions were not mushy, but more or less tough or dry and 
shrunken (PI. 12, D). The diseased area was dark brown in color, 
except when a fresh decay developed under favorable conditions deeper 
in the tissues. In the latter case it was practically of the same color as 
the normal flesh of the tuber, but soft and mushy in consistency. When 
such tubers were cut open and the cut surfaces exposed to the air, the 
diseased portions turned brown or even black. If the progress of the 
decay is completely anrested, the trouble may readil y be mistaken for 

Journal of Agricultural Research. Vol" ^^^^ ^°- ' 

Washington. D.C. Key No'.'g-»46 

(81) 



82 Journal of Agricultural Research voi. xxii. no. 2 

an inactive stage of jelly-end rot or for an after effect of freezing injury, 
particularly if the disease has made but little headway. The true 
nature of such obscure cases of the disease may be revealed with cer- 
tainty only by a series of cultural studies, coupled with experimental 
work and field observations. 

CAUSAL ORGANISM 

Isolations were made from every tuber of each of the four samples 
received from the West in 191 7-18. The results were surprising. In 
no case was Fusariu^n radicicola obtained; only one tuber yielded F. 
trichothecioides Wollenw. (from Nevada), two yielded Rhizoctonia 
(from Nevada), two F. oxysporum (from California), and a few gave 
miscellaneous, apparently saprophytic, fungi. Bacteria, on the other 
hand, were constantly present in the cultures, even when slightly acidu- 
lated potato agar was used. Carpenter (2)^ noted the presence of 
bacteria in jelly-end rot material, but he regarded these organisms as 
saprophytic, as they probably were. In the writers' cultures, however, 
the constant prevalence of one type of bacterial colony in the dilution 
plates was significant and warranted a detailed study of this organism. 
In the subsequent inoculation experiments with pure cultures it proved 
to be strongly pathogenic and produced a progressive decay of the 
tubers as well as a disease of the stems. A study of the cultural and 
biochemical features of the organism showed them to be fully in accord 
with the published description of the blackleg bacillus (5). 

MORPHOLOGY 

Short rod with rounded ends, also short chains; 0.5 to 0.9 Xi.c to 
2.2 ;u, average 0.6 X i-S /x; flagella few, peritrichiate; no endospores and 
no capsules; stains well in aqueous gentian violet, aqueous methylene 
blue, aqueous fuchsin, anilin water gentian violet, alkaline methylene 
blue, and carbol fuchsin. 

CULTURAL FEATURES 

Agar stroke. — Growth moderate, filiform, flat to slightly raised, 
glistening, smooth, slightly opalescent; white, no odor; consistency slimy 
to butyrous; one strain distinctly viscid at first, but after a few replat- 
ings it lost its viscidity. 

Potato. — Growth moderate to abundant; filiform at first then spread- 
ing, slightly convex changing to flat, glistening, smooth to slightly ru- 
gose, yellowish white or dirty white; a decided odor of decayed potatoes 
on the third to fourth day at 22° to 25° C. ; consistency somewhat slimy; 
medium slightly grayed at first, changing later to either plainly gray, or 
purplish, or brown, or a combination of these shades. 

' Reference is made by number (italic) to " Literature cited," p. 91-92. 



Oct. 8, 192 1 Blackleg Potato Tuber-Rot under Irrigation 83 

Agar stab. — Growth somewhat best at top, abundant, spreading, 
filiform to slightly papillate. 

Gelatin stab. — Growth best at top, filiform along the line of punctm-e, 
liquefaction, beginning on the first day at 20° C, varying in shape from 
crateriform or funnel-shaped to saccate and broadly infundibuliform, 
complete in 7 to 12 days. 

Nutrient broth. — Usually slight ring and slight granular pellicle in 
young cultures, clouding moderate to strong, persistent; medium not 
discolored, odor absent; sediment compact, granular, somewhat dirty 
white; one strain decidedly viscid at first, but losing this character after 
a few replatings. j-mi/ibtia Jrff-io 

MiivK. — Coagulation and extrusion of whey at 25° C, beginning on the 
fourth day; coagulum not digested; one strain extremely viscid at first, 
but losing its viscidity in later replatings; medium not discolored. 

Acid production in milk. — A slight increase of acidity in milk cultiu-es 
was noticeable after 24 hours. Two series of tests were made at certain 
intervals within the period of 20 days, two to three cultures being used 
on each day for every strain. The average progress of the acidity of three 
western strains was as follows : 

AGE OP REACTION IN 

CULTURE. fuller's SCALE. 

1 day ,, +12. 28 

2 days .'. .: '. +13-49 

3 days +22. 13 

5 days +28. 00 

10 days +32. 90 

20 days .,;..„ . .^,. , i: K-. +41- 12 

The average reaction of the control tubes was +11. 8 Fuller's scale. 
Two strains received from Dr. W. J. Morse, of the Maine Agricultural 
Experiment Station, were tested along with the western strains and gave 
similar reactions, one ("B. sol.") showing 37.0 and the other ("IIIA") 
40.75 acidity on the twentieth day. The cultures were grown at 22° to 
25° C. 

Litmus milk. — ^At 22° to 25° C. bleaching was complete at the end of 
three weeks; thorough reddening was accomplished in seven weeks. 

Gelatin colonies. — Growth rapid, form round, edge entire, lique- 
faction saucer-shaped. 

Agar colonies. — ^Surface colonies; growth rapid, usually round, but 
occasionally somewhat irregular, flat to slightly raised, entire to slightly 
undulate, finely granular with an internal ring surrounded by radiate 
striations; color pearly white, bluish opalescent by transmitted light; 
maximum diameter of colonies after 2 days 2 mm., after 3 days 4 mm., 
after 7 days 7 mm., after 14 days 9.5 mm. Buried colonies lens-shaped 
to nearly spherical, edge entire, color slightly yellow under hand lens. 

Fermi's solution. — Moderate clouding in 2 -day cultures; later 
growth becomes copious. ii^ mc*^": 

54818*— 21 8 



84 Journal of Agricultural Research voi. xxii, no. 2 

Corn's SOLUTION. — No growth. 

Uschinsky's solution. — Growth was somewhat irregular in the ordi- 
nary Uschinsky's solution but was uniform and copious in the modified 
Uschinsky's solution, clouding being very strong on the fifth day. 

Sodium chlorid in bouillon. — Growth slightly inhibited by 3 per 
cent and more so by 4 per cent; no growth appeared in 5 per cent tubes 
until the third day, and only occasional tubes containing 6 per cent were 
clouded after 5 days. Morse reports no clouding for Bacillus atrosepti- 
cus Van Hall in concentrations higher than 5 per cent. At the end of 
two months, when conditions remained unchanged, transfers were made 
from 6 per cent and 7 per cent sodium-chlorid cultures of the western 
strains to sterile broth. In 48 hours all the transfers from 6 per cent 
solutions showed growth, and in three days clouding appeared in the 
majority of the transfers from the 7 per cent solutions, the remainder 
being dead. 

Growth in bouillon over chloroform. — Growth somewhat re- 
strained at first, but increasing gradually. On the fourth day there was 
a strong and uniform clouding in all cultures. 

Best medium for long-continued growth. — Morse considers that 
in the case of Bacillus atrosepticus neutral beef bouillon is best for this 
purpose. In the western strains the writers observed that the organisms 
can live even longer on the agar than on the broth when grown at ordinary 
laboratory temperature of 22° to 25° C. Their death on agar appears 
to be primarily associated with drying of the medium, while in broth it 
seems to be due to certain chemical changes in the substratum and takes 
place sometime before the liquid dries up completely. Six series of 
parallel broth and agar cultures were made and tested at diff"erent inter- 
vals, from 8 to 36 weeks, by making transfers to tubes of sterile broth. 
It was found that occasional broth cultures showed a somewhat weak- 
ened vitality, as demonstrated by retarded clouding, at the age of 16 
weeks; some died after the expiration of 20 weeks, and none lived beyond 
26 weeks. On the other hand, in no case was the agar culture dead 
before 26 weeks, and some remained alive even after 36 weeks. The 
experiment was carried on with 10 cc. of medium in each test tube. 

PHYSICAL and biochemical FEATURES 

Fermentation tubes. — Gas and acid production as well as growth 
in the closed arm was observed with dextrose, lactose, and saccharose. 
No acid and no gas with glycerin in cultures 1,3, and 5 days old. 

Ammonia production. — Feeble (tested by Folin's aspiration method). 

Nitrates in nitrate broth reduced to nitrites. 

Indol production. — Positive, but very feeble both in young and old 
cultures. 

Toleration of hydrochloric acid and sodium hydrate. — The 
writers' western organism grew in tubes having an initial reaction before 



Oct. 8, 1921 Blackleg Potato Tuber-Rot under Irrigation 85 

final sterilization of +20 and —20, Fuller's scale, but not in those 
adjusted to +30 or to —30. Uninoculated tubes held as controls and 
titrated at the close of a 24-day incubation period showed marked 
changes from the original reaction, due doubtless in considerable measure 
to the absorption of gases with resulting chemical change. Tubes cal- 
culated for an initial reaction of -f 30 showed a final reaction of from +20 
to +25; those originally +20 were about +15; those —30 were about 
— 10; and those —20 were about —6. Transfers from inoculated tubes 
calculated for an initial reaction of + 30 and above and — 30 and below 
made 24 days after inoculation developed growth in some cases in tubes 
from + 30, but not in those from more acid reactions nor from the alka- 
line broths. 

Vitality on culture media. — ^Long on bouillon, but still longer on 
agar. 

Temperature relations. — In freshly inoculated broth cultures 
exposed 10 minutes, occasional retardation of clouding began at 45° C. ; 
occasional growth was noted at various points between 46° and 50°; 
and in no case was growth present after heating above 50°. Optimum 
temperature for growth about 25°. Maximum temperature for growth 
between 33° and 35°. Minimum temperature for growth below 5°. 

Effect of sunlight. — Thinly sown agar plates exposed on ice for 30 
minutes the latter part of March in Washington, D. C, resulted in 100 
per cent killed. 

Cytase production. — Five-day-old 30-cc. broth cultures in 300-cc. 
Erlenmeyer flasks were precipitated by 160 cc. of 80 per cent alcohol, 
filtered, and the precipitate dried promptly in the air. The papers con- 
taining the dried precipitate were washed with 30 cc. of water, and the 
washings were received in a flask to which a few drops of toluene and three 
raw Irish potato disks 15 by 2 mm. were added. The disks gradually 
assumed a soft, cheesy consistency but did not entirely disintegrate. 
Microscopic examination showed the cells had lost coherence through 
softening of the middle lamella. The cellulose lamella and the starch 
content of the cells showed no evidence of change. Controls with 
uninoculated broth did not soften the disks. 

GROUP number 22 1. 1 II 3033 

The last three points in this group number differ from those given by 
Morse (5) but coincide with the respective figures in Jennison's {3) 
revision, as reported by him at the fourth annual meeting of the Pacific 
Division of the American Phytopathological Society. The writers feel, 
however, that this may be largely a matter of interpretation of certain 
results and not necessarily an indication of actual difference in the 
organisms. Jennison studied 12 different strains of the blackleg bacillus, 
includingseveralof Morse's strains; but the results he obtained, apparently, 



86 Journal of Agricultural Research voi. xxh. No. » 



were identical for all strains. The writers regard their western strains 
as nonchromogenic, although a certain yellow discoloration in cooked 
potato cultures might be taken as a suggestion of yellow pigment. They 
obtained no evidence of diastatic action on potato starch, nor of acid 
production with glycerin. 

It may, therefore, be concluded on the basis of the characters described 
that the pathogenic bacillus isolated by the writers from a peculiar soft 
decay of western potato tubers is essentially identical with the organism 
causing the blackleg disease of potatoes for which Appel's binomial 
Bacillus phytophthorus is regarded to be correct by Smith (7) . Morse (5) , 
who was unable to obtain an authentic culture of Appel's strain for his 
comparative studies of various blackleg organisms, believed that 
B. atrosepticus should be chosen in preference to other names he had 
under consideration, but stated that — 

There is nothing in the data here presented which bears on the relation between 
the organism originally described by Dr. Appel (/) as B. phytophthorits and the other 
strains of blackleg bacteria. 

EXPERIMENTAL WORK 

The pathogenicity of the bacterial organism described above was 
established by means of the following laboratory, greenhouse, and field 
experiments. 

PLANTING OF THE ORIGINAL MATERIAL 

Preliminary to the inoculation work some of the diseased western 
material was planted in the greenhouse as soon as the isolations were 
completed. Four tubers were selected and cut in halves so as to make 
eight seed pieces. Each piece was planted in a separate large pot filled 
with sterilized soil. Of these seed pieces one decayed completely in the 
soil before germination, six produced diseased plants, and one produced 
a plant considerably weakened though not clearly diseased. Some of 
the affected plants decayed while very young, others grew up to practi- 
cally normal size, developing blackening of the stem above ground and 
brown to black lesions on the underground portions. The lower leaves 
turned yellow, but the upper leaves wilted while green. In two cases 
the blackening of the stems was very intense (PI. 13, A) while in the 
remainder the appearance was less typical of the familiar field symptoms 
of the disease as it occurs in the eastern sections of the country. Black- 
ening of the pith of the stem developed to the very top in one case. 
When an affected plant was removed from the pot and the soil was care- 
fully washed off, it could be seen that the infection had spread from the 
seed piece to the stem (PI. 13, B). No tubers were produced in this 
experiment. Healthy sprouting Irish Cobbler tubers were replanted in 
these pots, but no infection was contracted by this new set of plants. 



Oct 8, 1941 Blackleg Potato Tuber-Rot under Irrigation 87 

INOCUIrATlON OF HEALTHY TUBERS IN THE LABORATORY 

Over 60 tubers both new and old of the Netted Gem as well as of the 
Irish Cobbler varieties were inoculated in small lots at dififerent times 
with three strains of the western decay bacillus. When inoculations 
were made in wounds of any kind, whether on the side of the tuber or at 
either end of it, the results were invariably positive. The progress of 
the decay was much slower when uninjured potatoes were inoculated. 
In these latter cases the organism penetrated either through the eyes or 
through the young growing sprouts. If the infected potatoes are re- 
moved from the moist chamber after the decay has made considerable 
headway and are exposed to the dry air of the laboratory, the diseased 
tissues become shriveled and folded, resembling very closely the original 
specimens of natural infection (PI. 14, A, C). Ordinarily if the infected 
material is kept in moist chambers the decay is soft, mushy, spreading 
either equally throughout the tissue or sometimes more on the surface 
of the tubers, and is not confined to their piths as is usual in typical 
cases of blackleg. The color of the decaying areas ranges from that of 
the normal flesh to light or dark brown, often with blackish streaks or 
stripes in younger portions nearer to healthy tissues, but never black 
throughout. The margin is usually well defined, and there is no gradual 
transition from dead to sound tissues. This internal appearance changes 
considerably when tubers are taken from the moist chamber and are 
exposed to drying. The decay of the bark is then more or less arrested, 
and the disintegration centers mainly in the pith, so that a more or less 
sound shell surrounds the centers of the active decay. The diseased tis- 
sue is brown to black, the older regions becoming slimy (Pi. 14, B, D). 
In all cases the decay gives off a very strong putrefactive odor. 

PLANTING OF ARTIFICIALLY INOCULATED TUBERS IN THE GREENHOUSE 

Seven Irish Cobbler tubers inoculated with the western bacterial or- 
ganisms and partly decayed were planted in sterilized soil in pots. Four 
showed subsequently a stem decay and three remained apparently un- 
affected. One plant became girdled and died early. The disease ap- 
peared first on the remaining three plants in the form of black streaks in 
various positions on the stems, particularly at the leaf petioles. Later 
on in some instances the entire stalk became black at the base. Tuber- 
rot did not appear except on one tuber in one of the diseased pots. In 
this case it was a soft, watery decay, light in color, not typical for black- 
leg. The causal organism, identical with the original strains, was, how- 
ever, recovered from this area. Healthy sprouting Irish Cobbler tubers 
were immediately planted in the same pots in which these specimens 
were grown. The new plants were very vigorous, and none of them con- 
tracted the disease. 



88 journal of Agricultural Research voi. xxn. No. a 



INOCULATION OF HEALTHY STEMS IN THE GREENHOUSE 

The stems of four young healthy potato plants were inoculated with 
24-hour-old broth cultures of the three western strains of the bacillus 
injected by means of a hypodermic needle. A severe decay with an 
accompanying blackening resulted in all cases. 

FIELD EXPERIMENTS 

These experiments were conducted for two successive years at Arling- 
ton Farm, Va. In 1 9 19 Netted Gems and Irish Cobblers were used. The 
tubers were inoculated with the western strains of the blackleg organism 
a few days before planting. Six whole tubers and 20 halves of the first 
variety and 9 whole and 19 halves of the second variety were planted. 
In addition a number of uninoculated pieces of each variety were planted 
for controls. Planting was done on May 5. One half-tuber seed piece 
of each variety decayed in the ground. On July i one hill from the cut 
seed of Irish Cobblers was noted to show secondary symptoms character- 
istic of blackleg — ^namely, yellowing and rolling of the leaves. There 
was no blackening of the stem above ground. The underground portions, 
however, showed brown lesions and a brown rot of the stem at the point 
of attachment to the seed piece and somewhat above it. None of the 
remaining plants showed symptoms of the disease. At digging time, on 
September 15, no decay of the tubers was found, with the exception of 
one very small tuber of the Netted Gem variety which showed a soft bac- 
terial decay at the stem end. The progress of the decay, however, was 
checked, and the affected portion fell off, leaving only the sound part, 
so that the recovery of the causal organism was not possible. 

Since the hot weather after May 5 might have had something to do 
with the slight progress of the disease in 19 19, two sets of plantings were 
made on another piece of ground on the same farm in 1920 — one on 
April 8 and the other on May 6. Only Irish Cobblers were used this time. 
Twelve tubers were cut in halves through the inoculated wound so as to 
make 24 seed pieces for each of the two series. Inoculations were made 
a few days before planting. Eight tubers were inoculated with the three 
western strains of the blackleg organism and 4 with the "B. sol." strain 
received from Dr. Morse. Up to July 20 four hills out of 16 inoculated 
with the western strains in the earlier planting and 2 out of 8 inoculated 
with " B. sol." in the same series showed typical field symptoms of black- 
leg, including an intense blackening of the base of the stem. On the 
other hand, no hill of the series inoculated with the same organisms and 
on the same plan, but planted one month later, showed any signs of 
infection. At harvesting time, on July 20, a number of tubers in the 
planting of April 8 showed blackleg-rot, and in the later planting only 
2 tubers were found showing the same decay. It appears, therefore, 
that the earlier planting, when the soil and the air temperatures were 



Oct. 8, 1921 Blackleg Potato Tuber-Rot und^r Irrigation 89 

lower and the soil moisture was more abundant, greatly facilitated the 
development of blackleg. 

FIELD OBSERVATIONS 

The typical case of blackleg-rot on round varieties in the East has been 
figured in publications and charts issued by several agricultural institu- 
tions. As a rule, decay begins at the stolon end of the tuber with a 
comparatively small amount of rot visible on the outside or often only a 
small, black, circular opening. This opening leads to the interior of 
the tuber, where a progressive decay develops in the form of an irregular 
black, soft, or slimy hollow until nearly all of the tuber is consumed 
(PI. 15, D, B). However, the development of the disease may deviate 
from this type even in eastern and northern sections of the United 
States, when conditions are abnormal and favorable to the disease, such 
as those in moist places or in wet seasons. Morse stated (4) with refer- 
ence to blackleg in Maine that — 

When this disease occxirs on a field it doubtless is responsible for much of the soft 
rot of the tubers observed in wet seasons. 

It appears from certain observations made by the writers that under 
conditions of excessive soil moisture the bacteria in stems or seed tubers 
may be carried at least to the adjoining tubers of the same hill. The 
latter then become infected from the outside, or, if they are already 
infected through the stolons, the infection spreads in moist surround- 
ings more rapidly on the outside over the surface of the tuber, or evenly 
throughout the flesh. Specimens of this sort were observed on the 
Eastern Shore of Maryland and Virginia (PI. 15, G) and in Wisconsin, 
Minnesota, and Washington (PI. 15, F) on various round types of pota- 
toes. More accentuated symptoms of this order were found in irrigated 
sections of Colorado. The most peculiar manifestations of the blackleg 
tuber-rot were seen in the Snake River Valley of Idaho, where the 
Netted Gem variety is grown on a large scale. The following forms were 
observed there dm-ing a field survey arranged by the Office of Cotton, 
Truck, and Forage Crop Disease Investigations in 1920. 

I. The stem-end rot of pointed-end Netted Gems. The external 
appearance of this form is extremely misleading (PI. A, 1-4). It becomes 
prevalent in southeastern and eastern Idaho during the latter part of 
the season, shortly before the harvest. The relatively low temperature 
prevailing at this time of year is, no doubt, an important factor in the 
rapid progress of the disease. If the soil has plenty of moisture, freshly 
dug affected tubers show no shrinkage and preserve their natural shape 
(PL 15, A-C). In the course of two weeks the shrinkage is evident and 
the decay takes on an inward trend (PI. A, 4). By another two weeks, 
drying and folding of the decayed tissues become very pronounced, and 
the external appearance at this stage of the decay may well pass as an 



90 Journal of Agricultural Research voi. xxn, ko. n 



illustration of any of a number of stem-end tuber-rots (PI. i6, A, B). 
When a Netted Gem tuber of pointed shape affected with this type of 
the disease is cut open longitudinally as soon as it is removed from the 
ground, four distinct regions of decay may, as the rule, be seen: (a) 
The extreme stem-end region is usually decayed throughout ; it is mushy 
or slimy in consistency and dark brown to black in color ; in the field this 
seldom extends deeper than the outer demarkation line of the decay, 
but in storage under favorable conditions the disintegration advances 
more rapidly in the inner tissues of the tuber, taking on a cup-like shape 
and leading ultimately to the formation of a slimy cavity (PL A, 4; 14, B, 
D; 16, B). (6) An area of fresh decay appears within the core just 
beneath the first region; it is practically colorless, though it occasion- 
ally contains dark or black streaks, and in the very early stages it has 
the consistency of hardened butter (PI. A, 4). (c) The cambium layer 
shows a brown discoloration extending sometimes close to the eye end; 
in advanced stages a portion of this region nearest to the stem end is 
more or less disintegrated and forms a channel attentuating toward the 
eye end until it gradually transforms into a mere browning of the vas- 
cular network which also gradually loses its intensity and finally disap- 
pears altogether; this condition is very distinct with some freshly dug 
tubers, but later on with the inward progress of the decay it becomes 
less pronounced (PI. A, i, 4; 16, B), {d) The decay of the outer layer 
develops in the bark region, is soft but not mushy in consistency and 
more or less dark-brown in color; it frequently extends over the tuber 
much farther than the pith decay, but not always as far as the cambium 
discoloration; its progress is checked after tubers are dug and exposed 
to drying (PI. A, i, 2; 16, B). 

2. The shallow stem-end rot of round-shaped tubers. This type was 
observed mostly on the Idaho Rurals. Under conditions of abundant 
moisture the bacterial infection spreads from the stem end over the 
surface of the tuber and penetrates into the bark region, though not 
very deeply. When such tubers are taken out of the ground and exposed 
to the sun, as happens at digging, the infected areas dry up very promptly 
and form hard, black, shallow patches (PI. 16, E). The condition may 
easily be mistaken for the black fieldrot described by Pratt (6) and 
attributed to Fusarium radicicola. If, however, the tubers are again 
transferred to a moist place with a moderate temperature, a soft, mushy 
bacterial decay is likely to develop beneath these dry areas. On the 
contrary, under conditions unfavorable to the blackleg decay the status 
may either remain unchanged or become complicated by the entrance 
of various Fusaria and other rot-producing fungi. In the latter case it 
is impossible to determine the original cause of the disease. 

3. Siderot of either round or long potatoes. It may penetrate inside 
of the tuber to a considerable depth, and when a freshly dug diseased 



oct.8.r«>4i Blackleg Potato Tuher-Rot under Irrigation 91 

tuber is cut open it reveals a colorless buttery or mushy decay with 
black streaks, usually on the border line of the diseased and healthy 
tissues. If exposed to drying the decayed areas may become spongy 
and very much resemble the texture which is usually observed in Fusa- 
rium rots (PI. 16, D). In many instances, however, if the decayed region 
is sufficiently deep to prevent complete drying, sections through such 
tubers may show an inner layer of active bacterial decay. As is the case 
with the other forms of blackleg-rot this form, too, may become further 
invaded with various rot-producing or saprophytic fungi. The writers 
had under their observation a tuber of this type with a copious growth 
of Rhizoctonia on the outside all over the diseased area, while soft 
bacterial decay was still progressing within the tuber even in the dry 
laboratory atmosphere (PI. 16, C). 

SUMMARY 

(i) An organism isolated from western stem-end rotting potatoes is 
identical with Bacillus phytophthorus Appel in all the essential charac- 
ters commonly considered in the determination of bacterial species. 

(2) It is pathogenic to the potato, and inoculations of healthy stems 
or tubers with pure cultures produce, respectively, a rapid, soft decay 
of stems or a tuber-rot. 

(3) Blackleg tuber-rot under the field conditions in certain irrigated 
sections of the West, particularly in pointed-end Netted Gems, takes on 
a form atypical of the familiar manifestation of this disease in the East. 

(4) The external appearance becomes especially confusing when the 
affected areas dry up and shrivel in storage, but usually the trouble 
may be identified by cultural work or by planting diseased tubers under 
control conditions. 

LITERATURE CITED 
(i) Appel, Otto. 

1903. UNTERSUCHUNGEN USER DIE SCHWARZBEINIGKEIT UNO DIE DURCH 
BAKTERIEN HERVORGERUFENE KNOLLENFAULE DER KARTOFFEt. In 
Arb. K. Biol. Anst. Land. u. Forstw., Bd. 3, Heft 4, p. 364-432, 15 fig., 
pL 8 (coL). 

(2) Carpenter, C. W. 

1915. some potato tuber-rots caused by species op fusarium. in jout. 
Agr. Research, v. 5, no. 5, p. 183-210, pL A-B (col.), 14-19. Litera- 
ture cited, p. 208-209. 

(3) JENNISON, Harry Milliken. 

192 1. BACILLUS ATROSEPTICUS VAN HALL, THE CAUSE OF THE BLACKLEG DIS- 
EASE OF IRISH POTATOES. (Abstract.) In Phytopathology, v. 11, 
no. 2, p. 104. 

(4) Morse, W. J. 

1910. CERTAIN DISEASES OF MAINE POTATOES AND THEIR RELATION TO THE 

SEED TRADE. Maine Agr. Exp. Sta. [Misc. Publ.] 375, 12 p. 



92 Journal of Agricultural Research voi. xxn. No. 2 

(5) Morse, W. J. 

19x7. STUDIES UPON THE BLACKLEG DISEASE OF THE POTATO, WITH SPECIAL 
REFERENCE TO THE RELATIONSHIP OF THE CAUSAL ORGANISMS. In 

Jour. Agr. Research, v. 8, no. 3, p. 79-126. Literature cited, p. 124-126. 

(6) Pratt, O. A. 

I916. A WESTERN FlELDROT OF THE IRISH POTATO TUBER CAUSED BY FUSARIUM 

radicicola. In Jour. Agr. Research, v. 6, no. 9, p. 297-310, pi. 34-37. 

(7) Smith, Erwin F. 

1920. AN INTRODUCTION TO BACTERIAL DISEASES OF PLANTS. XXX, 688 p., 453 

fig. [pi.] Philadelphia and London. Literature at end of most of the 
chapters. 



Blackleg Potato Tuber-Rot under Irrigation 





> 





Journal of Agricultural Research 



Vol. XXIi, N 



rivATE A 

Types of blackleg potato tuber-rot on pointed-end Netted Gem from Idaho, showing 
external as well as internal appearance. The photograph was taken two weeks after 
the tvibers were removed from the ground. The same tubers are shown in Plate 15, 
A, C, on the first day after digging. 



PLATE 13 



Forms of blackleg tuber-rot in the West. 

A-C. — Specimens received in August, 1918, from Fresno, Calif. 

D- — Specimen received in December, 1917, from Fallon, Nev. 



I 



Blackleg Potato Tuber-Rot under Irrigation 



Plate 12 




Journal of Agricultural Researcii 



Vol. XXII, No. 2 



Blackleg Potato Tuber-Rot under Irrigation 



Plate 13 




Journal of Agricultural Research 



Vol. XXII, No. 2 



PLATE 13 

Blackleg on stems resulting from planting the diseased western material. 

A. — Appearance of plant above the grotmd. 

B. — Spread of the infecion from the diseased seed piece to the new stem. 



1 



PLATE 14 

Result of inoculation of healthy tubers with the bacterial organism isolated from 
the western diseased material. 
A, B. — Netted Gem variety. 
C, D. — Irish Cobbler variety. 



Blackleg Potato Tuber-Rot under Irrigation 



Plate 14 




Journal of Agricultural Research 



Vol. XXII, No. 2 



Blackleg Potato Tuber-Rot under Irrigation 



Plate 15 




'^t 




*f 



\ \ 



Journal of Afrricultural Research 



Vol. XXII, No. 2 



PLATE IS 

Different types of blackleg tuber-rot. 

A-C— Characteristic appearance on fresh specimens of the Netted Gem variety 
from Idaho. 
D, E. — T3rpical development on round varieties in the East. 

p^ G.— Other forms occurring on round varieties in various sections of the coimtry. 
Arrows in A, B, and C indicate the border line of decay. 



PLATE i6 

Confusing forms of blackleg potato tuber-rot in the West (all specimens collected 
in Idaho). 

A, B. — ^Netted Gem variety one month after digging. Tuber B as it appeared on 
the first day after digging is shown in Plate 15, B. 

C. — Long Idaho Rural, showing secondary growth of Rhizoctonia on the outside 
and active bacterial decay in the inside. 

D. — Idaho Rural with deep side infection of blackleg decay which became dry 
and spongy on exposure to the sun. 

E. — Shallow surface infection which became dry and black when exposed to the 
sun after digging. 



Blackleg Potato Tuber-Rot under Irrigation 



Plate 16 




Journal of Agricultural Researcli 



Vol. XXII, No. 2 



MICROSCOPIC STUDY OF BACTERIA IN CHEESE 

By G. J. HucKBR 

Associate in Research, New York Agricultural Experiment Station 

INTRODUCTION 

Heretofore cultural methods have ordinarily been used in the study 
of cheese flora, but the usual routine technic has given only an inade- 
quate conception of the number and types of bacteria present. Although 
these cultural methods were employed in order to secure an idea of the 
number and varieties of organisms occurring in cheese, they have failed 
to establish the relative abundance of each type of microorganism in 
the cheese. This objection remains valid in the light of both quantita- 
tive and qualitative studies. 

The so-called "dilution technic," which involves the mass action of 
the organisms, has proved valuable in determining the type or group of 
organisms which predominates in a given sample ; but it does not furnish 
information regarding the general flora. This method often tends to 
give erroneous results, especially when the dilution medium used favors 
the growth of special groups. Under such conditions the results are 
influenced by the selective action of the medium. This has been true 
where milk was used as a dilution medium in examining cheese. The 
milk favored the growth of the lactic acid group; while the inert and 
nonlactose fermenting types or slow-growing cocci were overgrown, due 
in a large measure to the selective action of the medium. This procedure 
has well served its purpose in assisting to isolate the organisms for 
which it has a special adaptation. 

In general, cultural methods are preferable to a microscopic examina- 
tion because cultures can be isolated and studied independently — a 
feature which will always remain the outstanding advantage of these 
methods. 

Although subject to the same limitations as any microscopic method, 
the following method has been successfully used in this laboratory and 
has been employed in a routine way in determining the number of bac- 
teria in cheese. 

HISTORY 

Johan-Olsen {Sy, working with the molds which ripen "Gammelost" 
(a Norwegian cheese), mentions a sectioning method and implies that 
it resembles the usual histological technic but does not outline the 
procedure in detail. 

' Reference is made by number (italic) to "Literature cited," p loo. 

Journal of Agricultural Research, Vol. XXII, No. i 

Washington, D. C Oct. 8, 1921 



(93) 
54818°— 21 4 



Key No. N. Y. (Geneva) ,-6 



94 Journal of Agricultural Research voi. xxii. no. 2 



Troili- Peterson {11), in discussing tlie bacterial flora of Swedish 
"Giiterkase," mentions the microscopic examination of cheese as a 
control for the cultural procedure but does not give the technical details. 
She presents photomicrographs of cheese sections and states that some of 
the preparations were stained in methylene blue and that a few were 
examined unstained. 

Gorini (j), in studying the distribution of the bacteria in Grana cheese, 
presents the details of a method by which he prepared sections for 
microscopic examination. In his procedure he fixed and dehydrated 
samples of cheese by passing them through a series of alcohols of increasing 
concentrations until a strength of 95 per cent was reached. The usual 
histological methods of sectioning were followed, and the sections were 
stained in an aqueous solution of methylene blue. 

In the following year Rodella (9) reported a method used in his labo- 
ratory for preparing sections used in the direct examination of cheese 
samples. With his technic the samples were dehydrated and fixed by a 
method similar to that of Gorini and sectioned in the usual way. He 
found, however, that carbol-thionin gave better results as a stain than 
did methylene blue. 

Harrison (<5) outlined in detail a method for embedding and sectioning 
cheese which is similar to the common histological method, but like his 
predecessors he made no estimate of the number of bacteria present. 

During the year in which Rodella (9) presented his paper, Troili- 
Peterson {12) and Gorini (4) published notes discussing the question of 
priority raised by the practically simultaneous publication of their 
papers. It appears that the methods followed by Trioli- Peterson were 
similar to those of Gorini, but that she did not feel the necessity of pre- 
senting the technical details because of the universal knowledge of the 
common embedding methods. 

No results have been obtained in any of this work that permit a com- 
parison between counts made by the plate method, so commonly used 
in floral studies of cheese, and counts made by direct microscopic examina- 
tion. Following the method outlined below, comparatively accurate 
counts have been made by the direct method, and the number of the 
different types of bacteria have been determined as they actually exist 
in the cheese mass. 

TECHNIC 

EMBEDDING AND SECTIONING 

The samples of cheese were embedded by the usual histological technic 
and sectioned with a Minot rotary microtome. In sectioning, the micro- 
tome was so adjusted as to give sections 5 ^i thick. The sections were 
stained by the Gram method and with an aqueous solution of methylene 
blue. 



Oct. 8,1921 Microscopic Study of Bacteria in Cheese 95 

In order to determine the effect of the embedding process upon the 
cheese, small measured cubes of cheese were subjected to the routine 
procedures. Only a slight shrinkage was found, indicating that the 
volume of embedded cheese when examined is approximately the same 
as that of the fresh sample. 

MICROSCOPIC EXAMINATION 

The preparations were examined with an oil immersion lens and a 
high power ocular, the most satisfactory combination being a 1.9-mm. 
fluorite objective with a numerical aperture of 1.32. Where a thick 
coverslip was used it was necessary to have a 3-mm. apochromatic 
objective with a numerical aperture of 1.4. Greater depth can be secured 
with compensating oculars than with the ordinary Huygenian oculars. 

The method, although at first used only for determining the types of 
organisms present in the samples and as a check on the usual plate 
method, was found useful as a means of determining the number of 
organisms present. In order to make such a computation the micro- 
scope was so standardized as to allow an estimate of the number of 
organisms per gram when only a small amount of the original section 
was examined. This computation is similar to that used in the direct 
method of counting bacteria in milk described by Breed and Brew (2). 
This was accomplished by measuring both the diameter of the micro- 
scopic field and the thickness of the section from which the amount of 
cheese actually seen in each field examined was determined. Knowing 
the volume and specific gravity of the cheese examined, the total number 
of organisms per gram can readily be computed. With the diameter 
of the field measuring 0.14 mm. (140 fx), the microtome so adjusted as 
to cut sections of a thickness of 0.005 nim. (5 ij.) , and a specific gravity of i, 
the amount of cheese examined per microscopic field would be 1/13,000,000 
gm. — that is, each organism observed in a single microscopic field repre- 
sents 13,000,000 per gram. 

This factor may be computed by the following formula, in which any 
measure may be substituted : 

1,000 , . , 
— -2 — = factor per gram. 

In the above formula, 

r = the radius of the field examined in millimeters as determined by 
actual measurement. 

a = the thickness of the section in millimeters. 

b = the specific gravity of the cheese. 

The radius of the field, as has been stated, is determined by measure- 
ment with a stage micrometer and varies with the magnification and with 
the tyoe of ocular used. However, it was found advisable to adjust the 



96 Journal of Agricultural Research voi. xxii. no. , 



draw tube of the microscope so that the field would be of the greatest 
possible diameter without losing definition, as the greater the diameter 
of the field the less the increment of error in the total counts. 

The thickness of the section is controlled by adjusting the microtome 
to cut sections of a desired and known thickness. If all the adjustments 
on the microtome are firm and a sharp knife is used, sections can be cut 
of uniform thickness with surprising accuracy. The thickness of the 
sections can also be remeasured with the fine adjustment screw on the 
microscope. Although not perfect, this method of measurement serves 
as a check upon the accuracy of the sectioning. The measurement is 
accomplished by focusing with the graduated fine adjustment screw on 
both the upper and lower surfaces of the section and noting the differ- 
ences in the readings between the two levels. The difference can be 
read in microns where graduations are given on the fine adjustment 
screw. 

To convert the per-cubic-centimeter counts into numbers per gram, 
the specific gravity of the cheese must be considered. As the specific 
gravity of all samples has been assumed to be approximately i, the 
counts are interchangeable. This assumption in regard to the specific 
gravity is arbitrary, but the variations in the specific gravity of cheddar 
cheese are so slight that the total count is not affected to any appreciable 
degree. Accurate determinations did not seem practicable, as the speci- 
fic gravity varies with the fat content and with the moisture and general 
consistency of the cheese. 

With the measurements and adjustments used in this laboratory the 
per-gram formula resolves itself into the following: 

r = 0.07 mm. (70 fx). 

a = .005 mm. (5 /x). 

6 = 1.0. 

1 ,000 

7—- X I = approximately i-; ,000,000. 

3.1416 X 0.0049 X 0.005 

APPLICATION OF THE METHOD 

It is evident that this microscopic technic is subject to the limitations 
of any direct method of examination, many of which are unavoidable 
and are due to mechanical limitations or to the human error, which 
enters in when counts or estimates are made. 

QUALITATIVE EXAMINATION 

As previously stated, cheese has been examined microscopically by 
many investigators. The possibility of error is not as great when samples 
are examined to determine the types of organisms present as when 
total count is made, which is true of any microscopic work. Our present 
staining methods make possible a direct visualization of the microorgan- 



oct.8, igji Microscopic Study of Bacteria in Cheese 97 



isms together with their morphological and other general characteristics, 
but an attempt to enumerate these types involves other difificulties. 

The direct examination of cheese in the different ripening stages is 
advantageous and important, since the different groups of organisms 
can be studied as they actually occur in the cheese mass, and their 
groupings and relative relationships noted. The grouping may be 
especially important when considered in relation to the number present. 
For example, an organism may be present in large numbers during the 
early stages of ripening, but appear in scattered and isolated groups 
containing only a few individuals. In some instances only single bac- 
teria were found through the mass. In such cases the total number of 
this group by the plate count may be large, but the grouping, as deter- 
mined by direct examination, may demonstrate that they are not ac- 
tively growing and playing a part in the ripening of the cheese. On 
the other hand, the presence of large clumps of organisms, with the size 
of the clumps increasing during ripening, indicates that such groups are 
developing in the cheese mass and are probably playing an important 
r61e in the changes involved. 

That this grouping of the organisms actually occurs can be seen in 
Plate 17, A. In this photomicrograph are shown the types and groupings 
of organisms found in a very green cheese, showing that the Streptococcus 
lactis-like organisms predominated and were scattered in pairs over the 
field. Any migration of these bacteria through the cheese mass appears 
to have been impossible, and one is impressed with the fact that growth 
and reproduction could not have been taking place rapidly or the number 
of individuals per group would have been larger. In Plate 17, B, which 
represents a section from a cheese 5 months older than that shown in 
Plate 17, A, the organisms are found in larger clumps with many of them 
so massed that accurate counting is impossible. From the examination 
of a series of sections from cheeses of varying ages, it has been found that 
the clumps increase in size as the cheese ripens, reaching a limit after 
seven to eight months. It is evident that the organisms in the clumps, 
mostly cocci and a few rods, are thriving and reproducing and must, 
therefore, change the surrounding medium as they utilize it for food. 
It is not within the scope of this paper to discuss the significance of this 
occurrence but only to point out that such variations are found when 
samples are examined directly. 

QUANTITATIVE EXAMINATION 

An objection often made to counting organisms in microscopic prepa- 
rations of dried liquids is the uneven thickness of the resultant dried 
film. This objection is eliminated when parafiin sections are used, as 
such sections are uniform in thickness and the organisms remain in their 
natural relationships. Boekhout and DeVries (z) at one time endeavored 



98 Journal of Agricultural Research voi. xxii. no. 2 

to show that the scattered organisms in cheese sections were due, in a 
large measure, to the breaking up and scattering of the clumps by the knife 
edge. This explanation will hardly appear plausible to anyone familiar j 
with the perfection of delicate histological sections prepared with a sharp 
knife. 

The grouping and clumping of the organisms often cause difficulty in 
accurately determining the number of organisms in the cheese sections. 
This is especially true in sections of old cheeses in which the bacteria 
tend to clump in large masses. The error can be overcome to a large 
degree by counting or estimating a large number of fields, the larger the 
number examined the smaller being the error in the final estimate. 

In a sample of green cheese where the organisms appear in large num- 
bers, but are evenly scattered, it is impracticable to count the entire 
field, and an ocular disk divided into quadrants may be inserted in order to 
facilitate accurate counting. 

In all cases 20 or more fields should be counted, and especially where 
the organisms are unevenly distributed. In such instances, typical 
fields which represent the general flora should be located by studying the 
entire section. 

COMPARISON OF DIRECT AND PLATE COUNTS 

Table I gives a few representative comparisons between direct micro- 
scopic and plate counts made from cheese samples in various stages 
of ripening. The plate counts average approximately one-twelfth the 
direct count, but no common ratio has been found to exist between the 
results obtained by the two methods. Wide variations in the ratios be- 
tween the counts were found, but in general the ratios from green cheese 
appeared to be larger than those from cheese more advanced in ripening. 

The above plate counts compare well with those found by other observ- 
ers who have examined cheddar cheese, Russell (10) found from 62 to 
665 million per gram. Harrison and McConnell (7) found the count to be 
as high as 625 million per gram in the earlier stages of the ripening, while 
Harding and Prucha (5) observed from 37 to 177 million per gram. 

Several explanations may be offered to account for the apparent 
discrepancy between the results obtained by the two methods. The 
plate count is an estimate based on observations of the growth of organ- 
isms on some particular medium which, in cheese investigations, usually 
contains lactose. Lactose has been generally used because media con- 
taining this particular carbohydrate have been found to allow the develop- 
ment of a larger number of colonies than do sugar-free media. Investigators 
have based their cultural methods upon media giving the largest counts 
rather than upon media which might serve as an index to the relative 
number of types present. In comparing the microscopic counts with 
results obtained with the plate method, it may be noted that the types 



Oct. 8, 1921 



Microscopic Study of Bacteria in Cheese 



99 



present in the cheese, as seen by direct examination, are not present 
in the same proportions on the plates, because those types which grow 
abundantly in the presence of lactose have outnumbered all groups 
which do not grow as readily on such media. 

Table I. — Relation between microscopic and plate counts obtained from cheese 





Approximate age 
of cheese. 


Miscroscopic count (millions per gram). 


Plate coimt (millions 
per gram). 


Cheese. 


Cocci. 


Short 
rods. 


Yeast. 


Strep- 
tococ- 
cus 
lactis. 
(Lister) 


Total. 


Strep- 
tococ- 
cus 
lactis. 


Miscel- 
laneous. 


Total. 


OOCIIXII 2.... 




100 
14,650 
312 
1,690 
962 
338 


600 
S, 200 

780 
260 
26 


26" 


4.500 
21,450 
143 
6,760 
988 
3.029 


5,200 

41,300 

481 

9.230 

2,210 

3.393 


46 
264 
132 
251 
306 


9 

284 
90 
41 

121 


55 
548 


OOCI3 XII 2 . . . 




6. 4,11 




BCII 




292 

427 
702 


6. 26, II 


5 months, 23 days. 

Unknown; a p - 

peared giecn. 


OO38I8 









Results of plate counting may also be lower because of insufficient 
grinding and emulsifying of the cheese sample previous to plating. This 
appears to be especially significant in cases where investigators grind the 
sample with sterile quartz or sugar and suspend the ground mass in 
sterile water in preparation for plating. If the sample is not well ground, 
small particles of cheese remain in the emulsion, and the individual bacteria 
are not separated so as to allow them to grow into separate colonies on 
the artificial medium. Emulsions examined under the microscope often 
show comparatively large masses of cheese which have not been affected 
by the grinding process. 

CONCLUSION 

The microscopic examination of cheese embedded and sectioned by 
the usual histological method is a valuable and satisfactory method for 
studying the different stages of cheese ripening. Such a direct method 
of examination may be used to determine the number of organisms 
present in the sample. It also serves as an index to the types of organisms 
present and makes possible a study of the organisms as they actually 
exist in the cheese mass, allowing observations on the groupings and 
relationships during cheese ripening. 

The cultural methods do not yield as high a count as the microscopic 
method, due primarily to the selective action of the medium used and the 
difficulty of liberating the organisms from the cheese mass previous to 
plating. 

A combination of microscopic and cultural studies yields a far more 
complete picture of what takes place in cheese ripening than can be 
obtained by the use of either method alone. 



loo Journal of Agricultural Research voi. xxii, no. a 

LITERATURE CITED 
(i) BoEKHODT, F. W. J., and Ott de VriBS, J. J. 

1899. UNTERSUCHUNGEN t)BER DEN RElFUNGSPROZESS DES EDAMER KASES. 

In Centbl. Bakt. [etc.], Abt. 2, Bd. 5, No. 9, p. 304-307. 

(2) Breed, Robert S., and Brew, James D. 

I916. COUNTING bacteria BY MEANS OP THE MICROSCOPE. N. Y. State AgT. 

Exp. Sta. Tech. Bui. 49, 31 p., 5 fig., i col. pi. Bibliographical 
footnotes. 

(3) Gorini, Costantino. 

1904. SULLA DISTRIBUZIONE DEI BACTERl NEL FORMAGGIO DE GRANA. In 

R. 1st. Lombardo Sci. Let. Rend, ser, 2, v. 37, fasc. 2, p. 74-78, 
2 fig. (on I pi.). 

(4) 

1906. ZUR PRIORITAT DER METHODE DER KASEUNTERSUCHUNG DURCH MIKRO- 

SKOPISCHE SCHNITTPRAPARATE. In Centbl. Bakt. [etc.], Abt. 2, Bd. 
16, No. 1/3, p. 66. 
(,5) Harding, H. A., and Prucha, M. J. 

1908. THE BACTERIAL FLORA OF CHEDDAR CHEESE. N. Y. State Agf. Exp. 

sta. Tech. Bui. 8, p. 120-193. 

(6) Harrison, F. C. 

1906. the distribution of lactic acid bacteria in curd and cheese of 
THE CHEDDAR TYPE. In Rev. G6n. Lait, v. 5, no. 18, p. 409-415, 9 
fig. (on 4 pi. ) References, p. 413. 

(7) and CoNNELL, W. T. 

1903. A COMPARISON OP THE BACTERIAL CONTENT OF CHEESE CURED AT DIF- 
FERENT TEMPERATURES. In Rev. Gen. Lait, v. 3, no. 4, p. 80-85; 
no. 5, p. 103-111; no. 6, p. 126-137, 1903 • "O- 7> P- i50~i55; iio- 8, 
p. 173-180, 1904. 

(8) Johan-OlsEn, Olav. 

1898. DIE BEi kasEREifung wirksamen pilze. In Centbl. Bakt. [etc.], 
Abt. 2, Bd. 4, No. 5, p. 161-169, 17 fig. (on pi. 4-9, 4 and 5 col.). 

(9) RODELLA, A. 

1905. EINIGES t)BER DIE BEDEUTUNG DER DIREKTEN MIKROSKOPISCHEN PRA- 

PARATE FtJR DAS STUDIUM DES KASEREIFUNGSPROZESSES. In Centbl. 

Bakt. [etc.], Abt. 2, Bd. 15, No. 4/5, p. 143-153- 5 ^g- (o" i pl-)- 

(10) Russell, H. L. 

1896. THE RISE AND FALL OF BACTERIA IN CHEDDAR CHEESE. In Wis. Agr. 

Exp. Sta. 13th Ann. Rpt. i^g^jgb, p. 95-111, fig. 25-26 (26 on col. 
fold, pi.) 

(11) Troili-Peterson, Gerda. 

1903. STUDIEN t;BER DIE MIKROORGANISMEN DES SCHWEDISCHEN GCTERKASES. 

In Centbl. Bakt. [etc.] Abt. 2, Bd. 11, No. 4/5, p. 120-143; No. 6/7, 
p. 207-2151, 5 fig. (on 3 pi.). 

(12) 

1905. BEMERKUNGEN ZUR DER ARBEIT VON A. RODELLA " EINIGES t)BER DIE 
BEDEUTUNG DER DIREKTEN MIKROSKOPISCHEN PRAPARATE FCR DAS 

STUDIUM DES KASEREIFUNGSPROZESSES." In Centbl. Bakt. [etc.], 
Abt. 2, Bd. 15, No. 13/14, p. 430. 



PLATE 17. 

A. — Section of cheddar cheese i month old, stained with an aqueous solution of 
methylene blue, showing isolated pairs of Streptococcus lactis Lister throughout the 
field. X 500. 

B. — Section of cheddar cheese 6 months old, stained as in A. X 800. 



Microscopic Study of Bacteria in Clieese 



Plate 17 




Journal of Agricultural Research 



Vol. XXII, No. 2 



FURTHER STUDIES ON RELATION OF SULPHATES TO 
PLANT GROWTH AND COMPOSITION 

By Harry G. M1L1.ER 
Chemistry Department, Oregon Agricultural Experiment Station 

This is a continuation of an investigation, part of the results of which 
were reported in a former pubHcation {4}} As stated in the earher paper, 
the addition of the different forms of sulphur caused a marked increase 
in the dry weight of red clover, and beneficial results were obtained with 
oats and rape. One very noticeable result observed in the former work 
was the high nitrogen content of the clover grown on soils in the green- 
house receiving sulphur fertilizer, compared to that of clover receiving 
only the residual sulphur of the soil. 

This great increase in nitrogen assimilation by the clover where sul- 
phates were applied, and under the conditions described, led the writer 
to believe that the sulphates favorably influenced the activity of the 
legume bacteria. Especially did this appear true where the beaverdam 
soil was used. This soil contained o. 18 per cent sulphur with appreciable 
quantities of sulphate sulphur in the soil extract, and no beneficial result 
from sulphur fertilizer was expected. Oats did not respond to sulphur 
with this soil, although the sulphur content of oats (j) and the amount of 
sulphur removed by one crop of oats is as large as with a red clover crop. 
From present data, the responses of red clover so often obtained with 
gypsum compared to cereals can not be explained through a difference 
in sulphur requirement. With alfalfa the amount of sulphur removed is 
so large compared to the cereals and red clover that the addition of 
sulphates would apparently function directly as a plant food where in- 
creased growth results. An example of the latter would be the enormous 
increases in the yield of alfalfa obtained in southern Oregon (7) where 
sulphur fertilizers were applied to soils with a very low sulphur content. 
These authors, however, mention the favorable action that sulphur ferti- 
lizers had on the root development and nodule production of alfalfa. 
Duley (2) reports increase nodule production on red clover where sul- 
phiu- was added to soils. Pitz (5) observed increased nodule production 
and root development with red clover by applying gypsum to soil cultures. 

As far as the writer can ascertain, no correlation has been shown be- 
tween nodule production and nitrogen content of the plant, by influencing 
the development of the former, with ordinary sulphur fertilizer com- 
pounds. In this paper a study has been made of the effect of different 

' Reference is made by number (italic) to "Literature cited," p. no. 

Journal of Agricultural Research. Vol. XXII, No. 

Washington, D. C. Oct. 8, 1921 

zt Key Na Greg. 7 

(lOI ) 



I02 



Journal of Agricultural Research 



Vol. XXII, No. 2 



concentration of sulphates on growth and nitrogen assimilation, and also 
the relation of total sulphur content of the plant as influenced by available 
nitrogen. The red clover and rape were used in this work. With clover, 
the initial concentration of legume bacteria has been varied by inocu- 
lating certain cultures, while others were not inoculated. 

Table I. — Analytical results with red clover on Medford loam soil 



Treatment. 



Weight 
of air- 
dried 
clover. 





N 








Total 

N. 


in- 
solu- 
ble in 
acet- 
ic- 
acid. 


Total 
S. 


Sul- 
phate 
S. 


Or- 
ganic 
S. 


P.cl. 
2. 88 


P.ct. 

2- 13 


P.ct. 
0. 20 


P.ct. 
0.06 


P.ct. 

0. 14 


^■l'^ 


2 


10 




20 


.04 




It) 


2.8l 


2 


02 




18 


.07 




II 


3-31 


2 


10 




16 


. 02 




14 


3- 30 
3-36 


2 
2 


15 
22 




26 

28 


. 10 

. 12 




16 
16 


3-58 
3-48 


2 

2 


31 
19 




40 

34 


•25 

• 14 




IS 

20 


3-46 
3-38 


2 
2 


39 

30 




28 
34 


• 13 

•IS 




15 
19 


3-34 


2 
2 


28 

33 




41 


.27 

•IS 




14 
18 



Sin- 
S in soluble 
acetic- in 
acid acetic- 
solu- acid 
tion. I solu- 
tion. 



Weight 
of air- 
dried 
roots. 



Total 
S in 
roots. 



Total 
Nin 
roots. 



Ratio 
of 

tops 

to 

roots. 



2. 64 

2. 77 



2. 74 
2- 73 



Control: 

Uninoculated . . 

Inoculated 

NaNOs: 

Uninoculated . . 

Inoculated 

NaNOa and HaSOt: 

Uninoculated . . 

Inoculated 

Na!S04: 

Uninoculated . . 

Inoculated 

CaSOi: 

Uninoculated . . 

Inoculated 

NaNOs and CaSOi: 

Uninoculated . . 

Inoculated 



Gm. 

15-84 

31. 24 

17. 70 
21- 29 

18.82 

23-35 

15. 00 
20. 31 

16.49 

16. 6s 

14. 18 
16.89 



.08 
.07 



• 15 
-17 



09 



Gm. 

6. 00 

7- 6s 

6.45 
7.80 

7. 60 
7. 10 

8.60 
8.20 



6. 10 
6. 20 



P.ct. 



34 


2. 


28 


I. 


23 


I. 


60 


I 


72 


I 


59 


I 


53 


3 


50 
61 


2 
2 


•55 


I 


-55 


2 



2. 39 

3. 92 



In the first experiment Medford loam soil, designated as soil B in the 
previous publication (4), was used. This soil was heated in an electric 
oven where the temperature was gradually raised to 120° C. and main- 
tained for six hours. This was to destroy the legume bacteria present 
in the soil. Four kgm. of soil, after being mixed with the different fer- 
tilizers, were placed in paraffined clay pots and carefully seeded to red 
clover. One series was inoculated with Bacillus radicicola Bey. The solu- 
tion for inoculation was prepared by removing the growth of organisms 
from a culture and mixing with water. Each inoculated soil culture re- 
ceived a definite number of cubic centimeters of the bacterial solution, 
concentrated where the seeds were placed. The amount of different fer. 
tilizers added per pot was as follows : Sodium sulphate (NajSO^) , 3 gm. ; 
sodium nitrate (NaNOj), 2 gm. ; calcium sulphate (CaS04.2 H3O), 3.75 
gm. ; and calcium carbonate (CaCOg) , 3 gm. The cultures were placed 
in the greenhouse on October 16 and harvested on March 18. Ten plants 
were allowed to grow in each pot. The cool temperature in the green- 
house did not permit rapid growth, and the plants were cut before ma- 
turity was reached. The treatment and analytical results are given in 
Table I. The cultures were weighed every other day, and the moisture 
was maintained at 20 per cent. In removing the roots the soil was 
shaken out of the pot and carefully loosened. The roots were then sepa- 
rated out and washed. After drying they were weighed, and the non- 



Oct. 8,1921 Relation of Sulphates to Plant Growth and Composition 103 

volatile matter was determined by ashing a ground sample representing 
each cultm-e. This was done to correct for any excess weight due to 
adhering soil particles. 

The total sulphur was determined by the sodium-peroxid method. 
The sulphate sulphur was extracted by taking 2 gm. clover and 150 cc. 
of water and digesting on the steam bath for three hours. It was then 
slightly acidified with hydrochloric acid, and after standing for an hour 
the extract was filtered. Five cc. of 10 per cent barium-chlorid solution 
was used to precipitate the sulphate sulphur in the hot solution. After 
standing overnight the barium sulphate settled to the bottom of the 
beaker in all cases, and no particles of precipitate could be detected in 
the supernatant liquid. This liquid was carefully decanted off; and the 
white precipitate was washed on a Gooch crucible, dried, and weighed. 
Several of these precipitates were ignited, but since no appreciable loss 
in weight was detected this method appeared perfectly reliable for 
comparable results on sulphate sulphur in the different samples of plant 
material. For the determination of total nitrogen insoluble in acetic 
acid, i-gm. samples were digested with about 150 cc. of water on the 
water bath for two hours. The extract was then acidified with dilute 
acetic acid and filtered after standing about 30 minutes. The total 
nitrogen was determined on the precipitate by the Kjeldahl method. 
The filtrate was made alkaline with sodium carbonate, evaporated to a 
few cubic centimeters in volume, transferred to a nickel crucible, and 
total sulphur determined by the sodium-peroxid method. 

The dry weights of plant material produced show no increase in pro- 
duction that can be attributed to presence of sulphates. This result is 
different compared to the noticeable increase reported with the same 
soil before (4). As mentioned above, conditions were very unsatisfactory 
for growth, and the plants were cut before maturity five months after 
planting. In the former work reported, conditions permitted rapid 
growth, and the plants, though not mature, were harvested two months 
after seeding. As the soil used in this later work had been heated, there 
was perhaps some change in degree of solubility of soil minerals and in 
the biological flora. The concentration of added mineral salts was also 
greater in this work. 

Examination of the roots showed that all plants had become infected. 
In the uninoculated series, roots from cultures i and 2 contained very 
few nodules compared to the roots grown in the soil receiving sulphate 
fertilizer. This remarkable difference in nodule formation no doubt 
accounts for the low nitrogen content of the clover plants in pots i 
and 2. That these plants became infected without any artificial inocu- 
lation is not surprising. Wilson (8) found that — 

of fifteen legumes grown in Volusia silt loam soil, only one, Trifolium pratense, 
developed nodules without artificial inoculation. 



I04 Journal of Agricultural Research voi. xxii. No. a 

During the growth of the plants the inoculated series showed greater 
development, which is apparent upon examining the dry weights. 
Cultures i and 2 of the inoculated series show no effects from lack of 
sulphates, and all the sulphur-fertilized pots contained numerous well- 
developed nodules. 

The total nitrogen insoluble in acetic acid was no greater in some 
of the cultures receiving added sulphates than in the controls, so no 
statement can be made that sulphate addition caused this fraction to 
become larger. The nitrogen content of the fraction soluble in acetic 
acid is lower in i and 2 of the uninoculated series. The increase in 
percentage of sulphur caused by fertilizer treatment is accounted for 
generally by higher sulphate content. Although the organic sulphur 
is apparently higher in some, the results are not consistent with the total 
sulphur, to state that the former results from increased sulphur assimi- 
lation in this experiment. Total sulphur in the acetic-acid extracts 
runs parallel with sulphate sulphur results and is slightly higher, showing 
that there is some sulphur in the organic form not accounted for in the 
precipitate from acetic-acid solution. This was also found to be true 
with clover grown in other pots which was harvested while in blossom. 

The ratio in weight of tops to roots is greater in the inoculated sulphur- 
fertilized cultures than in the uninoculated sulphur-fertilized cultures. 
This difference does not appear to be due to inoculation alone, for this 
does not hold true in comparing i and 2 of both series, while the ratio of 
tops to roots in i and 2 of the uninoculated series is greater than the 
remaining four where there is heavy nodule growth. Amy and 
Thatcher (/) report a greater ratio in weight of tops to roots where 
inoculation was made with alfalfa and sweet clover. 

The sulphur content of the roots is larger than in the other portion 
of the plant, whereas the opposite is true in percentage of nitrogen. 

The second part of this work was carried on with beaverdam soil and 
red clover. Each pot contained 7 kgm. of soil, and the following amount 
of fertilizers were added as indicated in Table II: 12 gm. of calcium 
sulphate, 10 gm. of sodium sulphate, 2 gm. of sulphur, and 6 gm. of 
sodium nitrate. Two gm. of potassium chlorid and 10 gm. of calcium 
carbonate were added to all the soil cultures. Twenty red clover plants 
were allowed to grow in each pot, and the moisture content was kept at 
40 per cent. The first crop grew at the same time as the clover on the 
Medford loam soil and was also cut before the blossoming stage. Three 
other crops were grown on these same cultures. The first was harvested 
on March 24, the second on May 20, the third on July 9, and the fourth 
on August 17. The last three crops were cut during the blossoming stage. 
The results are given in Table II. 



Oct. 8,1921 Relation of Sulphates to Plant Growth and Composition 105 



TablS II. — Analytical results with red clover on beaverdam soil 



Treatmeat. 



Control 

CaSOi and NaNOs. 

CaS04 

NaN03 

NasSOi 

Na2S04 and NaNOs 

S 

SandNaNOs 



Crop I. 



Weight 
of air- 
dried 
clover. 



Gm. 
46 



46 



Total 

N. 



P. a. 

3-28 
3- SO 
3-41 
3-25 
3-47 
3- 12 
3-42 
3-58 



Total 
S. 



P.ct. 



Crop II. 



"Weight 
of air- 
dried 

clover. 



Gm. 



102 
91 

80 
83 



Total 

N. 



P.ct. 



2. 70 
2. so 
2. 40 
2-53 
2.49 
2.47 
2. 56 



Total 
S. 



P.ct. 



Cropni. 



Weight 
of air- 
dried, 
clover. 



G.m 

69 
66 



Total 

N. 



Total 

S. 



Sul- 
phate 
S. 



P.ct. 

o. 02 



Or- 
ganic 
S. 



P.ct. 

O. 12 



•15 
. 12 
•13 



Treatment. 



Control 

CaS04and NaNOs.. 

CaS04 

NaNOs 

Na2S04 

Na2S04 and NaNOs. 
S. 



S and NaNOs. 



Crop IV. 



Weight 
of air- 
dried 
clover. 



Gm. 



Total N. 



Per cent. 



3- 16 



Total S. 



Per cent. 
18 



Sulphate 
S. 



Per cent. 
03 
15 
12 
06 



Organic 
S. 



Per cent. 
IS 
IS 
IS 
16 

.16 
16 
14 

.16 



Bxammation of the foregoing data shows no result from sulphate 
application in the first two crops. There is no increase in dry weight 
in the pots receiving sulphur fertilizers, and in some crops the yield on 
the control is greater. In the third and fom'th crops, one distinctive 
difiference appears in the lower nitrogen content of the clover grown on 
the control soil cultures. The fact that the clover grown under condi- 
tions supplying more sulphate sulphur has a higher nitrogen content 
adds to the data already obtained pertaining to the favorable influence 
that sulphates have upon legume bacteria, the action of which results 
in a higher nitrogen content of the clover. According to experimental 
evidence, sulphates do not increase nodule production on all legumes. 
Wilson (8) reports that certain sulphates depressed nodule formation 
on the soybean. On the other hand, Prucha (6) mentions magnesium 
sulphate and calcium sulphate as exerting a beneficial influence on nodule 
development of the Canada field pea. 

The low nitrogen content did not appear to be due to the absence of 
sulphates in the plant tissue, thus limiting protein synthesis, as sulphates 
were present in all samples. However, the percentage of sulphate 
sulphur was lower in clover grown on the control pots. 

That the presence of available nitrogen or nitrogen assimilation by 
the plant tends to control or limit the total sulphur assimilation is illus- 
trated by data in Table III, 



io6 



Journal of Agricultural Research 



Vol. XXII. No. a 



In this experiment sea sand was used which had been washed with 
dilute hydrochloric acid and large volumes of distilled water. The sand 
still contained sulphur compounds, but no sulphates soluble in dilute 
hydrochloric acid. Six kgm. of sand were mixed with lo gm. of calcium 
carbonate and placed in paraffined clay pots. Where elemental sulphur 
was used, 0.75 gm. was added at the same time. The other nutrients 
added were applied in solution form through a period of 70 days; the 
growing period was 80 days. The total amount of sodium sulphate 
which was added varied in the different cultures. Concentration i 
denotes 1.55 gm., concentration 2, 3.10 gm., etc. The same figures apply 
to calcium sulphate too. All cultures, with the exception of No. 3, 
13, 20, and 21, received 3.9 gm. of sodium nitrate, and they each re- 
ceived 0.75 gm. Every culture received 2.6 gm. of potassium dihy- 
drogen phosphate and 1.3 gm. of magnesium chlorid. Twenty plants 
grew in each pot, and the moisture content varied fron 15 to 20 per cent 
in the different cultures. The weights recorded are the average of 
duplicates, and the analyses were made on a sample obtained by mixing 
the duplicates. The 22 cultures from 11 to 21, inclusive, were inocu- 
lated with legume bacteria and 5 gm. of beaverdam soil added to the 
same culture. 

Table III. — Analytical results of clover grown on sand receiving a nutrient solution 



Uninoculated series. 



Cul- 
ture 
No. 



Form of sulphur 
added. 



NasS04 




....do 


3 


....do 


Na2S04 and CaSO^ . . 
NajSOi 


2 


Control 




S 




CaS04 




....do 




....do 




NajSOi and CaSO* . . 


2 



Sul- 
phate 
con- 
cen- 
tra- 
tion. 



Weight 
of air- 
dry 
clover. 



Gm. 
9.65 
10. 7 

3-95 

8.7s 
6.9 
4- IS 
6.62 
7. 02 
S-I 



Total 
S. 



Per ct. 
o. 26 
.42 
•30 
•30 
•34 



Total 

N. 



Per ct. 
3-76 
3-77 
2. 00 
3- 50 
3-69 
3-90 
3-75 
3-8s 
3- 80 



Inoculated series. 



Cul- 
ture 
No. 



013 
14 
tS 
j6 
J7 
18 
19 
a 20 
a ji 



Weight 
of air- 
dry 
clover. 



Gm. 

10. 00 
7. 00 
4.80 
6.4s 
6.8s 
7-3 
7- 77 
7-3° 
5.90 
7-5° 
S-07 



Total 
S. 



Perct. 

o. 27 
30 



Total 

N. 



Sul- 
phate 
S. 



Perct. 
3.62 
3-71 
3.49 
3- 70 
3.60 
3-90 
3- 69 
3- 70 
3- 90 

2. 42 
2.46 



. 12 
.08 
.18 

■ I."; 
.08 
• 17 
.18 



.09 
.14 



Or- 
ganic 
S. 



Perct 
IS 



" Low nitrate. 



In the foregoing data the low sulphur contents occur in the clover 
grown in the pots receiving less nitrate nitrogen. In the inoculated series 
the sulphur content of the clover does not appear to increase by increasing 
the sulphate sulphur of the nutrient media. On the other hand, where 
there is a reduction in nitrate added, there is an appreciable reduction in 
the sulphur content of the clover. In the other series, where the concen- 
tration of legume organisms was not as great at the start, the percent- 
ages of sulphur generally run higher. In No. 3 the percentage of 



Oct. 8,1921 Relation of Sulphates to Plant Growth and Composition 107 

sulphur is not lower than in some of the others, but here the total yield 
is small, and this often accounts for higher percentages of certain 
elements. The jdelds in No. 3 and 6 are about the same; but the 
sulphur content is higher in 6, although this culture depended only 
upon the sulphur in the sand. No. 2 shows response in sulphur con- 
tent to the higher concentration of sulphates in the media. The cor- 
responding culture 12 in the other series does not show higher sulphur 
content; and as the average sulphur content is lower in this uninoculated 
series, it appears that the legume organisms might have some effect on 
limiting the quantity of sulphur present in the clover hay. 

Table IV. — Data showing the sulphur-nitrogen relation in the portion insoluble in 

dilute acetic acid 



CaS04 

and 
NaNOs. 



In soil treated with- 



NaNOa. 



Na2S04. 



NajSOi 

and 
NaNOs. 



Average. 



Ratio 
of Nto 
S in the 
insolu- 
ble por- 
tion. 



Aver- 
age 
ash 
con- 
tent. 



Crop I : 

Percentage of N insoluble in 

acetic acid 

Percentage of S soluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 

Crop II: 

Percentage of N insoluble in 

acetic acid 

Percentage of S soluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 

Crop III: 

Percentage of N insoluble in 

acetic acid 

Percentage of S soluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 

Crop IV: 

Percentage of N insoluble in 

acetic acid 

Percentage of S soluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 

Inoculated series, Medford loam: 
Percentage of N insoluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 

Uninoculated series, Medford loam : 
Percentage of N insoluble in 

acetic acid 

Percentage of S insoluble in 

acetic acid 



Per a. 
2. 26 

.08 
•13 

I. 70 
. II 
.09 

1.78 

•13 

.09 

2.05 
.19 
. II 



Per ct. 

2. 24 

.07 
. 12 

1-57 
.06 
.08 

I. 76 

•05 
.09 

1.97 
.09 
•13 



Per ct. 

2.58 

.09 



I. 64 
.07 
.09 

1.87 



I. 



.16 
. II 



Per ct. 
2.15 

.07 
•15 

1-73 
. II 
.09 

1. 85 
. II 
.09 

2. 12 
.18 
. II 



Per ct. 
2.31 



• 13- 
1.66 



m-1 



1. 81 



>i9. 4 



•93 
2.03 



•"5 

2. 20 
■125 

I. 21 
. II 



ni-s 



► 17. 6 



Per ct. 



9.94 



7-31 



6-95 



54818°— 21- 



io8 



Journal of Agricultural Research 



Vol. XXII, No. 2 



To say that percentage of sulphur will not increase regardless of sul- 
phate concentration in the nutrient media without increasing the avail- 
able nitrogen would not be in accordance with data already obtained. 
It does appear, though, that when the lack of nitrogen is sufficient to 
lower the nitrogen content compared to the normal nitrogen content of 
the clover there is a tendency toward decreased sulphur assimilation. 
It is interesting to compare the rape plant with the clover in this respect. 

The figures given in Table IV show that the clover cut before the blos- 
soming stage not only contains a higher percentage of total nitrogen but 
also a higher percentage of nitrogen insoluble in acetic acid. As the per- 
centage of nitrogen removed by this fraction varies, so also does the per- 
centage of sulphur. There appears to be a definite relationship between 
the sulphur and nitrogen content in this insoluble portion, thus adding 
more significance to this fraction in regard to quality and perhaps rep- 
resenting the true protein of the clover hay. No difference in ash con- 
tent caused by variation in fertilizer treatment was observed in the 
diflerent pots. The ash content of the different crops did vary however, 
as is sho\vn in Table IV. 



EXPERIMENTAL WORK WITH THE RAPE PLANT 

The Medford loam soil used in the first part of this work was used in 
this experiment. After the clover roots were removed the soil was re- 
turned to the pots and seeded to rape. Three gm., of sodium nitrate 
were added to those cultures which had received nitrate nitrogen in 
the clover experiment. The plants were harvested after a growing period 
of 50 days. At the end of this time there had been a cessation of growth, 
and the basal leaves dried up and fell off. The results appear in Table V. 

Table V. — Analytical results obtained with rape 



Treatment. 



Control , 

NaNOg 

NaNOg and Na, SO4. . 

Na2 SO4 

NaNOj and CaSO^ 
CaS04 



Number 

of 
plants. 


Weight. 


Total N. 


N insol- 
uble in 
acetic 
ac'd. 


Total S. 


Sulphate Org 
S. S 


anlc 
5. 




Gm. 


Per cent. 


Per cent. 


Per cent. 


Per cent. Per 


cent. 


"7 


12. 


1-93 


0. 90 


0.35 


0. 16 


19 


7 


S-o 


2.88 


I 


02 


•39 


. 12 


27 


7 


12 2 


2-35 




93 


. 22 


•03 


19 


07 


10. I 


4. 14 




95 


. 16 


. 02 


14 


S 


"•3 


3-36 




71 


I. 46 


I. 09 


37 


06 


14. 


2. 72 




88 


1.38 


•97 


41 


7 


2.4 


2.57 




91 


3-^Z 


2. 76 


37 


a 8 


30 

14-45 


2-59 
2.30 






2,41 
I. 26 


2.05 
■83 


3b 
43 


6 




83 


6 


16.35 


2.27 




66 


1-31 


•93 


38 


07 


2.4 


2. 69 




88 


3.28 


2.98 


30 


8 


1.8 


2.66 




2. 70 


2-37 


33 







Ash. 



Per 



cent. 
13.6 
19. 

IS- 



3 
9 
o 

18.3 
17-7 
22. o 

19.4 
14.7 

16. 2 
31. 9 
21. 9 



« Pots from the inoculated series. 



The presence of added sulphates appeared to have retarded growth of 
the rape, as there is a greater dry-weight yield on the controls. With 



oct.8,i92i Relation of Sulphates to Plant Growth and Composition 109 

the cultures receiving nitrate fertilizer the addition of sulphate sulphur 
apparently caused increased growth or had just the opposite efifect. 
If the concentration of the sulphates was great enough to produce a 
toxic effect, the sodium nitrate may have counteracted this action. 

The wjiter has observed just the opposite effect with clover seedlings 
growing on agar agar-mineral salt nutritive media. One gm. of sodium 
nitrate per liter had a noticeably toxic effect, while the same concentra- 
tion of sodium sulphate produced no noticeably injurious effect. In the 
cultures containing both the same concentration of sodium nitrate and 
sodium sulphate there was an improvement in growth over the former 
sodium-nitrate cultures. 

Application of nitrates produced very good 5delds on a comparative 
low sulphur assimilation by the plants. The question naturally arises 
whether the rape does not absorb sulphur, if present, far in excess of that 
required for carrying on the synthesis of its organic compounds. This 
appears so noticeable in comparing the figures in Table V. Of course 
it is realized that the optimum concentration of nutrients for plant 
nutrition has always been a problem. The acetic acid-insoluble nitrogen 
is higher in the rape grown on the soil receiving nitrate fertilizer only, 
compared with that in the rape which received both nitrate and sulphate 
fertilizer. There seems to be a tendency of the sulphates to decrease 
this form of nitrogen. Sulphate application increased the organic 
sulphur and total sulphur content of the rape, while at the same time the 
presence of these sulphate compounds retarded growth where no nitrates 
were added. The extremely high sulphate content is very obvious in 
these samples of rape. This may account for the high ash content. The 
percentage of ash in the samples of rape varies considerably, depending 
upon the fertilizer treatment and magnitude of growth. Such a variation 
did not occur with the clover. 

SUMMARY 

Sodium sulphate and calcium sulphate had a beneficial effect on 
nodule development and nitrogen assimilation of the red clover grown 
on previously sterilized soil. On a similar series which was artificially 
inoculated with Bacillus radicicola at the time of seeding, sulphates 
caused no increase in nodule development. 

When a soil of high sulphur content was used, the nitrogen content in 
clover of the third and fourth crops was lower on the control pots than 
where either sulphur, calcium sulphate, or sodium sulphate was applied. 
As sulphate sulphur was present in all plants, the low nitrogen content 
could not be explained by a cessation in protein synthesis due to the 
absence of sulphates. 

This again shows the relation of sulphates to nitrogen assimilation and 
the favorable influence of sulphates on the legume bacteria or on some 
other agency controlling nitrogen assimilation. 



no Journal of Agricultural Research voi. xxii, no. 2 

The ratio of nitrogen to sulphur in the portion of the clover plant 
insoluble in dilute acetic acid remains about the same, regardless of the 
stage in the development of the plant. This gives further support to the 
view that the nitrogen insoluble in acetic acid represents protein nitrogen. 
The total nitrogen and total nitrogen insoluble in acetic acid wias higher 
in those plants cut before the blossoming stage. 

With clover growing on sand cultures, it was possible, by reducing the 
available nitrate, not only to limit the growth and nitrogen content but 
also to decrease the sulphur assimilation. So, while sulphates appar- 
ently cause greater nitrogen assimilation through their beneficial effect 
on nodule development, the amount of sulphur taken up by the plant is 
limited by the total nitrogen absorbed. 

The rape plant assimilated a large amount of sulphur, although the 
presence of sulphates reduced the yield compared to the control soil 
cultures. Sulphate plus nitrate caused increased yields compared with 
those secured when nitrate was added alone. There does not appear 
to be any direct relation between nitrogen and sulphur assimilation in 
the rape plant. 

LITERATURE CITED 
(i) Arny, a. C, and Thatcher, R. W. 

1915-17. THE EFFECT OF DH'FERENT METHODS OP INOCULATION ON THE YIELD 
AND PROTEIN CONTENT OP ALFALFA AND SWEET CLOVER. In JOUT. Amer. 

Soc. Agron., v. 7, no. 4, p. 172-185, 1915; v. 9, no. 3, p. 127-137. 
1917. 

(2) DULEY, F. L. 

1916. THE RELATION OF SULFUR TO SOIL PRODUCTIVITY, /n JoUT. Amer. Soc. 

Agron., V. 8, no. 3, p. 154-160. 

(3) Hart, E. B., and Peterson, W. H. 

191 1. SULFUR REQUIREMENTS OF FARM CROPS IN RELATION TO THE SOIL 

AND AIR SUPPLY. Wis. Agr. Exp. Sta. Research Bui. 14, 21 p. 

(4) Miller, H. G. 

I9I9. RELATION OF SULFATES TO PLANT GROWTH AND COMPOSITION. In Jour. 

Agr. Research, v. 17, no. 3, p. 87-102, pi. 9-12. Literature cited, 
p. 100-102. 

(5) PiTz, Walter. 

1916. EFFECT OF ELEMENTAL SULFUR AND OF CALCIUM SULFATE ON CERTAIN 

OF THE HIGHER AND LOWER FORMS OF PLANT LIFE. In JoiW. AgT. 

Research, v. 5, no. 16, p. 771-780, pi. 56. 

(6) Prucha, Martin J. 

1915. PHYSIOLOGICAL STUDIES OP BACILLUS RADICICOLA OF CANADA FIELD PEA. 

N. Y. Cornell Agr. Exp. Sta. Mem. 5, 83 p. Bibliography, p. 79-83. 

(7) Reimer, F. C, and Tarter, H. V. 

I919. SULFUR AS A FERTILIZER FOR ALFALFA IN SOUTHERN OREGON. Oreg. 

Agr. Exp. Sta. Bui. 163, 40 p., 9 fig. Bibliography, p. 39-40. 

(8) Wn^ON, J. K. 

1917. PHYSIOLOGICAL STUDIES OF BACILLUS RADICICOLA OF SOYBEAN (SOJA 

MAX piper) AND OF FACTORS INFLUENCING NODULE PRODUCTION. 

N. Y. Cornell Agr. Exp. Sta. Bui. 386, p. 363-413, fig. 80-94. Bibli- 
ography, p. 412-413- 



SOYBEAN MOSAIC ' 

By Max W. Gardner, Associate in Botany, and James B. Kendrick, Assistant in 
Botany, Purdue University Agricultural Experiment Station 

In a small field of Hollybrook soybeans in West La Fayette a typical 
mosaic disease was found August 25, 1920. A rather low percentage of 
the plants were affected, and the disease was more or less confined to 
one quarter of the field adjacent to which were several rows of garden 
beans affected with mosaic to a considerable degree. In another larger 
field of soybeans in the same locality no mosaic was found. Leafhoppers 
were very prevalent on the soybeans. The impression was gained that 
the disease might have spread from the garden beans to the soybeans, 
but as yet no evidence to support such a theory has been obtained. 

Clinton^ found soybean mosaic in 19 15 at Mount Carmel, Conn., and 
under the name of chlorosis or crinkling has given an excellent account 
of the leaf symptoms along with a good illustration. He found the 
disease on the varieties Medium Green, Wilson, Swan, Kentucky, Wing's 
Mikado, and Hollybrook, and states that the Hollybrook showed the 
most marked symptoms. He found the chlorosis without the crinkling 
on the varieties O'Kute, Ito San, and Manhattan. C. R. Orton ^ has 
reported the occurrence of mosaic in a field of Ito San soybeans at 
Girard, Pa., July 30, 1920. 

SYMPTOMS 

The mosaic symptoms on the soybeans were conspicuous and unmis- 
takable, resembling those characteristic of mosaic diseases in general. 
Affected plants were stunted, and petioles and intemodes were shortened 
to some extent. The leaflets were stunted, greatly misshapen, and puck- 
ered with dark-green puffy areas along the veins (PI. 18, A, C, D, B). 
Between these puffy areas the leaf tissue was etiolated. Affected leaflets 
tended to be asymmetrical, twisted, and curled downward about the 
margins (PI. 18, D, E). As in other mosaic diseases, the young, rapidly 
growing leaves showed the most severe effects, and in some cases whole 
leaflets or portions thereof were extremely stunted or killed outright by 
the disease (PI. 18, B). The mosaic symptoms were readily distinguish- 
able from a uniform crinkling of the leaflets which was rather common 
in this field and apparently attributable to insect injury. 

The pods on mosaic plants were stunted and flattened, less pubescent, 
and more acutely curved than those on normal plants (PI. 19, C, D). 

' Contribution from the Botanical Department of Purdue University Agricultural Experiment Station, 
I^a Fayette, Ind. 

2 Clinton, G. P. notes on plant diseases of Connecticut. In Conn. State Agr. Exp. Sta. Ann, 
Rpt., 1915, p. 446-447, pi. 23a. 1916. 

3 Fromme, F. D. diseases of cereal and forage crops in the united states in 1920. In U. S. 
Dept. Agr. Bur. Plant Indus. Plant Disease Bui., Sup. is, p. 173- 1921- Mimeographed. 

Journal of Agricultural Research, Vol. XXII, No. a 

Washington, D. C. Oct. 8, 1921 

zu Key No. Ind. -10 

(III). 



112 Journal of Agricultural Research voi. xxii, No. 2 

Those borne at the upper nodes were more severely affected. The yield 
of seed was very materially reduced (PI. 19, A, B), since a considerable 
proportion of the pods contained no germinable seeds and the remainder 
as a rule not more than one or two seeds (PI. 19, D) . Even the germinable 
seeds were in general undersized. 

Observations made a month later showed that the mosaic plants were 
remaining green longer than the normal plants, so the disease evidently 
delayed maturity. 

FIELD INOCULATIONS 

In another field of soybeans in which no mosaic was present inocula- 
tions were made August 2 7 by rubbing the young intemodes with cotton 
soaked in the juice from crushed mosaic soybean leaves and then wound- 
ing these intemodes with a needle. One hundred and fourteen plants were 
thus inoculated, but no mosaic developed. Fifty- two plants were simi- 
larly inoculated, except that the juice of leaves from mosaic garden beans 
was used as inoculum, and none developed the disease. Forty-six garden 
bean plants were also inoculated in a similar manner with the virus from 
soybean mosaic, and none developed mosaic. 

SEED TRANSMISSION 

To determine whether or not the disease was seed-borne, a quantity 
of seed was saved from mosaic and healthy plants early in October for 
subsequent tests in the greenhouse. On October 25, 150 seeds from 
mosaic plants were planted in 25 pots of sterilized soil, 6 in each pot. 
By December 15, 124 plants had come up, and 18 showed unmistakable 
mosaic symptoms. None of the 148 controls grown from seed from 
normal plants showed mosaic. 

In a second trial about 180 seeds from mosaic plants were planted 
December 9 in 59 pots of sterilized soil. February 3, 192 1, 11 out of the 
106 plants which were up showed mosaic. None of the 38 controls 
grown from seed from normal plants showed the disease. As a result 
of these two tests it is evident that about 13 per cent of the seedlings 
from seed produced on mosaic plants developed the disease. 

The mosaic seedlings were spindling (PI. 18, F, G), and the j&rst pair of 
true leaves were characterized by downward, longitudinal curling or 
rolling, a crinkling, and a faint etiolation or mottling. These leaves 
turned yellow prematurely. The leaves subsequently formed were 
greatly stunted and showed the mottling and crinkling more conspicu- 
ously than the first leaves. 

GREENHOUSE INOCULATIONS 

From these mosaic seedlings the disease was transmitted to healthy 
soybean seedlings. Several methods of inoculation proved successful. 
A number of inoculations made early in January yielded only negative 
results, but later better success was obtained. 



Oct. 8,1921 Soybean Mosaic 113 

On January 26, twenty-five plants were inoculated by pricking with a 
needle at the nodes and rubbing the wounded areas with cotton soaked 
in the juice from crushed mosaic leaves. Eight plants used as controls 
were similarly treated, except that sterile water was substituted for 
the mosaic virus. Because of the unfavorable greenhouse conditions 
the plants made slow growth during the winter, so that the mosaic 
symptoms were very slow in developing. On March 5 two plants 
showed mosaic mottling on the young leaves. On March 25 two more 
showed mosaic, and on April 7 seven out of the 25 plants had developed 
the disease. The controls developed no mosaic. 

A number of inoculations were made March 2. In one series crushed 
mosaic tissue was inserted into slits made with a scalpel near the grow- 
ing points and on the petioles. On March 15 two of the seven plants 
thus inoculated showed mosaic symptoms on the young leaves, and on 
April 7 five had developed mosaic. 

In a second series of inoculations made the same date by cutting off 
one leaf at each node and smearing these wounded surfaces with crushed 
mosaic tissue, three out of eight plants showed mosaic symptoms on 
the new leaves March 15, or 13 days after inoculation, and on April 7 
six plants had developed mosaic. 

In a third series five plants were inoculated by a combination of the 
two methods above described. Thirteen days later three showed mosaic 
mottling, and by April 7, or 37 days after inoculation, four had developed 
the disease. 

In a fourth series, five plants w^e inoculated by rubbing the under 
surfaces of the leaves with slightly crushed mosaic leaves forcibly enough 
to cause slight abrasions. On March 15, four of these plants showed 
the disease, and on April 7 all showed typical mosaic. 

None of the five control plants inoculated by one or the other of these 
methods without the application of mosaic tissue developed mosaic. 
At no time was there any spread of the disease in the greenhouse. 

In these inoculations the symptoms became evident only on the 
young leaflets. These in some cases developed distinct mottling, and 
in other cases they exhibited a slight degree of etiolation and the char- 
acteristic downward, longitudinal rolling. The incubation period 
under the conditions of this test was 13 days. 

Preliminary cross inoculations to garden beans and cowpeas have 
given negative results. Further tests are being made. 

Soybeans, therefore, are subject to a destructive mosaic disease which 
greatly reduces the yield of affected plants. The disease is transmis- 
sible from plant to plant and also is seed-borne. 



PLATE i8 

A. — Typical mosaic leaf showing darker green puffy areas along the veins, 
B. — Mosaic leaf showing extreme sttmting of terminal leaflet. 
C. — Normal leaflet. 

D. — Mosaic leaflet showing longitudinal rolling. 
E. — ^Typical mosaic leaflet. 

F. — Mosaic seedlings from seed from a mosaic plant, showing stunting of the plant 
and longitudinal rolling of first leaves. 
G. — Normal seedlings from seed from a mosaic plant. 



Soybean Mosaic 



Plate If 









Journal of Agricultural Research 



Vol. XXII, No. 2 



Soybean Mosaic 



Plate 19 





D 




Journal of Agricultural Research 



Vol. XXII, No. 2 



PLATE 19 

A. — Upper nodes of a normal plant, showing yield of pods. 
B. — Upper nodes of a mosaic plant, showing effect of the disease on the yield. 
C. — Normal pod. 

D. — Type of pods produced by a mosaic plant. 
54818° — 21 6 



ADDITIONAL COPIES 

OF THIS PUBUCATION MAT BE PKOCUKED FEOM 

THE StTPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTING OFFICE 

WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 
SuBSCEiPTiON Price, $3.00 Per Yeab 

A 



BBMBaHnBMBaai 



iP-r — 

Vol. XXII OCTOBER 15, 1921 No. 3 

JOURNAL OP 

AGRICULTURAL 

RESEARCH 



CONXKNXS 

Page 

Influence of the Plane of Nutrition on the Maintenance 
Requirement of Cattle - - - - - - -115 

F. B. MUMFORD, A. G. HOGAN, and W. D. SALMON 

( Contribution from Missouri Agricultural Experiment Station ) 

Turnip Mosaic - - - - - - - - 123 

MAX W. GARDNER and JAMES B. K^NDRICK 

( Contribution from Indiana Agricultural Experiment Station ) 

Hydrocyanic Acid in Sudan Grass - - - - - 125 

C. O. SWANSON 

( Contribution from Kansas Agricultural Experiment Station ) 

Nutrient Requirements of Growing Chicks: Nutritive 
Deficiencies of Corn - - - - - - - 139 

F. E. MUSSEHL, J. W. CALVIN, D. L. HALBERSLEBEN 
and R. M. SANDSUEDT 

( Contribution from Nebraska Agricultural Experiment Station ) 

Aecial Stage of the Orange Leafrust of Wheat, Puccinia 
triticina Eriks. - - --- -- - 151 

H. S. JACKSON and E. B. MAINS 

( Contribution from Bureau of Plant Industry and Indiana Agricultural Experiment Station ) 

A Transmissible Mosaic Disease of Chinese Cabbage, 
Mustard, and Turnip ------- 173 

E. S. SCHULTZ 
(Contribution from Bureau of Plant Industry) 



PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICCLTDRE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



WASHINOTON, D. C. 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 

J. G. LIPMAN 

Dean, State College of Agriculture, ottd 
Director, New Jersey Agricultural Etperi- 
ment Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS 

Dean, School of Agriculture, and Direct, 
Agricultural Experiment Station, The 
Pennsylvania State College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles- from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultiu-al Experiment Station, New 
Brunswick, N. J. 



JOmAL OF ACRiaiTIIAL RESEARCH 

Voiv. XXII Washington, D. C, October 15, 192 1 No. 3 



INFLUENCE OF THE PLANE OF NUTRITION ON THE 
MAINTENANCE REQUIREMENT OF CATTLE' 

By F. B. MuMPORD, Dean of the College of Agriculture, A. G. Hogan, of the Depart- 
ment of Animal Husbandry, and W. D. Salmon, Graduate Student in Animal Hus- 
bandry, College of Agriculture, University of Missouri. 

In 19 1 4 an investigation was begun at the University of Missouri to 
study some of the effects of underfeeding. Calves of beef-breeding 
stock were secured, and they were placed on three planes of nutrition. 
Group I was fed to grow rapidly, but not to become fat. Group II was 
placed on a lower nutritive plane and was fed to gain about }4 pound per 
day. Group III was placed on a still lower nutritive plane and fed to 
gain about V3' pound per day. At the present time three animals remain 
that were started on the investigation in 19 14. Seven others were added 
in 191 7. The older animals, therefore, have been under observation for 
seven years and the younger animals for four years. Under these cir- 
cumstances it seemed desirable to make a study of the maintenance 
requirement of steers at different ages and on different planes of nutrition.^ 

The ideal method of conducting an investigation of this kind would 
require a respiration calorimeter. Since that was impossible, the alter- 
native was to calculate the energy value of the feed consumed and cor- 
rect this for the estimated value of the gains or losses in body weight. 

The net energy of the feed consumed was calculated in accordance 
with procedures developed by Armsby.^ The energy value of the changes 
in body weight were calculated from the composition of steers that had 
been analyzed at this Station by the Department of Agricultural Chemis- 
try. So far as possible steers were selected as controls for this purpose 
that were of similar age, measurements, and weight and that had received 
similar treatment. 

' The data for this paper were taken from the thesis of W. D. Salmon, presented at the University of 
Missouri, as partial fulfillment of the requirements for the degree of Master of Arts. The investigation 
was initiated by F. B. Mumford, Dean of the College of Agriculture, and by P. F. Trowbridge, at that 
time Chairman of the Department of Agricultural Chemistry. Since September 1918, E. A. Trowbridge, 
Chairman of the Department of Animal Husbandry, has had general supervision of the project. This 
article was prepared by A. G. Hogan, who has been in immediate charge since September, 1920. A large 
number of workers have contributed to the success of the experiment. 

' The original data will be reproduced in detail in a subsequent publication. 

8 Armsby, H. p., and Fries, J. A. net energy values for ruminants. Pa. Agr. Exp. Sta. Bui. 
142, 20 p. 1916. 

Journal of Agricultural Research Vol. XXII, No. 3 

Washington, D. C. Oct. 15, 1921 

zv Mo. -J 

(115) 



ii6 Journal of Agricultural Research voi. xxn. no. 3 

METHOD OF THE EXPERIMENT 

RATIONS 

The concentrate consisted of the following mixture: Com chop, 60 
per cent; wheat bran, 30 per cent; linseed meal, 10 percent. The rough- 
age fed to the 3 old steers. No. 528, 589, and 585, from the beginning of 
the experiment until July 20, 191 7, was timothy. For the next 10 days 
a mixture of 5 parts timothy, 3 parts alfalfa, and 2 parts oat straw was 
fed. Following this the roughage consisted of a mixture of 60 per cent 
alfalfa amd 40 per cent oat straw. The animals were fed twice daily and 
had access to water at all times. Salt was accessible at feeding time. 

PERIODS 

The calculations are made for periods of 180 days, with the exception 
of the first period for the 3 older steers, which was as follows: No. 528, 
130 days; No. 579, 142 days; No. 585, 150 days. The warm months 
of the year were selected for these periods to avoid a possible disturbing 
effect of low temperatures in the winter months. 

WEIGHTS 

The steers were weighed each morning after feeding but before water- 
ing. The weight given for the beginning of a period is the average of the 
10 preceding days. The weight given at the end is an average of the last 
10 days of the period 

ENERGY INTAKE 

The amount of dry matter consumed was calculated from the weight 
and composition of the feed consumed. The net energy was computed 
from this by the use of factors reported by Armsby and Fries. ^ For the 
concentrates the value 83.82 therms per 100 pounds dry matter was used. 
This is the factor given for Armsby's grain mixture No. 2,- which approxi- 
mates the grain mixture used in this experiment. For timothy hay the 
value 48.63 therms per 100 pounds dry matter was used. The factor for 
the roughage mixture used in the latter part of the experiment was cal- 
culated from the Armsby values, for alfalfa, 34.10 therms and for oat 
straw, 26.03 therms per 100 pounds dry matter. A mixture of 60 parts 
alfalfa and 40 parts oat straw would have a value of 30.87 therms per 100 
pounds dry matter. The calculations of the energy value of the milk are 
based on factors published by Armsby.^ There are 29.01 therms per 
100 pounds whole milk (4.4 per cent) and 14.31 therms per 100 pounds 
skim milk (0.2 per cent). From these values factors were computed for 
the different grades of milk used. 

'Armsby, H. P., and Fries, J. A. op. cit. 

2 Annsby's grain mixture No. 2, 60 per cent com meal, 30 per cent crushed oats, 10 per cent O. P. linseed 
meal. Our grain mixture, 60 per cent com meal, 30 per cent wheat bran, 10 per cent O. P. linseed meal. 
^Armsby, Henry Prentiss, thb nutrition op farm animals, p. 719. New York, 1917. 



Oct. IS, 1921 ' Influence of Plane Nutrition on Maintenance 



117 



CHANGES IN BODY WEIGHT 

In order to obtain data concerning the maintenance requirement of 
these steers, it is necessary to calculate the energy gained or lost through 
changes in body weight. Our calculations are based on analyses previ- 
ously made by the Department of Agricultural Chemistry, University of 
Missouri.^ Control animals were selected from those on which analyses 
were available, on the basis of similar weights and measurements, and 
when possible of similar ages, daily gains, and daily consumption of dry- 
matter. In some cases suitable control animals were not available, and 
the composition of steers for those periods was estimated by interpola- 
tion, with the exception of the last period for steer No. 528. In this case 
a value published by Armsby^ was used. The average energy values of 
a pound gain as calculated by this method for steers in the three groups 
are given in Table I. For purposes of comparison the values given by 
Armsby are shown in the same table. 

Table I. — Energy values of a pound gain 



Approximate age (months). 



6. 
18 
36 
54 
66 
78 



Group I. 



Therms. 

o- 95575 

1. 0918 
I. 7136 

2. 1993 
2. 50 
3.00 



Group II. 



Therms. 

o- 95575 
I- 0583 
I. 1608 
I. 4104 

I- 5352 
I. 660 



Group III. 



Therms. 
o- 8343 
•9445 
I. 0548 
I. 1013 
I. 4790 
I. 6490 



Annsby's values. 



Age. 



Months. 

I 

2 to 3 

5 to 6 

II to 12 

18 to 24 



Energy. 



Therms. 

1. 170 
1-374 

1.680 

2. 292 

3. 000 



Armsby's values are consistently higher, as is to be expected. Our 
animals were thin and contained less than the usual amount of fat in the 
gain. .; „.j'f u.-A ..■...■.:■' f.,^.. vjorf 

In calculating the maintenance requirements per 1,000 pounds live 
weight, Moulton's ^ formula was used. He has shown that the surface 
areas of thin cattle are proportional to the ^ power of the live weight. 
The results of this calculation, on the basis of dry matter consumed, are 
given in Table II. 

The net energy required for maintenance was also calculated by 
another method,* based on the digestible organic matter of the feed. 

The following factors are given for the metabolizable energy of digesti- 
ble organic matter consumed: Roughage, 1.588 therms per pound; grains 
and similar feeds with less than 5 per cent digestible fat, 1.769 therms per 
pound. In the same publication the "Average energy expenditure by 
cattle per 100 pounds of dry matter eaten" is given. 

1 These have not yet been published. 

* Armsby, Henry Prentiss, op. cit. 

' MOULTON, C. R. THE AVAtLABILITY OF THE ENERGY OF FOOD FOR GROWTH. In Jour. Biol. Chem., V. 

31, no. 2, p. 390. 1917. 

* Armsby, H. P., and Fries, J. A. op. cit. 



ii8 



Journal of Agricultural Research voi. xxii. no. 3 



Table II. — Average daily maintenance requirement as calculated from dry matter con- 
sumed 



Steer No. 


Number of periods averaged. 


Therms of net energy per i.ooo pounds, 
based on H power of live weight. 




Group I. 


Group II. 


Group III. 


528 

577 

571 

579 

578 

573 

585 

575 

574 

572 


6 


5. 870 
5.280 
5-730 






3 

3 

e 










4.920 
3-830 
4.409 




1. 






•J 






c 




4. 221 
4.041 
4.302 
3-250 


T. 






7. 






•J 












Average of all animals for all periods 


5-523 


4-485 


3-830 



Table III. — Energy expenditure by cattle per loo pounds dry matter consumed 



Ration. 



Roughage : 

Timothy hay 

Alfalfa hay 

Oat straw 

Concentrate : 

Grain mixture No. 2 



Energy ex- 
penditure. 



Therms. 
35-47 
53-03 
46. 00 

51-76 



The coefficients of digestibility used in these calculations were derived 
from digestion trials conducted under similar conditions at this Station. 
These indicated that the digestibility of the ration varied with the rela- 
tive amounts of hay and grain fed. The factors used are given in 
Table IV. 

Table IV. — Digestion factors for organic matter 



Ratio of grain to 
hay. 


1. 1 


2-3 


I. 3 


1.3.4. or 5 


1.6 or 7 


1. 8,9, or 10 


Hay only. 


Factor 


.6956 


.6695 


•6434 


.6340 


. 6229 


.6030 


0. 5832 



Inasmuch as the thermal value of a pound of organic matter from 
grain differs from that of a similar weight of organic matter from rough- 
age, the Armsby factors ^ previously quoted in this paper could not be 
directly applied to the values obtained with the foregoing digestion 
coefficients. Those factors would not provide for the widely varying 
proportions of grain and hay. The following method, therefore, was 



'Armsby, H. P., and Fries, J. A. op. cit. 



Oct. IS. 1921 Inf,uence of Plane Nutrition on Maintenance 



119 



used in computing tlie energy intake on the basis of digestible organic 
matter consumed. By use of the factors in Table IV, the weight in 
pounds of digestible organic matter in the mixed ration was determined 
for each period. This was multiplied by 1.588, the Armsby factor for 
metabolizable energy in a pound of digestible organic matter from hay. 
The thermal value of digestible organic matter from grain is 1.769, 
however, or 0.181 therms more. Therefore, each pound of digestible 
organic matter derived from grain was multiplied by 0.181, and the 
product was added to the result obtained by multiplying the total 
digestible organic matter by 1.588. This gave the total metabolizable 
energy in both the hay and grain. The digestibility of the organic 
matter of the grain was estimated by difference. This ranged closely 
around 80 per cent. The factors for energy expenditure are given in 
Table III. 

It seemed impracticable to calculate the net energy of the milk con- 
sumed on the basis of digestible organic matter, so the calculation was 
based on the quantity consumed, as previously described. Since the 
amount was small, however, the method of calculation would have 
little effect on the final result. [i;i> 0. 

The method used in correcting for changes in body weight has already 
been described, and the maintenance requirement as calculated on the 
basis of digestible organic matter consumed is given in Table V. 



Table v.- 



-Average daily maintenance requirement, as calculated front digestible organic 
matter consumed 



Steer No. 


Number of periods averaged. 


Therms of net energy per i.ooo pounds 
based on 5/8 power of live weight. 




Group I. Group II. 


Group III. 


528 

577 

571 

579 

578 

573 

585 

575 

574 

572 


6 . . . . 


6. 261 
5-412 
5-174 




















5. 260 
4. 192 
4-893 




















4.725 
4-454 
4-591 
3-649 


■2 ... 






-2 . 






•2 . 






Average of all animals for all 
periods 








5-777 


4.869 


4. 408 









In determining the maintenance requirement on the basis of digestible 
organic matter, the calculations were based on digestion coefficients ob- 
tained at this Station under similar conditions. This method is probably 
more accurate than that of calculation on the basis of dry matter con- 
sumed, and for the animals concerned it gives a result about lo per cent 
higher. 



I 



120 



Journal of Agricultural Research 



Vol. XXII. No. 3 



In calculating average results, obtained by both methods, four periods 
in which there were losses in live weight were omitted. The results for 
those periods were low, and we were uncertain as to whether the result 
was approximately correct or whether it was due to an incorrect assump- 
tion as to the energy value of the loss in weight. Most of the dry matter 
of the loss was probably fat, and if so, our calculation of its energy value 
was too low and so made our calculation of the maintenance requirement 
too low. 

One steer, No. 585, had a navel infection during the first period, 
accompanied by a very high maintenance requirement. This period also 
was discarded in calculating averages. 

INFLUENCE OF NUTRITIONAL PLANE 

There is a close parallel between the intake of net energy and the 
maintenance requirement of the animal. The record of steer 574 illus- 
trates that tendency. For the first period the average daily intake of 
net energy was 3.884 therms per 1,000 pounds, based on the 5/8 power 
of the live weight; and the maintenance requirement was 3.818 therms. 
For the second period the energy intake was increased to 5.783 therms, 
and the maintenance requirement increased to 5.1 19 therms. In the 
third period the energy intake was 5.253 therms, and the maintenance 
requirement was 4.836 therms. 

Table VI. — Dailymaintenance requirements of cattle — Net energy 

RESPIRATION EXPERIMENTS 



Num- 
ber of 


Investigator. 


Condition of ani- 
mal. 


Therms per i.ooo pounds live weight. 


ments. 


Maximum. 


Minimum. 


Average. 


22 


Armsby and Fries ^ . 


Medium 

do 


7-430 
6. 780 

8.871 


4-723 
4-921 
7-319 




7 


Kellner ^ 


5-99^ 




do 


Fat 


5- 742 
7.946 









LIVE-WEIGHT EXPERIMENTS 



Armsbv ' . . 

do'. ... 

Haecker * . . 
Eward ^ . . . 
Eckles^ .. 
Shirky 2. . . 
... .do. . .. 
Our results . 
....do. ... 
....do. ... 



Thin 

do. . . 

Medium... 

do.... 

do.... 

do3.. 

Thin*.... 
Group I . . . 
Group II.. 
Group III. 



7.044 
6.039 
5.676 
7.850 
7.079 



5- OQ59 
7.380 
5- 724 
5- 217 



6. 136 

4- 713 
4. 662 
6.450 
5.841 



4-953 
4-915 
3.809 
3.276 



50s 
423 
021 
180 
173 
732 
0245 

777 



5 
5 

4. 869 
4. 408 



' Armsby, Henry Prentiss, op. cit., p. 291. 

2 Shirkey, S. B. extent to which growth retarded during the early life of the beef animal 
CAN BE LATER REGAINED. Univ. of Mo. thesis, 1919. (Unpublished ) animal 

' Corresponds to group I of this experiment. 
* Corresponds to group 11 of this experiment. 



Oct. 15. 1921 Influence of Plane Nutrition on Maintenance 121 



In comparing the maintenance requirements of the three groups it 
should be kept in mind that group I does not represent a high plane of 
nutrition. The aim was to secure maximum growth with no considerable 
fattening. Their maintenance requirements as computed in this paper 
correspond closely to the average of 22 respiration experiments by Armsby 
and Fries ^ and of 7 by Kellner,^ on cattle in medium condition. 
A comparison of our results, and of those obtained by other investiga- 
tions, is given in Table VI. 

INFLUENCE OF AGE 

The ages represented in this experiment vary from 30 days for some of 
the calves at the beginning of the first period to more than 6 years at the 
close of the seventh period. Apparently there was no relation between 
the age and the maintenance requirement of these animals. Some of the 
steers showed a gradual decrease in the maintenance cost from the begin- 
ning to the end of the experiment. In such cases it was found that the 
energy intake per 1,000 pounds had also decreased. On the other hand, 
steers with an increasing energy intake showed an increased maintenance 
requirement. Maintenance trials on young animals usually give higher 
results than have been obtained with mature animals, but if age does 
influence the maintenance requirement the effect is too slight to be shown 
in a live- weight experiment of this kind. 

SUMMARY AND DISCUSSION 

There is a close relation between the amount of net energy consumed 
and the maintenance requirement. Periods of high energy intake were 
apparently periods of high maintenance cost, while periods of low energy 
intake were accompanied by a lowered maintenance requirement. 

The averages of the periods discussed show the following daily main- 
tenance requirements per 1,000 pounds live weight, calculated on the 
basis of digestible organic matter, and in terms of net energy: Group I, 
5.777 therms; group II, 4.869 therms; and group III, 4.408 therms. If 
the maintenance requirement of group I is 100 per cent, that of group II 
is 84.4 per cent, and that of group III is 76.3 per cent. 

The calculations on the basis of dry matter consumed indicate even 
greater differences. The maintenance requirements as derived by this 
method may be compared as follows: Group I, 100 per cent; group II, 
81 per cent; group III, 69.3 per cent. The estimated maintenance re- 
quirement of group I, as calculated by this method, is 30 per cent greater 
than the total net energy intake of group III. 

There is no apparent relation between the age of the animals and the 
amount of energy required for maintenance. 



' Armsby, Henry Prentiss, op. cit. 



Turnip mosaic ^ 

By Max W. Gardner, Associate in Botany, and James B. Kendrick, Assistant in 
Botany, Purdue University Agricultural Experiment Station 

In one comer of a small field of turnips near South Bend, Ind., Octo- 
ber 12, 1920, a considerable percentage of the plants were found affected 
with an unmistakable mosaic disease. The symptoms were typical of 
mosaic diseases in general. The leaves were stunted, misshapen, and a 
lighter green with dark green blisters or puffy areas. Many of the leaves 
were extremely distorted by crinkling and folding (PI. 20, A). The dis- 
ease seemed to be confined to one area in the field, to some extent coin- 
cident with a heavy infestation of tarnished plant bugs. 

Several diseased plants were transplanted to pots in the greenhouse, 
where they continued to form new leaves during the winter. The mosaic 
symptoms exhibited by the new foliage formed under greenhouse con- 
ditions were not quite so extreme as had been noted in the field. One 
of these plants, with mottled and spindling leaves, is shown in Plate 20, 
B, as it appeared in December. 

Inoculation of a number of potted turnip and radish seedlings was made 
by breaking off a leaf and rubbing the wound with crushed leaf tissue 
from one of the mosaic plants. Out of 21 turnip seedlings inoculated 
early in January, 13 developed characteristic mosaic symptoms. The 
first symptoms were noted 26 days after inoculation. The turnips inocu- 
lated showed some varietal difference from the plants collected in the 
field in that the leaves were much less distinctly pinnatifid. Out of 46 
radish seedlings, including both white and red varieties, similarly inocu- 
lated, none developed mosaic symptoms. 

A later series of inoculations was made January 26 by wounding the 
plants with a needle and rubbing the wounded areas with a piece of cotton 
soaked in the juice from mosaic leaves ground up in a mortar. Ten out 
of 14 turnip plants thus inoculated developed the mosaic disease. The 
first symptoms were noted 16 days after inoculation. No mosaic devel- 
oped among 13 control plants similarly treated except that sterile water 
was substituted for the mosaic virus. Twenty-two radish plants were 
also inoculated, and none of these developed the disease. Subsequent 
reinoculation of turnip plants from one of these radish plants produced 
no mosaic. The mosaic disease of turnips is therefore readily transmis- 
sible to turnips but not to radishes. 

' Contribution from the Botanical Department of Purdue University Agricultural Experiment Station, 
I,aFayette, Ind. 

After this article was prepared it was learned that Eugene vS. Schultz, of the Bureau of Plant Industry, 
United States Department of Agriculture, was also working on this disease. 

Journal of Agricultural Research, Vol. XXII, No. 3 

Washington, D. C. Oct. 15, 1921 

zw Key No. Ind.- 1 1 

(123) 



PLATE 20 

A. — Leaves from mosaic turnip plants collected October 12, tqso. 
B. — Mosaic turnip plant transplanted to a pot in the greenhouse. Photographed 
December 20, 1920. 

(124) 



Turnip Mosaic 



Plate 20 




Journal of Agricultural Research 



Vol. XXII, No. 3 



HYDROCYANIC ACID IN SUDAN GRASS' 

By C. O. vSwANSON ^ 
Professor of Agricultural Chemistry, Kansas State Agricultural College 

In a previous paper ^ it was shown that hydrocyanic acid (HCN) is 
obtained from green Sudan grass by macerating, digesting in water, and 
distilling into a dilute solution of sodium or potassium hydroxid. Sev- 
eral experiments reported in that paper made it clear that this acid does 
not exist free in Sudan grass and is obtained only if the conditions of the 
determination are favorable to enzym action. It appears to be a common 
belief that hydrocyanic acid is developed by freezing. This merely bursts 
the green cells and thus performs the same function as maceration, with 
the result that the hydrocyanic acid is rapidly lost from frosted grass. 
It was also shown that while in some cases poisoning had been reported 
from pasturing Sudan grass, under normal conditions no poisoning took 
place either before or after the grass was frozen. It was suggested that 
when frozen the hydrocyanic acid had been liberated and then evap- 
orated as the grass dried. 

Because of the importance of the subject it was thought worth while 
to make further investigations. During the summer of 1920, material 
was obtained from a i/20-acre plot of Sudan grass grown by the Depart- 
ment of Agronomy of the Kansas State Agricultural College. The 
Sudan grass had been planted early in June in rows about 2 feet apart. 
On June 22, when the experiments were begun, the grass was about 6 
inches high. These experiments were continued during the summer and 

early fall. 

METHOD OF DETERMINING HYDROCYANIC ACID 

At present there are no satisfactory quantitative methods for esti- 
mating hydrocyanic acid obtained from organic material. All are open 
to some objection. After considering several, the Prussian-blue method 
was adopted as best suited for the purposes of the present investigation. 
Because of simplicity in manipulation it is possible to run a large number 
of determinations at the same time. The amount of hydrocyanic acid 
obtained from the different samples was estimated colorimetrically, 
using standard solutions containing known amounts of potassium cyanid. 
One objection to the colorimetric measurements was the difficulty in 

' Contribution No. 92 from the Department of Chemistry, Agricultural Experiment Station of Kansas 
State Agricultural College. 

* Credit is due Mr. Carl M. Conrad for efficient assistance in making the determinations reported in this 
paper. 

' SWANSON, C. O. HYDROCYANIC ACID IN SUDAN GRASS AND ITS EFFECT ON CATTLE. Ill JoUr. Amer. 

Soc. Agron., v. 13, no. i, p. 33-36. 1921. 

Journal of Agricultural Research Vol. XXII, No. 3 

Washington, D. C. Oct. 15, 1921 

7x Key No. Kans.-J7 

(125) 



126 Journal of Agricultural Research voi. xxii. no. 3 



obtaining a uniform blue color. Very often the precipitate was decidedly 
green. It was found that by warming and letting the precipitate 
stand for some time in loosely stoppered bottles a uniform blue color 
could be obtained. The use of nitric or sulphuric acid instead of hydro- 
chloric acid or the addition of potassium fluorid, all of which have been 
suggested by other workers, did not seem to eliminate, entirely, the 
green color. While the defects of the Prussian-blue method are fully 
realized, it compares favorably with other methods.' In no sense are 
the values reported in this paper to be regarded with the same degree of 
accuracy as a protein or even a crude-fiber determination. For this 
reason no conclusions should be drawn from the results unless the figures 
presented are uniformly consistent or the differences large. 

The calculations in this paper are based upon approximately 200 gm. 
of green material. When the grass was wilted or dry the weight of sample 
used was proportionately lessened. It is impracticable to secure green 
samples of uniform weights of dry matter, particularly if they are 
gathered during different hours of the day and throughout several weeks 
and months. Then, as will be shown in what follows, the hydrocyanic 
acid is localized in the plant, being present in the largest amounts in 
those portions of the plant possessing the greatest vegetative activity. 
For this reason leaves were separated from the stems whenever these 
were present. The amount of hydrocyanic acid obtained is small in 
proportion to the total weight of samples used. It was seldom more 

than 0.015 per cent. 

EFFECTS OF MACERATION 

The first sample was collected June 22, when the grass was about 6 
inches high. This was cut into pieces about % inch long and digested for 
three hours in water at room temperature. Less than i mgm. hydrocy- 
anic acid was obtained. Another sample, taken the next day, was cut 
and thoroughly macerated by pounding in an iron mortar and was then 
digested in water. This sample gave 27 mgm. hydrocyanic acid. On 
June 28 a sample was secured and divided into two equal portions. One 
portion was cut and macerated as described above, and the other was 
cut and macerated with coarse, sharp sand. Both were digested in 
water for the same length of time. The portion macerated with sand 
gave 26 mgm. hydrocyanic acid, and the other gave 36 mgm. It ap- 
peared from this that maceration with sand was not necessary and might 
result in a loss. Subsequent experiments showed that as soon as the 
grass is macerated the hydrocyanic acid is liberated and for this reason 
may be lost. On August 18 a sample was divided into four portions 
and, after the preliminary treatment mentioned, was digested overnight, 
with the result given in Table I. 

* ViEHOVER, Arno, and Johns, Carl O. on the determination of small quantities op hydrocy- 
anic ACID. In Jour. Amer. Chem. Soc, v. 37, no. 3, p. 601-607. 1915. 



octis. I92I Hydrocyanic Acid in Sudan Grass 127 

Table I. — Effect of maceration on liberation of hydrocyanic acid 



Sam- 
ple 
No. 



Treatment. 



HCN. 



No cutting or maceration 

Cut in feed cutter {}^ to ys inch) 

Cut and macerated slightly 

Cut and macerated thoroughly . . . 



Mgm. 

o 

10 

10 

II 



From this it appeared that if the time of digestion is sufficiently long 
the amount of maceration is less important, provided, however, that the 
plant tissue is cut fairly fine. The smaller amount obtained from the 
grass on August 18, as compared with that obtained in June, is in accord 
with a general observation made during the summer, that as the season 
advanced smaller amounts were obtained from the 200-gm. portions. 

TIME REQUIRED FOR DIGESTION •tM,;;-n 

The time required for digestion in order to obtain the maximum amount 
of hydrocyanic acid was determened. A sample collected on June 28 was 
divided into three portions and similarly treated, except for the time 
allowed for digestion. The results obtained are given in Table II. 





Table II. — Effect of time of digestion on liberation of hydrocyanic acid 




Sample 
No. 


Treatment. 


HCN. 


I 


Digested 3 hours 


Mgm. 
18 


2 


Digested 6 hours 


32 
32 


3 


Digested 24 hours 





This experiment seemed to show that digesting 3 hours was not long 
enough, while 6 hours was as effective as 24. On August 28 a similar 
experiment gave the results shown in Table III, the results in each case 
being an average of duplicate samples. 

Table III. — Effect of tim,e of digestion on liberation of hydrocyanic acid 



Sample 
No. 



Treatment. 



HCN. 



Digested two days. . . 
Digested three days . 
Digested four days. . . 
Digested seven days . 



Mgm. 

10 

10 

10 

O 



The portion digested seven days developed a very bad odor. Because of 
these results, the usual procedure with experiments reported in this paper 
was to macerate the sample and then digest at room temperature over- 
night. 



128 



Journal of Agricultural Research voi. xxn. No. 3 



LOCALIZATION OF HYDROCYANIC ACID IN THE PLANT 

At three different times the grass was divided into leaves and stems. 
From 200-gm. portions the number of milligrams of hydrocyanic acid 
shown in Table IV were obtained. 

Table 1\ . — Hydrocyanic acid in leaves and stem s of Sudan grass 



Date collected. 



June 30 
July 7.. 
July 21. 




stems. 



Mgm. 

Trace. 



The immaturity of the sample collected June 30 accounts for the moder- 
ate amount obtained from the stems. At a later date two tests were made 
on immature heads. No hydrocyanic acid was found. In the following 
tests reported in this paper leaves only were used unless otherwise stated. 

INFLUENCE OF STAGE OF GROWTH 

Since grass was cut almost every week throughout the summer there 
were afforded several opportunities to test the comparative amounts 
present in various stages of growth and development. The shorter grass 
was obtained from plants which had been cut once or several times. 

Table V. — Hydrocyanic acid in Sudan grass at different dates and stages of growth 



Date collected. 



Portion used. 



Average height. 



HCN. 



July 22. 



26. 

Aug. 12 
27. 
30 



Whole plant. 

....do 

....do 

Leaves 

....do 

....do 

Whole plant. 

do 

do 

Leaves 

do 

do 

do 

I Whole plant. 
Leaves 
do 

(Whole plant, 
do 
Leaves 
do 

fWhole plant. 
\Leaves 

(Whole plant, 
do 
Leaves 
do 



6 inches 

8 inches 

12 inches 

18 inches 

24 inches 

30 inches 

4 inches 

6 inches 

12 inches 

Beginning to head 

Partly headed 

Fully headed 

Partly dead 

4 inches 

16 inches 

Headed 

4 inches 

10 inches 

15 inches 

24 inches 

5 inches 

20 inches 

2 inches 

12 inches 

Ready to head. .. . 
Blooming 



Mgm. 
27 

19 

7 
24 
20 

9 
40 

32 
10 
12 

17 
20 
6 
II 
18 

19 

40 

12 

16 

10 

10 

II 

5 

5 

9 

10 



Oct. 15, I93I Hydrocyanic Acid in Sudan Grass 129 

The results show that more hydrocyanic acid is found in the whole 
plant in the earlier stages of growth and less as the season advances. 
The difference is perhaps due to the large proportion of stems in the 
latter part of the season, since if leaves only are compared there is very 
little difference except where they are from mature plants. This indi- 
cates that most of the hydrocyanic acid is obtained from those parts of 
the plants where the vegetative activity is most pronounced. This 
agrees with the results obtained by Menaul and Dowell ^ at the Okla- 
homa Agricultural Experiment Station. These observations support the 
theory that hydrocyanic acid is an intermediate product between the 
nitrates and the amino acids. ^ 

DISAPPEARANCE FROM MACERATED MATERIAL 
As soon as the grass is macerated the hydrocyanic acid begins to pass 
off. This was demonstrated several times by suspending small pieces 
of sodium-picrate paper above some macerated grass in stoppered flasks. 
The paper very soon assumed a brown color. The quantitative deter- 
minations given in Table VI were made on samples macerated July 10 
and treated as indicated. 

Table VI. — Disappearance of hydrocyanic acid in macerated grass 

_ 

^X!'^ Treatment. | HCN. 



Digested in water two days 

Placed without added water in covered mason jar for two days then 
small amount of water added and distilled 

Placed in flask two days so that the hydrocyanic acid could escape 
only into the receiving flask, after which water was added and 
distilled 

4 Left in open jar for two days, digested and distilled 

5 Repeat of 3 but kept in flask overnight only 



Mgm. 
25 



26 
Trace. 
26 



DISAPPEARANCE FROM GRASS AFTER CUTTING 

In a previous paper ^ it was stated that tests made on partially wilted 
grass may be worthless. In the experiments made at that time, the 
amount of sulphuric acid added was not carefully enough controlled. It 
will be shown in the following paragraphs that if acid is added beyond 
certain limits no hj^drocyanic acid will be obtained from either green or 
partially wilted grass. In each of the determinations given in Table 
VII the grass was macerated after the treatment stated and then di- 
gested in water overnight. 

1 Menaul, Paul, and DowEi,!,, C. T. cyanogenesis in sudan grass: a modification of the franos- 
CONNELI, METHOD OF DETERMINING HYDROCYANIC ACID. In Jour. Agr. Research, v. i8, no. 8, p. 447-450. 
1920. 

2 Ravenna, C, and Zamorani, M. ntjove ricerche sulla fxjnzione fisologica dell, acido ciani- 
DRICO NEL SORGHUM VULGARE. In Atti R. Accad. Lincei, Rend. Cl. Sci. Fis., Mat. e Nat., v. 18, sem 
2, no. 8, p. 283-287. 1909. Abstract in Chem. Abs., v. s, no. 6, p. 1113. 1911. 

' SWANSON, C. O. HYDROCYANIC ACID IN SUDAN GRASS AND ITS EFFECT ON CATTLE In Jour. Amer. 

Soc. Agron. , v. 13, no. i, p. 33-36. 1921. 



I30 



Journal of Agricultural Research voi. xxii. no. 3 



TablR YII. — Disappearance of hydrocyanic acid from Sudan grass after cutting 



Date collected. 



Sample 
No. 



Treatment. 



HCN. 



June 29 

July 7. 

Aug. 12 
Sept. I. 



\Vilted in shade 

Green , control sample 

Wilted in sun for three hours 

Dried in sun from morning till evening, outdoors over 

night 

Dried in shade for same length of time as 2 

Dried outdoors two days and nights 

Dried in the shade two days and nights 

Dried in shade tliree da)'s and nights 

Dried in the shade five days 

Dried in the shade two days 



Mgm. 



28 
36 



15 
24 

IS 

7 

20 

32 
6 



While these results are not uniform, they do show conclusively that 
hydrocyanic acid can be obtained from wilted grass. Because of this 
result an attempt was made to determine more accurately the amount of 
hydrocyanic acid that may be obtained from wilted and dried grass. 

EFFECTS OF KEEPING GREEN GRASS MOIST AFTER IT IS CUT 

A large sample of grass collected June 29 was placed stems down in a 
large bottle so that about one-fourth was immersed in water. At the 
end of different periods of time 200-gm. portions of the leaves were mac- 
erated and digested in water overnight. The amount of hydrocyanic 
acid obtained is given in Table VIII. 





Table VIII. 


— Effect of keeping grass moitt after cutting 




Sample 
No. 


Length of treatment. 


HCN. 




6 hours 


Mgm. 
32 
20 


2 


22 hours ' 


3 

4 


30 hours 


8 


48 hours 


2 







The results indicate that hydrocyanic acid slowly disappears from the 
grass after it is cut, but also that the grass may be kept for a while in the 
green condition without much loss of the hydrocyanic acid. Control 
samples taken at this time gave 30 mgm. hydrocyanic acid. 

When the grass was wholly covered with water or when the air was 
excluded the results were different. In each case in the experiment re- 
ported in Table IX, unless otherwise stated, the grass was macerated and 
digested at the end of the treatment given. 

This shows that the presence or absence of air has an intimate relation 
to the evolution of hydrocyanic acid. Experiments were performed in 
which the grass was kept in an atmosphere of carbon dioxid and also of 



Oct. IS, 1921 



Hydrocyanic Acid in Sudan Grass 



131 



hydrogen. No hydrocyanic acid was obtained from the grass kept in an 
atmosphere of hydrogen, whereas from that kept in carbon dioxid con- 
siderable amounts were obtained. Some macerated grass was also placed 
in a desiccator from which the air was exhausted continuously. This 
did not seem to affect the amount of hydrocyanic acid obtained, but the 
experiment was not satisfactory. The effect of keeping the grass in 
different atmospheres needs further study. 

Table IX. — Effect of different treatments after cutting on hydrocyanic acid content 



Date collected. 



July 8. 



July 10. 



Treatment. 



'Placed uncut in bottles and covered with water two days 

Distillate from this water 

Placed uncut in sealed mason jar with small amount of water 
two da^'s 

Placed uncut in sealed mason jar with small amount of chlo- 
roform two days 

Placed uncut in sealed mason jar for 2 days, no water 

Placed uncut in bottle 2 days, covered with water 

Obtained by distilling water from this 

Macerated and digested 2 daj^s in water 

<! Macerated and placed in bottle 2 days then water added and 

distilled 

[Macerated and placed in open pan 2 days 



HCN. 



Mgm. 



12 
Trace. 



EFFECT OF HOT WATER 

To determine this relation, enough grass was cut to make twenty-four 
200-gm. portions of leaves. After the preliminary treatments as indi- 
cated in Table X, one set of 12 samples was digested in cold water and 
another set of 12 samples in hot water. 

Table X. — Effect of adding hot water on amount of hydrocyanic acid obtained 



Time of 

drying in 

shade. 

(hours). 



7- 
28 



Treatment before digestion. 



Uncut 

Cut in feed cutter , 

Macerated 

Uncut 

Cut in feed cutter . 

Macerated 

Uncut 

Cut in feed cutter. 

Macerated 

Uncut 

Cut in feed cutter. 
Macerated 



Time of digestion. 



5 hours. . . . 

do 

do.... 

24 hours. . . 

do.... 

do.... 

Overnight. 

do.... 

do.... 

30 hours. . . 

do.... 

do.... 



HCN obtained after 
adding — 



Water at 
room tem- 
perature. 



Mgm. 



10 

7 

14 
34 

8 
12 

19 

4 

10 

19 



Boiling 
water 



Mgm. 



Trace. 



Trace. 
2 
2 



65508°— 21- 



132 



Journal of Agricultural Research voi. xxn.No. 3 



This shows that it is possible to obtain some hydrocyanic acid from the 
uncut green grass if the time of digestion is sufficiently long. In every 
case more was obtained when the material was cut in the feed cutter and 
still more when it was macerated. Hot water placed on the green mate- 
rial entirely prevented liberation. The small amount obtained from the 
partially wilted grass when the hot water was added was probably in a 
free condition at the time of adding the hot water. Almost as much 
hydrocyanic acid was obtained from the grass that was wilted seven hours 
as from the fresh grass if digested in water at room temperature suffi- 
ciently long. 

This experiment as well as several others show that under some 
circumstances it is possible to obtain hydrocyanic acid from wilted or 
dried grass both with and without digestion in either hot or cold water. 
To investigate this further the following experiment was planned and 
executed. Five sets of 12 samples were secured and treated as follows: 
(i) Dried in the sun; (2) dried in the shade; (3) exposed in the sun, 
but kept moist by frequent sprinkling with water; (4) exposed in the 
shade but kept moist by sprinkling with water; (5) frozen in an ice 
machine and then exposed in open pans in the shade. The duration 
of these treatments was for 4, 8, 24, 31, and 48 hours, respectively. 
Six of the samples from each set were macerated after the period of 
the preliminary treatment, and hot water was poured on and distilled 
at once. The other six were digested in cold water overnight and then 
distilled. The results are shown in Table XL 

Table XI. — Rate of disappearance of hydrocyanic acid from Sudan grass after it is cut 

and variously handled <i 



Time of treatment. 


Hours 
of 

pre- 
limi- 
nary 
treat- 
ment. 


Dried in sun, 
treated with — 


Exposed in 
sun but kept 

wet and 
treated with— 


Dried in shade 

and 
treated with — 


Exposed in 

shade but 

kept wet and 

treated with — 


Frozen before 

exposed in 

shade and 

treated with— 




Hot 

water. 


Cold 
water. 


Hot 
water 


Cold 
water 


Hot 
water 


Cold 
water. 


Hot 
water. 


Cold 
water 


Hot 
water 


Cold 
water 


9 a. m first day 



4 
8 

24 

31 
48 



6 
5 
6 

'A 
Trace. 


16 
12 
10 
10 
3 

Trace. 







4 
I 

6 

5 


16 
14 

lO 



5 
7 





14 


3 
Trace. 


I 




g 


9 a. m. first day to i 
p. m. first day 


Trace. 

8 
I 

Trace. 




4 
8 
4 
6 
Trace. 


8 


9 a. m. first day to s 
p. m. first day 






6 


9 a. m. first day to 9 
a. m. second day 

9 a. m. first day to s 
p. m. second day .... 

9 a. m. first day to 9 
a. m. third day 


9 

8 
4 


5 
8 
10 


Trace. 

Trace. 

10 



" The figures indicate milligrams of hydrocyanic acid from 200 gm. of grass and are averages of several 
determinations. 

The results show that no hydrocyanic acid is obtained from green 
material when treated with hot water very soon after cutting and mac- 
erating, but that when the grass is wilted as much as four hours in the 
sun, considerable hydrocyanic acid is obtained by treating with hot 
water immediately after maceration. The amount of hydrocyanic acid 



Oct. IS. I92I Hydrocyanic Acid in Sudan Grass 133 

obtained was not greater when the grass was wilted for a longer time. 
Less hydrocyanic acid is obtained from grass that is kept moist while 
in the sun than from grass that is allowed to dry rapidly. According 
to Ravenna and Zamorani ^ the nitrogen passes through the following 
stages in the plant: Nitrate -^ hydrocyanic acid — > amino substance 
— ^ protein substance. According to this theory the cells which con- 
tinue to be active use the hydrocyanic acid for the building of protein 
substance, and as more nitrates from the soil are not supplied for man- 
ufacture of more hydrocyanic acid, the potential amount present when 
the plant is cut is soon exhausted. 

When the grass was dried slowly in the shade the hydrocyanic acid 
disappeared more slowly than when it was dried in the sun, and the 
amount obtained from the hot-water treatment became approximately 
equal to that obtained from the longer digestion in cold water. This 
seems to mean that when the plant wilts the hydrocyanic acid is split 
off from glucocids and held in such loose combination that it can be 
set free by hot water and that practically all the hydrocyanic acid is in 
such combination, since additional amounts can not be obtained by 
further digestion. Splitting off begins as soon as the plant is cut. 
Determinations made on grass kept moist in the shade appear to show 
that after 24 hours all the hydrocyanic acid not otherwise used by the 
cells is in such a condition that it is soluble in water. 

In the test in which the grass was frosted the hydrocyanic acid dis- 
appeared very rapidly, though the results were not very consistent. 

EFFECT OF ACIDS 

On June 24 a sample of grass was placed in a flask after maceration, 
covered with water, and sulphuric acid was added to acid reaction. 
After it was digested and distilled as usual only a trace of hydrocyanic 
acid was obtained. From a sample of like material and similarly 
treated, except that no acid was added, 27 mgm. were obtained. On 
June 29 this experiment was repeated with the result that 8 mgm. 
were obtained when acid was used and 26 mgm. when it was not used. 
On June 30, i and 28 mgm. were obtained by these respective treat- 
ments. These experiments clearly indicate that the presence of acid 
has a very importance influence on the amount of hydrocyanic acid that 
may be obtained. To test the effect of the amount of acid used, four 
samples were prepared on July 14 and digested overnight in the fol- 
lowing: (i) water; (2) N/o.i sulphuric acid (HjSOJ; (3) N/0.2 sul- 
phuric acid; (4) N/i sulphuric acid. No hydrocyanic acid was ob- 
tained from any of the treatments with sulphuric acid, whereas the 
water digestion gave 30 mgm. On August 4 this experiment was 
repeated, using a weaker acid solution. Digestion in water gave 10 

1 Ravenna, C., and Zamorani, M. nuove ricerche sdlla funzione fisiologica dell' acido cian- 
iDRico NEL SORGHUM VULGARE. In Atti R. Accad. Lincei, Rend. Cl. Sci. Fis., Mat e Nat., v i8, sem. 2, 
no. 8, p. 283-287. 1909. Abstract in Chem. Abs. v. 5, no. 6, p. 1123. 1911. 



134 



Journal of Agrictdtural Research 



Vol. XXI. No. 3 



mgm. of hydrocyanic acid; N/o.oi sulphuric acid, 1 1 mgm. ; and in N/ 0.02 
sulphuric acid, 4 mgm. On August 2 three samples were prepared and 
digested in N/i vSulphuric acid; in N/0.2 sulphuric acid; and in N/0.05 
sulphuric acid. Just before distillation, sodium hydroxid was added to 
almost neutral reaction. From the N/0.05 sulphuric acid 18 mgm. of 
hydrocyanic acid were obtained; a trace was obtained from the N/0.2, 
and none from normal. The weakest of the acid solutions gave no 
more than water alone. The smaller amounts obtained from the water 
treatments at the later date is in accord with the general observation 
that as the season advanced less hydrocyanic acid was present. It 
was planned to determine the exact hydrogen-ion concentration at 
which the hydrocyanic acid is most easily split off, but time did not 
permit. It is hoped that this may be determined in the future. 

It was shown in connection with the hot-water treatment that when 
grass dries the hydrocyanic acid is changed into a free condition, so that 
simply adding hot water and distilling will drive off the hydrocyanic acid. 
To see if more would be driven off if acid was also present the following 
experiment was performed. Six samples of leaves were placed in the 
open in clear weather from 9 a. m. till 9 a. m. the next day. Then they 
were macerated and digested in water and in different concentrations of 
sulphuric acid. The results are given in Table XII. 

Table XII. — Effect of acid solutions in formation of hydrocyanic acid 



HiS04 added. 



Nil... 
NI0.2. 
NI0.05 



HCN ob- 
tained. 



Mgm. 



H2S04 added. 



NI0.O2 
NJO.OI 

Water. 



HCN ob 
tained. 



Mgm. 

Trace. 
10 
18 



Thus, it appears that sulphuric acid is unfavorable to the liberation of 
the hydrocyanic acid even in the wilted material. 

To determine whether hot sulphuric acid would liberate the hydro- 
cyanic acid, hot water and sulphuric acid of varying normalities were 
added to green material immediately after maceration on July 16. The 
results are shown in Table XIII. 

Table XIII. — Effect of hot sulphuric acid on formation of hydrocyanic acid 



H2SO4 added. 



Nil.. 
NI0.5 
NI0.2 
NIo.i 



HON ob- 
tained. 



Mgm. 

Trace. 



Trace . 



H2SO4 added. 



NI0.05 

NI0.02 

NIo.oi. .. . 
Hot water 



HCN ob- 
tained. 



Mgm. 

O 

Trace. 
Trace. 



Oct. 15, 1921 



Hydrocyanic Acid in Sudan Grass 



135 



This shows that the use of hot acid is similar to that of hot water and 
that acid has no power to spht off the hydrocyanic acid, at least in the 
concentration used. The traces obtained in some cases were no doubt 
due to liberation of hydrocyanic acid during maceration. It would, 
appear, however, that hot water was less destructive than hot acid. 
I/ike experiments with hydrochloric acid were performed with similar 

results. ,,^.. ^.r>{' .r..>!!;.-:i r..,^ 

On July 2 1 eight samples were prepared and digested at room tempera- 
ture in phosporic acid (Table XIV) . 

Table XIV. — Effect 0/ phosphoric acid on liberation of hydrocyanic acid 



H3PO4 added. 



Nji.. 
NI0.5 
NI0.2 
Njo.i 



HCN ob- 
tained. 



Mgm. 



H3PO4 added. 



NI0.05 
N 1 0.02 
Njo.oi 
Water. 



HCN ob- 
tained. 



Mgin, 



The results indicate that the inhibiting power of phosphoric acid 
(H3PO4) was somewhat less than that of hydrochloric (HCl) or sulphuric 
acid. This would be expected since the degree of ionization of phosphoric 
acid is less than that of hydrochloric or sulphuric acid. Experiments 
with tartaric acid gave similar results. 

EFFECT OF DIGESTING IN ALKALINE SOLUTION 

On July 20, 16 samples were prepared and digested in sodium-hydroxid 
(NaOH) and sodium-carbonate (Na2C03) solutions, respectively (Table 
XV). 

Table XV. — Effect of alkaline solution on formation of hydrocyanic acid 



NaOH added. 



Nji. .. 
NI0.5. 

NjO.2. 

NIo.i. 
Njo.05 
N 1 0.02 
Njo.oi 
Water. 



HCN 
obtained. 



Mgm. 



19 



Nas CO3 added. 



Nil... 
NI0.5. 
NI0.2. 
NIo.i . 
NI0.05 
NI0.02 
NIo.oi 
Water . 



HCN 
obtained. 



Mgm. 

o 

o 

o 

o 

Trace. 

17 

17 

20 



The results show the same general effect as that secured with acid 
solutions. 



136 



Journal of Agricultural Research voi. xxii. No. 3 



EFFECT OF ACID OR ALKALI ON HYDROCYANIC ACID AFTER IT IS 

LIBERATED 

An experiment was performed to show what effect acid or alkaline 
solutions have on the hydrocyanic acid after it is liberated. The green, 
macerated material was digested overnight in measured amounts of 
water. Enough standardized acid or alkali was then added to give the 
normality desired, and distilled. The results are given in Table XVI. 

Table XVI. — Effect of acid and alkali on hydrocyanic acid after it is liberated 



Solution added. 



H2SO4 
Hcl . . . 
H3PO4 
NaOH 



HON obtained after treatment with solutions of- 



Nlr. 



Mgm. 



9 
16 



Nlo.g. 



Mgm. 



N/o.i. Njo.os. 



Mgm. 



16 

20 

2 



Mgvi. 



14 

18 

18 

O 



Water. 



Mg7n. 



20 
20 



No hydrocyanic acid passed over in the first distillate from the sodium- 

hydroxid solution. The mixture was acidified with sulphuric acid and 

then distilled with the results given in Table XVI. While the results 

obtained in this experiment are not very uniform, they do show that 

hydrocyanic acid can be obtained from acid and alkaline solutions if the 

hydrocyanic acid is in a free condition before the acids are added. The 

experiment also appears to show that the addition of acid or alkali 

resulted in diminishing the amount of hydrocyanic acid obtained. The 

experiment was also tried by digesting the grass in sulphuric acid and 

sodium hydroxid of the normalities N/i, N/0.2, N/o.i, and N/o.o^ and 

then neutralizing before distilling. In no case was any hydrocyanic acid 

obtained. 

INFLUENCE OF WEATHER 

From a sample taken June 23, when there had been no rain for three 
weeks, 27 mgm. of hydrocyanic acid were obtained. On June 30, after 
a heavy rain and a week of good growing weather, during which there 
was plenty of moisture, 30 mgm. were obtained. On July 24, when 
there had been a period of dry weather, the amount obtained was 7 mgm. 
The next day, following a rain during the night, the amount was 16 mgm. 
Several experiments indicated that the largest quantity was obtained 
when the plant was in the most vigorous growing condition. This is 
contrary to a common belief that stunting has some effect in increasing 
hydrocyanic acid. On the contrary, the potential amount may be 
lessened. Determinations were made on samples collected at sundown 
and also before sunrise. The data obtained were not conclusive in deter- 
mining the effect of light on the potential amount of hydrocyanic acid 
present. 



Oct IS. 1021 Hydrocyanic Acid in Svdan Grass 137 

HYDROCYANIC ACID IN SUDAN HAY 

Two samples were taken from the outside of a stack of Sudan hay 
and two from the inside. No hydrocyanic acid was found. 

AMOUNT OF HYDROCYANIC ACID IN OTHER SORGHUMS 

On July 23 a sample of kafir was taken and separated into leaves 
and stems. From the leaves were obtained 16 mgm., and from the 
stems 10 mgm. of hydrocyanic acid. The kafir stems were very little 
developed. Sudan grass, tested the same day, gave 8 mgm. of the acid 
from the same weight of material. On July 26, just after a heavy rain, 
following a period of dry weather, a sample of kafir gave 72 mgm. and a 
sample of sorgo (cane) 42 mgm. of hydrocyanic acid. Sudan grass 
6 inches high, tested on that date, gave 32 mgm. of hydrocyanic acid. 

On August 7 a sample of second-growth sorgo (cane) was received 
from LaHarpe, Kans. About one-fifth was quite dry, two-fifths were 
wilted and yellow, and two-fifths were green. The sample was some- 
what moldy. One portion digested in the usual manner gave 13 mgm. 
hydrocyanic acid. Another portion distilled at once from hot water 
gave 24 mgm., showing that the hydrocyanic acid was in free condition. 
Another sample of sorgo was sent in from Seneca, Kans. This was 
reported to have killed six cows. From the portion distilled from hot 
water 20 mgm. were obtained and from the portion digested in the 
usual way 36 mgm. 

On September 2 a quantity of Red Amber kafir was collected, and 
six portions were prepared, and treated with the results given in Table 

XVII. 

Table XVII. — Hydrocyanic acid in Red Atnber kafir 



Sample 

No. 



Treatment. 



Left in flask 15 minutes after maceration, after which hot water 

was added and distilled 

Digested overnight in water ^^,.^.,^,^. 

Digested overnight in N/o.^ H2SO4. ...-.'.?..'.. 

Digested overnight in Njo.i HjSO^ 

Digested overnight in N/o.oi H2SO4 

Digested overnight in N/o.oi NaOH 



Mgm. 

32 
119 

None. 

None. 
40 
36 



Part of this experiment was repeated by putting macerated sorgo into 
boiling water at once. This gave 8 mgm. hydrocyanic acid, while that 
digested overnight gave 96 mgm. Another portion was divided into five 
portions. After maceration they were all digested in water overnight. 
Then to these portions standardized sulphtuic acid was added so as to 
make the normahties indicated. The results are given in Table XVIII. 

These determinations show without a doubt that sorgo and kafir con- 
tain much larger amounts of hydrocyanic acid than does Sudan grass, 
and also that the conditions for obtaining it are very similar. 



138 Journal of Agricultural Research voi. xxii. No. 3 



Table XVIII. — Effect of different treatments on liberation of hydrocyanic acid in sorgo 



Sample 

No. 



Treatment. 



HCN. 



Digested in water and distilled 

Digested in water and distilled 

Digested in water and distilled from Njo.^ H2SO4. . . 
Digested in water and distilled from NIo.i H2SO4. . . 
Digested in water and distilled from NI0.02 H2SO4. . 



Mgni. 

72 
80 
64 
72 
89 



EFFECT OF HYDROCYANIC ACID FROM GREEN SORGO ON A HORSE 

Ten-pound portions of green sorgo, testing the amount of hydrocyanic 
acid given in Table XVIII, were fed to a horse. No effect on respiration, 
pulse, or temperature could be observed by Dr. H. F. Lienhardt, of the 
Veterinary Division, who made the observations. Data presented in 
this paper show that such a degree of acidity as is found in the stomach 
of a horse would prevent liberation of hydrocyanic acid from the green 
material. Feeding wilted sorgo was not tried. 

SUMMARY 

(i) In this paper are presented data givmg the results of tests made 
on Sudan grass for hydrocyanic acid during the summer and early fall of 
1920. 

(2) The maximum amount of hydrocyanic acid was obtained by mac- 
erating the material and digesting in water at room temperature for about 
six hours or overnight. 

(3) Practically all the hydrocyanic acid was found in the leaves. In 
well-developed stems none was found. 

(4) More hydrocyanic acid was found in younger plants than in those 
more mature. This is due mostly to stem development. If leaves only 
are used the dififerences are small, except when the plants approach 
maturity. More was found in the summer than in the fall. 

(5) Hydrocyanic acid does not exist as free HCN in the growing plant. 
It begins to be liberated as soon as the plant is macerated or undergoes 
wilting. 

(6) Liberation of hydrocyanic acid is intimately associated with enzym 
action. If this enzym action is inhibited by addition of hot water or 
acids, no hydrocyanic acid will be liberated. Hydrocyanic acid was 
obtained from wilted grass when hot water was added, because during 
the wilting process hydrocyanic acid was set free. 

(7) Hydrocyanic acid can not be set free from the green material by 
acids. 

(8) The action of strong alkali is similar to that of acids. 

(9) Most hydrocyanic acid is present when the plant is in a vigorous 
growing condition. 

(10) Sudan grass contains less hydrocyanic acid than sorgo or kafir. 



NUTRIENT REQUIREMENTS OF GROWING CHICKS: 
NUTRITIVE DEFICIENCIES OF CORN^ 

By F. E. MUSSEHL, Professor of Poultry Husbandry, J. W. Calvin, Associate Chemist, 
Nebraska Agricultural Experiment Station, with the cooperation of D. L. Halber- 
SLEBEN and R. M. Sandstedt 

Investigators in the field of nutrition have noted that chickens behave 
unlike rats and swine when limited to rations of com or wheat grains and 
their products. This fact has made necessary the planning and execu- 
tion of experimental work having for its object a determination of the 
values and deficiencies of our common feeding stuffs when used for poultry 
and egg production. The results of a series of experiments carried on at 
this Station with this objective are reported in this paper. 

From the experience of investigators ^ who have worked with other 
species, mainly rats and swine, it has seemed that systematic inquiry 
should be made into the (a) ash re- 
quirements, (b) protein requirements 
(quality and quantity), and (c) food 
accessory requirements. Earlier in- 
vestigational work with chicks by 
Osborne and Mendel ^ and Hart, 
Halpin, and Steenbock* indicates that 
another element, (d) the physical 
factor, is also of fundamental impor- 
tance and must be considered in any 
complete study of the nutritive values 
of a particular grain or ration. 

In our work lo-day-old vS ingle-Comb 
White Leghorn chicks were used, spe- 
cial care being taken to select for vigor, 
vitality, and uniformity in each lot. 
Nine chicks per lot were used for the 
first series of experiments. Chicks were weighed individually every 
seven days, and the growth curves selected are typical of each lot (fig. 
i-ii). They show the weight of the chicks at the beginning of the 
experiment and the change in weight thereafter. Records of the feed 

1 Published with the approval of the Director of the Nebraska Agricultural Experiment Station. 

'McCoLLUM, E. v., SiMMONDS, N. , and PiTz, W. the relation of the UNroENTiFiBD dietary fac- 
tors, THE fat-soluble a, AND WATER-SOLUBLE 6, OF THE DIET TO THE GROWTH-PROMOTING PROPERTIES 
OF MILK. In Jour. Biol. Chem., v. 27, no. i, p. 33-43, 6 charts (1-3, 6 in text). 1916. 

'Osborne, Thomas B. , and Mendel, Lafayette B. the growth of chickens in confinement. In 
Jour. Biol, chem., v. 33, no. 3, p. 433-438, pi. 4-6. 1918. 

^Hart, E. B., Halpin, J. G., and Steenbock, H. use of synthetic diets in the growth of baby 
CHICKS. A study of LEG WEAKNESS IN CHICKENS. In Jour. Biol. Chem., v. 43, no. 2, p. 421-442, 2 pi. 1920. 




o 
100 



—- 


» — 


1125 




— * 


X 






/oo 







■:HiaK 1 

//oa I 




X 








,3S ^2 



Fig. I.— Graph showing unsatisfactory results 
from feeding ration of 100 parts yellow com 
and calcium carbonate grit ad libitum to 
chicks of lot II. The time at which chicks 
died is indicated by X. 



Journal of Agricultural Research, 

Washington, D. C. 

zy 



(139) 



Vol. XXII. No. 3 
Oct. 15, 192 1 
Key No. Nebr.-4 



140 



Journal of Agricultural Research voi. xxii. No. 3 



(300 




































y 




































y 


y 


































/ 


































< y 


/ 


































f/ 


V 










300 






















A 


H/ 


y 












Jj 700 

\ 

\ 600 

SOO 




















a 


4 


y 






























//^i 


'} 


/ 






























d 


y 




















300 














J- 


X 






























/ 


X'' 
































^ 


-^ 

































^ 


-^ 




































Fio. a.— Graph showing slow but continuous growth of chick in lot 211, fed ration of 95 parts yellow com 

and 5 parts ash mixture 



Nutrient Requirements of Growing Chicks 141 



/.soo 



A 200 



/,/oo 



/.ooo 



900 





































/ 


































/ 


z' 
































/^ 


/ 






























^ 


/ 
































y 


/ 
































/f'^ 


/ 






























, ( 


i'^ 


/ 






























ii/i 


^ 


/ 




























f 


i^ 


3> 




























.( 


)f^(i 


J^ 


^ 






























^ 


i^ 




























_,--- 


^ 


^' 






























^ 





































5<?(S' 



I 

eoo 



300 



2O0 



fOO 



700 


































^ 


^ 
































^ 


>*^ 
































.-^ 


:^ 




























r.K 


2jj 


^ 


^ 






























0\ 


'■^rf 


lA 


^ ^ 




















^J 


^ 








,a£^ 




u> 























— 


I*""*' 

































O 7/^2/ Ze 3S -^2 4& se 63 70 77 S4 S/ S8 /OS //2 //^ /Z6 

Fig. 3. — Graph showing growth of chicks in lot 213, fed ration of 80 parts yellow com, 15 parts casein, 

and 5 parts ash mixture. 



142 



Journal of Agricultural Research voi. xxu,no.3 



/,ooo 

9CX) 

soo 

700 

600 

t 500 



10 ^oo 



300 



200 



/OO 





























/ 


























y 


/ 
























/ 


/ 
























y 


/ 






















if?' 


ly 


/ 




















ki 


rjt^ 


0^ 




















A 1 


j\ 


c^ 


I^ 




















403 


J^ 


■y 






















^ 


i^ 



















































/OO 

o 



CH/CK 2736 




O 7 /4 2/ 2e 3S ^^2 ^& S6 63 70 77 S'^ S/ S3 

Fig. 4. — Graph showing detrimental results of adding 5 parts butter fat to ration of 65 parts yellow com, 
15 parts casein, 5 parts ash mixture, and 10 parts starch for chicks in lot 215. 



Oct. IS, 1921 



Nutrient Requirements of Growing Chicks 



143 



/ooo 
soo 

TOO 

^00 

I 

^00 
ZOO 

too 
o 























































/^ 


/■ 
























/ 


/ 
























y 


/ 






















.i(^^ 


f^ 


/ 




















.A 


t' 


^ 






















(?/?! 


?i^ 


y 


















1/ 


)(^t 


^ 


^ 




















*^^ 


J^ 


'y 




















































O 7 /^ 2f 23 3,^ *Pe '^9 c5<5 G3 70 77 S^ S/ &3 

Fig. 5.— Graph showing that the addition of 20 parts com gluten did not improve ration of 65 parts yellow 
com, 5 parts ash mixture, and 10 parts starch for chicks of lot 208. 



144 



Journal of Agricultural Reserach voi. xxn. no. 3 



I 



900 

aoo 

700 

600 
\soo 

\fOO 

300 

200 

/oo 

o 



























y 


y 
























/ 


/ 






















, 


■y 


/ 






















w 


/^ 


'/ 






















A 


^ 






















aI"^ 


^ 


v> 




















^iQ 


^ 


^ 




















^^ 


^ 


P 






















•^ 


.*-' 



























300 



^ zoo 



O 7 /"f 2/ £3 3S ^2 -^9 S6 63 70 77 3^ 9/ 93 

Fig. 6.— Graph showing that the addition of 5 parts butter fat (fat-soluble A) did not improve ration of s 
parts yellow corn, 20 parts corn gluten, 5 parts ash mixture, and 5 parts starch for chicks of lot 209. 































X- 


j/r 


K 


P7f 


j^.. 








— 




-*- 








6/ 




rff< 


27. 


'^z' 





















Oct. IS, I93I 



Nutrient Requirements of Growing Chicks 



145 



'POO 



200 



/oo 



A 

I o 
%zoo 












_ 




-X 


"^ 


TWCK 


27^ 











consumption of each lot were also obtained. Clean wood shavings 
were used for litter, and each lot was confined to a pen 2 by 8 feet 
in size until the chicks were 8 weeks old, when the near normal lots 
were given a yard 4 by 8 feet 
in size. 

Methods of feeding are known 
to have considerable influence on 
the efficiency of a ration; so a 
standard policy was established of 
dividing the ration into two parts — 
the scratch or coarse feed and the 
mash or fine feed. The rations 
were mixed so that equal quantities 
of mash and scratch feeds were 
provided. The mash feed con- 
tained all the supplemental ingre- 
dients, such as ash, butter fat, and 
purified casein. 

The casein used in the rations was 
purified by extracting repeatedly 
with distilled water slightly acid- 
ified with acetic acid . After extrac- 
tion the casein was drained and 
dried in an air oven at 70° to 100° C. and then ground. The butter fat 
was purified by melting at 40° in a water bath and was then centrifuged 
to remove ash, casein, and other material. The ash mixture^ used in 
our rations was composed of the following ingredients, parts by weight: 

Bone ash .jV^/:'^: 't';^;! . V'^'v'. .'. 50 

Calcium carbonate 

Sodium chlorid 

Dipotassium phosphate 

Calcium lactate 

Magnesium sulphate 

Sulphur 

Iron sulphate 




^i9 



Fig. 7. — Graph showing that the addition of is parts 
soybean meal did not improve ration of 80 parts 
yellow corn and 5 parts ash mixture for chicks of 
lot 210. The time at which chicks died is indicated 
by A'. 



JXi.i . U. 



"■^^rA. 



14 
IS 
10 

5 
3 
2 

I 



The results of our inquiries may briefly be summarized as follows: 
(i) Yellow com (maize) is deficient in several of the essential qualities 
necessary for the complete nutrition of growing chicks. A deficiency in 
the ash content of the yellow com kernel is no doubt responsible for the 
early failure of baby chicks when restricted to a ration of corn alone. 
Supplementing the corn kernel with 5 per cent of a complete ash mix- 
ture improved the ration so as to enable very slow but persistent growth. 

1 Philips, A. G., Carr, R. H., and Kennard, D. C. meat scraps versus soy-bean proteins as a 
StJPPLEMENT TO CORN FOR GROWING CHICKS. In Jour Agf. Research, v. i8, no. 7, p. 391-398, i fig. , pi. 50. 
1920 



146 



Journal of Agricultural Research 



Vol. XXII, No. 3 



o 

JOO 

% ° 
S 100 

o 

JOO 



•^ 


frH 




CHICK 

Vr\3 


— X 




C»/Ch 
Z770 


"■"^x 




CHICK 

zeas 


— X 



£>i9y.s 

Fig. 8.— Graph show- 
ing bad results from 
lack of roughage in 
ration of 65 parts 
yellow com, 15 parts 
gelatin, 5 parts but- 
ter fat, 5 parts ash 
mixture, and 10 
parts starch for 
chicks of lot 217. 
The time at which 
chick died is indi- 
cated by X. 



(2) Yellow corn is deficient in quality and quantity of protein required 
for normal growth of chicks. The addition of more com protein by 
including corn gluten in the ration did not markedly 
improve the efficiency of the ration. Compare growth 
curves, lots 208 and 211. 

(3) The addition of 15 per cent purified casein to a 
basal ration of yellow corn and ash did improve the 
ration decidedly. Compare growth curves, lots 211 and 
213. The amino acid deficiencies of the corn proteins 
are no doubt supplemented by the amino acid contribu- 
tions of the casein. 

(4) Supplementing the basal yellow com ration with 
certain other proteins, ^^g albumen, and gelatin, lowered 
rather than raised the efficiency of the ration. The poor 
results with rations 217 and 218 were probably due to 
a distinctly sticky physical quality which prevented 
normal nutrition. 

(5) The fat-soluble food accessory does not appear to 
be a limiting factor in a yellow com diet for baby 
chicks. The addition of butter 
fat to a yellow corn, casein, 

and ash ration (lot 215) did not improve but 
rather lowered the efficiency of the ration. A 
slight change in the physical condition of the 
ration may explain this lowered efficiency, 
though it is more probable that the butter fat 
addition temporarily stimulated growth so that 
the supply of some other essential accessory 
was exhausted earlier than would have been 
the case had the butter fat been omitted. It 
is apparent at least that the failure of chicks 
on ration 215 was not due to fat-soluble A 
starvation. 

(6) Green feeds make certain very valuable 
contributions to a ration for growdng chicks. 
The addition of wheat greens to a yellow com, 
casein, and ash ration effected a decided im- 
provement in the efficiency of the ration. An 
excess of the wheat greens was offered, and 
subsequent observations indicate that about 5 
per cent (dry matter basis) of this kind of green 
food are consumed when offered regularly in excess. The helpful influ- 
ence of the wheat greens may have been due to, first, an improvement 
of the physical condition of the ration; second, a food accessory con- 



zoo 



/oo 



o 
200 



/oo 

I o 
^200 

/oo 



o 
/oo 





_„--' 


5^' 


^0 


^A//'^ 














^x 


Thi^ 


$73? 













>v 


Thi^ 


7^ 




V 



^ 


Z73\ 


^X 





o 



/«? 2/ 



za 



Fig. 9.— Graph showing that be- 
cause proper physical quality 
was lacking the addition of 15 
parts egg albumen did not im- 
prove ration of 65 parts yellow 
com, 5 parts butter fat, s parts 
ash mixture, and 10 parts starch 
for chicks of lot 218. The time 
at which chicks died is indi- 
cated by X. 



Nutrient Requirements of Growing Chicks 



H7 



/.■300 



/.OOO 



900 



eoo 

<f) 700 

\ 

































1 — 




^ 


































^ 
































/ 






























/ 
































y 


/ 






























& 


/ 
































f> 


y 






























ndi 


if 


/ 




























A 


, 6 


^ 






























^LQ^ 


i^ 


y 






























y" 


i> 






























^ 


^ 






























^ 





































soo 



400 



200 



/OO 



eoo 



600 



.soo 



I 

Vi 300 



400 

































^ 


^ 


^- 


























<:'.. 


H 


^ 




























* 


♦;-^ 


/ 






























'f.i 


fy 


y 




























^ff 


f' 


^ 


-3 






















. o 


^$'i. 


W'^ 


^^ 


ii 
^ 


€^ 


c> 


















^ 


^ 


'0. 


l'^ 


i* d 






0' 






c 

















































zoo 

/OO 



~ O 7 /4 2/ 23 33 42 49 S6 63 TO 77 34 S/ S3 /OS //2 //S /ZS 

Fig. io. — Graph showing that the addition of excess of wheat greens improved ration of 8o parts yellow 
com, 15 parts casein, and s parts ash mixture for chicks of lot 227. 

65508°— 21 3 



148 



Journal of Agricultural Research voi. xxu,no.3 



/./oo 



/,ooo 



900 



800 



700 



(ft 600 

\ 

400 



300 



200 



/CO 

































/ 




























4 


/ 




























4 


7 




























4\ 


/ 


























( 


J 


/ 


























jH 


t^ 


y 


























t^ 


f 




























^ 


y 


























y 


^ 




























^' 


























^ 


































7 M 2/ Z8 3S f2 -^9 S6 63 70 77 S'^ 3/ 38 /OS //2 

Fig. II.— Graph showing normal growth produced by ration of 65 parts wheat, 15 parts casein, s parts 
butter fat, 5 parts ash mixture, 10 parts starch, and excess of wheat greens for chicks of lot 205. 
Birds were kept on this ration until they produced eggs, at age of about 200 days. 



Octis, I92I Nutrient Requirements of Growing Chicks 149 

tribution; third, an increased food consumption, due to a stimulatory 
effect on the appetite. A series of experiments to shed further hght 
on the specific contributions which green feeds make to the ration is in 
progress at this time. 

(7) It is possible to raise to normal maturity chicks confined to a small 
pen. Drummond ^ reports great difficulty in rearing chicks in confine- 
ment, and other investigators have noted some of the problems, especially 
leg weakness.' Our lot 205 grew to normal maturity, some of the pullets 
producing eggs when about 200 days old, though never having more 
range than was provided in a yard 4 by 8 feet in size. Ration 205, though 
not synthetic, is of interest because of its comparative simplicity. 

1 Drummond, Jack Cecil, observations upon the growth of young chickens under laboratory 
CONDITIONS. In Biochem. Jour., v. lo, no. i, p. 77-88, i pi. 1916. 

2 Hart, E- B., Hai,pin, J. G., and Steenbock, H. op. ai. 



AECIAL STAGE OF THE ORANGE LEAFRUST OF 
WHEAT, PUCCINIA TRITICINA ERIKS.^ 

By H. S. Jackson, Chief in Botany, and E. B. Mains, Associate Botanist, Purdue 
University Agricultural Experiment Station, and Agents, Office of Cereal Investiga' 
tions. Bureau of Plant Industry, United States Department of Agriculture ^ 

This paper presents, in part, the results of a study of the leafrusts 
of wheat, rye, barley, com, and related grasses which was begun in 191 8. 
One of the important phases of this investigation is the determination of 
the aecial relationships of the various races or species included in the 
collective species, Puccinia Clematidis (DC.) Lagerh. (P. Agropyri Ellis 
and Ev.), and other closely related forms. While a number of the rusts 
of this group which occur on wild grasses have been connected with 
aecia, their host limitations and interrelations are not well understood. 
This study is especially important in the case of the leafrust of wheat, 
P. triticina Eriks. So long as the aecial stage of this species was un- 
known, little progress could be made in developing our knowledge with 
reference to its origin, development, spread, and relation to other rusts. 
The results of the investigation of the aecial relationship of this rust 
are presented in the following pages. 

HISTORICAL REVIEW 

Three rusts are known to attack wheat : the black or stemrust, Puccinia 
graminis Pers. ; the stripe or yellow rust, P. glumarum (Schmidt) Eriks. 
and Henn. ; and the orange or leafrust, P. triticina. Of these the stem- 
rust is the only one for which the aecial stage has been determined. 
This rust was shown by De Bary to have its aecial stage on Berberis 
vulgaris L., and this relationship has since been demonstrated repeatedly 
by a number of workers in various parts of the world. The discovery 
of the place of Aecidium Berberidis Pers. in the life cycle of P. graminis 
caused De Bary (4, p. 207-211)^ to turn his attention to the study of 
other grass rusts having incomplete life cycles. This resulted in the 
discovery that P. rubigo-vera (DC.) Wint. (P. straminis Fckl.) on rye 
was connected with aecia on Anchusa officinalis and Anchusa arvensis. 
Sowings made with teliospores from rye resulted in the production of 

' Published with the approval of the Director as a contribution from the Department of Botany, Purdue 
University Agricultural Experiment Station. Cooperative investigation between the Purdue University 
Agricultural Experiment Station and the Office of Cereal Investigations, Bureau of Plant Industry, United 
States Department of Agriculture. 

' The writers wish to acknowledge their indebtedness to various pathologists throughout the country 
for aid in obtaining material for the cultural studies upon which this paper is based, and to Mr. Forest 
Fuller, temporary culture assistant during the spring of 1919, and Mr. Emile Mardfin and Miss Florence 
M. Smith, Scientific Assistants, Ofifice of Cereal Investigations, Bureau of Plant Industry, for assistance 
in carrying out the cultural investigations. 

' Reference is made by number (italic) to "Literature cited," p. 170-171. 

Journal of Agricultural Research, Vol. XXII, No. 3 

Washington, D. C Oct. 15, 1921 

zz Key No. G-347 



1^2 Journal of Agricultural Research v^oi. xxii. No. 3 

aecia on Anchusa, and when sowings were made with aeciospores from 
Anchusa, uredinia on rye were developed. Sowings of basidiospores from 
rye upon Berheris vulgaris Hoi., Rhamnus Frangida, Rhamus cathartica, 
Ranunculus acris, Ranunculus bulbosus, Taraxacum officinalis, and Urtica 
dioica were without result. Nielsen {20, p. 37) 10 years later reported 
obtaining infection with aeciospores from Anchusa officinalis on both rye 
and wheat. Plowright {21, p. 168) states that in the fall of 1885 he 
obtained aecia upon Anchusa arvensis by placing wheat straw rusted 
with P. rubigo-vcra near that host. 

At the time this work was carried out the name Puccinia rubigo-vera 
was used for the leafrusts of wheat, rye, and barley, as well as for similar 
grass rusts having globoid urediniospores and long covered telia. 
Eriksson and Henning (11, p. 197-203, 2^7-2^9) separated this species 
into two — Puccinia glumarum, the stripe rust, and Puccinia dispersa 
Eriks., the brown rust. Under the latter they included the rust of wheat 
as well as that of rye. The rust of wheat, however, was considered as a 
forma specialis, Tritici, of Puccinia dispersa. As the leafrust of rye had 
been shown by De Bary (4) to be connected with aecia on Anchusa, 
Eriksson (jo, p. 254-2^7) sought for the same connection for the leafrust 
of wheat. His sowings of basidiospores from wheat upon Anchusa 
officinalis and A. arvensis, however, produced no infection, as was also 
the case when aeciospores from Anchusa were sown on wheat. No results 
were obtained when basidiospores were sown on Nonnea rosea, Myosotis 
arvensis, M. alpestris, Symphytum asperrimum and Pulmonaria officinalis, 
species of Boraginaceae related to Anchusa upon which unconnected 
aecia were known to occur. As a result of these cultures, Eriksson {10, 
p. 270) concluded that the orange leafrust of wheat was a distinct species 
and gave it the name, Puccinia trificina. 

Klebahn (17, p. 85-86; 18, p. 246) made rather extensive cultures in 
an endeavor to discover the aecial host of Puccinia triticina. Besides 
sowing aeciospores of Aecidium Anchusae Eriks. and Henn. on wheat he 
made sowings of basidiospores on Anchusa arvensis and Anchusa officinalis 
without result. Sowings of basidiospores also were made without success 
upon Triticum vulgare, Ranunculus acer, Ranunaihis asiaticiis, Ranun- 
culus auricomus. Ranunculus bulbosus, Ranunculus Ficaria, Ranunculus 
flammula, Rammculus lanuginosus , Ranunculus repens, Anemone ranun- 
culoidcs, Aconitum Lycoctonum, Aconitum Napellus, Berberis vulgaris, 
Nasturtium sp., Barbaraea vulgaris, Melandryum albutn, Coronaria fl.os- 
cuculi, Agrostetnma Githago, Rhamnus cathartica, Lythrum Salicaria, 
Ribes Grossularia, Aegopodium Podagraria, Pastinaca sativa, Valeriana 
dioica, Knautia arvensis, Tussilago Farfara, Taraxacum, officinale, Cen- 
taurea Cyanus, Achillea Ptarmica, Campamila rotundifolia, Ligustrum 
vulgare, Phillyrea sp., Echium vulgare, Lithospermum purpureo-coeruleum, 
Myosotis sp., Symphytum officinale, Glechoma hederacea, Prunella vulgaris, 



Oct. IS, 1921 Aecial Stage of the Orange Leaf rust of Wheat 153 

Rumex acetosa, and Urtica dioica. These results substantiate those 
obtained by Eriksson and indicate that the positive results reported by 
Nielsen {20) and Plowright {21) were probably due to a mixture of rusts 
or of hosts. 

The failure to obtain infection on Boraginaceous hosts has influenced 
other workers to turn their attention to other families in a search for the 
aecial hosts. Arthur (j, v. 9, p. 304), largely as a result of morphological 
studies, reached the conclusion that Puccinia triticina was best con- 
sidered a race of P. Agropyri, and upon this basis Arthur and Fromme 
(j. P' 333~337) have placed it in the collective species Dicaeoma Clerna- 
tidis (DC.) Arth. Several races of this collective species had been shown 
by workers in Europe and America to go to species of Clematis. Arthur 
thought that the aecial host might be either Clematis ftamm.ula or C. 
vitalha as these were the only common species of Clematis found in the 
wheat-growing regions of southern Europe, northern Africa, and western 
Asia, a region which at that time was considered as the probable home 
of the original wild wheat. His culture with wintered telia of the leaf- 
rust of wheat on C. flammula, however, was unsuccessful. 

According to Butler (d, p. 75) Cunningham and Prain (9) considered 
that there was considerable ground for believing that an Aecidium on 
Launaea asplenijolia, one of the Cichoriaceae, was the aecial stage of 
Puccinia triticina, as it was found throughout the greater part of the 
wheat-growing area of India. Butler, however, sowed aeciospores from 
this host upon wheat without obtaining infection. 

These unsuccessful attempts to demonstrate an aecial stage for Puc- 
cinia triticina have resulted in the development of the idea that the 
aecial stage of this rust has been lost and that it is able to maintain itself 
without one. In this connection a number of important facts have 
been established and a number of interesting hypotheses proposed. It 
has been shown by BoUey (5, p. 13-14), Hitchcock and Carleton (15, p. 
1—2), Carleton {8, p. 21-22), and others that in certain regions, P. tri- 
ticina is able to overwinter by means of its uredinal mycelium and that 
no aecial host is necessary for the maintenance of this species. This 
does not appear, however, to be true for all regions where P. triticina is 
abundant {6, p. 11). A number of suggestions have been made to ex- 
plain the 5'early appearance of the rust in regions where the uredini- 
ospores or uredinial mycelium does not overwinter. It was considered 
possible that spores may be carried from other regions by the wind. 
The mycoplasm theory of seed transmissal has also been put forward as 
a possible explanation. Whatever may be the merits of these hypothe- 
ses, they have resulted in recent years in directing attention away from 
a search for the aecial host of this species. 



154 Journal of Agricultural Research voi. xxii, no. 3 



BASIS OF CULTURAL INVESTIGATIONS 

A study of Puccinia trUicina in comparison with other grass rusts with 
long covered teha shows that it can not be readily separated morpholog- 
ically from the leafrust of rye. The separation of this form as a species 
was made by Eriksson {10) because he obtained only slight infection on 
rye with urediniospores and was not able to obtain infection on Anchusa 
with basidiospores and because the teliospores germinated in the spring, 
while those of the rye rust germinated in the fall. The close morphological 
similarity, however, furnished considerable grounds for the assumption 
that the aecial host of leafrust of wheat was likely to be some species of 
Boraginaceae other than Anchusa, especially as another rust of this type, 
Puccinia hromina Eriks., has since been found to have its aecia on the 
Boraginaceous hosts Symphytum officinale and Pulmonaria montana, with 
very weak development of aecia on Anchusa {ig, p. 182-202). Unfin- 
ished investigations now being conducted in this laboratory strongly 
indicate that in America certain grass rusts having aecia on Boraginaceous 
hosts are very similar to the leafrust of wheat and rye. For these reasons 
it was considered desirable to test as many Boraginaceous hosts as were 
available, as possible aecial hosts for the leafrust of wheat. 

There is, however, still another group of grass rusts very similar to the 
orange leafrust of wheat to which Arthur (j, v. 9, p. 304) has called atten- 
tion. This group has aecia upon various Ranunculaceous hosts and in- 
cludes forms which have been separated from time to time, according to 
their aecial connection, together with slight morphological variation, into 
a number of species, including Puccinia persistens Plowr., P. perplexans 
Plowr., P. Agropyri, and P. alternans Arth, The writers felt from the 
beginning that the greatest possibility of success in the search for the 
aecial stage was to study thoroughly the genera of this family on which 
aecia were known to occur. 

The idea that Puccinia tritichia has lost its ability to develop an aecial 
. stage through long propagation by urediniospores, while admittedly pos- 
sible, was not considered to be fully substantiated. 

CULTURES MADE IN 1919 

With these considerations in mind rather extensive sowings were made 
in the spring of 19 19 upon a considerable number of species of the fami- 
lies Ranunculaceae and Boraginaceae and the closely related family 
Hydrophyllaceae. For this purpose, 20 collections of telia of Puccinia 
tritici'na were obtained from various sections of the country during the 
summer and fall of 1918 and placed outdoors to winter. Early in March 
these began to germinate. Ten of the 20 collections gave good germina- 
tion and were sown upon various species of the above-named families and 
upon Ornithogalum umbellatum L., Impaiiens sp., and Catnassia esculenta 
(Ker.) Robins. (Quamasia hyacinthina). The results obtained are given 
in Table I. 



Oct. IS. 1921 Aecial Stage of the Orange Leaf rust of Wheat 



155 



Table l.—Data obtained in igig from sowing teliospores of Puccinia triticina, from 10 
different localities, on various host plants, mostly of the families Ranunculaceae and 
Boraginaceae o- 



Host inoculated. 


No. ii8 
(Okla.). 


No. 218 
(Ala.). 


No. 418 
(Tenn.). 


No. 618 
(Ga.). 


No. 718 
(Ga.). 


No. 818 
(S.C). 


No. 918 
(Ind.). 


No. 

3518 

(Wis.). 


No. 

3818 

(Wis.). 


No. 

4Si8 

(Wis.). 








- 




















- 














_ 




- 






- 
















- 








- 


- 


_ 




■~ 


- 




- 




Anemone japonica Sieb. & 


- 










- 














- 


- 














_ 


Aquilegia canadensis L 


- 


~ 


■■■-'■' 


- 




- 




_ 




- 


_ 


Aquilegia glandulosa Fisch. . 


- 




= 




















1 


Aquilegia vulgaris 1, 




- 




- 


■:':;::;::::: 


_ 










Cimicifuga racemosa (L.) 
Nutt 


- 








- 


































- 
























_ 








- 










- 














- 




- 








- 








_ 


Clematis virginiana 1, 


~ 


■■'-"■ 


"" 


- 


- 










Delphinium "Belladonna." 
(Hort.).. 








- 




































- 
















" "— 


- 




















"" 


















- 


















- 


- 




- 


_ 


Mertensia virginica (L-) 
Link 


- 


- 


— 
























— 














_ 


Ornithogalum umbellatum. 
L 








- 


- 
















- 










- 










— 








Camassia esculenta (Ker.) 
















- 


- 












- 


_ 




- 






b "— 
b "— 






Ranunculus repens L 


- 






































- 




- 












- 








_ 




















Miihl 

















- 




















- 










— 










— 












i 









" — No infection. 
" Pycnia produced. 
^ Two sowings were made, 
the other. 



Pycnia were produced from only one sowing, no result being obtained from 



The telial collections used in the cultures were all obtained from Triti- 
cum aestivum (T. vulgare) in the following localities : 

118, from Stillwater, Okla., collected by J. D. Moore. 
218, from Flint, Ala., collected by McClellan. 
418, from Tennessee, collected by W. T. Evans. 
618, from Carrolton, Ga., collected by R. O. Bums. 
718, from Carrolton, Ga., collected by R. O. Burns. 



1:^6 Journal of Agricultural Research , ,l Voi. xxii, No. 3 

818, from Anderson, S. C, collected by R. O. Burns. 

918, from La Fayette, Ind., collected by E. H. Toole. 

3518, from Menah, Wis., collected by E. H. Toole. 

3818, from Wisconsin, collected by E. H. Toole. 

4518, from Superior, Wisconsin, collected by E. H. Toole. 
Negative results were obtained on all but two species of the hosts used. 
The collection from La Fayette, Ind. (No. 918), gave infection upon 
Thalictrum angustifolium and T. aquilegijolium, producing, however, only 
pycnia. It was impossible to carry this study further in 1919, as the 
above results were not obtained until late in the spring. 

The failure of aecia to develop from the two successful infections could 
be explained on either of two hypotheses. The conditions in the green- 
house may have been unfavorable, or the species of Thalictrum used may 
have been resistant. In either case, however, these results were inter- 
preted as indicating that the aecial host of the leafrust of wheat was some 
species of ThaHctrum. There was considerable basis for this assumption. 
All of the culture studies being carried on in this laboratory with the 
related rusts, occurring on wild grasses, and having aecia on members of 
the family Ranunculaceae, have indicated that while a given race may 
develop aecia on several species in one host genus with varying degrees 
of virulence it will not go to species of more than one genus. The rusts 
of this group show a very high degree of specialization. The two species 
of Thalictrum on which infection was obtained were foreign species, 
while the North American species, Thalictrum dioicum and T. polygarnum, 
were not infected. On this account it was thought that the susceptible 
aecial hosts for the leafrust of wheat probably were foreign species of 
Thalictrum. As the leafrust of wheat presumably is an introduced form, 
as explained in the following pages, this would be expected, and on that 
basis the species of Thalictrum should be western Asiatic or eastern 
European, corresponding to the region in which wheat is believed to have 

originated. 

CULTURES MADE IN 1920 

In preparation for cultural studies for the spring of 1920 an effort was 
made during the summer and fall of 19 19 to obtain as many species of 
Thalictrum as possible. It was impossible to obtain material from for- 
eign botanical gardens in time to be of use, and the best that could be 
done was to secure such species of Thalictrum as were carried by nursery- 
men in this country, together with such native species as could be obtained 
through collectors in various parts of the United States. As a result 14 
species were brought together. An appeal was also made to the plant 
pathologists in the various agricultural experiment stations tliroughout 
the country for aid in securing telial material of the leafrust of wheat. 
A very gratifying response to this appeal was made, and in this way 80 
collections of telia were obtained and placed out to overwinter. Of 
these, 51 collections germinated in the spring of 1920 and were sown. 
The number of collections was so great that it was not possible to sow 



Oct. IS, 1921 Aecial Stage of the Orange Leaf rust of Wheat 



157 



them on as large a number of species as was done in 19 19, and attention 
was devoted mainly to sowing upon species of Thalictrum. Of the 51 
collections used 9 were sown upon as many Thalictrum species as possi- 
ble in order to determine the relative susceptibility of these species. 
The results are given in Table II. 

Tabl:^ II. — Data obtained in ig20 from sowing teliospores of Piiccinia triticina from 
nine different localities on -various species of ThalictrumO' 



Host inoculated. 


Labor- 
atory 
No. 


No. 
5619 
(Ga.). 


No. 

6019 
(N. C ). 


No. 

7219 
(N.C). 


No. 

7819 

(Tenn.). 


No. 

8019 

(Mich.). 


No. 

8719 

(Miss.). 


No. 
11619 
(Pa.). 


No. 

12519 
(Idaho). 


No. 
15119 

(Nev.) 


Thalictrum angustifo- 


6 

S 

66 

65 

56 
16 
53 
17 

6s 

"5 

114 

19 

98 

55 


61 


I 
I 

61 









-(2) 



I 




- 



— 

-(2) 


61 







I 
bi 

I 


I 




Thalictrum aquilegifo- 


i>i 


Thalictrum aquilegifo- 




- 




Thalictrum dasycarpum 
Fisch. and LaU 




Thalictrum De lavayi 


01 


I 


I 
I 

-(3) 




Thalictrum diocium X, 




Thalictrum flavum I/. . . . 
Thalictrum minus 1, 


I 


- (2) 





I 


I 


Thalictrum, minus adi- 







Thalictrum occidentale 
Gray 




Thalictrum polycarpum 
S. Wats 






bi 



I 




Thalictrum polygamum 
Muhl 





- (2) 
I 


01 

I 




I 





-(2) 
I — 

















a — Noinfection, ■ . v, u'.vr^'! •,>'.t>.-.i, T 

Pycnia only produced. 

1 Aecia following pycnia. 

A numeral in parenthesis following the sign indicates the number of times the results were obtained. 
^ Although aecia were produced, the infection was weak. 

The following is a list of the sources of the telial material used in the 
cultures : 

5619, from Athens, Ga., collected by C. A. Ludwig. 

6019, from W. Raleigh, N. C, collected by Ludwig and Wolf. 

7219, from Hickory, N. C, collected by C. A. Ludwig. 

7819, from Tennessee, collected by C. A. Ludwig. 

8019, from Coldwater, Mich., collected by B. W. Mains. 

8719, from Canton, Miss., collected by C. A. Ludwig. 

11619, from State College, Pa., collected by J. T. Adams. 

12519, from Moscow, Idaho, collected by C. W. Hungerford. 

15 1 19, from Reno, Nev., collected by G. R. Hoemer. 
In addition to the sowings indicated in Table II, culture 8719 was 
sown upon Aquilegia glandulosa F'isch., A. olympica Boiss., Clematis 
heraclaefoHa DC, C. paniculaia Thunb., C. recta L., and Ranunculus acris 
L-, all without infection. 

An examination of Table II shows that 12 out of the 14 species of 
Thalictrum were infected, Thalictrum occidentale and T. aquilegifolium 
apparantly being immune. The species on which infection occurred 
showed varying degrees of susceptibility. Thalictrum dasycarpum and 



158 Journal of Agricultural Research voi. xxii. no. 3 

T. polygamum gave mostly negative results or the occasional production 
of pycnia. T. angustifolium, T. aquilegijolium (5), T. minus, T. minus 
adiantifolium, and T. polycarpum showed occasionally a weak develop- 
ment of aecia, but usually only pycnia developed or no infection occurred. 
T. dioicum, in one case, showed a moderate development of aecia; in 
all other cases only pycnia developed, or no infection resulted. T. 
Delavayi and T. sp. (98) (PI. 21, A, B) showed fairly vigorous infection, 
accompanied in most cases by more or less hypertrophy and usually by 
well-developed aecia. T. flavum (PI. 21, C) and T. sp. (55) (Pi. 21, 
D) showed a very vigorous infection accompanied usually by pronounced 
hypertrophy of the infected leaf and petiole tissue and practically always 
with the production of well-developed aecia. 

An attempt has been made to check the determination of the species 
of Thalictrum used in these studies, but this has been difficult because 
a number of them have produced neither flowers nor fruit, and the leaf 
characters in this genus are in most cases extremely variable. Specimens 
of most of the species have been sent to Mr. S. F. Blake, of the Bureau 
of Plant Industry, Washington, D. C, who has kindly compared them 
with specimens in the United States National Herbarium and has given 
his opinion as to the identity of our material. The following list gives 
the species used above, their sources, and native distribution as accurately 
as they could be determined. The accession number of this laboratory 
follows the name of each species. , 

Thalictrum angustifolium L. (6) . Source : Seed from Brooklyn Botanic 
Garden. Distribution: Central Europe and Asia Minor. 

Thalictrum aquilegifolium L. (5 and 66) . Source : Bobbink and Atkins 
Nursery Co. Distribution: Europe, Middle and Northern Asia. (No. 
66 was purchased for T. paniculatum.) 

Thalictrum dasycarpum Fisch. and Lall. (65). Source: Department of 
Botany, Michigan Agricultural College. Distribution: Northern and 
central United States and southern Canada. 

Thalictrum Delavayi Franchet (56). Source: Farr Nursery Co. Dis- 
tribution: Western China. 

Thalictrum dioicum L. (16). Source: LaFayette, Ind. Distribution: 
Eastern United States. 

Tlialictrum flavum L. (53). Source: Farr Nursery Co. Distribution: 
Europe, Western Asia, and Asia Minor. 

Thalictrum minus L. (17). Source: An American nursery. Distribu- 
tion: Europe, Asia, and eastern and southern Africa. 

Thalictrum m,inus adiantifolium (63). Source: Seed from Brooklyn 
Botanic Garden. Distribution: See T. minus. 

Thalictrum occidentale Gray (115). Source: Corvallis, Oreg. Distribu- 
tion: Mountains, California to British Columbia. 

Thalictrum, polycarpum S. Wats. (114). Source: Berkeley, Calif. Dis- 
tribution: California. 



Oct. IS, 1921 Aecial Stage of the Orange Leaf rust of Wheat 



159 



Thalictrum polygamum Muhl. 19. Source: Ithaca, N. Y. Distribution: 
Eastern United States. 

Thalictrum sp. (98). Source: Palisade Nursery Co. Distribution: 
Exotic. 

Thalictrum sp. (55). Source: Farr Nursery Co. Distribution: Exotic. 

From the data presented above it is evident that a number of species 
of Thalictrum are susceptible hosts for Puccinia triticina. As far as the 
host determinations are at all certain, the evidence would indicate that 
the most susceptible hosts are from western Asia and eastern Europe, 
and doubtless in this region other species will be found of as great or 
greater susceptibility. 

The remaining collections of telia showing good germination were 
sown on one or more of the susceptible species of Thalictrum in order 
to determine how uniformly Puccinia triticina from the United States 
would go to Thalictrum. Table III gives the results of these cultures. 

Table III. — Data obtained in ig20 from sowing teliospores of Puccinia triticina, from 
many different localities, on four especially susceptible species of Thalictrum o- 



Nmnber and source of telia. 



5119 Pa 

11719 Pa.. . . 
9819 W. Va. 
6319 N. C. . 
6419 N. C. . 
7319 N. C. . 
5819S. C. ... 
5019 Ga. . . . 
4119 Ala. . . . 
4519 Ala. . . . 
4419 Tenn.^. 
17419 Tenn. 
6519 Ky . . . . 
4819 Ind. . . . 
5319 Ind.... 
10119 Ind. . . 
10419 111 . . . 
12419 Minn. . 
3619 Iowa. . 
9219 Mo. . . . 
9319 Mo . . . . 

9519 Mo 

9619 Mo . . . . 

3819 La 

6719 Tex. .. . 
7019 Tex.c. . 
3219 Ariz. . . 
61 19 Calif. . 
122 19 Wash. 
165 19 Wash. 
199 19 Wash. 
I39i9 0reg. . 



T. sp. (ss). 



T Delavayi 
<S6). 



T. flavum 
(Si). 



T. sp. (98) 



I-(2) 



o — No infection. 

Pycnia produced. 

1 Aecia following pycnia. 

A numeral (in parenthesis) following the sign indicates the number of times the results were obtained. 
*> Sown also on T. angustifolium, producing pycnia, and on T. aquilegifolium without results. 
e Sown also on T. dasycarpum without results. 



i6o Journal of Agricultural Research voi. xxii.no. 3 

Source of telial material used in cultures : 

51 19, from York, Pa., collected by F. D. Kern. 

1 1 719, from Bradford Count}^ Pa., collected by E- T. Nixon. 

9819, from Morgantown, W. Va., collected by N. J. Giddings. 

6319, from Statesville, N. C, collected by C. A. Ludwig. 

6419, from Statesville, N. C, collected by C. A. Ludwig. 

7319, from Biltmore, N. C, collected by C. A. Ludwig. 

5819, from Clemson College, S. C, collected by C. A. Ludwig. 

5019, from Tifton, Ga., collected by C. A. Ludwig. 

41 19, from Bay Minette, Ala., collected by C. A. Ludwig. 

4519, from Auburn, Ala., collected by C. A. Lud\vig. 

4419, from Union City, Tenn., collected by Carl Kurtzweil. 

1 7419, from Johnson City, Tenn., collected by C. A. Ludwig. 

6519, from Lexington, Ky., collected by R. S. Kirby. 

4819, from Mount Vernon, Ind., collected by E- B. Mains. 

5319, from Washington County, Ind., collected by H. S. Jackson. 

10119, from La Eayette, Ind., collected by E. B. Mains. 

10419, from Bloomington, 111., collected by Koehler and Toole. 

1 24 1 9, from Wasioja, Minn., collected by G. W. Martin. 

36^9, from Ames, Iowa, collected by I. E. Melhus. 

9319, from Columbia, Mo., collected by W. E. Maneval. 

9419, from Columbia, Mo., collected by W. E. Maneval. 

9519, from Columbia, Mo., collected by W. E. Maneval. 

9619, from Columbia, Mo., collected by W. E. Maneval. 

3819, from Baton Rouge, La., collected by Thiel and Lud\\4g. 

6719, from Dallas, Tex., collected by W. H. Ballamy. 

7019, from San Antonio, Tex., collected by R. S. Kirby. 

3219, from Yuma, Ariz., collected by L. Y. Leonard. 

61 19, from Chico, Calif., collected by R. M. Kelia. 

122 19, from Dayton, Wash., collected by J. W. Hotson. 

165 19, from Colton, Wash., collected by J. W. Hotson. 

19919, from Puyallup, Wash., collected by G. R. Hoemer. 

13919, from Oregon, collected by G. R. Hoemer. 
The data in this table, taken with those in Table II, show that Puccinia 
triticina from Pennsylvania, West Virginia, North Carolina, South Caro- 
lina, Georgia, Alabama, Mississippi, Tennessee, Kentucky, Indiana, 
Michigan, Illinois, Minnesota, Iowa, Missouri, Louisiana, Texas, Arizona, 
California, Washington, Oregon, Idaho, and Nevada gave positive results 
when sown upon Thalictrum. 

The following collections showed some germination but produced no 
infection when sown on Thalictrum : 

11819, from Hopkinsville, Ky., collected by Carl Kurtzweil. 

9419, from Columbia, Mo., collected by W. E. Maneval. 

3919, from Fayetteville, Ark., collected by H. R. Rosen. 

5419, from Memphis, Tenn., collected by A. F. Thiel, 



Oct IS, 1921 Aecial Stooge of the Orange Leaf rust of Wheat 161 



7519, from Southampton, N. Y., collected by H. S. Jackson. 
8319, from Rocky Ford, Colo., collected by J. G. Leach. 
10019, from Buffalo, Minn., collected by G. W. Martin. 
105 19, from Plainview, Nebr., collected by H. W. Thurston. 
11019, from Vermillion, Minn., collected by G. W. Martin. 
1 12 19, from Newark, Del., collected by T. F. Manns. 
Besides the above, the following collections were wintered, but no 
germinating teliospores were found, and in consequence they were not 

sown. 

1419, from Santa Rosa, Calif., collected by H. S. Jackson. 

3319, from Sonora, Mexico, near Yuma, Ariz., collected by L. Y. 
Leonard. 

3419, from St. Louis, Mo., collected by E- B. Mains. 

3719, from Jackson, Tenn., collected by Kurtzweil and Thiel. .5 

4019, from Corvallis, Oreg., collected by G. R. Hoemer. ,i 

5519, from St. Paul, Minn., collected by A. F. Thiel. .^ | 

6619, from Hiawatha, Kans., collected by W. H. Ballamy.o\^ | 

6819, from Marshall, Mo., collected by R. S. Kirby. ) n 

6919, from Guthrie, Okla., collected by R. S. Kirby. 

71 19, from Wellington, Mich., collected by G. H. Coons. 

7919, from Nashville, Tenn., collected by C. A. Ludwig. 

91 19, from Madison, Wis., collected by E. B. Mains. 

108 1 9, from Manhattan, Kans., collected by L. E. Melchers. 

Ill 19, from Toledo, Iowa, collected by I. E- Melhus. 

11919, from Fort Collins, Colo., collected by J. G. Leach. 

12019, from East Lansing, Mich., collected by Acelia M. Leach. 

12 1 19, from East Lansing, Mich., collected by Acelia M. Leach. 

12319, from Pullman, Wash., collected by F. D. Heald. 

1 38 1 9, from Fort Collins, Colo., collected by J. G. Leach. 

21019, from Moscow, Idaho, collected by G. R. Hoemer. 

25519, from Murfreesboro, Tenn., collected by Carl Kurtzweil. 

25719, from Clarksville, Tenn., collected by Carl Kurtzweil. 
The accompanying map (fig. i) shows the source of collections used 
in the work together with the results obtained with them at La Fayette, 
Ind. This map shows that material from the States of Pennsylvania, 
West Virginia, Indiana, Illinois, North Carolina, South Carolina, Georgia, 
Alabama, Mississippi, Louisiana, and Texas gave germination uniformly 
and infected Thalictrum in all cases. A region represented by the States 
of Colorado, Oklahoma, Arkansas, Kansas, Nebraska, Minnesota, Iowa, 
and Missouri, with one arm running through Wisconsin into Michigan 
and another through Tennessee into Kentucky, gave material which 
usually did not germinate or, if germination was obtained, produced 
infection on Thalictrum in only a few cases. Whether this situation 
indicates the presence of another strain of the leafrust having different 
characteristics as regards its viability and power to infect Thalictrum, 
or whether it means that the season or climate was of such a nature that 



l62 



Journal of Agricultural Research voi. xxii. No. 3 



teliospores of a low vitality were produced, remains for future investiga- 
tion to decide. Telial material from the Pacific coast, while not viable 
in a number of cases, produced infection on Thalictrum in all cases where 
germination was obtained. 

The aecia produced from the following telial collections were sown 
back upon wheat: 

4519, from Auburn, Ala. 

4819, from Mount Vernon, Ind. 

5019, from Tifton, Ga. 

5619, from Athens, Ga. 

5819, from Clemson College, S. C. 

6019, from W. Raleigh, N. C. 

6319, from Statesville, N. C. 



/u_ p/ \ ■ — j — 

I T^ L 1 L 

V \ / /^i- 




2 


i 
.+ 


500~^ 


?pr+^ 


? 






V 


1. 
+0 

— < 

• 
\ 


\ \ /— ~— ^/ * 


0° 


vXT r 

I XI A/ A/OT SOi^A/. 





+ 


+ 

V 


,-> 


\ai4 





Flo. I, — Map showing results obtained at LaFayettc, Ind., with leafrust material collected in different 

parts of the United States. 

7219, from Hickory, N. C. 

7819, from Tennessee. 

8019, from Coldwater, Mich. 

8719, from Canton, Miss. 

9619, from Columbia, Mo. 

9819, from Morgantown, W. Va. 
10419, from Bloomington, 111. 
11619, from State College, Pa. 
1 24 1 9, from Wasioja, Minn. 
125 1 9, from Moscow, Idaho. 
151 19, from Reno, Nev. 
165 19, from Colton, Wash. 
1 7419, from Johnson City, Tenn. 



Oct. 15, 1921 



Aecial Stage of the Orange Leaf rust of Wheat 



163 



These were each sown upon the variety of wheat known as Dawson 
Golden ChaflF, and in all cases positive infection was obtained followed by 
the development of uredinia which were typical of Puccinia triticina. 

Sowings of aeciospores also were made upon a number of grasses. 
Aecia which were produced from telia obtained from Hickory, N. C, 
Canton, Miss., and Moscow, Idaho, were used and Table IV gives the 
results. 

Table IV. — Results obtained in ig20 from sowing the aeciospores of Puccinia triticina 
produced from telia obtained in three different areas, on wheat and related grasses 



Host inoculated. 



Arrhenatherum elatius (L.) Mert & Koch. . . 

Agropyron caninum (L.) Beauv 

Agropyon cristatum Beauv 

Agropyron desertorum Schult 

Agropyron in£rme (Schribn. & Sm.) Rydb . 

Agropyron repens (L.) Beauv 

Agropyron tenerum Vasey 

Elymus atistralis Schribn. & Ball 

Elynius canadensis L 

Elymus glaucus Buckl 

Elym^us triticoides Buckl 

Elymus virginicus L 

Hordeum caespitosum Schribn 

Hordeum gu^soneanum Pari 

Hordeum jubatum L 

Hordeum pusillum Nutt 

Hordeum inurinum I, 

Hordeum vulgare L 

Hystrix Hystrix (L.) Millsp 

Notholcus lanatus (L.) Nash 

Secale cereale L 



Sitanion Hystrix (Nutt.) J. G. Sm. 

Triticum aegilops Beauv 

Triticum aestivum L 



Number and source of aecia. 



No. 7219 
(N.C.). 



t Many. 



No. 8719 
(Miss.). 



t One ure- 
dinium. 



t Many. 



No. 12519 
(Idaho). 



t Few. 
t Many, 



— No infection. 



t Uredinia produced. 



Except for the one uredinium produced on Secale cereale, Triticum 
aestivum and T. aegilops were the only species infected 

DESCRIPTION OF AECIA 

The following description has been drawn from the aecia obtained 
in the cultures discussed above. 

Pycnia amphigenous, mostly epiphyllous, numerous, crowded upon 
more or less swollen reddish brown to yellowish areas 2 to 15 mm. in 
diameter, conspicuous, subepidermal, honey-yellow, globoid or flattened 
globoid, 80 to 145 /i broad by 80 to 130 fx high; ostiolar filaments 95 to 
190 jj, long, agglutinated to form a prominent, broad column. 
65508°— 21 4 



V 



164 Journal of Agricultural Research voi. xxii. no. 3 



Aecia hypophyllous, crowded in more or less swollen, gall-like, reddish 
brown or yellowish areas 2 to 15 mm. in diameter, cupulate or short 
cylindric, 0.2 to 0.6 mm. in diameter, up to 0.5 mm. high; peridium white 
or yellowish, the margin erose or somewhat lacerate, recurved; peridial 
cells oblong or somewhat rhomboidal in longitudinal radial section, 14 to 
19 by 18 to 29 IX, abutted or slightly overlapping, the outer wall 6 to 7 /x 
thick, transversely striate, the inner wall thinner 2 to 3 yLt, very coarsely 
verrucose; aeciospores angularly globoid or ellipsoid, 16 to 20 by 16 to 
26 /x; wall colorless, thin, i /i or less, very closely and finely verrucose. 

The pycnia and aecia usually were produced in definite galls or swellings. 
These galls apparently were formed by the excessive enlargement of the 
cells of the infected areas, especially those of the mesophyll. When 
infection took place in the young, rapidly developing tissue of the petiole, 
galls developed (PI. 21, D) which were 10 or 15 times as large as the 
normal petiole. A very noticeable odor, resembling that of the hyacinth, 
was often detected as the pycnia reached maturity. 

GENERAL DISCUSSION OF RESULTS 

The discovery that species of Thalictrum are the aecial hosts for 
Puccinia triticina goes to support Arthur's contention {i, v. 9, p. 304) that 
the leafrust of wheat is closely related to grass rusts of the type of Puccinia 
Agropyri, having aecia on species of the family Ranunculaceae. A num- 
ber of cultures have been made with rusts of this type, connecting them 
with various species of Thalictrum. Plowright, in England {21, p. j8i), 
connected aecia on Thalictrum flavum with a rust on Agropyron repens. 
To this rust he gave the name Puccinia persistens Plowr. He considered 
Aecidium Ranunculacearum 7 Thalictri flavi DC, and Aecidium Thalictri 
fiavi (DC.) Winter as synonyms, and describes the aecia as occurring 
on thickened spots with aeciospores subglobose 17 to 20 by 20 to 30 /i. 
Fischer (12, p. 37-63), in Switzerland, cultured a rust from Poa nemoralis 
var. firmula on Thalictrmn minus, T. aquilegijolium and T. foetidum. 
On account of the morphological similarity, he concluded that his material 
belonged to Puccinia persistens, although he made no cultures on either 
Agropyron repens or T. flavum. He describes the aecia (ij, p. 347-349) 
as having peridial cells with the outer wall 4.5 to 13.5 ^ thick and the inner 
2 to 6 fjL and aeciospores 10 to 21 /x broad and up to 28 /i long. An exam- 
ination of Sydow's Uredineen No. 725, issued as Puccinia persistens on T. 
aquilegifolium, shows the following measurements : Peridial cells, 18 to 23 
by 21 to 26 /x; outer wall, 7 to g fx; inner, 3 to 5 /x; aeciospores, 16 to 19 
by 19 to 26 n. 

Juel (j<5, p. 411), in Sweden, made cultures connecting aecia on Thalic- 
trum alpinum with a rust on Agrostis borealis and Anthoxanthum odor- 
atum. To this rust he gave the name Puccinia borealis Juel, and con- 
sidered Aecidium thalictri Grev. as a synonym. His description follows: 
Pycnia not present; aecia not causing hypertrophy of host tissue; aecio- 



Oct i^. 1921 Aecial Stage of the Orange Leaf rust of Wheat 165 

spores about 13 ju in diameter. The Sydows {23, p. 718-719) give the 
measurement of the aeciospores as 13 to 18 /x in diameter or 13 to 16 by 
18 to 20 M and note that no swelHngs are produced on the leaves of the host. 
An examination of Eriksson's Fungi Parasitici Scandinavici 432a, col- 
lected by Juel in Norway, gives the following measurements: Peridial 
cells, 16 to 19 by 19 to 29 /x; the outer wall, 10 /x; the inner, 3 to 4 ju; 
aeciospores 14 to 16 by 16 to 21 m- 

Rostrup (22, p. 269-273), in Denmark, obtained infection with aecio- 
spores from Thalictrum minus on Elymus arenarius and considered the 
rust to be Puccinia Elymi Westendorp. The writers have seen neither 
description nor material of these aecia. 

The Sydows (23, p. 827) mention that Lindroth in Finland connected 
an aecidium on Thalictrum m/ijus with a rust on Agropyron caninum. 
No description or material of this connection is available for study. 

In North America a number of connections have been established by 
the cultures of Arthur and of Fraser. Arthur (/, v, i, p. 248-249) reports 
culturing a rust found associated with aecia on Thalictrum sparsifiorum 
from Bromus Porteri to T. dioicum. To this he gave the name Puccinia 
alternans. He describes the aecia as having peridial cells 2 1 to 29 /x long 
with the outer wall 9 to 12 /x thick and the inner 5 to 7 /x and with aecio- 
spores 15 to 20 by 17 to 24 IX. A number of other species of Thalictrum 
are given as hosts. 

Arthur (j, v. 2, p. 226) also reports obtaining infection from telia on 
Agropyron resulting in aecia on Thalictrum alpinum but not on T. dioicum. 
This material he considered as belonging in Puccinia obliterata Arth., 
which he had previously shown as having aecia on Aquilegia. A study 
of the material obtained by this culture shows little or no hypertrophy 
of the host tissue. The peridial cells measure 16 to 21 by 24 to 32 /x, 
having the outer wall 7 to 9 ^u thick and the inner 3 to 5 /x. The aecio- 
spores measure 14 to 18 by 18 to 23 jx. 

Still another connection was obtained by Arthur (i, v. 8, p. 132-133) 
when he cultured a rust on Festuca Thurberi to Thalictrum dioicum, 
producing aecia. To this he later (2, p. 113) gave the name Puccinia 
Cockerelliana Bethel. He gives the peridial cells as 16 to 23 by 27 to 
36 /x with the outer wall 6 to 8 /x and the inner 2 to 3 a« and aeciospores 18 
to 24 by 20 to 29 IX with a wall 1.5 to 2.5 ix thick. The natural host for the 
aecia is given as T. Fendleri. 

Fraser {14, p. 131-133) reports sowing aeciospores from Thalictrum 
dasycarpwn on Elymus canadensis, E. virginicus, Agropyron tenerum, 
A. Richardsonii, Hordeum jubatum, Triticum vulgare, and Bromus 
ciliatus, obtaining infection on E. canadensis, E. virginicus, H. juba- 
tum, and B. ciliatus. When, however, the rust obtained upon B. cili- 
atus was sown on E. virginicus, A. tenerum., A. Smithii, A. repens, and 
H. jubatum no infection was obtained on these species. From these 
results Fraser concludes that two strains of Puccinia Agropyri KHis and E., 



1 66 Journal of Agricultural Research voi. xxn. no. 3 



were present in the aecial material on Thalictrum which he used for the 
culture. A study of the aecia used in these cultures shows the dimensions 
of the peridial cells to be 15 to 19 by 23 to 29 n with the outer wall 7 to 
10 /x and the inner 3 fx thick and the aeciospores 14 to 19 by 19 to 23 p. in 
diameter. 

A comparison of the foregoing description of the aecia of Puccinia 
triticina with the measurements given for the various grass rust aecia on 
Thalictrum shows surprisingly little variation. The aecia of P. Cockerel- 
liana show the greatest difference, having larger peridial cells and some- 
what larger aeciospores with much thicker walls than the aecia of P. triti- 
cina. Slightly smaller aeciospores occur in P. horealis and P. obliterata, 
and the aecial infection causes little or no hypertrophy of the host. 
The remaining aecia differ mainly in slightly thicker walls of the peridial 
cells. 

It is evident that Puccinia triticina is closely related to P. persistens. 
Whether the name Aecidium Tlialictri-flavi (DC.) Wint. should apply to 
the aecial stage of the former is a question which can not be answered 
with the available information. De Candolle (7, p. giy) described 
A. Ranunculacearum for aecia occurring on the family Ranunculaceae 
and as a variety of this gives Thalictri-flavi without further description. 
Winter (24, p. 269) raises this variety to specific rank and gives a 
description which, however, could apply to the aecia of either rust. 
As both P. triticina and P. persist-ens are common rusts throughout 
Europe, there is no way of determining definitely to what aecia the name 
was applied beyond the fact that they were on Thalictrum flavum. As 
it has been shown that at least some aecia on that host in England belong 
to P. persistens the name A. Thalictri-flavi should be retained for the 
present as a synonym of that species, at least until aecia can be found 
in Europe upon T. flavum which will produce the leafrust of wheat. 

Upon their grass hosts these rusts present a somewhat greater varia- 
tion. They all have uredinia with globoid or ellipsoid urediniospores 
with a varying number of scattered pores, usually more than six, and 
with few or no paraphyses. The telia are long, covered by the epidermis, 
usually with more or less stroma present, and the teliospores are cylindric, 
more or less flattened at the apex, and with a very short pedicel. Puc- 
cinia Cockerelliana differs most noticeably from P. triticina in that the 
teliospores are much longer and the telia do not remain entirely covered 
by the epidermis at maturity. P. Elymi differs especially in the thicker 
and darker walls of the urediniospore and in the longer teliospores which 
are often many-celled. P. alternans, P. borealis, P. obliterata, and P. 
persistens differ but little, mostly in the tinting of the urediniospore wall 
and a slight variation in pore number. 

Although the morphological differences between Puccinia triticina and 
the related rusts discussed above are not great, their biologic specializa- 



Oct. IS, 1921 Aecial Stage of the Orange Leaf rust of Wheat 167 

tion to their hosts is very pronounced. This appears to hold true for the 
aecial as well as the grass hosts. P. Cockerelliana and P. alternans go to 
Thalictrum dioicum, and P. Elymi to T. minus, as aecial hosts, neither of 
which is a favorable host for P. triticina. The rust of T. dasycarpum, 
used by Fraser {14) in his cultures, is on a host which was not infected 
by P. triticina. P. horealis and P. ohliterata on T. alpinum offer no com- 
parison, as P. triticina was not sown on that host. P. persistens, as 
cultured by Fischer on T. minus and T. aquilegifolium, is upon species of 
Thalictrum unfavorable for P. triticina, while P. persistens as originally 
cultured by Plowright upon T. flavum is on the most congenial host for 
the leafrust of wheat. It is very probable that Plowright and Fischer 
were working with two distinct biologic strains. Although T. flavum 
appears to be a favorable host for both P. persistens and P. triticina, and 
these two rusts are very similar in their morphology, the inability of the 
latter to infect Agropyron repens shows that it is biologically distinct 
from the former. A study is being made of the relationship of P. triticina 
to grass hosts other than wheat. From the data now at hand, it would 
appear that, in addition to the grasses listed in Table IV, species of 
Bromus, Festuca, Agrostis, Poa, and Anthoxanthum are immune from 
the leafrust of wheat. These results indicate that, as far as its telial 
host is concerned, P. triticina also is biologically distinct from other grass 
rusts having aecia on Thalictrum. 

A similar situation exists in the relationship of Puccinia triticina to 
rusts having aecia upon species of other genera of the Ranunculaceae. 
Slight morphological differences, such as urediniospore size, wall color, 
and pore number, exist among the different races producing aecia upon 
species of such genera as Actaea, Anemone, Clematis, Delphinium, etc. 
A similar, or perhaps greater, biologic specialization is also to be found 
among these races. The importance of these morphological characters 
and biological differences which occur among the members of this group 
can not be fully determined at present on account of our comparatively 
limited knowledge of but few races. Any final interpretation must 
await further study of a greater number of such races. On the basis of 
our present knowledge, the disposition of P. triticina must depend largely 
upon the species concept held. In Europe there is a tendency among 
certain students of the rusts to consider as species those rusts showing 
distinct biologic specialization regardless of the absence of morphological 
difference. In this country, on the other hand, the general tendency is 
to include in a single species all closely related forms having but little 
difference in their morphology. Forms limited to a definite host, or 
hosts, are considered as races of such species. On the former basis, P. 
triticina would be considered a distinct species comparable to P. Elymi, 
P. Agropyri, P. persistens, etc., while with the latter concept it would be 
united with all or part of these, each being considered a race of a 



168 Journal of Agricultural Research Voi. xxii. no. 3 



collective species to be designated, according to the limitations of 
the species concept held and the system of nomenclatm-e used, as 
P. Agropyri B. and E. (j, v. 9, p. 304), P. Clematidis (DC.) Lagerh., or 
Dicaeoma Clematidis (DC.) Arth. (j, p. 333-337). 

The close biological specialization of Puccinia triticina to wheat is of 
considerable significance with respect to the bearing it has upon the 
possible origin of this rust and of wheat itself. Since wheat is an intro- 
duced plant, it is logical to assume that a rust showing such close biolog- 
ical specialization to it is also introduced and of foreign origin. 

It is generally recognized among students of the rusts that a high de- 
gree of host specialization must have been acquired in certain groups of 
species at a very early stage in the evolutionary history of this group of 
fungi. It is also recognized that the host is the most important factor 
in the evolution of highly specialized pai-asitic fungi. As the higher 
plants have gradually developed during geological times, their rust 
parasites have developed with them. It therefore appears reasonable to 
assume that Puccinia triticina, which shows such a high degree of spe- 
cialization to wheat at the present time, had its origin as a distinct strain 
comparatively early in the development of the group of grasses from 
which our cultivated wheats have originated. The original distribution 
of the rust presumably would coincide with the distribution of the 
ancestral wheats. 

A study of the relative susceptibility of various species of Thalictrum 
to infection by this rust is of interest in this connection. The four most 
susceptible species of Thalictrum encountered in this investigation are 
all of foreign origin. The most susceptible of our native North American 
species, Thalictrum dioicum, does not compare in susceptibility with 
these four foreign species— T. flavum, T. Delavayi, T. sp. 55, and T. sp. 
98 — but is comparable to the resistant foreign species such as T. minus. 
That these foreign susceptible species of Thalictrum are also to be con- 
sidered as indicating a foreign origin of the rust would appear to follow if 
the nature of aecial infection is considered. Heteroecious rusts in most 
cases infect their aecial hosts only for a comparatively short period of 
tune while the teliospores are germinating in the spring. The infection 
produced, not being able to propagate itself upon such hosts, causes little 
or no damage, and they are in most cases soon able to outgrow it. On 
this account it is hardly to be expected that a natural selection of resist- 
ant strains of aecial hosts takes place in nature comparable to that which 
occurs where the host is killed or prevented from producing seed. Should 
this occur in heteroecious rusts which are not able to survive adverse 
conditions in winter or summer by means of urediniospores, such a selec- 
tion would be fatal to the rust itself. For this reason the susceptibility 
of the aecial hosts of P. triticina may be taken as indicative of its origin. 
It is true that susceptibility of a host species does not necessarily indi- 
cate that such a species was a native host of the rust nor does resistance 



Oct. IS, i92t Aecial Stage of the Orange Leaf rust of Wheat 169 



of some one species denote that the rust is not to be found in the habitat 
of such a resistant species, for susceptibility or resistance is not dependent 
upon the presence or absence of the rust but may develop with the species 
in any region. It is regarded as significant, however, that of the species 
of Thalictrum tested the most susceptible are exotic. This fact, taken 
with the foreign origin of wheat itself, is confirmative of the foreign 
origin of the rust. 

The native habitats of two of these species of Thalictrum are known 
with some degree of certainty. Thalictrum flavurn is found throughout 
Europe, western Asia, and Asia Minor. T. Delavayi is given by the Index 
Kewensis ^ as occurring in western China, probably indicating a distri- 
bution in the little-known mountainous regions of Tibet and Chinese 
Turkestan. These two species, taken together, would therefore indicate 
as the most probable original distribution a region in which the two 
Thalictrum species may border or overlap, such as that of southwestern 
Asia. Such an origin would indicate a like origin for wheat itself, 
which, we believe, would agree with the latest theories advanced as to 
the original home of wheat. 

Concerning the occurrence and distribution of the aecia of Puccinia 
triticina but little can be said with the data at hand. It is also probable 
that the aecial stage occurs, and probably assumes greater importance, 
in other regions than it may in either Europe or North America, where the 
rust is known to overwinter in its uredinial stage. Thus in such countries 
as India, where Butler has shown there is no oversummering of the rust, 
the Thalictrum species of the foothills may be of importance in starting 
the rust the next season. The question of the role which the aecia of the 
leafrust of wheat plays in its life history and distribution must, however, 
be left for future research to solve. Whether native species of Thalictrum 
serve as aecial hosts in North America and, if so, whether they serve as 
important factors in the development of the leafrust of wheat and 
whether there is more than one race of the leafrust, as indicated by the 
results obtained from the Great Plains area, or whether these results 
were due to other causes, such as climatic or seasonal effects weakening 
the vitality of the teliospores, are all questions on which further investi- 
gation is planned. Other species of Thalictrum from foreign botanic 
gardens also v/ill be studied in regard to their susceptibility to the orange 

leafrust of wheat. 

SUMMARY 

(i) The aecial stage of Puccinia triticina has been produced in green- 
house cultures upon several species of Thalictrum. 

(2) The various species of Thalictrum show varying degrees of sus- 
ceptibility to the rust. Thalictrum occidentale was apparently immune. 
Upon T. dasycarpum and T. polygamutn an occasional devlopment of 

> INDEX KEWENSIS PLANTARUM PHANEROGAMARUM. SUPPI.EMENTUM PRIMUM . . . CONFECERUNT 
THBOPHIlrUS DURAND ET B. DAYDON JACKSON. p. 42^. BniXCllis I9OI-06. 



lyo Journal of Agricultural Research voi. xxn. no. 3 



pycnia took place. When T. angustijoliuvi, T. aqtiilegifolium, T. dioicum, 
T. minus, T. minus adiantijolium, and T. polycarpum were inoculated 
usually only pycnia resulted, with an occasional weak development of 
aecia, while in other cases no infection occurred. Two undetermined 
species of Thalictrum, as well as T. Delavayi and T. flainmi, when inocu- 
lated, showed a vigorous development of aecia, increasing in suscepti- 
bility in the order named. 

(3) Puccinia triticina is apparently limited to species of the genus 
Thahctrum, no infection being obtained upon species of Aconitum, 
Actaea, Anemone, Aquilegia, Cimicifuga, Clematis, Delphinium, Echium, 
Ilepatica, Hydrophyllum, Impatiens, Mertensia, Myosotis, Ornithogalum, 
Phacelia, Camassia, Ranunculus, or Trollius. 

(4) On account of the morphology and host relationships, Puccinia 
triticina is considered to be very closely related to P. persistens, P. 
borealis, P. alternans, P. ohliterata, P. Elymi, and P. Agropyri, but is 
separable from these rusts by its sharp biologic limitation to wheat. 

(5) Puccinia triticina is considered to be of foreign origin, because 
wheat, for which it shows close specialization, is an introduced host, and 
because the most susceptible species of Thalictrum which serve as aecial 
hosts also are exotic. 

LITERATURE CITED 
(i) Arthur, Joseph Charles. 

1909-17. CtliTURES OF UREDINEAE IN 1908, I909, 1915, 1916 AND 1917. In 

Mycologia v. i, no. 6, p. 225-256; v. 2, no. 5, p. 213-240; v. 8, no. 3, 
p. 125-141; V. 9, no. 5, p. 294-312. 
(2) 

1919. NEW SPECIES OF UREDINEAE — XI. In Bui. Torrey Bot. Club, v. 46, 

no. 4, p. 107-125. 

(3) and Fromme, Fred. Denton. 

1920. DICAEOMA ON POACEAE. In North American Flora, v. 7, pt. 4-5, p. 

269-341. 

(4) Bary, Anton de. 

1866. NEUE UNTERSUCHUNGEN ubER urEdinEEn. In Monatsber. K. Preuss. 
Akad. Wiss. Berlin, 1866, p. 205-215, i pi. 

(5) BoLLEY, H. L. 

1889. WHEAT rust. Ind. Agr. Exp. Sta. Bui. 26, 19 p., 9 fig. 

(6) Butler, E. J., and Hayman, J. M. 

1906. INDIAN wheat rusts. In Mem. Dept. Agr. India, Bot. Ser., v. i, no. 2, 
p. 1-52, I fig., 5 pi. (1-4 col.). 

(7) CandollE, a. p. de. 

1815. FLORE FRANfAisE. • • t. 5 (v. 6). Patis. 

(8) Carleton, Mark Alfred. 

1899. CEREAL RUSTS OF THE UNITED STATES. A PHYSIOLOGICAL INVESTIGA- 
TION. U. S. Dept. Agr. Div. Veg. Physiol, and Path. Bui. 16, 74 p., 
I fig., 4 col. pi. Bibliography, p. 70-73. 

(9) Cunningham, D. D., and Prain, D. 

1896. A note on INDIAN WHEAT-RUSTS. In Rec. Bot. Survey India, v. i, 
no. 7, p. 99-124. 



Oct. IS, 19" Aecial Stage of the Orange Leaf rust of Wheat 171 

(10) Eriksson, Jakob. 

1899. NOuvELLES ^TudES sur LA rouillE brunE dES c^r^alES. In Ann. 

Sci. Nat. Bot. ser. 8, t. 9, no. 2/4, p. 241-288, pi. 11-13 (col.). Lit- 
tdrature citee, p. 286-287. 

(11) and Henning, Ernst. 

i8g4. DIE hauptrESULTATE EinER nEUEN unTERSUCHUNG ubeR die GETREI- 
DEROSTE. In Ztschr. Pflanzenkrank., Bd. 4, p. 66-73, 140-142, 197- 
203, 257-262. 

(12) Fischer, Eduard. 

1898. ENTWICKLUNGSGESCHICHTLICHE UNTERSUCHUNGEN UBER ROSTPIUZE. . . 

X, 120 p., 16 fig., 2 pi. (Beitrage zur Kryptogamenflora der Schweiz, 
Bd. I, Heft. I.) 

(13) 

1904. DIE UREDINEEN DER SCHWEIZ. xciv, 590 p., 342 fig. Bern. Literatur- 
verzeichniss, p. 558-576. (Beitrage zur Kryptogamenflora der vSchweiz 
Bd. 2, Heft. 2.) 

(14) Eraser, W. P. 

1919. CULTURES OF hETEroecious RUSTS IN 1918. In Mycologia, v. 11, no. 3, 
p. 129-133. 

(15) Hitchcock, A. S., and Carleton, M. A. 

1894. SECOND REPORT ON RUSTS OP GRAIN. Kans. Agr. Exp. Sta. Bui. 46, 9 p. 

(16) JuEL, H. O. 

1894. MYKOLOGISCHE BEITRAGE I. ZUR KENNTNISS EINIGER UREDINEEN AUS 

DEN gebirgsgEgendEn skandinaviens. In Ofvers. K. Svenska 
Vetensk. Akad. Forhandl., Arg. 51, no. 8, p. 409-418. 

(17) Klebahn, Heinrich. 

1900. beitrage zur kenntnis der getreideroste. II. In Ztschr. Pflanzen- 

krank., Bd. 10, Heft. 2, p. 70-96, I fig. 

(18) 

1904. DIE wirtswechselnden rostpilze. . . 447 p. Berlin. Literatur, p. 
ix-xxxvii. 

(19) MtJLLER, Fritz. 

1901. BEITRAGE ZUR KENNTNIS DER GRASROSTC. In Bot. Centbl., Beihefte. 
Bd. 10, Heft 4/5, p. 181-212, I fig.. 

(20) Nielsen, p. 

1877. bemaerkninger om nogle rustarter, navnlig om en genetisk 

FORBINDELSE MELLEM AECIDtUM TUSSILAGINIS PERS. OG PUCCINIA POARUM 

N. SP. In Bot. Tidsskr., Raekke 3, Bd. 2, p. 26-42, 3 fig. 

(21) Plowright, Charles B. 

1889. A monograph of the BRITISH UREDINEAE AND USTILAGINEAE . . . vii, 

347 p., 13 fig., 8 pi. London. Authors quoted, p. 309-315. 

(22) ROSTRUP, E. 

1898. ET NYT VAERTSKIFTE HOS UREDINACEERNE OG KONIDIER HOS THECAPHORA 

CONVOLVULI. In Overs. K. Danske Vidensk. Selsk. Fordhandl., 1898, 
No. 5, p. 269-276. 

(23) Sydow, p., and Sydow, H. 

1904. MONOGRAPHiA uredinearum . . . V. I. Lipsiae. 

(24) Winter, Georg. 

1884. die PILZE DEUTSCHLANDS, OESTERREICHS UND der SCHWEIZ. ABT. I. 
SCHIZOMYCETEN, SACCHAROMYCETEN UND BASIDIOMYCETEN. In Raben- 

horst, L. Kryptogamen-Flora von Deutschland, Oesterreich und der 
Schweiz. Aufl. 2. Bd. i, Abt. i. Leipzig. 



PLATE 21 

A.— Infection produced upon Thalictrum sp. (98) inoculated with Puccinia triticina 
from Colton, Wash. (16519). 

B. — Infection produced upon Thalictrum sp. (98) inoculated with Puccinia triticina 
from Canton, Miss. (8719). 

' C. — Infection produced upon Thalictrum flavjim (53) inoculated with Puccinia iriii- 
ana from Hickory, N. C. (7219). 

D. — Infection produced upon Thalictruin sp. (55) inoculated with Puccinia triticina 
from Hickory, N. C. (7219). 

(172) 



Aecial Stage of the Orange Leafrust of Wheat 



PLATE 21 







Journal of Agricultural Research 



Vol. XXII, No. 3 



A Transmissible mosaic disease oe Chinese 

CABBAGE, MUSTARD, AND TURNIP 

By E. S. ScHUivTz 

Pathologist, Office of Cotton, Truck, and Forage Crop Disease Investigations, Bureati of 
Plant Industry, United States Department of Agriculture 

In the fall of 19 19, while the writer was selecting different kinds of 
plants for inoculation experiments with mosaic of Irish potatoes (Sola- 
num tuberosum Linn.), Dr. W. A. Orton called his attention to mottling 
in plants of Chinese cabbage (Brassica pekinensis (Lour.) Gagn.), mus- 
tard {B. japonica Coss.), and turnip {B. rapa, Linn.). The mottling 
resembled that of mosaic plants of other species, such as potato and 
tobacco. Diseased and healthy individuals were found in the same plot; 
the former appeared in groups in some parts of tlie field, suggesting an 
infectious character of this malady. Evidence bearing upon the nature 
of this disease, its symptoms, and means of transmission is presented in 

this paper. 

SYMPTOMS 

Mosaic of Chinese cabbage, mustard, and turnip produces a distinct 
mottling of the leaves, very similar to that of mosaic diseases of the 
Solanaceae. This mottling is produced by the appearance of irregular 
light green and dark green areas on the leaves (PI. B ; 22, D, E ; 24, A, B). 
These light green areas usually adjoin the veins, from which they may 
extend so as to include a considerable area of the leaf surface between 
the veins. Another very common macroscopic symptom of this disease 
is the characteristic ruffling and distorting of the leaf surface (PI. 24, 
A, B). On the raised areas the dark green patches appear. The leaf 
margins frequently are much more irregular than in healthy plants, 
causing some of the leaves to appear somewhat unsymmetrical. In addi- 
tion to these common abnormalities on the leaves the entire plant may 
be dwarfed, and the flower stalk and number of blossoms may be con- 
siderably reduced (PI. 22, B; 23, B). 

OCCURRENCE IN THE FIEED 

Since mosaic individuals appeared among Chinese cabbage, mustard, 
and turnip plants growing in adjoining plots, interspecific susceptibility 
was suggested. Furthermore, it was found that a large percentage of 
the plants were infested with aphids, Myzus persicae Sulz.,^ one of the 

1 Identified by Dr. A. C. Baker, Entomologist, Bureau of Entomology, United States Department of 
Agriculture. 

Journal of Agricultural Research, Vol. XXII, No. 3 

Washington, D. C. Oct. 15, 1921 

aaa Key No. G-248 

(173) 



174 



Journal of Agricultural Research voi. xxn. N0.3 



casual agents in the transmission of mosaic and leafroll of Irish potato.^ 
In view of these field observations experiments on this disease were 
conducted in the greenhouse at Washington, D. C, during the winters of 
1919-20 and 1920-21. 

TRANSMLSvSION WITH PLANT JUICE 

Chinese cabbage, mustard, and turnip plants showing mosaic mottling 
were taken from the field and planted in pots in the greenhouse. Only 
a small percentage of these mature and mosaic plants survived trans- 
planting, so that the supply of mosaic material for inoculations was thus 
considerably reduced, and therefore only a small number of healthy 
plants were inoculated at one time. 

Inoculations with juice were made by rubbing the leaves between the 
fingers so that considerable sections of the leaflets were crushed, apparently 
permitting the applied juice to be absorbed by such areas of the leaf as 
still remained free or partly free from mutilation. Such operations were 
performed chiefly upon the youngest leaves, the first applications being 
made when the plants had developed about five leaves. In Table I the 
results of these inoculations are presented. 

Table I. — Inoculations with juice from mosaic plants 



Variety and species 
inoculated. 



Time of inocu- 
lation. 



Source of juice. 



Southern 

turnip. 

Do. . 

Do. . 



Prize 



Mustard . 
Do. 



Do. 
Do. 



Pe-tsai or Chinese 
cabbage. 

Do 

Do 



Dec. 4, 1919 . 

do 

Dec. 6, 1919 . 



...do 

/Mar. 8, 192 1 

\Mar. 21, 192 1 

fMar. 9, 192 1 

\Mar. 21, 1921 

Jan. 15,1921 

Dec. 4, 1919 



....do 

Dec. 6, 1919 



Mosaic Southern 

Prize tiu-nip. 
Healthy turnip. . . . 
Mosaic Green Moun- 
tain potato. 
do 

}....do 

[Mosaic mustard. . . . 

Mosaic pe-tsai or 

Chinese cabbage. 

do 

Healthy 

Mosaic Green Moun- 
tain potato. 



Number 

of plants 

inoculated. 


Number 
of plants 
mosaic." 


9 


6 


9 

5 






5 





6 





8 


5 


5 


3 


8 


6 


5 
4 







Per cent- 
age mosaic. 



67 



63 
60 

1^ 



<» Date of last observation April 2, 1920 and 1921. 



The data indicated in Table I disclose the fact that mosaic mottling 
was obtained only when juice from a mosaic plant was introduced into 

I ScHULTz, E. S., FoLSOM, Donald, Hildebrandt, F. Merrill, and Hawkins, Lon A. investigations 
ON THE MOSAIC DISEASE OF THE IRISH POTATO. In Jour. AgT. Research, v. 17, no. 6, p. 247-274, pi. A-B 
(col.), 25-30. 1919. Literature cited, p. 272-273. 

ScHULTz, E. S., and Folsom, Donald, transmission of the mosaic disease of irish potatoes. In 
Jour Agr. Research, v. 19, no. 7, p. 3x5-338, pi. 49-56. 1920. 

ScHULTZ, E. S., and Folsom, Donald, leafroll, net-necrosis, and spindling-sprout of the irish 
POTATO. In Jour. Agr. Research, v. 21, no. i, p. 47-So, pi. 1-13 1921. Literature cited, p. 7S-80. 



Oct. IS, I92I Transmissible Mosaic Disease of Cabbage 175 

a plant of the same or a closely related species.^ No mosaic mottling 
appeared on any of the cruciferous plants inoculated with juice from 
mosaic potato. With a more adequate supply ofcrucifer mosaic mate- 
rial and repeated applications it is probable that every plant treated 
would have developed mosaic mottling, such as has frequently been 
obtained with mosaic potato juice inoculations on the Irish potato.^ 

The first mosaic mottling was observed from 20 to 30 days after inoc- 
ulation, which also corresponds very closely with the incubation period 
for mosaic of Irish potato. The results in Table I also disclose successful 
inoculations on plants in different species of Brassica. Further evidence 
on this interspecific infection is presented in Table II on transmission by 

means of aphids. 

TRANSMISSION WITH APHIDS 

Since aphids were found on every mosaic plant examined in the field 
and on account of the fact that these insects have been found to transmit 
mosaic of tobacco,^ spinach blight/ and mosaic of potato,'' experiments 
were carried on with these insects. Aphids belonging to Myzus persicae 
Sulz. were used in this investigation. These insects were originally 
collected from the morning-glory and transferred to healthy turnip and 
mustard plants on which they were cultured while confined under cages 
until needed for inoculation. Neither the morning-glory nor the turnip 
or mustard plants on which these insects fed before being transferred 
to mosaic Chinese cabbage and turnip developed mosaic mottling. 
This indicates that the morning-glory apparently was free from mosaic, 
at least from the type which could infect the crucifers used in this 
experiment. 

When the healthy plants for inoculation had developed from five to 
eight leaves, aphids were transferred from the cultures to mosaic plants, 
where they were allowed to feed for a few days before they were intro- 
duced to the healthy plants. All inoculated plants also were confined 
in cages so as to prevent dispersal from one species to another. After 
the aphids had fed from 7 to 14 days on the inoculated plants they were 
killed by tobacco fumigation in a fumigation chamber. These plants 
were now allowed to grow without cages in a greenhouse where fumiga- 
tion was practiced at regular intervals for the control of aphids. Since 
mosaic mottling developed from 12 to 30 days after these insects were 
killed by fumigation, mosaic mottling can not be attributed simply to 
the mechanical injury produced by the aphids. This fact is further 

J Gardner, Max W., and Kendrick, James B. turnip mosaic. In Jour. Agr. Research, v. 22, no. 3, 
p. 123-124, I pi. 1921. 

- ScHXJLTz, E. S., Foi^oM, Donald, Hudebrandt, F. Merrill, and Hawkins, I,on A. op. ot. 

' Allard, H. A. THE MOSAIC disease OF TOBACCO. U. S. Dept. Agf. Bul. 40, 33 p. , 7 pi. 1914. 

* McCuNTOCK, J. A., and Smith, Loren B. true nature of spinach-blight and relation of insects 
TO ITS transmission. In Jour. Agr. Research, v. 14, no. i, p. 1-60, pi. A (col), i-ii. 1918. 

6 Schultz, E. S., Folsom, Donald, Hildebrandt, F. Merrill, and Hawkins, Lon. A. op. ciT. 



176 



Journal of Agricultural Research voi. xxn. no. 3 



confirmed by the control plants which remained free from mosaic mottling 
after aphids taken from healthy plants had fed upon them. The results 
secured from inoculation with aphids are presented in Table II. 

Table II. — Transmission of mosaic of m,ustard, pe-tsai, and turnip by means of aphids 



Variety inoculated. 


Date of inocu- 
lation. 


Ap- 
proxi- 
mate 
num- 
ber of 
aphids 
trans- 
ferred. 


Source of aphids. 


Num- 
ber of 
plants 
inocu- 
lated. 


Date of first 
symptoms. 


Num- 
ber of 
plants 
mo- 
saic. « 


Per- 
centage 
mo- 
saic. 


Southern Prize tur- 
nip. 

Do 


Jan. 12,1920 

Jan. 26, 1920 
Jan. 12,1920 
Jan. 22,1920 

Feb. 24, 1920 

Mar. 5, 1920 
Mar. 10,1920 
Jan. 12,1920 
Jan. 19, 1920 
do 


5° 

SO 
5° 
12 

100 

100 
50 

50 

100 

100 

50 

so 
50 
25 

50 

2S 
12 

2S 
25 
SO 


Mosaic Southern 

Prize turnip. 
do 


2 

3 
3 
3 

2 

3 
I 
I 
3 

I 
3 
I 

s 

3 

S 
3 
3 

I 
5 
S 


Feb. 3 , 1920 
Feb. 20, 1920 


3 

3 

3 

3 

2 
I 
I 
3 


3 



5 

3 

3 

2 

3 

I 
3 



100 


Do 


Healthy turnip. . . 
Mosaic Southern 

Prize turnip. 
Mosaic mustard. . . 

do 




Purple Top turnip.. 

Seven Top or South- 
Prize turnip. 
Do 


Feb. 14, 1921 

Mar. 20, 1930 

Mar. 31,1920 
Apr. 2,1920 
Feb. 11,1920 
Feb. 19, 1920 


100 
100 


Do 


do 






do 




Do 


do 




Do 


Healthy mustard . 
Mosaic mustard . . . 
Healthy mustard. . 
Mosaic mustard . . . 
Mosaic Southern 

Prize turnip. 
do 




Do 


Jan. 26, 1920 
. . .do 






Do 






Do 


Mar. 12,1921 
Jan. 29,1921 

Feb. 19, 1921 
Feb. 8, 1921 
Jan. 15, 1921 

Jan. 29, 1921 
Feb. 19,1921 
Mar. 2,1921 


Apr. 2,1921 
Feb. 20, 192 1 

Mar. 25,1921 




Do 




Do 


40 


Do 


do 


Do 


Mosaic Chinese 

cabbage. 
do 


Feb. 14,1921 

Feb. 20,1921 
Mar. 38,1921 




Do 




Do 


do 


60 


Do 


Mosaic Green 
Mountain potato. 











« Date of last observation, Apr. 2, 1920 and 1921. 

From the data indicated in Table II it is evident that aphids transmit 
mosiac of the crucifers in question between different species as well as 
among plants of the same species, as was suggested in Table I on juice 
inoculations by means of rubbing. It will also be noted that the period 
in which the first mosaic mottling appeared corresponds very closely to 
that obtaining with mosiac diseases of other plants. As with the plants 
inoculated by rubbing, the plants inoculated by means of aphids devel- 
oped the mosaic symptoms only on the younger leaves. Mosaic symp- 
toms on the inoculated plants were like those which were observed on 
mosaic lots taken from the field. 

Since turnips from mosaic plants taken from the field continued to 
produce mosaic foliage it is apparent that such plants become a source 
of infection if planted near susceptible varieties. Mustard seed from 
mosaic mustard plants apparently develop healthy seedlings. This was 
observed in loo seedlings, which were grown from seed from mosaic 
mustard in the fall of 1920; in this test every seedling was free from 
mosaic mottling. 



oct.is. I92I Transmissible Mosaic Disease of Cabbage 177 

SUMMARY 

From these preliminary observations and experiments it appears that 
the crucifers here mentioned may be added to the Hst of plants suscepti- 
ble to mosaic, a disease whose cause has not been discovered but which 
can be transmitted from mosaic to healthy plants by direct transfer of 
juice as well as by means of aphids which apparently are very effective 
natural agents in the dissemination of this disease. 



PLATE B 

I. — Leaf from healthy turnip, control to mosaic turnip in figure 2. 
2. — Leaf from mosaic turnip, mosaic induced by aphids transferred from mosaic 
turnip plant. 

3. — Leaf from healthy mustard, control to mosaic mustard in figure 4. 

4. — Leaf from mosaic mustard, mosaic produced by aphids from mosaic mustard. 

(178) 



A Transmissible Mosaic Disease of Chinese Cabbage 



Plate B 




Journal of Agricultural Research 



k 



Vol. XXII, No. 3 



Transmissible Mosaic Disease of Cabbage 



Plate 22 





Journal of Agricultural Researcli 



Vol. XXII, No. 3 



PLATE 2 2 

A. — Healthy turnip plant, control to B. Aphids from healthy turnip were allowed 
to feed on this plant. Planted the same time as B. 

B. — Mosaic on turnip plant, variety Seven Top or Southern. Mosaic mottling 
appeared 26 days after the introduction of aphids from a mosaic Southern turnip 
plant. 

C. — Leaf from A, healthy. 

D, E. — Two mosaic leaves from B. Mosaic mottling and ruflSing apparent on the 
diseased leaves. 

65508°— 21 5 



I 



PLATE 23 

A. — Healthy mtistard plant, control to B. Planted the same time as B. 
B.^ — Mosaic on mustard plant produced by transferring aphids from mosaic mustard. 
Distinct mosaic mottling was noted 28 daj^s after introduction of aphids. 



Transmissible IVlosaic Disease of Cabbage 



Plate 23 




Journal of Agricultural Research 



Vol. XXII, No. 3 



Transmissible Mosaic Disease cf Cabbage 



Plate 24 




Journal of Agricultural Research 



Vol. XXII, No. 3 



PLATE 24 

Leaves from plants shown in Plate 23, A, B. 

A, B. — Mosaic leaves showing mottling and ruffling. 

C— Healthy leaf. 



ADDITIONAL COPIES 

OF THIS PUBLICATION MAT BE PEOCUKED FROM 

THE SUPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTING OFFICE 

WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 
Subscription Price, $3 Per Year 

V 



Vol. XXII OCTOBER 22, 1921 No. ^ 

JOURNAL OP 

AGRICULTURAL 
RESEARCH 



coN'rE>N'rs 

Page 

Flora of Corn Meal --------179 

CHARLES THOM and EDWIN LeFEVRE 

( Contribution from Bureau of Chemistry ) 

Hopkins Host-Selection Principle as Related to Certain 
Cerambycid Beetles ------- 189 

F. C. CRAIGHEAD 

( Contribution from Bureau of Entomology) 

Notes on the Organic Acids of Pyrus coronaria, Rhus 
glabra, and Acer saccharum - - - - - - 221 

CHARLES E. SANDO and H. H. BARTLETT 

( Contribution from Bureau of Plant Industry and University of Michigan ) 

Fertility in Shropshire Sheep - - - - - -231 

ELMER ROBERTS 

( Contribution from Illinois Agricultural Experiment Station ) 



PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



^W:A.SHlNaTON, D. c. 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALIyEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATIOII 

J. G. UPMAN 

Dean, State College of Agriculfurei and 
Director, New Jersey Agricultural Experi- 
ment Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS 

Dean, School of Agriculture; and Diredor: 
Agricultural Experiment Station; Tht 
Pennsylvania State College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellennan, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultxiral Experiment Station, New 
Brunswick, N. J. 




joim OF AGrtrmL re 



Vol. XXII Washington, D. C, October 22, 1921 xM'^j^^ 4 

FLORA OF CORN MEAL '^ >".:-.*. 

By Charles Thom, Mycologist in Charge, and Edwin LeFevre, Scientific Assistant, 
Microbiological Laboratory, Bureau of Chemistry, United States Department of Agri- 
culture 

INTRODUCTION 

Corn meal as it comes from the mill carries the mycelia of certain fungi 
which infect unground grain. In addition, numerous species of molds 
and bacteria, present in spore form as contaminations upon the surfaces 
of sound kernels or as saprophytes in partially spoiled grains, are recov- 
erable by routine cultural examination of the finished meal. Many 
experiments, extending over several years and including the work of vari- 
ous members of the Microbiological Laboratory, show that certain groups 
of organisms are practically always abundant in such cultures. Other 
species are usually present, but in smaller numbers, and many forms are 
obtained occasionally as accidental contaminations. In undertaking to 
study this complex flora, it may be possible to determine by routine cul- 
ture the species represented and something of their relative abundance in 
the sample, but the list so obtained gives little information as to the 
relative importance of the individual species as causes of spoilage in the 
product. 

The culture media commonly used in such routine examination of food- 
stuffs present conditions for the growth of microorganisms which differ 
greatly from those found in corn meal. The nutrients used in preparing 
such media are selected because they are readily assimilable to most 
organisms. These nutrients appear in solution or in jelly-like masses 
which contain high percentages of moisture. Corn meal, on the other 
hand, presents a range of composition, according to Winton and his 
associates {8),^ approximately as follows: Moisture, 10 to 18 per cent, 
but under usual commercial practices ranging from 12 to 15 per cent; 
protein, 5 to 10 per cent; fat, i to 5 per cent, according to the method of 
milling; nitrogen-free extract, including starch and sugar, 68 to 78 per 
cent. Of the nitrogen-free extract, sugars constitute perhaps 3 per 
cent, and gums and dextrin, some of which are readily fermentable,, 
perhaps an equal quantity. In dealing with this product as a substratum 
for organisms, the percentage of water found is an important limiting 

I Reference is made by number (italic) to " lyiterature cited," p. i88. 



Journal of Agricultural Research, Vol. XXII, No. 4 

Washington, D. C. Oct. 22, 1921 

aab Key No. E-16 

(179) 



i8o Journal of Agricultural Research voi. xxii, no. 4 

factor. Obviously this product, 'even at its maximum moisture content, 
presents a marked contrast to laboratory media as usually prepared. 
Nevertheless, corn meal has been so often found an unstable product that 
it is commonly milled only for consumption within a few weeks or by 
methods intended to eliminate the most readily fermentable portions of 
the grain. 

Under ordinary conditions of handling, spoilage in this product ap- 
pears in one of the following forms: Souring, rancidity, mustiness, the 
formation of clumps or balls, extensive concretions which may involve 
the solidification of an entire bag, or the formation of a hard, cylindrical 
outer mass with the center loose and mealy. Heating occurs only in 
the wettest samples. Much com meal, if held beyond a very short 
period, develops a musty, moldy, or sour odor and shows occasional 
balls or masses of meal held together by mold, which bring about losses 
in palatability and market quality in the product. Such changes as 
rancidity and the formation of extensive concretions into moldy masses 
are so obviously due to high moisture content and involve such losses 
that they have been almost eliminated from commercial practice. 
When losses occur the meal is found to carry more than a critical mois- 
ture percentage. This may be due either to milling com which is in- 
sufficiently dried or to the storage of the meal under conditions which 
will maintain a moisture content above the danger point. For the 
samples used in all series reported here this figure was approximately 

13 per cent (2). 

CULTURAL EXAMINATION 

In routine cultural examination reported here, plain agar was used 
for bacterial counts, wort agar for mold counts, and dextrose-litmus 
shake agar to determine acid, gas, and anaerobic growth. The presence 
of particular organisms was determined by tlie use of special methods 
on special media. Experimentation covered a range wide enough to 
justify the restriction of routine cultures to the media already noted. 

After comparative study of many series of cultures. Table I is intro- 
duced as giving a group of cultural results fairly typical for commercial 
meal in sound, merchantable condition. The nine samples reported 
were purchased in different retail stores of Washington, D. C, during 
October and November, 1920. Four of them were yellow and fairly 
coarsely ground. The white meals were softer or more finely ground. 
All were bolted. All showed by microscopic examination traces of 
both bran and germ, although these portions were scanty in certain 
samples. The history of the samples was not obtained. 

These samples were sound in appearance and odor. There was no 
evidence of the multiplication of microorganisms. Among the bacterial 
colonies micrococci, members of the mesentericus and of the colon- 
aerogenes groups were characteristically present. Special tests in cab- 



Oct. 22, I92I 



Flora of Corn. Meal 



I8i 



bage juice showed, in four of the nine samples, the presence of lacto- 
bacilH with the morphology and cultural characters of the organism of 
pickle and sauerkraut fermentation. N>) bacterial colonies were obtained 
in plain agar from two of the samples. A duplicate of sample 9 proved 
equally negative. Mold colonies were obtained in all samples. These 
represented in varying proportions Aspergillus repens De Bary, A. niger 
Van Tieghem, A. fiavus hink, Fusarium, various mucors, and unidentified 
colonies. 

Table I. — Results of cultural examination of commercial corn meals 



Sample No. 


Bacteria 

per gram 

on plain 

agar. 


Molds per 

gram on 

wort agar. 


Bacteria 
per gram 

on 
dextrose- 
litmus 
agar. 


Acid 
colonies. 


I 


10, 000 


TO. 000 


16, 000 


Per cent. 
50 




5, 000 I, 000 

t;=;, 000 T3. 000 


•^ 


42, 000 

13, 000 

10, 000 

8, 000 


60 




60, 000 

70, 000 

5,000 


10, 000 
400, 000 

20, 000 
3,000 

11, 000 
5,000 


50 
60 




6 


7 




8 


10, 000 


4, 000 
3,000 







30 







A more extensive series of studies was conducted in cooperation with 
the Plant Chemical Laboratory of this bureau. The general results 
of this experiment are described elsewhere (2). In brief, during the 
spring of 1920, a series of bags of meal were prepared for this storage 
experiment from com bought by the mill in the regular course of business. 
This grain, while sold as No. 2, was obviously wet and barely passable 
as a fair product. Infected and even badly decomposed ears were not 
uncommon among the ears of corn received in bulk. Although the lots 
of meal included were milled at water contents varying from 12.7 to 
16.18 per cent, the conditions of storage were such that no spoilage 
determinable by the senses took place. Cultures were made from the 
meal as freshly ground in April, then, beginning May 5, once each week 
until July. In all these cultures no evidence of multiplication of either 
mold or bacteria was found. It was, therefore, possible to follow the 
relative numbers of viable organisms in the various groups from the 
time of grinding through the four months of storage. 

In the freshly milled samples the average count of colonies of bacteria 
upon plain agar was about 1,000,000 per gram of meal, with variations 
from 600,000 to 1,600,000. Upon wort agar the count of mold colonies 
averaged about 100,000 per gram of meal, with variations in different 
samples from 70,000 to 160,000. Of the bacterial colonies observed 
about 60 per cent were acid producers. 



1 82 Journal oj {Agricultural Research voi. xxn, no. 4 

For comparison a special series of samples were prepared by adding 
5 per cent of meal made from com markedly rotted with Diplodia and 
Fusarium. In the freshly grouted meal of this series the bacterial count 
upon plain agar was about 2,600,000. The count of mold colonies upon 
wort agar was about 1 10,000. About 70 per cent of the bacterial colonies 
were acid producers. 

After storage for approximately one month (May 20 and 21) samples 
from a particular lot of five bags of the regular meal showed an average 
count of 108,000 bacterial colonies and 15,000 molds. Samples from the 
same bags on June 30 showed an average count of 12,600 bacterial 
colonies, and 7,600 mold colonies. Without placing emphasis upon 
exact figures, these cultural results are fairly typical of the mass of 
figures obtained from cultures made weekly from representative samples 
involving the whole series of 88 bags of meal. These figures are readily 
comparable with those obtained from commercial samples (Table I). 
Discrepancies which occur may perhaps be accounted for by the fact 
that samples 3, 4, and 5 were evidently the product of local mills, sold 
fairly quickly after milling, while samples 2, 7, and 9 were clearly the 
product of special processes and handled under conditions involving 
much slower distribution. 

In this lot of meal, therefore, the conspicuous change due to storage 
was the drop in the number of viable organisms to about i per cent of 
the original number of bacteria and perhaps 10 per cent of the original 
number of molds. The larger part of this decrease occurred during the 
first six weeks, with a slow reduction throughout the succeeding periods. 

In connection with the study of these figures, data obtained by Thom 
and Stiles (unpublished) in examining Winton's {8) samples ^ in 19 14 
were restudied and compared with the results here considered. Win- 
ton's com meal varied in initial moisture content from 19.27 to 10.79 
per cent. In those lots of meal (A, B, and C) carrying moisture mark- 
edly above 13 per cent, the evidence of multiplication of molds and 
bacteria was clearly discernible. Musty odors and balls of meal held 
together by mold were present in every sample. In cultures, the count 
of colonies of molds and bacteria reached 13 million in the wettest lot. 
Of these several million were Aspergillus flavus. The predominant or- 
ganisms were molds rather than bacteria, but there was fairly clear 
evidence of some bacterial multiplication at the higher water percentages. 

In the roller-gi'ound samples of lots D, E, F, which did not spoil and 
whose water percentage was near to or less than 13, the total counts 
found by Stiles approached very nearly those already given in this paper. 
These examinations began too late in the storage period to show that 
part of the bacterial flora which dies off rapidly. The stored samples 
still showed some acid organisms, but micrococci and aerobic spore 



' Samples of the meal studied were examined bacteriologically by G. W. Stiles, formerly of the Bureau 
of Chemistry, and for mold activity by Charles Thom, then in the Bureau of Animal Industry {8, p. zs)- 



Oct. 22, 1921 Flora of Corn Meal 1 83 

formers of the mesentericus group formed the majority of the bacteria 
obtained. 

In the lots with moistm^e content decreasing toward 13 per cent there 
was progressive reduction in the number of active species of molds. 
Bxtensive experimentation showed clearly that Aspergillus repens was 
the agent which formed the balls of meal loosely held together with 
mold hyphae, which characterized meal containing barely enough 
water to start spoilage. In another series of experiments A. flavus 
began to be active only in samples containing about 16 per cent of water. 
Yeasts, mucors, and Penicillia were reported by Stiles only in the sample 
carrying about 19 per cent of water. 

During the examination of the preliminary samples in the 1920 ex- 
periment, an effort was made to identify the groups or actual species 
represented. As a matter of routine, inoculations were made from each 
flask prepared for diluting plates (consisting of 5 gm. of the meal to 
45 cc. of sterile water) into the following media: Plain milk, gelatin, 
and litmus lactose broth. Smears were also made on Endo's agar in 
each instance. In every case there was prompt coagulation of the milk, 
with extrusion of whey, but no digestion of curd. Pink rings formed 
near the surface. Gelatin was liquefied in every instance, and acid and 
gas formed in all broth tubes. Growth in Endo's media indicated the 
presence of Bacterium aerogenes Escherich. Further cultural studies 
showed that Bad. aerogenes was the predominant bacterial species pres- 
ent in all these samples. This predominance was maintained throughout 
the series of examinations made. Microscopical examinations of smears 
made in each case, however, showed the presence of spore-bearing bac- 
teria, especially the mesentericus group, and micrococci of various kinds. 
Dextrose agar tubes often contained colonies growing deep in the media, 
indicating the presence of anaerobic bacteria. Yeasts were found in all 
samples, their growth being largely of the mycoderma type. The plates 
showed many mold colonies. Various mucors, species of Fusarium, 
Aspergillus flavus, A. niger, and occasional green Penicillia were ob- 
served. The species of molds present on the plates varied from period 
to period and with the sample. Molds were always more numerous on 
plates made from meal to which Fusarium and Diplodia had been added, 
but growth on these plates did not show dominance of these particular 
forms. 

Evidence of the effect of bolting upon the abundance of organisms 
was furnished in the 1920 experiment by the examination of samples 
of two series of five bags each, representing a single lot of meal, one-half 
of which was bolted and the other half unbolted. The bolting to which 
these samples were subjected removed a considerable part of the bran 
but little of the germ from the meal. After one month of storage, the 
bolted meal showed an average of 34,000 bacterial colonies and 20,000 
mold colonies. The unbolted samples showed 108,000 bacterial colonies 



184 Journal of Agricultural Research voi. xxii. no. 4 

and 15,000 molds. This observation was confirmed by a restudy of 
Stiles's unpublished examination of Winton's (8) samples. Of every 
lot of com handled, part was ground in a stone mill without sifting or 
bolting and part was carefully " degerminated " and "roller" ground. 
In the bolting process all of the bran was taken out, and many of the 
samples consisted almost completely of homy endosperm. In that part 
of this series made up of meals in which no multiplication of micro- 
organisms occurred, bolting consistently reduced the cultural count of 
microorganisms below that of the stone-ground meal. Frequently the 
number found in the bolted meal was less than one-tenth of that in the 
stone-ground meal. 

By removing the bran, bolting takes away the largest area of contami- 
nation with saprophytic organisms. The tip of the kernel and the 
germinal area carry the majority of the infections found in com. Study 
of many samples of com over a period of years shows that invasion of the 
germinal area by molds is not uncommon in com which has not been 
fully matured or has not been promptly and thoroughly dried. Sam- 
ples have frequently shown the invasion of the germ in every kernel by 
Aspergillus repens. Recently samples representing a bulk shipment 
have shown nearly every grain to contain one or the other of two species 
of Penicillium. Meal therefore may be so milled and sifted or bolted 
as to remove the larger part of all contaminations, as well as those mold 
infections which do not involve general disintegration. The cleaning 
process before milling removes the grains thoroughly rotted by Fusarium 
and Diplodia. Corn has still been seen going into the rolls of a mill 
in which the low grade of the stock could not have been concealed if it 
had passed through a stone mill without being bolted. The product, 
however, was going into human food without showing tangible evidence 
of the low quality indicated by the unground grain. In other words, 
the fractional milling of low-grade grain makes possible such separation 
as turns the infected portions of the grain into oil stock or cattle feed 
and the solid or homy portions which are less obviously damaged into 
meal. 

The literature of maize deterioration is reviewed by Alsberg and Black 
up to 19 13 (7). The activity of Fusarium and Diplodia as causes of 
rotting in ear com was discussed by Burrill and Ban-ett (5) and that of 
Diplodia alone by Heald, Wilcox, and Pool (4). 

More recently McHargue (6) has studied the activities of certain fungi 
and their relation to commercial conditions in the handling of the product. 
Excessive moisture in the grain is regarded as the limiting factor in most 
cases of such spoilage. The factor of temperature must not be over- 
looked. The moisture content limit may be materially increased during 
the winter without evidence of the activity of microorganisms. The 
agents of spoilage in all the cases under review were primarily molds. 
The results already given in this paper harmonize in general with those 



Oct. 22, I92I Flora of Corn Meal 185 

of McHargue. It has been possible, however, to go farther and indicate 
more clearly the groups of organisms regularly present and to record 
the conditions under which certain of them become active factors in 
spoilage. 

Routine mass or dilution cultures show that certain molds are recov- 
erable from practically all samples of meal. Among these are Rhizopus 
nigricans Ehrenberg and some of the mucors which frequently overgrow 
plate cultures within two days of incubation, although they probably 
are present only in spore form in the meal. Syncephalastrum, belonging 
to the same group, is not uncommon. Aspergillus flavus and A. niger 
are only occasionally visible factors in the infection of the unground 
grains, but they always appear as rapidly growing colonies in the mass 
or dilution cultures made. The brown masses of A. tamari Kita are 
commonly found with A. flavus. A. fumigatus Fres. and A. terreus Thorn. 
are frequently present but are quickly overgrown by the more active 
species already mentioned. A. repens, though practically always pres- 
ent, can be found only by careful search in the presence of these rapidly 
growing forms. 

Several strains of Penicillium are found in meal cultures. Peniciilia 
of the group with submerged orange mycelia and of the Citromyces group 
are probably most common. Penicillium expansum Link is reported by 
McHargue. P. oxalicum. Thom and Currie is found in many samples of 
meal, but rarely in miscellaneous cultural work. Strains related to 
P. luteum Zukal and P. purpurogenum O. Stoll are frequently present but 
usually indicate soil contamination rather than active growth in the corn 
or meal. One sample of corn rotted by a member of this series has been 
examined, but the conditions shown clearly indicated that the product had 
contained high percentages of moisture at the time the rotting occurred. 

Colonies of Fusarium develop from almost every sample of meal. 
Infections of this group are so abundant that conidia or grains of meal 
containing living hyphae are rarely absent. Cladosporium and Alter- 
naria are frequently found but represent spore contamination rather than 
infection. The other organisms observed in culture from time to time 
appear to represent excessive contaminations with spores due to unfavor- 
able conditions in the handling of the product, or, in certain species, to 
actual infection of the grain locally by the mold. 

The bacteria found in the fresh samples here considered were pre- 
dominantly Bacterium aerogenes. Certain other organisms have been 
regularly obtained in culture. When the necessary moisture is present, 
souring is so characteristic of the product that Round and Gore (7) found 
the addition of 3 per cent of fresh meal an adequate starter to insure the 
dominance of lactic acid fermentation in potato silage. Lacto-bacilli 
were present in four of the nine lots reported in Table I. According to 
unpublished records in the Microbiological Laboratory, Round found 
organisms of this group abundant also in fresh meal, but occasionally 



1 86 Journal of Agricultural Research voi. xxn. no. 4 



absent in old meal or meal made from old and thoroughly dried corn. 
Micrococci are constantly encountered in culture but have not been 
typed. Aerobic spore formers of the mesentericus group are always 
present, and in spore form they constitute the larger part of the living 
bacteria in some meals after long storage. 

This was clearly demonstrated by a series of experiments upon the 
possibility of producing a sterile meal with steam, dry heat, or both 
(unpublished cultural results of Ruth B. Edmondson). The spores of 
this group survived more heating than could be applied under practical 
working conditions to the product. Aside, however, from meal so wet 
as to be unmarketable, these experiments show no evidence of bacterial 
activity. One sample of apparently sound yellow meal showed the 
presence of Bacillus niger Migula in such extensive numbers that masses 
of meal placed upon culture media were promptly overgrown and with 
the agar turned bluish black with this species. The meal was contributed 
by Dr. S. S. Adams, of Washington, D. C, who reported the feces of a 
child apparently well to have been blue when fed this meal. 

When, however, com or meal is bottled and incubated at laboratory 
temperature (20° to 30° C), those species capable of developing under 
the conditions presented show active growth. In the authors' series 
such growth was not detected by physical appearance in products carry- 
ing less than 13 per cent of moisture. Certain stone-ground samples of 
Winton's series (8) showed some evidence of mold activity below that 
figure. Measurable changes in quality certainly occur in such meals 
during storage. Some experimental results have suggested the possibility 
that these changes in such meal are due to the distribution of infected 
material throughout the mass by the grinding of infected corn. This 
conflicts with the current trade belief that the natural enzyms of the 
germinal area are the chief causes of such deterioration, but reflects the 
findings of HofTer (5) and his coworkers that even selected seed com 
may be extensively infected. Examinations of commercial samples in 
the Microbiological Laboratory have shown extensive development of 
molds within the grain itself in com of other than the higher grades. 

In samples carrying 14 to 15 per cent of water the formation of balls 
and concretions in the meal begins to be evident. The principal agent 
in their formation appears to be Aspergillus repens, although many diffi- 
culties are encountered in fixing a minimum moisture percentage for 
the activity of this species. Changes involving the development of 
mold mycelium in the meal begin within the limit of 13 to 15 per cent 
of moisture. Incubation at 20° to 30° C. merely accelerates changes 
which would progress more slowly in colder places. Moist chamber 
experiments with meal inside this range of water content show the 
presence of active mycelia of more than a single species, but principally 
Aspergillus repens. When the percentage of moisture reaches 16, sev- 
eral species are clearly able to grow. Special studies with Aspergillus 



Oct. 22, I92I Flora of Corn Meal 187 

flavus show that very little development of this species occurs below 16 
per cent, but that from 16 per cent upward development of this species 
rapidly increases and the number of forms capable of growing rapidly 
rises. Among the characteristic saprophytic molds observed under 
these conditions, in about the order of their abundance under the con- 
ditions, are Aspergillus repens, Aspergillus flavus, Actinomyces sp., 
Penicillium sp. and Citromyces sp., Fusarium sp., Aspergillus candidus, 
Aspergillus ochraceous Wilhelm, Aspergillus iamari, and Aspergillus niger. 

Bacterial activity appears to be a concomitant of the disintegration 
due to mold action in such rotting processes as this. As indicated by 
Bailey and Thom (2, Table I), active disintegration by molds is 
accompanied by an increase in the water percentage of the sample. 
Bacteria follow rather than initiate the process in the samples studied, 
thus becoming a small factor in the merchantable product. 

Throughout this investigation a close correspondence has been observed 

between the flora of deterioration in unground com and the flora of the 

milled product. 

SUMMARY 

In seeking possible causes for the well-recognized instability of com 
meal, cultures show considerable numbers of molds and bacteria to be 
generally present. Among these the following species of molds were 
characteristic of many series of cultures: Fusarium sp., Aspergillus 
repens, A . flavus, A . tamari, A . niger, Citromyces (or Penicillium section 
Citromyces) sp., Penicillium oxalicum, P. luteum varieties, Mucor sp., 
Rhizopus nigricans, and Syncephalastrum sp., together with various 
yeasts and yeast-like fungi. Among bacterial groups, the colon- aerogenes 
group and lacto-bacilli were most abundant in fresh meal. Aerobic 
spore formers and micrococci were always present and persisted in the 
stored product. 

Within the range of composition found in merchantable meals, no 
bacterial activity was detected. Only one grade of unbolted meal 
showed signs of mold development below 13 per cent of moistiu-e. Above 
13 per cent moisture, Aspergillus repens begins to be an active agent 
of spoilage somewhere between 13 and 15 per cent of moisture, varying 
with the form of milling practiced. Several other species of molds are 
active in meal containing 16 per cent moisture; and numerous forms, 
including some bacteria, develop when 18 to 20 per cent of moisture is 
found. 

Many samples of corn are found to carry extensive infections with 
Fusarium, Diplodia, Aspergillus repens, or Penicillium, especially in the 
germinal area and in the tip of the kernel. These sections of the kernel 
are removed in varying degrees by different milling systems. The 
bolted meals examined show a corresponding reduction in count of 
viable organisms as shown by culture. 



1 88 Journal of Agricultural Research voi. xxu, no. 4 

LITERATURE CITED 
(i) Alsberg, Carl L., and Black, Otis F. 

19 13. CONTRIBUTIONS TO THE STUDY OF MAIZE DETERIORATION. BIOCHEMICAL 
AND TOXICOLOGICAL INVESTIGATIONS OF PENICILLIUM PUBERULUM ANI> 

PENiciLLiUM STOLONiFERUM. U. S. Dept. Agr. Bur. Plant Indus. 
Bui. 270, 48 p., I pi. Bibliographical footnotes. 

(2) Bailey, L. H., and Thom, C. 

1920. SOME observations of corn meal in storage. In Operative Miller, 
V. 25, no. 12, p. 368-371, chart A-D. 

(3) BuRRiLL, Thomas J., and Barrett, James T. 

1909. ear rots op corn. 111. Agr. Exp. Sta. Bui. 133, p. 63-109 incl. pi. 
i-ii, I col. pi. 

(4) Heald, F. D., Wilcox, E. M., and Pool, Venus W. 

1909. THE life-history AND PARASITISM OF DIPLODIA ZEAE (SCHW.) LEV. Itl 

Nebr. Agr. Exp. Sta. 22nd Rept. [1908], p. 1-19 incl. 10 pi. Biblio- 
graphy, p. 7. 

(5) HoFFER, George N., and Holbert, J. R. 

1918. SELECTION OF DISEASE-FREE SEED CORN. Ind. Agt. Exp. Sta. Bui. 224, 
16 p., 20 fig. 

(6) McHargue, J. S. 

1920. THE cause OF DETERIORATION AND SPOILING OF CORN AND CORN MEAL. 

In Jour. Indus, and Engin. Chem., v. 12, no. 3, p. 257-262. 

(7) Round, L. A., and Gore, H. C. 

1916. A PRELIMINARY REPORT UPON THE MAKING OF POTATO SILAGE FOR CATTLE 

FOOD. In Proc. 3rd Ann. Meeting, Potato Assoc. America, p. 75-79. 

(8) WiNTON, A. L., Burnet, W. C, and Bornmann, J. H. 

I915. COMPOSITION OF CORN (mAIZE) MEAL MANUFACTURED BY DIFFERENT PRO- 
CESSES AND THE INFLUENCE OF COMPOSITION ON THE KEEPING QUALI- 
TIES. U. S. Dept. Agr. Bui. 215, 31 p. 



HOPKINS HOST-SELECTION PRINCIPLE AS RELATED 
TO CERTAIN CERAMBYCID BEETLES 

By F. C. Craighead 

Specialist in Forest Entomology, Bureau of Entomology, United States Department of 

Agriculture 

INTRODUCTION 

In connection with the reported dying of lodgepole pine (Pinus con- 
torta Loud.) over extensive areas in northeastern Oregon caused by the 
mountain pine beetle {Dendroctonus moniicolae Hopk.) and the threat- 
ened invasion by this beetle of the adjacent areas of yellow pine {Pinus 
ponderosa Laws.), detailed investigations were made by the Bureau of 
Entomology under the direction of Dr. A. D. Hopkins. Manuscript 
reports of these investigations, submitted in the summer of 19 lo, showed 
that the infestation by the beetle in the lodgepole pine was so extensive 
that there was no hope of controlling it, but that the comparatively small 
amount of infestation in the valuable stands of yellow pine was such as 
to warrant the undertaking of control, provided the beetle did not 
migrate from the lodgepole pine to the yellow pine. 

In a letter from Dr. A. D. Hopkins under date of July 30, 19 10, relating 
to a manuscript report of Mr. H. E. Burke, the following statement 
occurs which appears to be the first written reference to the host-selection 
principle : 

The more I consider the various features of the problem, the more I am convinced 
that it is entirely practicable to protect the yellow pine, even if we leave all but the 
immediately adjacent lodgepole pine to take care of itself. This is based on my 
belief that the majority of the broods of the beetles which have been breeding in the 
lodgepole will continue to confine their attack to that species, and gradually diminish 
with the reduced supply and their increased struggle to adapt themselves to the 
yellow pine. I may be wrong in this, but it is a matter worthy of careful considera- 
tion. Remember, that in all these years, there has been no marked or general migra- 
tion of beetles from lodgepole to the yellow pine. Therefore, it appears that the 
broods which are most dangerous to the yellow pine are those which have been breed- 
ing in it, and that these are the broods we will have to deal with mainly in our efforts 
to protect the best bodies of yellow pine. 

The control operations that were carried on during the following year, 
191 1, were confined mainly to the yellow pine area. In manuscript 
reports by Messrs. W. D. Edmonston and George Hofer on a special 
examination of the yellow pine and lodgepole pine areas in the summer 
and fall of 1913, it is stated by Edmonston: 

In 1912 the examination of the areas on which insect control work was carried on 
during April, May, and June, 191 1, showed an average reduction of the infestation 
on the entire area, 76,430 acres, of close to 85 per cent. 

Journal of Agricultural Research Vol. XXII, No. 4 

Washington, D. C. Oct. 22, 1921 

aac (1S9) Key No. K-ioa 



190 Journal of Agricultural Research voi.xxn.No. 4 



Examinations made this season, 1913, show a still greater reduction in the infesta- 
tion; in fact, the infestation is so light tliat it is actually less on the treated areas 
than it is throughout any other area on this Forest. 

and — 

There was no reoccupation of the treated areas by broods from the lodgepole infested 
trees at higher elevations. 

and by Hofer — 

As we reached the summit near the North Powder Peaks we attained an altitude of 
8,000 feet; the elevation at the Sheep Ranch is about 4,000 feet. From the summit 
of this divide for a distance of 10 miles north, 10 miles east, and 16 miles west we 
noted heavy infestation, both old deadings and the new work also, in both the lodge- 
pole pine and white bark pine, especially on both slopes of Antone Creek. 

No new infestation was found on the treated areas on Anthony Creek, Camp area. 

This seemed to furnish substantial evidence that the principle would 
hold. 

The principle as defined by Dr. Hopkins ^ is that an insect — 

species which breeds in two or more hosts will prefer to continue to breed in the host 
to which it has become adapted. 

In order to secure further evidence relating to this principle, the writer, 
after consulting with Dr. Hopkins, began a series of experiments in 19 14 
with insects which infest two or more species of wood. The wood-boring 
Cerambycidae, or long-homed beetles, offered material which was very 
well adapted to the conduct of such experiments. Many species were 
easily available which exhibit great diversity in their selection of hosts 
in nature, as illustrated by those breeding exclusively in a single species 
of plant and those apparently attacking almost any wood. This varia- 
tion in host habits at once brought up the following questions : Will those 
species confined to a single host live in any other, and do the individuals 
coming from a certain plant of those species breeding in a variety of hosts 
select the same species of plant on which to oviposit ? Again, if such is 
the case, how do these host strains originate in nature? 

As these experiments progressed new problems came up demanding a 
broadening of the experiments from- year to year until, during the season 
of 1918, over 100 individual experiments were in progress. Fourteen 
species of insects and 21 species of plants were used, combining to form 
45 host strains. It was thought desirable to conduct experiments on 
more species rather than more intensive experiments on a few species. 
It will be noted that certain experiments were not carried as far as others, 
due to the fact that time was not available or due to the absence of the 
writer at the critical time. At present several points remain to be con- 
clusively settled, and investigation of these will be continued another 
year or so. Nevertheless it is believed that sufiicient data have been 
accumulated to show definitely the extent to which the influence of the 
host applies to these insects. 

1 Hopkins, A. D. economic investigations of the scox,ytid bark and timber beetles of north 
AMERICA. In U. S. Dept. Agr. Program of Work, 1917, p. 353. 1916. 



Oct. 22. I92I Hopkins Host-Selection Principle 191 

HISTORICAL 

Very few references to the adaptation of insects to their host plants 
or the variation in their selection of host plants can be found. The 
most important paper dealing with the subject is that by Pictet.^ This 
author shows by several examples, (Ocneria) Porthelria dispar for 
one, that caterpillars of the second and third generations may be made 
to change their preferred food plants and that the adults reared from 
them exhibit changes in size and coloration. This paper is reviewed, 
and supplemented with reports of corresponding observations, by 
Schroder,^ who in a previous article ^ showed that even nidification (in 
Gracilaria stigmatella F.) and habits of feeding, combined with 
changes in reproduction (in the beetle Phratora vitellinoe L.). can be 
changed and that these acquired characters are transmitted spontane- 
ously from the third generation. 

In 1907 and 1908 Paul Marchal ^ succeeded in transferring numerous 
specimens of Lecanium corni Bouche from the peach (Amygdalus 
persica Linn.) to the black locust {Rohinia pseudacacia Linn.). 
Eggs hatched and larvae developed on the new host plant, spreading out 
over the leaves in large numbers, and in the fall migrating from the leaves 
to the wood for hibernation. In the summer of 1908 the insects com- 
pleted their development and had then the large size, deep coloration, 
and characteristic appearance of the insect described by Douglas as L. 
robiniarum, the attacks of which on the black locust had been severe 
in several European localities. This indicated that L. robiniarum was 
only a race of L. corni, resulting from individuals that had become 
transferred in some manner from the peach to the introduced American 
black locust. Dr. Marchal found great difficulty in reestablishing on 
the peach individuals of L. corni produced on the black locust. 

There are other records showing the acquired adaptation of certain 
species to new host plants, similar to those here cited. The practical 
application of such phenomena, however, has, so far as can be ascer- 
tained, first been recognized by Dr. A. D. Hopkins (referred to on p. 189 
of this article) and presented by him in concrete form. 

In a paper prepared by M. Joseph Capus ^ on a nematode disease of 
peas in the Gironde and read by Paul Marchal at the session of July 10, 
19 1 8, of the French Academy of Agriculture there is a record of injury to 
peas by a fungus (Fusarium vasinfectum var. pisi van Hall, considered as 
the conidial form of Necosmopora vasinfecta E. F. Smith, accompanied by 

1 PicTET, Arnold, influence de l'alimentation et de l'humidite sur la variation des papillons. 
In Mem. Soc. Phys. et Hist. Nat. Genfeve, v. 35, fasc. i, p. 45-127, pi. 2-5. 1905. 

2 Schroder, Chr. die literatur uber die FARBtmo der insekten des jahres 1905. In Ztschr. 
Wiss. Insektenbiol., Bd. 3, p. 162-164. i907- 

3 i'BER experimentall erzielte instinktvariationen. In Verhandl. Deut. Zool. GeseU., 

Jahresversamml. 13, p. 158-166. 1903. 

* Marchal, Paul, le lecanium du rohinia. Compt. Rend. Soc. Biol. [Paris], t. 6s, p. 2-5. 1908. 

^ Capus, Joseph, and Marchal, Paul, sur la maladie vermiculaire des pois dans la gironde. In 
Compt. Rend. Acad. Agr. France, t. 4, no. 25, p. 712-716. 1918. 



192 Journal of Agricultural Research voi. xxii, no. 4 

the nematode Heterodera schachti Schmidt). After pointing out the 
interdependence and relation of the two, M. Capus says: 

One might ask himself why this species, everywhere known for its injury to beets, 
does not establish itself on this plant in the Gironde and appears so abundantly 
on peas. 

Following M. Capus's explanations of this phenomenon. Dr. Marchal 
observed : 

Among the very interesting facts pointed out by M. Capus in his note I wish to 
call attention to the following: That injury to beets by Heterodera in the Gironde is 
not constant, is rare. We should recall in this connection the observations, already 
old but interesting, of the Dutch naturalist Ritzema-Bos. He has shown that when 
nematodes multiply in course of years without interruption on the same host, biologic 
races are formed adapted to this host which later pass to other vegetation with greatest 
difficulty, even when these are of those preferred by the species. It must be, there- 
fore, that, by virtue of the conditions of pea culture in the Gironde, a race of Heterodera 
schachti was formed especially adapted to peas and to the attack of which beets are 
resistant up to a certain point. There is no doubt that it will adapt itself to beets 
cultivated for a number of years in succession in the same soil infested with H. schachti. 

The experiments conducted at the Gipsy Moth Laboratory^ show that 
of the many plants tested a decided variation was found in regard to 
the susceptibility to attack by this insect. The plants were divided into 
four groups: I, favored species; II, favored food species after early 
stages; III, species on which a small proportion may develop; IV, species 
that are unfavored food. These results show that, although this insect 
has a wide variety of hosts on which it is capable of feeding, certain ones 
are selected in preference to others in the natural forests. As far as 
known, no observations have been reported showing whether or not 
several years' feeding on any particular host produces a strain which 
selects that in preference to others. 

In a recent paper Dr. C. T. Brues^ writes as follows (on p. 328-329): 

It has been claimed that the food habits may be modified experimentally, in that 
caterpillars reared on a strange plant (where they could be induced to select it) give rise 
to moths whose progeny more readily accept the new plant. It is very difficult to 
accept such evidence, at least as having any general application, without very clear 
and incontrovertible proof. If such transformations can occur so easily and become 
hereditary so quickly they should have entirely destroyed the coherent habits now 
existent, during the enormous period which has elapsed, for example, since the 
violet-feeding Argynnids were differentiated, since the holarctic and nearctic Vanes- 
sids have been separated, or while the world-wide Aristolochia-feeding Papilios were 
attaining their present distribution. That such a change has actually occurred in 
the caseof other groups seems equally evident, althoixgh, as has been shown, we can 
more easily believe that they may have arisen through mutations in maternal instinct 
not incompatible with larval tastes and then only in extremely rare cases and con- 
fined to certain groups. 

METHODS OF CONDUCTING EXPERIMENTS 

In connection with the experiments by the writer several types of 
cages, the particular type determined by the amount of material handled 

> MOSHER, F. H. FOOD PLANTS OF THE GIPSY MOTH IN AMERICA. U. S. Dept. Agr. Bui. 250, 39 p., 6 pi. 

2 Brues, Charles T. the selection of food-plants by insects, with speoal reference to lepi- 
DOPTEROUS LARVAE. /» Amer. Nat., V. 54, no. 633, p. 3i2-33"2. 1920. 



Oct. 22, 1921 Hopkins Host-Selection Principle 1 93 

and the exact conditions required, have been used in confining the colonies 
of beetles. It is essential to duplicate as closely as possible the conditions 
in which the insects are found in nature. 

For the larger logs and for experiments in which a large amount of 
material was used, an open wire insectary was constructed. This 
insectary is 40 feet long by 10 feet wide by 7 feet high. The foundation 
is of concrete, the side walls and top of i8-mesh galvanized wire screening, 
and over all a removable lattice- work roof was placed. This roof was 
adjusted to simulate shade conditions in the woods. It was removed 
in winter and replaced in summer. The floors were made of ashes to 
give good drainage. Cross partitions divided the insectary into seven 
compartments of different sizes. One room was entirely boarded in 
and roofed over. It was used for seasoning wood. Another was lined 
with cheesecloth, which was used for holding different cuts of wood under 
natural conditions until desired for use. In the other compartments 
were placed logs containing various species of insects. Where no danger 
is present of any infestation from the original host wood into cuts of 
different wood, it was possible to place several beetle species in the same 
compartment and continue their breeding in the same host from year to 
year. In this way forms such as Callidium in pine {Pinus spp.) , Neodytus 
capraea Say in ash {Fraxinus spp.), and Cyllene pictus Drury in hickory 
{Hicoria spp.) were placed together. 

The smaller insects, especially those in twigs and branches, were 
confined in glass museum cylinders of various sizes. The tops were kept 
in place so that a very constant degree of humidity could be maintained. 
This cage was found to give best results for the development of the larvae 
and, as no sand was needed, the adults were easily found in the cages. 
These jars were kept under a roof all the year in another insectary. 

Each of these insects has a particular preference for a certain condition 
or seasonal cut of wood. Also in some species the adults require food 
before ovipositing, consisting of green bark from twigs, leaves, or fungus 
s'pores. In the latter case the pustule of the chestnut blight {Endothea 
parasitica (Murr.) P. J. and H. W. Anderson) was used. Again, some 
require much moisture, others rather dry surroundings. The determi- 
nation of these factors sometimes delayed the successful continuance of 
a species for a year or more. When a new colony was collected from 
nature it was ascertained as nearly as possible when the tree died and the 
condition of the wood, also what degree of humidity was desirable. For 
instance, those insects naturally feeding in dead branches of a standing 
tree required drier conditions than those attacking branches fallen to the 
ground. 

In order to meet these conditions, wood of the various species used was 
cut every month or every other month of the year and stored under 
different conditions. Part was placed in the dry shed for dry seasoning, 
part hung or stood up in the open-air cage for normal air seasoning, and 



194 Journal of Agricultural Research voi. xxii, no. 4 

part laid on the ground in the wire cages for wet seasoning. The con- 
dition of the wood on which the insects were first found ehminated the 
use of certain of these periodic cuts and conditions of seasoning. How- 
ever, the first-year adults were usually caged with the choice of many of 
these cuts and the one infested most heavily was considered as the 
optimum condition and used afterwards for continuation of the succes- 
sive broods. The optimum cut could only be determined when sufficient 
material was given for the number of insects present, as an unfavorable 
cut may be attacked when the adults are confined on it without sufficient 
optimum material. 

Wood used a month or two after being cut is spoken of as green or 
freshly cut material. 

In many cases wood from several individual trees was used to avoid 
any possibility of offering an undesirable individual. 

To illustrate the variation in optimum conditions of wood, several 
examples are given: Callidium antennatum Newm. requires wood dry- 
seasoned over winter; Neoclytus capraea, wood cut during the late winter 
with the inner bark still sappy ; Liopus alpha Say in hickory, twigs cut 
in the early fall, air-seasoned for a while and then left on the ground 
over winter so that the inner bark sours somewhat. (This condition is 
brought about by the girdling habit of Oncideres cingulata Say.) 

Several terms which may need explanation are used in reference to 
the species of host wood: Primary host, or original host, refers to the 
wood in which the insect is found in nature and first caged in these experi- 
ments ; as secondary host is understood wood in which a colony has been 
successfully produced in the experiment, but it may or may not be 
recorded as a host in nature; an unfavorable host is one not recorded 
from nature and in which attempts to produce a colony have not been 
entirely successful. 

All experiments conducted are here given, although a few have been 
unsuccessful or have given no results. Occasionally failure to continue 
a colony is recorded. In all cases an explanation can not be given. It 
may be because of an improper cut of wood or of a peculiarity of the 
individual host. In one case partial failure was due to a nematode 
parasite causing sterility of the females; in another, the parent insects 
were entangled in spider webs and killed before ovipositing. 

Reference is made to larval transfers from one host to another. This 
is accomplished by making a smooth cell through the bark of the new 
host, partially filling it with frass from the larval mines of the original 
host, then placing the larva in this cell and finally tightly fixing a piece 
of bark over the cell. Such transfers do not injure the larva or affect 
its development. Many cases of transfer to the same host resulted in 
the survival of every larva. 

These experiments were conducted at tlie Eastern Field Station, East 
Falls Church, Va., and all flight dates of the adults and times of cutting 
of the wood refer to this locality unless otherwise stated. 



Oct. 22, I92I Hopkms Host- Selection Principle 195 

OUTLINE OF EXPERIMENTS ON EACH SPECIES 
XYI^OTRECHUS COI,ONUS. EXPERIMENT I 

Xylotrechus colonus Fab. is found in nature in a wide variety of hosts. 
In fact, it feeds in nearly all hardwood deciduous trees of the eastern 
and central United States. It shows little or no preference for any 
exact condition of the wood, except that it will not attack perfectly 
seasoned material. The larvae can be found in dying standing trees or 
in logs felled in any month of the year provided they still contain a 
certain amount of moisture. 

The first flight of the year occurs in the last week of May or first week 
of June, reaching the maximum in about two weeks. A few adults 
emerge sporadically throughout the summer. From eggs deposited in 
June a few adults usually emerge in September, but the main brood 
remains as larvae until the next spring. These fall adults have never 
oviposited under confinement. 

The larvae feed entirely beneath the bark, or in the bark if it is thick. 
The pupal cell is made in the outer sapwood or in the bark. 

The wood of all species for this experiment was cut on April 15 unless 
otherwise stated. The colony was started by felling a red oak tree in 
March, 19 14. The wood was attacked during that June, caged soon 
afterwards, and the colony has since been maintained in red oak. From 
the original oak form colonies were secured in hickory (Hicoria), chestnut 
{Castanea dentata {M.QXsh..) Borkh.), locust {Robinia pseudacacia Linn.), 
red maple {Acer rubrum Linn.), and ash (Fraxinus sp.), in the following 
manner : 

QuERCUS. Experiment I. — During May, 1915, hickory logs were 
placed in the cage with oak intended to carry on the colony. Many 
adults were present, somewhat over 100, and the hickory as well as the 
oak was subsequently found infested. 

In June, 19 16, in the same cage stocked with oak for continuing the 
colony, chestnut and hickory wood was placed. There was again an 
overabundance of adults and all woods were infested. 

In June, 191 7, oak was placed in this cage to continue the colony 
and also ash, chestnut, locust, hickory, and red maple logs, all cut in 
February except the hickory, which was cut in April, 191 7; extra pieces 
of chestnut and maple, cut in November, 1916, and September, 1916, 
respectively, were also placed in the cage. There was an abundance of 
adults. In July these logs were examined and it was found that the oak 
was heavily infested; the chestnut and hickory were lightly infested; 
the ash, maple, and locust had no infestation. This same year, 191 7, 
adults were isolated on ash (I^), maple (I*), and locust (I^) with results 
as described in later paragraphs. 
65583°— 21 2 



196 Journal of Agricultural Reserach voi.xxn. no. 4 

In June, 19 18, two pairs of adults from oak were caged on a collection 
of oak, hickory, ash, maple, and chestnut logs, all cut April 15, and of 
about equal size. Examination in July showed that the oak was heavily 
infested (over 50 larvae present), the chestnut contained 10 larvae, the 
hickory 7 larvae, and the ash and the maple none. 

At the same time a similar cage was prepared, and six pairs of adults 
were placed in it to test the influence of a greater number of beetles 
on the selection of hosts. The results showed the same relative propor- 
tion of infestation except that ash also was attacked. The maple was 
not infested. 

In June, 1919, this experiment was repeated with the same conditions 
except that the hickory sticks were accidentally omitted. They were 
examined in July and the infestation was as follows : The woods in the cage 
of two pairs of adults contained 28 larvae in oak, 22 in chestnut, and none 
in ash and maple; that of six pairs contained over 50 in oak, 19 in chest- 
nut, and none in ash or maple. 

HicoRiA. Experiment I ^ — May, 1915, hickory logs were placed (as 
before described) in the oak cage with the wood intended to carry on 
the colony. They were infested and in subsequent years kept isolated 
and continued as the hickory form. 

In June, 191 6, oak was placed in this cage, together with the hickory 
to continue the colony, and was subsequently found heavily infested. 

In June, 19 17, together with the hickory for reinfestation, chestnut 
and locust were placed in the cage. An examination in July showed 
that the chestnut contained a few larvae and the locust none. 

In June, 1918, two pairs of adults were isolated in a cage containing 
oak, hickory, ash, chestnut, and maple, all cut April 15, and of equal 
size. The results showed that the hickory was heavily infested by over 
50 larvae, the oak contained 7 larvae, the chestnut i larva, and the maple 
and ash none. 

In June, 19 19, selection tests and selection quantity tests were carried 
out with this strain. The quantities and cuts of wood were the same 
as before, except that oak was accidentally omitted. In one cage two 
pairs of adults were isolated, the resulting infestation being hickory 
18 larvae, chestnut 12 larvae, maple and ash none. In another cage six 
pairs of adults were isolated, the resulting infestation being hickory over 
50, chestnut 40, and maple and ash none, 

Castanea. Experiment I ^. — In May, 1916, chestnut logs were placed 
(as before described) in the oak cage with the wood intended to carry on 
the colony. The wood was heavily infested, and these individuals have 
since been confined to chestnut. 

In May, 191 7, together with the chestnut, hickory was placed in this 
cage. The hickory was lightly infested. 

In June, 19 18, two pairs of adults were isolated in a cage containing 
oak, hickory, ash, chestnut, and maple, all cut April 15 and of equal 



Oct. 22, 1921 Hopkins Host-Selection Principle 197 

size. An examination in July showed that the oak and chestnut were 
equally well infested, the hickory contained one larva, and the maple 
and ash none. 

RoBiNiA, Experiment I ^. — In June, 19 1 7, eight adults from oak were 
isolated on black locust cut in February, 191 7. The adults laid eggs, but 
all died later. 

In June, 19 18, the experiment was repeated with 15 adults and wood 
cut April 15. Many young larvae entered the bark, but by August 15 
nearly all had died, and none lived to transform the next spring. The 
experiment was not repeated in 19 19. 

Acer rubrum. Experiment I*. — In June, 191 7, eight adults from 
oak were isolated on a piece of red maple cut February i, 19 17. A few 
larvae lived and three adults (two males and one female) emerged in 19 18. 
They were isolated in a cage containing oak, chestnut, hickory, ash, and 
maple, all cut April 15, but no infestation occurred in any wood. During 
August, 19 18, twenty larvae were transferred from oak to maple to con- 
tinue the species in this host. A few adults emerged in 1919, and were 
recaged on maple to develop a larger colony, which will be continued 
several years before testing the selection again. 

Fraxinus. Experiment I ^. — In June, 191 7, eight adults from oak 
were isolated on a piece of ash cut January i, 19 17. A fair infestation 
occurred, but the larvae developed slowly, and in May and June, 1918, 
only seven adults (both sexes represented) emerged, while many larvae 
were still in the wood. These adults were transferred to a cage contain- 
ing oak, chestnut, hickory, ash, and maple, all cut April 15, but no infesta- 
tion occurred in any wood. Larvae were again transferred to ash, and a 
few adults emerged in 19 19. These were recaged on ash, and several 
adults emerged in June, 1920, but failed to develop any larvae in the new 
wood. 

I, I\ I^, P. — During June, 1920, adults emerged from the oak, hickory, 
chestnut, and maple; strains and adults from all were recaged on the 
same wood and produced new colonies. No selection tests were made, 
and these strains will be continued for several years in the same wood 
before similar experiments are again attempted. 

CONCLUSIONS 

The original oak strain of Xylotrechus colonus shows a decided prefer- 
ence for a few woods, notably oak, chestnut, and hickory. Two years' 
trial failed to produce larvae capable of completing their development in 
locust, while the ash and maple colonies were maintained with difficulty. 
In nature these woods (ash, maple, and locust) have been found contain- 
ing thrifty colonies of this species. 

Originally the oak strain showed little preference as between oak, hick- 
ory, and chestnut; yet, after several years, strains were developed in each 
wood that showed a growing preference for the given wood. 



198 Journal of Agricultural Research voi. xxn. no. 4 

The number of insects present under identical conditions influences 
their selection of hosts. When few are present they concentrate on 
original or favored hosts; when more than can successfully oviposit on 
original hosts are present, less favored hosts are taken. 

CYIvI,ENE PICTUS, HICKORY HOST STRAIN. EXPERIMENT II 

The larvae of Cyllene pictus feed almost exclusively in hickory. A few 
specimens have been taken in wild grapevine (Vitis sp.), mulberry {Moriis 
rubra Linn.), osage orange (Toxylon pomiferum Raf.), and hackberry 
{Celtis occidentalis Linn.), but such instances are rare and of very local 
occurrence. In one locality near Harrisburg, Pa., all except one of these 
unusual food plants have been recorded. This borer is found generally 
distributed east of the Mississippi River. The optimum condition of 
wood is that cut during the winter, preferably in January, and left lying 
on the ground. November cuts are sometimes attacked, but no wood is 
suitable unless the inner bark is still sappy. Sticks cut at the time of 
emergence are too green for attack. 

The first emergence occurs about the middle of April and continues 
for three weeks. By September the larvae are full grown and have con- 
structed their pupal cells in the wood. They soon pupate, and in this 
stage they overwinter. The larvae feed about equally beneath the bark 
and in the wood. 

These experiments were started in April, 19 15, when adults were 
found ovipositing on a hickory log cut during the winter at Falls Church, 
Va. The strain has since been continued in January and February cuts of 
this wood, and other host strains have been attempted with varying suc- 
cess in grape {Vitis sp.), locust (Robinia pseudacacia) , ash {Fraxinus sp.), 
and mulberry {Morus rubra). Experiments were conducted as follows: 

ViTis. Experiment II ^ — In April, 1917, a piece of grape, cut in 
January, was placed in the same cage with much hickory used for the 
continuation of the hickory form. This grape was not infested. 

June 26, 19 1 7, sixteen larvae, about half grown, were transferred from 
hickory to grape cut in January. Nearly all these larvae lived, and the 
following spring 12 adults emerged. They were isolated in a cage con- 
taining several pieces of grape and one of hickory, both cut in February. 
Examination in June showed the grape to be heavily infested while the 
hickory contained no larvae. 

In April, 191 8, a large number of adults emerged from the grape. 
Two pairs were isolated in a cage containing one piece of grape 2 inches 
in diameter and 2 feet long, and one piece of hickory of the same size, 
both cut in January. Examination in July showed the grape to be very 
heavily infested, whereas the hickory contained only a few larvae. 

Robinia. Experiment II ^ — April 21, 191 7, three females and two 
males from hickory were caged on a piece of locust cut a month pre- 
viously. The females laid all their eggs on the locust and the young 



Oct. 22, 1921 Hopkins Host-Selection Principle 199 

larvae bored through the bark, but by the middle of June all had died. 
June 15, twelve larvae 5 mm. long were transferred to locust and these 
all died by July 11, when three more, over half grown, were transferred. 
These lived to construct pupal cells and pupated, but all the pupae died 
during the winter. 

In April, 1918, five adults (three females and two males) from hickory 
were caged on locust cut during January, 19 18. The females laid all 
their eggs, but only a few larvae lived. These constructed pupal cells 
and pupated beneath the bark. In locust the larval mines are not 
normal, lying in almost all cases immediately beneath the bark instead 
of extending deep into the wood. About half the larvae made pupal cells 
in the outer sapwood while the others pupated beneath the bark instead 
of, as normally, deep in the wood. 

April 20, 19 19, a total of six adults had emerged and two pairs were 
caged on pieces of locust and hickory cut in January, 19 19. An exami- 
nation July 16 showed no infestation in either. 

Fraxinus. Experiment II ^. — April 24, 191 7, three females and two 
males were isolated on ash cut during January. The females laid all 
their eggs and the young larvae bored through the bark, but all died 
before June 15. At this time fourteen larvae 5 to 7 mm. long were trans- 
ferred to the same ash, and all died within a month. July 15, five more, 
over half grown, were transferred. They mined extensively beneath the 
bark, but all died before the end of September without pupating. 

July 24, 19 18, fifteen larvae, one-half to three-fourths matured, were 
transferred to ash cut in January. 

April 21, 19 19, a total of eight adults had emerged. One pair was 
caged on the January cut of ash, and two pairs were caged on January 
cuts of ash and hickory. 

July 16 the wood was examined, but in no case was it infested. 

MoRUS. Experiment II*. — April 29, 1918, three females and two 
males from hickory were caged on mulberry cut in January. The females 
laid eggs, and a very heavy infestation was secured. They developed 
normally and suffered little more than the normal rate of mortality 
experienced in hickory. 

April 21, 19 19, a total of 17 adults had emerged; two pairs were trans- 
ferred to a cage containing two pieces of mulberry and one piece of 
hickory cut during January, 19 19. In another cage containing the same 
quantity of wood four pairs of adults were transferred. In neither case 
was the quantity of mulberry sufficient to permit the development of 
all the larvae. Each piece was 2 inches in diameter and 14 inches long. 

July 16 the cages were examined, and that containing two pairs of 
adults was infested as follows: Hickory 6 larvae, mulberry over 30; 
that containing four pahs, hickory 13 larvae, mulberry very heavily in- 
fested, over 40. 



200 Journal of Agricultural Research voi. xxn.No.4 

Seasoned Hicoria. Experiment II*. — ^Attempts were made in 
April, 191 7, to develop a colony adapted to seasoned wood by caging two 
females and one male on wood cut in October and dry seasoned. Eggs 
were laid and larvse entered the bark but developed very slowly, never 
entering the wood. Only three lived to pupate, and these made their 
pupal cells between the bark and wood. All three pupae died during 
the winter. 

In May, 1918, the experiment was repeated with three females and 
three males and wood cut in November, 191 7. In the fall of 19 18 a 
number of larvse lived and pupated, but all were below normal size. 
Only a few adults emerged in April, 191 9, and these were below normal 
size. 

1 1^. — Dr. A. D. Hopkins, in 19 16, recorded a dying hickory tree heavily 
infested by Cyllene picius with no evidence of primary injury from other 
causes. 

This suggested that a strain capable of attacking living trees might 
be produced, and attempts were made to secure a colony in such a tree. 
A small hickory 3 inches in diameter was selected and entirely stripped 
of leaves August 11, 1916. April 30, 191 7, it was again defoliated, and 
80 adults were caged on it. The adults laid eggs and the young larvae 
entered the bark, causing sap to flow from the wounds. However, all 
died after growing to 3 mm. in length. 

In April, 19 18, the tree was again defoliated, and 156 adults were caged 
on it. The same results were observed. 

During both years the tree put out healthy foliage after artificial defo- 
liation, but it died in August, 19 17. In no case did the Cyllene larvae hve 
to mine more than % inch beneath the bark. 

QuERCUs. Experiment II^ — In transferring adults during the 
spring of 1918 to new hickory wood to continue a large colony, a piece of 
oak was unintentionally left in the cage. This cage contained six large 
hickory logs 4 to 6 inches in diameter and 5 feet long. The oak log 
was 3 inches in diameter and 4 feet long. 

During September, 1918, work of Cyllene was noticed on this piece of 
oak, and in the spring of 1919 it was separately caged. Five adults 
emerged in April — all very small, much below normal size. 

Two females and one male were transferred to a cage containing only 
oak; one pair to a cage of oak and hickory. These cages were examined 
July 16. Neither wood of the selection test was infested, but the oak 
wood on which two pairs were caged contained a few very small larvae. 

11^^ — ^To test the influence of host selection on the condition of host. 
In April, 19 18, two males and two females from hickory were isolated in 
a cage containing a piece of grape and a piece of hickory of equal size — 
the grape of optimum cut, January, 19 18, the hickory less favorable, 
November, 191 7. Examination in July showed the grape to be heavily 
infested while the hickory contained very few larvae. 



Oct 22, 1921 Hopkins Host-Selection Principle 201 

11^^. — To test the influence of an overabundance of adults and scarcity 
of the primary host on the host selection. Three pairs of adults from 
hickory were caged on a small piece of grape and a small piece of hickory 
(each 2 inches in diameter and i inch long), each cut during January, 
19 1 8. Examination in July showed both woods to be infested, the grape 
containing a few more larvae than the hickory. 

In 1920 no adults emerged. 

II, 11^. — In 1920 only two strains were continued, those in hickory and 
those in mulberry. No attempt was made to reestablish the others that 
failed. 

CONCLUSIONS 

This species, although most commonly found in hickory, will readily 
adapt itself to several other plants, notably mulberry and grape, both of 
which are recorded as natural hosts. 

In some unfavorable hosts, or in an optimum host hi an unfavorable 
condition, the larvae may become established, but the mortality is high 
and the progeny seem to be sterile. 

After one year's feeding in a new host the larvae may select that host in 
preference to others. 

The selection of a host is influenced by the number of adults present 
and the quantity of the primary host, in that adults will prefer a sec- 
ondary host to overinfesting the original host. 

The selection of a host is influenced by the condition of the host, a 
favorable condition of secondary host being preferred to an unfavor- 
able condition of the original host. 

The optimum condition of any host capable of properly supporting 
growth of the larvae is of very restricted limits. 

CYI,I,ENE PICTUS, GRAPK HOST STRAIN. EXPERIMENT III 

This is the same species as previously discussed, having the same 
biological habits except that this host strain in grape (Vitis) was taken 
in nature at Hummelstown, Pa., in January, 1916, by J. N. Knull. 
Since then the colony has been continued at East Falls Church, Va., in 
grape cut in January or February. Other host strains have been pro- 
duced and experiments conducted as follows : 

HicoRiA. Experiment III ^ — When these adults emerged from 
grape (May, 19 16), three females and two males were isolated in a cage 
containing 10 pieces of grape and i piece of hickory cut in February, 19 16. 
No eggs were laid on the hickory. April 17, 191 7, two pairs of adults 
were isolated on hickory cut in January, 19 17. Eggs were laid and the 
larvae developed but not so rapidly as in the grape. June 15 they were 
under normal size. By September only two larvae were alive. One of 
these pupated but died during the winter. 



202 Journal of Agricultural Research voi. xxn, N0.4 

Many adults emerged from the grape in 19 18 and five pairs were caged 
on three small pieces of grape i}4 inches in diameter and 2 feet long and 
one piece of hickory about the same size, both cut in Januar}^ 1917- In 
July they were examined, and both hickory and grape were heavily in- 
fested. 

Quantity selection. Experiment III^. — To again test out the 
effects of host selection when an insufficient amount of wood is given than 
that required for the number of adults present, in April, 19 18, two pairs 
from grape were caged on grape and hickory cut in January, 191 8. The 
piece of grape was 2 inches in diameter and i foot long, the hickory 2 
inches in diameter and 2 feet long. An examination in July showed both 
grape and hickory infested. 

This experiment was repeated in April, 1919, using two pieces of grape 
and one piece of hickory, all of optimum cut and equal size. In one case 

1 pair of adults was isolated, in another case 3 pairs were used. The wood 
on which i pair was caged contained 5 larvae in hickory and 3 larvae in 
grape ; that on which three pairs were caged contained 4 larvae in hickory 
and over 25 in grape. 

The grape colony was not continued in 1920. 

CONCLUSIONS 

This host variety from nature had acquired a decided preference for 
grape. 

The selection of a host is influenced by the quantity of wood present for 
a given number of adults, in that the adults will select a new host in 
preference to overinfesting the original host. 

The tendency in this species in nature to confine itself to a certain host, 
either hickory or grape, is not as marked as in some other species. 

CYIvLENE PICTUS, HICKORY STRAIN II X GRAPE STRAIN III 

In order to determine whether crossing of these two host strains would 
influence the progeny in the selection of the host, males and females were 
isolated from their pupal cells in the spring of 191 7. April 17 three 
females from hickory and two males from grape were isolated in a cage 
containing hickory. Two females from grape and two males from hickory 
were isolated on grape. In neither case did mating occur as readily as 
when both sexes from the same host were paired. The sexes often ap- 
proached each other and moved away before finally copulating. 

Good infestations were secured in both cases. In April, 1918, one pair 
from hickory was caged on equal amounts of grape and hickory cut in 
January, 19 18. Only the hickory was infested. Two females and two 
males from grape were isolated on the same amount of grape and hickory 
cut in January, 19 1 8. Both woods were infested. These pieces were all 

2 inches in diameter and 1 8 inches long. 



Oct. 22, 1921 Hopkins Host-Selection Principle 203 

CONCIvUSIONS 

This crossing of the two host forms had no influence on the selection of 
hosts. 

The amount of wood and number of adults present influence the host 
selection, as shown when one female and two females were given the same 
amount of wood. 

CYI,I,SNE CRINICORNIS. BXPERIMENT XI,I 

Cyllene crinicornis Che v., found in the southwestern United States, is 
known to feed only in mesquite (Prosopis juliflora (Swartz) de C.) and 
occasionally on an allied legume, paloverde {Parkinsonia microphylla 
Torr.). In general its biology is similar to that of C. pictus, and it pre- 
fers the same conditions of wood. Adults begin emerging in the natu- 
ral range during late February, and part of this generation emerges in 
September. 

Mesquite infested with these larvae was sent to Falls Church, Va., by 
T. E. Snyder from San Antonio, Tex., April 27, 191 7. Adults emerged 
at Falls Church the following May and were caged on mesquite cut in 
March, 1918. A good infestation was secured and has since been con- 
tinued on mesquite. 

RoBiNiA. Experiment XLr. — In May, 19 18, two pairs were isolated 
on locust cut February, 19 18. May 31 the females were dead, and the 
abdomens were dissected and found to contain eggs. Probably no eggs 
were laid, and in July no evidence of larval work could be found. 

July 24, 19 1 8, seven nearly matured larvae were transferred from the 
mesquite to locust cut January, 1918. During May, 1919, five adults 
emerged. One pair was caged on locust cut in January, 1919, and one 
pair on both locust and mesquite. The mesquite was cut October, 19 18. 

July 16, 1919, these cages were examined, and the selection test showed 
that mesquite was not infested, whereas the locust contained several 
small larvae. The cage containing only locust was lightly infested. 
None of these larvae transformed in 1920. 

CONCLUSIONS 

The locust was such an unfavorable host that the adults would not 
oviposit on it, but larvae may live and transform for one or two genera- 
tions when forced to take it. 

CALUDIUM ANTENNATUM. EXPERIMENT IV 

Some confusion exists as to the taxonomy of the blue species of Calli- 
dium allied to Callidium antennatum. A number of species have been 
described of questionable validity. Two species have been experi- 
mented with — C. antennatum and C janthinum Lee. These two adults 
are easily separable, and their habits are also quite distinct. The 



204 Journal of Agricultural Research voi. xxn, no. 4 

former, so far as the Forest Insect records are concerned, feeds only in 
pine (Pinus) and spruce (Picea), and for this discussion will be confined 
to the form occurring in the northeastern United States. It shows a 
decided preference for a certain condition of the host, requiring wood 
that has been cut in the early fall or winter and is well seasoned. When 
the inner bark is still sappy the insects will not normally make their 
attack. 

Adults first appear about the middle of April, and the flight period 
continues about a month. One year is required to complete the life 
cycle. The larvae feed beneath the bark until half grown, then enter 
the wood to construct a long pupal excavation, at the end of which the 
pupal cell is chambered off. 

These experiments were started in December, 19 16, when infested 
Virginia pine was caged. In 19 16 and the following years the colony 
was continued in seasoned pine, and a form was also developed in spruce 
and freshly cut or green pine. Unsuccessful attempts were made to 
produce a juniper (Juniperus) strain. This insect has been reported as 
feeding in juniper (Juniperus) and maple (Acer). 

Juniperus. Experiment IV^ — In April, 19 16, juniper was placed 
in the cage together with pine; in addition, about 20 adults were isolated 
on a stick of juniper (both woods were cut in January, 19 16). In neither 
case was the juniper attacked, and the females isolated on juniper failed 
to oviposit. 

The same test was repeated in 191 7 with juniper cut in October, and 
the same results were obtained. During June and July, 20 larvae, from 
small to over half grown, were transferred to juniper. All finally died, 
some living a month. None increased in size before death. 

Picea. Experiment IV^. — April 12, 1916, a piece of seasoned 
spruce was placed in the cage, together with the pine, for reinfestation. 
An examination in June showed only one larva in the stick, and this was 
far below the normal size of those in pine. By July 11 this larva had 
died. It is possible that more eggs were laid on the spruce but the larvae 
died earlier. 

June 29, 1916, nine larvae about half grown were transferred to the 
same spruce wood. July 11, 191 6, three larvae were alive and 17 more 
were transferred. From these larval transfers 10 adults were secured 
in April, 191 7. They were caged on four small pieces of seasoned spruce 
and a piece of seasoned pine placed in the cage for one week, both cut in 
October, 19 16. Examination in July showed the spruce well infested, 
but only 4 larvae were found in the piece of pine. 

In 19 1 8 the same experiment was repeated, four females and three 
males from spruce being caged on four sticks of spruce and one of pine 
cut in September, 191 7. An examination in July showed that the spruce 
contained many larvae, but none were found in the pine. 



Oct. 22, 1931 Hopkins Host-Selection Principle 205 

In 19 1 9 this was again repeated, but the results showed that pine was 
infested while the spruce contained no larvae. Both woods were of No- 
vember cut, but it is not likely that this would have such a decided in- 
fluence. However, six adults were isolated on spruce to continue the 
colony, and this wood was very lightly infested. No explanation can be 
offered as to the reason for this discrepancy from former results unless the 
spruce wood was in an unfavorable condition. 

During the same season another cage was prepared of pine cut during 
November, 1918, and green spruce cut in April, 1919. The pine was 
heavily infested ; the spruce contained no larvse. 

Experiment IV^. To test effects of condition of wood on selection. — 
April 23, 1920, one pair of adults from spruce was caged on a favorable 
cut of pine (November) and freshly cut spruce (April) . When the woods 
were examined on July 5, 1920, the pine contained many larvae while the 
spruce contained none. At the same time a pair was caged on November 
pine and November spruce, both optimum cuts. Several days later 
the female was found dead in the cage and had laid no eggs. 

Acer. Experiment IV^. — ^As maple has been reported as a host of this 
species, attempts were made in 19 16 to start a colony in this wood. On 
July 6, six half- grown larvae were transferred to a seasoned piece of 
wood, but by July 21 all but one had died and it was smaller than when 
transferred. This one died soon after. None of the larvae fed on the 
maple. 

Green pine. Experiment IV^. — In 1916 some of the wood used to 
carry this colony along was cut March i, and consequently little seasoned. 
It was very unfavorable for the ovipositing of the adults, but some eggs 
were laid on the pieces. The larvse developed slowly and at the time 
of pupation were below normal size. The adults secured in 1 9 1 7 averaged 
about one-half normal size. They were caged again on wood cut in 
March, 191 7. Adults were secured in 1918 and again caged on the same 
condition of wood, and a good infestation was secured. While the author 
was absent for a month from the field station in the summer of 19 18 
these larvae were all killed by a fungus. The sticks were on the ground 
and so were caught in a period of rainy weather and were water soaked. 

Experiment IV, IV ^. — In 1920 the pine and spruce strains were 
continued in the same wood. 

CONCI^USIONS 

The pine form shows a decided preference for that host. 

It can live in spruce and then shows a decided preference for that 
host. 

It will not live in juniper or maple. 

In producing a new host strain a high mortality occurs in the young 
larvse. 



2o6 Journal of Agricultural Research voi. xxn. No. 4 

A colony can be produced in a host which is in an unfavorable condition, 
but the resulting adults are below normal size. Owing to the failure to 
continue the colony it can not be stated whether or not such a strain 
would show preference for the new condition of the host by selecting it 
voluntarily. 

CALLIDIUM JANTHINUM. EXPERIMENT V 

Under Callidium antennatum reference was made to C. janthinum 
Lee. It is distinguishable from the former by its smaller size, shining 
surface, and bluish green color of the adult, by the fact that the larva 
feeds only in juniper, and that the adults emerge about four weeks later 
in the spring. It requires wood which was cut during the late fall and 
which has not seasoned in contact with the ground. It will oviposit in 
greener wood than C. antennatum although the inner bark should not 
be sappy. 

The first flight occurs during the first to third week in May and con- 
tinues about two weeks. One year is required to complete the develop- 
ment. The larvae feed beneath the bark until half grown, then excavate 
long pupal chambers, the ends of which are plugged off for the trans- 
formation cell. 

These experiments were started with a lot of infested juniper branches 
from Hummelstown, Pa., collected by J. N. Knull in April, 1916. May 2, 
six adults emerged and were caged on juniper cut in April and rapidly 
dried in the house. Since then the colony has been continued each 
year in September and November cuts of juniper, which are preferred. 

PiNUS. Experiment V^ — May, 191 7, two pairs were isolated on pine 
cut in November, 19 16. A few eggs were laid from which larvse hatched 
and entered the bark. By July 10 all had died. The same test was 
repeated in 19 18 with similar results. 

CONCLUSIONS 

This species shows a decided preference for jumper and will not develop 
in pine from early stages. Larval transfers to pine with nearly matured 
larvae were not made. 

CAIvWDIUM ANTENNATUM AND C. JANTHINUM. VXIV; IV XV 

Because of the taxonomic confusion between these species (cited pre- 
viously) and with the idea that a crossing of these two forms might possi- 
bly influence the selection of a host, attempts were made to cross the 
species. 

VxIV. — In May, 19 17, four females of the juniper form were crossed 
with two males of the pine form and were caged on juniper. Both 
species had been previously isolated from the pupal cells to avoid all 
possibility of mating. These insects immediately mated, and the eggs 
were laid on the juniper from which a good infestation was secured. 



Oct 23, 1931 Hopkins Host-Selection Principle 207 

April 17, 19 1 8, the first adult emerged from this cross. The second 
adult emerged April 25. May 3 and 4 eight adults emerged. All the 
19 1 8 adults resembled the juniper form in color. These adults were all 
isolated in a cage containing juniper and pine cut in November, 19 17. 
An examination in July showed only the juniper to be infested, but by a 
very light brood. During the remainder of the summer all died except 
three larvae which constructed pupal cells. May 5, 19 19, two males and 
one female emerged and one pair was caged on juniper cut in November. 
A light infestation occurred. May 12, 1920, five adults emerged and 
were recaged on juniper. The selection of pine and juniper was not 
again tested. t;"- ,.,) hf>q<f:?" 

IV X V. — ^These same species were mated in 19 17 by making the recip- 
rocal cross (males from juniper and females from pine) and caged on 
pine. The sexes did not mate readily, not noticing one another for some 
hours after being caged together. However, several matings finally took 
place and eggs were laid on the pine. Young larvae developed but all died 
later. The same experiment was repeated in 19 18 with the same results. 
Conditions were similar to those in the previous experiment. 

CONCLUSIONS 

The crossing did not influence the selection of a host in the first genera- 
tion of resulting adults. 

These two forms, even though they may be crossed successfully, should 
be regarded as distinct species based on adult characters and biological 
differences. 

The successful cross-mating produced progeny in the first generation 
that emerged over the period of emergence of both parents — a few early 
when the pine form emerges, the remainder some two weeks later when 
the juniper form appears. In later years they emerge as the juniper 
form. 

The juniper color pattern of the adults is dominant. 

HYLOTRUPES UGNEJUS, JUNIPEIR FORM. EXPERIMENT VIII 

The adult forms generally included under Hylotrupes ligneus Fab. show 
a great variation of color patterns. Many of these varieties have been 
described as distinct species by Col. T. L. Casey. In the experiments 
conducted all color varieties, however, have been kept distinct only 
by the host in which they were found in nature and not by the color 
variations. The experiments were primarily conducted to test these 
variations in color patterns, but certain results bearing on the host- 
selection principle were obtained and are here described. 

Hylotrupes ligneus, juniper form, has a wide selection of hosts. Speci- 
mens in the Forest Insect Collection of the Bureau of Entomology have 
been recorded from all genera of coniferous trees indigenous to North 
America. It uniformly prefers wood that has not seasoned a great deal. 



2o8 Journal of Agricultural Research voi. xxn, no. 4 

Later winter or fall cuts in which the inner bark has remained sappy are 
most suitable. Species of wood which season more slowly, due to thick 
bark, must be cut earlier. 

The time of first emergence varies greatly with locality, but the species 
is everywhere one of the first cerambycid beetles to fly in the spring. 
The larvse in all cases feed immediately beneath the bark, only entering 
the sapwood in late summer to make a shallow pupal cell. Pupation 
and transformation to the adult usually take place in the fall. 

The present experiment was started at Kanawha Station, W. Va, Dr. 
Hopkins felled a juniper (Juniperus) in October, 1914. This tree was 
infested the following spring and shipped to East Falls Church, Va. 
The colony has since been continued in juniper and one host strain has 
been produced in Douglas fir (Pseudotsuga) . This particular color 
variety had never been recorded from Douglas fir. 

Pseudotsuga. Experiment VIII^— April 11, 191 7, three females and 
two males were isolated on a piece of Douglas fir cut April i . Eggs were 
laid and young larvse entered the bark, but many died during the summer 
and only two constructed pupal cells. One adult was secured next 
spring. The fact that this wood was too green and that it seasons very 
slowly may have caused a higher mortality than would otherwise have 
occurred. 

May 29, 191 7, twelve larvae and June 15, nine larvse were transferred 
to this host, the wood then being better seasoned. March i, 1918, five 
adults — four females and one male — were removed from pupal cells. 
The remainder of the larvae had died. Two females and one male were 
used to continue the colony by caging on Douglas fir, cut in October, 
19 1 7. A good infestation was secured. A piece of juniper cut during 
January, 1918, had also been placed in tliis cage but was not infested. 
In January, 1919, four adults were removed from the logs — three males 
and one female. The remainder had all died and these were very weak 
and below normal size. One pair was recaged on juniper cut in January 
and Pseudotsuga cut in November. July 30, 19 19, the sticks were 
examined, but no infestation was found in either wood. 

CONCLUSIONS 

This juniper form of Hylotrupes ligneus, after feeding part of a year in 
a new host, showed a preference for the new host. 

A high percentage of mortality occurred in producing the new host 
strain, which finally died out. 

HYEOTRUPES EIGNEUS, PSEUDOTSUGA FORM. EXPERIMENT XXXV 

This form of Hylotrupes ligneus is much darker and more hairy than the 
preceding. It has been recorded only from Douglas fir. Its biology is 
essentially similar to that of the juniper form except that the adults 
emerge somewhat later. It is known from the Rocky Mountain region. 



Oct. 22. I92I Hopkins Host-Selection Principle 209 

The colony was started from a small tree collected at Colorado Springs, 
Colo., and shipped to Falls Chm'ch. This tree had been killed by Scolytus 
in the fall of 19 16 and infested by Hylotrupes in the spring of 191 7. 

In 19 1 8 adults did not emerge until May and were caged on Douglas 
fir to continue the colony. They were recaged on Douglas fir in 1919, 
but all the larvae died from a fungus attacking the bark. 

HYI,OTRUP:eS WGNBUS, PSEUDOTSUGA STRAIN VIII ^ AND PSEUDOTSUGA 

STRAIN XXXV 

Two females from VIII ^ (the juniper form in Douglas fir) were held 
in a cool cellar until adults of this XXXV variety emerged. May 29, 
1918, they were caged on Pseudotsuga with two males from the true 
Douglas fir form (XXXV) . The two sexes absolutely avoided each other 
and were never observed to mate. The females died without laying 
eggs. Many attempts were also made to mate the original juniper form 
with the Douglas fir form but without success. 

HYLOTRUPES UGNEUS, SEQUOIA FORM. EXPERIMENT XL 

The form of Hylotrupes ligneus occurring in sequoia is slightly larger 
but otherwise resembles that in juniper very closely, although the speci- 
mens reared in the experiments show a much greater variety of color 
pattern than do those from juniper. 

April 2, 1 91 8, a large series of these adults were removed from their 
pupal cells in Sequoia sempervirens (Lamb.) Endl. and isolated in small 
vials by F. B. Herbert at Laurel, Calif. April 13, 191 8, they were re- 
ceived at Falls Church, Va. 

JUNIPERUS. Experiment XL ^ — Three prominent color forms were 
paired and each was caged on a piece of juniper cut in January, 191 8, 
since no sequoia was on hand. They all oviposited, but about half of 
the larvae died by July. The remainder made pupal cells and emerged. 
The strain has since been continued in juniper. 

Pseudotsuga. Experiment XL^. — April 20 one pair was caged on 
a piece of Douglas fir cut in October, 191 7. Eggs were laid and a better 
infestation secured than with the juniper form (VIII). All larvae died 
and no adults were secured in the spring of 191 9. A fungus growth 
under the bark was responsible in a large measure. 

hylotrupes ligneus, JUNIPERUS STRAIN VIII X SEQUOIA STRAIN XL 

April 13 several males from redwood (XL) were separately caged 
with females from juniper (VIII) held over in a cool cellar since they 
were isolated from the cells. One of these males mated with two females 
(first and third tried) immediately on being isolated with them. This 
same male would not mate with the second female tried, nor would any 
males of XL mate with females of VIII. Many juniper (VIII) males 



2IO Journal of Agricultural Research Voi. xxn. no. 4 

were isolated with redwood (XL) females, but in no case did copulation 
take place. 

The females of the j'uniper form mated with males of the sequoia form 
were caged on juniper cut in January, 19 18, and good infestations were 
secured. 

CONCLUSIONS ON THE ENTIRE HYLOTRUPES EIGNEUS GROUP 

The experiments on the Hylotrupes ligneus group, as mentioned above, 
were conducted primarily for the study of its color variation, and not a 
great deal of attention was devoted to the host-selection principle. The 
experiments cited show that among all the color varieties of this group 
there are probably two good species, the darker and more hairy Douglas 
fir form representing one species and all the other forms another. These 
two species absolutely refused to mate, but the varieties from sequoia 
and juniper were successfully crossed. 

NEOCLYTUS CAPRAEA. EXPERIMENT VI 

Neoclytus capraea is known to inhabit the eastern and central western 
United States, extending its range south and west into Arizona. It has 
been recorded from only two hosts, ash (Fraxinus) and white oak 
(Quercus alba of the Rocky Mountains). In the eastern United States 
it has never been found in oak. The condition of the wood necessary 
for oviposition by these beetles must be exactly right. It must have 
been freshly cut and the inner bark must be still moist and sappy. Should 
this inner bark be slightly dried the females will not oviposit on it unless 
forced to do so. Logs cut about two months before the flight period 
are preferred to older cuts or those cut during flight. Trees cut as 
early as November 15 are sometimes infested, but not commonly. 

The adults fly very early in the spring in this locality (Falls Church, 
Va.), about the last week in March and the first two weeks of April. The 
larvse feed chiefly in the wood proper. Mining beneath the bark for a 
short time, they then enter the sapwood and later the outer heartwood, 
extensively honeycombing it. Pupation and transformation to the adult 
take place in the early fall. 

VI.— March 26, 1915, twelve adults were taken as they emerged from 
an ash log and were caged on freshl}' cut wood. A good infestation was 
secured, and the colony has since been continued in ash. 

In the spring of 19 19 no adults emerged. All the larvae remained 
over as larvae in their pupal cells until the fall of 19 19, when they trans- 
formed to adults and emerged in 1920. No explanation for this can be 
offered unless the logs were too moist in the early part of the summer 
so that the larvae did not develop properly. Excess humidity or exces- 
sive desiccation have both been found to produce retardation in devel- 
opment of larvae in small isolated cages. This insect is one of the most 
regular of those reared, in the time of emergence and development of 



Oct. 22, I92I Hopkins Host-Selection Principle 211 

the broods. Attempts have been made to start colonies in hickory and 
white oak. 

HicORiA. Experiment VI ^— March 31, 19 17, seven adults (four 
females and three males) from ash were isolated on hickory cut Feb- 
ruary I, 191 7. No infestation occurred. May 31 of the same year 
fifteen larvae, 2 to 4 mm. long, were transferred to hickory. Again 
on June 15, seven larvae, 4 mm. long, were transferred to the same 
piece of wood. July 11 one larva was living and five more, over half 
grown, were transferred. 

April 6, 191 8, three adults (two males and one female) emerged from 
the hickory. They were caged on hickory and ash cut January, 19 18. 
These adults were very weak and inactive, not at all characteristic of 
normal adults. 

An examination in July showed neither wood to be infested. 

July 24, 19 1 8, twenty larvae, one-half to three-fourths grown, were 
transferred from ash to hickory cut April 15, 19 18. 

April II, 19 19, one female emerged, one adult had died in its pupal 
cell, and the remainder of the larvae had died before pupating. This 
female was mated with a male from ash and caged on hickory and ash 
of optimum cuts. An examination in July showed no infestation in the 
hickory, but the ash contained a few larvae. These died later in the 
summer. 

QuERCUs AiyBA. EXPERIMENT VI ^. — ^April I, 1917, four pairs of 
adults from ash were isolated on white oak cut in March, 191 7. Eggs 
were laid on the wood, and the small larvae bored through the bark, but 
all died before May 31. On this date fifteen larvae, 2 to 4 mm. long, 
were transferred to white oak. July 1 1 one larva was living. Septem- 
ber 17 all were dead. 

In April, 19 18, three pairs were caged on wood cut in January, 19 18. 
July 18 many larvae were still alive but under size. Several lived to 
pupate, but all died before the following spring. 

Seasoning. Experiment VI. ^ — ^April 4, 19 17, four pairs of adults 
were isolated on ash cut September i, 19 16, and white oak cut in March. 
The females laid eggs on the white oak, but the larvae did not live. On 
Ma7 3 1 neither wood contained larvae. 

CONCLUSIONS 

The foregoing experiments show that this species feeding in ash 
(Eraxinus) has become decidedly accustomed to that host. Several 
attempts, both by oviposition and larval transfers, to produce strains in 
Quercus alba Linn, and Hicoria have resulted in failure. In Hicoria 
the few adults secured were incapable of continuing the colony, and in 
both woods a high or total larval mortality occurred. 

Even with this decided preference for a host, the adults laid eggs on 
a new host rather than on an unfavorable cut of the normal host. 
65583°— 21 3 



212 Journal of Agricultural Research voi. xxn, no. 4 

MOLORCHUS BIMACUI,ATUS. EXPERIMENTS IX, X, AND XXXVI 

Two forms included under Molorchtis bimaculahis Say have been 
caged in these experiments, a large form from hackberry {Celtis occi- 
dentalis Linn.), and a smaller form from dogwood (Cornus florida Linn.) 
and maple (Acer) . They both are found throughout the eastern half of 
the United States. From the observations on the biology of these two 
forms they are regarded by the writer as distinct species. Both forms 
prefer early fall cuts of wood, but the Celtis form requires much drier 
seasoned material. 

MOLORCHUS BIMACULATUS, CORNUS FORM. EXPERIMENT IX 

The Comus form feeds in a great variety of eastern hardwoods. It 
has been reared from Hicoria, Acer, Juglans, Ouercus, Liriodendron, 
Comus, Cercis, and Castanea. The larva feed beneath the bark, making 
a long, curved pupal cell in the wood. By September they have trans- 
formed to adults, which emerge early in May at Falls Church, Va. The 
flight is very regular, nearly all emerging at the same time. The adults 
are much smaller than those of the hackberry form. 

In May, 19 16, adults were reared from dogwood collected at Falls 
Church, Va. They were recaged on dogwood cut in April, but a poor 
infestation was secured from which only five adults emerged in 191 7. 
These five adults were caged on September and November cuts of dog- 
wood and redbud (Cercis canadensis Linn.) . A good infestation occurred 
in the dogwood, but no larvae were found in redbud. 

May I, 191 7, five adults were isolated in a cage containing November 
cuts of dogwood and maple. The maple was not infested, but many 
larvae were found in the dogwood. 

In April, 1918, 1919, and 1920, the colony was continued only in dog- 
wood. No selection tests were made. 

MOLORCHUS BIMACULATUS, ACER FORM. EXPERIMENT XXXVI 

This form in all respects is similar to the dogwood variety IX. 

Infested limbs collected at Falls Church, Va., were caged in the sum- 
mer of 1916. 

May I, 191 7, five adults were caged on branches of maple and dog- 
wood cut in September and November. The maple was infested but no 
larvae entered the dogwood. 

In 19 1 8 many adults emerged from the maple and were recaged on 
October cuts of maple and dogwood. Eggs were laid on the maple, but 
the cage unfortunately was overlooked and became so dry that none of 
the eggs hatched. 

MOLORCHUS BIMACULATUS, CELTiS FORM. EXPERIMENT X 

The form in hackberry, in which the adults are much larger, has been 
reared only from this host. The larvae feed as in the dogwood or maple 



oct.22, I92I Hopkins Host-Selection Principle 213 

forms, but only about half the brood emerges at the end of the first year, 
the remainder going over in the larval stage to the following season. The 
adults emerge about a month earlier, April i to 10. 

Infested hackberry branches from Hummelstown, Pa., were collected 
and sent to Falls Church, Va., in December, 1915, by J. N. KnuU. 

In April, 19 16, 20 adults were caged on January cuts of hackberry, 
dogwood, and redbud, but only the hackberry was attacked. 

In April, 191 7, eight adults were isolated on sticks of redbud, dogwood, 
and maple, all cut in September and November. No eggs were laid in 
any of these woods. The form has since been continued in hackberry. 

CONCLUSIONS 

A very decided predilection for the original host is exhibited by the 
host strains of this species. It is not surprising in the case of the hack- 
berry form, as this is the only host from which it has been found. How- 
ever, this form would not even lay eggs on any hosts other than the 
original. In the dogwood strain adults were not isolated on maple alone, 
nor were adults of the maple strain isolated on dogwood alone. If this 
had been done, it is veiy likely that infestations would have resulted. 

NEOCIvYTUS ERYTHROCEPHAI^US. EXPERIMENTS XI, XII, AND XIII 

The adult and larva of Neodytus erythrocephalus Fab. are quite different 
from those of Neodytus capraea, but the range and habits are much the 
same. The species attacks wood in a greater variety of conditions, but 
the most favorable condition is an early spring cut. It has been collected 
in almost all eastern hardwoods. 

The first flight occurs at Falls Church, Va., in late May or early June; 
consequently, that the wood may be sappy for infestation it must be 
cut during April. The species overwinters in the larva stage, pupation 
not taking place until early April. Farther south two or more genera- 
tions occur each season. 

Three host strains were collected in nature and experimented with. 

NEOCLYTUS ERYTmiOCEPHALUS, HICORIA FORM. EXPERIMENT XI 

June 9, 19 16, adults emerging from hickory at Falls Church, Va., were 
recaged on wood cut in late March. A good infestation was secured. 

June 8, 191 7, the colony was continued in April cuts of hickory. Two 
pairs were isolated in a cage containing hickory and redbud cut in April 
and dogwood and tulip {Liriodendron tulipifera Linn.) cut in May. In 
July an examination showed hickory to be the only wood infested. 

May 23, 19 18, two pairs were isolated on hickory, dogwood, and redbud 
cut on April 15. When examined on July 18 hickory was found to be 
lightly infested, dogwood heavily, and the redbud contained no larvae. 

May 24, 19 18, six pairs of adults were caged on two pieces of hickory 
and one of dogwood, cut April 15, of the same size as those of the 



214 Journal of Agricultural Research voi. xxn.No. 4 

experiment of May 23, 19 18. When examined on July 18, both woods 
were heavily infested. Redbud was unintentionally omitted. 

This experiment was repeated in 19 19, two pieces of hickory, one of 
dogwood, and one of redbud being used. Two cages were prepared; in 
one, a single pair was isolated, the resulting infestation being, hickory 
heavily infested, dogwood and redbud uninf ested ; in the other cage three 
pairs were isolated, the resulting infestation being, hickory and dogwood 
both heavily infested, redbud uninf ested. 

NEOCLYTUS ERYTHROCEPHALUS, CORNUS FORM. EXPERIMENT XII 

June 13 to 15, 1916, adults emerging from dogwood at Falls Church* 
Va., were recaged on this wood cut in April, 1916. A good infestation 
was secured. June, 1917, the colony was continued in dogwood and two 
pairs of adults were isolated in a cage containing dogwood and tulip cut 
May 30 and hickory and redbud cut April 18. 

In July it was found that both redbud and dogwood contained few 
larvae while hickory and tulip contained none. 

For some unknown reason the larvae continued in dogwood did not 
develop very well, and in 1918 only one female emerged. May 25, 1918, 
this female was mated with a male from hickory and isolated in a cage 
containing dogwood, hickory, and redbud cut April 15. 

July 18, 1 91 8, the dogwood was heavily infested, the redbud lightly, 
and the hickory contained one larva. 

In June, 1919, one pair was caged on pieces of dogwood, redbud, and 
hickory. An examination in July showed dogwood to be very heavily 
infested, the redbud and hickory containing seven and six larvae, respect- 
ively. 

NEOCLYTUS ERYTHROCEPHALUS, CERCIS FORM. EXPERIMENT XIII 

Redbud infested with this species was collected at Hummelstown, 
Pa., by J. N. Knull and sent to Falls Church, Va., in April, 1916. Adults 
emerged in June and the colony was continued in redbud. June, 1917, 
the colony was again continued in redbud, and two pairs of adults were 
isolated in redbud and hickory cut in April and tulip and dogwood cut 
in May. 

An examination in July showed the redbud to be heavily infested; 
the dogwood and hickory contained several larv^ae, and the tulip none. 

In May, 191 8, two pairs were again caged on redbud, dogwood, and 
hickory, all cut April 15. In July it was found that the redbud and the 
dogwood were heavily infested while the hickory contained but three 
larvae. 

The same experiment was repeated in 1919, and the results showed the 
redbud to contain eight larvae, the dogwood five, and the hickory two. 

The selection tests of 191 7 were all carried out with the same quantity 
of wood ; in each case the pieces were i }{ inches in diameter and i foot 
long. Bach cage contained two pieces of the wood from which the 



Oct. 22. I92I Hopkins Host-Selection Principle 215 

adults emerged and only one each of the others. This amount of the 
original host for two females was considered sufficient for oviposition 
without bringing in the quantity factor. 

These adults are extremely active and run rapidly over logs when 
ovipositing in nature. They have very long hind legs. It was noticed 
that in the glass cylinder used for cages in 19 18 these long legs were a 
disadvantage. The adults could not get a foothold on the glass and 
had difficulty in climbing up on the wood from the glass surface. They 
crawled awkwardly about and when coming in contact with any stick 
maneuvered until they managed to get on it. Such conditions may 
have influenced the wood selected, as the adults could only with difficulty 
go from one stick to another. In 19 19 wire boxes were used, the wood 
lying flat on the bottom. In 1920 only ash and dogwood strains were 

continued. 

CONC1.US10NS 

These experiments up to 19 19 did not seem to show results in any defi- 
nite direction. Selections of the various host strains occasionally gave 
results in conformity with those generally obtained, while again just 
opposite results were recorded. 

The experiments of 19 19 showed results in closer conformity to those 
of other species. This may have been due to the different method of 
caging, which gave the adults more opportunity to move about and select 
the host. 

UOPUS AIvPHA. EXPKRIMBNTS XXV AND XXX 

Two color forms of Liopus alpha have been experimented with, a brown 
form from sumac (Rhus) and a gray form from hickory (Hicoria) . These 
color forms are very distinct and easy to recognize as adults. They are 
not known from any other hosts. The sumac form has been collected 
throughout the eastern United States and as far west as the Rocky Moun- 
tains. The hickory form follows the range of the hickory trees. 

The adults fly in late May and continue flying through June at Falls 
Chiu-ch, Va. One year is required to complete the life cycle. The larvae 
feed beneath the bark and pupate in the wood. They are found only in 
small branches. 

LIOPUS ALPHA, RHUS FORM. EXPERIMENT XXV 

The sumac form prefers branches cut in the early fall and dried standing 
in the air, although it will attack later cuts, provided they have dried 
considerably. 

April 26, 1916, Mr. Champlain sent from Long Island, N. Y., a lot of 
infested sumac twigs which were caged at Falls Church, Va. In June the 
first adults emerged, and 20 were caged on sumac cut in November, 1915. 
Into the same cage were placed chestnut, hickory, and wild cherry twigs 
cut during the winter, but none of these latter woods were infested. Since 
then it has been continued in sumac. 



2i6 Journal of Agricultural Research voi. xxn.No4 

CastanEa. Experiment XXV^ — ^September 20, 1916, fifteen larvae, 
one-half to nearly full grown, were transferred from sumac to chestnut 
cut durmg March. July 10, 191 7, one adult emerged, the only one 
from these transfers. 

HicoRiA. Experiment XX V^ — July 25, 19 16, eleven larvae about 
half grown were transferred from sumac to hickory. August 9, eleven 
more were transferred. The larvae seemed to do quite well and by winter 
many had made pupal cells. 

During June, 191 7, twelve adults emerged and were caged on pieces 
of hickory cut the preceding June, August, April, and February. No 
infestation occurred in any of the wood. The cage accidentally dried for 
a two- week period while the adults were ovipositing and this may account 
for the failure of infestation, as they require considerable moisture. 

In June, 191 7, adults from sumac were isolated in various cuts of 
hickory but no infestation occurred. 

July 23, 191 7, twenty-seven larvae were transferred from sumac to 
hickory cut in September, 19 16, and March, 191 7. The larvae did well 
and the following May and June 10 adults were reared and caged on 
hickory sticks cut in September, 19 17. On several of these sticks bands 
of thin outer bark of sumac were tied. 

The adults oviposited only on those sticks and at those places where 
the sumac bark was tied. July 30 they had not yet bored beneath the 
hickory bark proper, but by fall nearly all had entered the bark. Only 
one larva transformed to an adult in the summer of 19 19. One adult 
emerged in 1920. Several larvae did not transform but continued feeding 
beneath the bark during the summer of 19 19. 

LIOPUS ALPH.\, HICORIA FORM. EXPERIMENT XXX 

The Hicoria form was not successfully continued in confinement until 
the summer of 191 7. It requires wood cut in August, dried in the air for 
a month or so, and then placed on damp earth over winter. In addition 
the adults must be well fed on fungus spores {Endothea parasitica was 
used) before they will oviposit. 

It was again continued in hickory in 1918, 191 9, and 1920. During 
June, 19 19, many adults were caged on sumac branches and eggs were 
deposited. Three larvae lived to construct mines under the bark, but 
these died before November. 

CONCLUSIONS 

From the foregoing experiments and the fact that each of these two 
color forms has been taken only in the host given, it is evident that each 
has become restricted to that host and shows a strong predilection for 
it. Even after having fed for one year in a new host (Hicoria) adults 
developing from them showed a preference, in their oviposition, for that 
part of the hickory twig surrounded by Rhus bark. A fairly high mortal- 
ity of larvae occurred after transfer to the new host. 



Oct. 22, iQji Hopkins Host-Selection Principle 217 

HYPERPI/ATYS MACUIvATUS. EXPERIMENTS XXVI, XXVIII, AND XXIX 

Hyperplatys maculatus Hald. occurs throughout the eastern United 
States and west through the Rocky Mountain region. Two very similar 
species have been described, H. maculatus Hald. and H. aspersus Say, 
but the distinction is not drawn here, as each has many variations in 
color and maculation. It feeds on a great variety of hardwood deciduous 
trees. Probably any wood is attacked, provided it is in the proper 
condition for infestation. Smaller twigs and branches are usually pre- 
ferred. Those that have died during the fall and lain on the ground so 
that a certain amount of fermentation has taken place in the bark give 
the optimum condition. 

The larvae feed entirely beneath or in the bark, only entering the sap- 
wood to make a very shallow pupal cell. Adults fly in the early summer, 
late May, and early June. Two distinct variations occur in the length of 
the seasonal history. One form takes an entire year to complete the 
development, only one generation appearing each year. Another ma- 
tures from one-half to three-fourths of the brood in August and Sep- 
tember, the adults emerging and infesting new wood. This may be a 
basis on which to separate the two confused species. Four host strains 
have been experimented with, chestnut (Castanea dentata), gooseberry 
(Ribes), dogwood (Cornus florida Linn.), and yellow poplar (Liriodendron 
tulipifera lyinn.). 

HYPERPLATYS MACULATUS, LIRIODENDRON HOST STRAIN. EXPERIMENT XXVIII 

The colony was started by collecting infested tulip branches in No- 
vember, 1916, at Falls Church, Va. The following June adults emerged 
and were isolated in a cage containing yellow poplar, maple, dogwood, 
chestnut, and gooseberry cut in the fall of 1916. The original host, 
yellow poplar, was well infested, and a few larvae were found in goose- 
berry, but no other woods were attacked. In 1918, 1919, and 1920 the 
colony was continued in yellow poplar; the selection was not repeated. 

Only one generation of this form occurs each year. 

Castanea. Experiment XXVIII ^ — ^June 4, 1917, ten adults from 
yellow poplar were caged on chestnut cut in November, 1916. A very 
good infestation was secured, forty-five adults emerging in 1918. Eight 
of these adults were isolated in a cage containing yellow poplar and 
chestnut cut in November, 191 7. 

Examination in August showed the yellow poplar to be heavily 
infested, while no larvae were present in the chestnut. 

June I, 1 919, two pairs from yellow poplar were isolated on chestnut, 
and in 1920 thirty-six adults emerged. Ten were caged on optimum 
cuts of chestnut, and the yellow poplar was heavily infested. 

HYPERPLATYS MACULATUS, RIBES HOST STRAIN. EXPERIMENT XXVI 

This colony was started in December, 191 5, with infested gooseberry 
stems sent to Falls Church, Va., from Colorado Springs, Colo., by G. 



2i8 Journal of Agricultural Research voi. xxn,Na4 

Hofer. It has since been continued in gooseberry cut at Colorado 
Springs in the fall and shipped to Falls Church; in addition, several 
other host strains were produced. Only one generation of adults occurs 
each year. 

May 20 to June 10, 1916, adults emerged and 47 were isolated in a 
cage containing gooseberry, chestnut, and wild cherry, all cut in the 
preceding fall. The gooseberry stems were heavily infested, a few larvae 
were present in the wild cherry, but none were found in the chestnut. 
The colony has since been continued in gooseberry. 

Prunus. Experiment XXVI ^ — The infested wild cheiry twigs 
(described above) were caged separately, and in June, 191 7, four adults 
emerged. These were caged again on a fall cut of wild cherry. The 
infestation was not very good, and only six adults were secured in 19 18; 
these were recaged on the same wood, but no infestation occurred. 

Castanea. Experiment XXVI^ — As previously stated, the chestnut 
sticks were not infested in 1916 when caged with gooseberry. In June, 
191 7, nine adults were isolated on chestnut cut in November, 19 16, and a 
good infestation was secured. June, 1918, nine adults emerged and were 
isolated in a cage containing chestnut and gooseberrry cut in November, 
1917. Later examination showed only the gooseberry to be infested. 

LiRiODENDRON. EXPERIMENT XXVIl — ^June 5, 191 7, eight adults 
from gooseberry were isolated on tulip cut in November, 19 16. Five 
adults emerged from these sticks in 19 18 and were isolated in a cage 
containing tulip and gooseberry cut in November, 191 7. Neither wood 
was infested. 

HYPERPLATYS MACULATUS, CASTANEA HOST STRAIN. EXPERIMENT XXIX 

In April, 19 16, at Falls Church, Va., branches of chestnut (Castanea) 
containing larvse in the pupal cells were collected and caged. Some of 
the adults emerging in June were isolated with chestnut cut in March, 
1916, and the others isolated in a cage containing chestnut and dogwood 
(Comus) branches cut in March, 19 16. Those isolated on chestnut 
alone attacked this wood although it was a late cut. Those isolated on 
the two woods infested both, but the dogwood more heavily. Nothing 
more was done with the chestnut form. Many adults emerged that fall. 

CoRNUS. Experiment XXIX ^ — ^The dogwood sticks were then caged 
separately and adults secured in September, 1916, and more of them in 
June, 191 7. Those emerging during the latter period were recaged on 
August and November cuts of dogwood, but no infestation occurred. 

CONCLUSIONS 

In Hyperplatys maculatus host selection occurs to a certain degree ; 
but this beetle behaves differently from most of the other species 
tested. Thus the tulip form (experiment XXVIII) in 1 9 1 7 chiefly selected 
the same host, but it also oviposited on gooseberry. This gooseberry 
colony, however, was weak, and a high mortality in larvae occurred. 



Oct. 22. X92t Hopkins Host-Selection Principle 219 



Furthermore, although not selecting chestnut when the other host was 

present, they produced a good colony when isolated on it; but in 19 18 

these adults again selected tulip in preference to chestnut. The same was 

true with the original gooseberry form which was transferred to chestnut 

(experiment XXVI ^), for in 19 18 it returned to gooseberry in preference 

to chestnut. 

SUMMARY OF RESULTS 

(i) In practically all the species experimented with the adults show a 
marked predilection for the host in which they have fed as larvae, 
provided they are not deterred by other factors, such as the unfavor- 
able condition or the small quantity of the host. 

(2) There is considerable variation in the degree of preference for the 
original host, as between different species. Thus — 

(a) Certain species are capable of living in only one genus or species of 
plant, which consequently they select. 

(6) Certain species, chiefly those living in nature in several hosts, can 
be forced to adopt a new host. 

(c) Certain species, chiefly those feeding in nature in a great variety 
of plants, show little discrimination in the selection of hosts. 

(d) Certain species feeding in nature in a great variety of hosts often 
show a preference for a few of these. 

(3) In forced transference of individual adults of a species to a new 
host, a high mortality of the broods usually occurs, especially in the case 
of eggs laid by beetles emerging from the original host, in which case 
the mortality is often total. One-half to full-grown larvae, however, 
usually can be successfully transferred to a new host and live and trans- 
form to adults. 

(4) With some species that can be reared in a secondary (new) host, 
by the larvae feeding one or part of one year, preference for that host is 
shown by the resulting adults. 

(5) In general, the fewer the hosts in nature, the more marked the 
predilection for a particular host, and vice versa. 

(6) Continued breeding in a given host intensifies the preference for 
that host. 

(7) The condition of the host has a great influence on host selection, in 
that every species prefers an optimum condition of the host which it 
selects and will choose a new host in the optimum condition in preference 
to an old host in which the conditions are unfavorable. 

(8) The quantity of wood at the disposal of the ovipositing adults 
may influence the insects in their choice between different kinds of host 
wood, in that, if there are many adults to a limited amount of the primary 
host, some species will select a secondary host if such is available. If 
this is done, however, the resulting brood is weakened. 

It is altogether possible that these experiments may indicate the origin 
of certain closely related species or varieties of insects. For instance, a 
species restricted to a very few plants may accidentally be forced to 



220 Journal of Agricultural Research voi. xxn. no. 4 



take a new host (as actually happened in the experiments with Cyllene 
ia oak). A few individuals may survive and continue the strain so 
that it becomes, after a time, at least physiologically different and may 
also develop correlated differences of color or structure. It can hardly 
be said that such forms are much less distinct than in the case of the two 
species Callidium antennatum in pine and C. janthinum in juniper; for 
even though these have a slight color distinction and each is absolutely 
restricted to its own host, they interbreed. On the other hand, in the 
different forms of Hylotrupes ligneus, of which the eastern form in juniper 
is constant in marking, the western form in redwood is quite variable, 
as is also the Rocky Mountain form in Douglas fir. The juniper and 
redwood forms interbreed, but all attempts to mate either of these with 
the Douglas fir form have failed. All these forms can be furnished with 
substitute hosts, but in the experiments in which this has been done the 
original color pattern has resulted thus far. 

The grape and hickory strains of Cyllene pictus, although showing no 
color differences, do not readily mate. Two species of Cyllene, C. pictiis 
and C. rohiniae, are separable only as adults, by a slight difference in the 
color pattern, yet in seasonal and biological habits they are strikingly dif- 
ferent. It is conceivable that one of the two species originated through 
the adoption of a new plant and continuous breeding in that plant. 

It may be asked. If one or two years' feeding in a new host results in 
individuals which prefer that host, thus giving rise at least to new physio- 
logical varieties, why does not this occur more frequently in nature? 
That it does occur must be granted, as we have species living in many 
host plants as well as those restricted to a species or genus, but that it 
is not of more common occun-ence is believed to be due to the high mor- 
tality in first-stage larvae in a new host rather than to absence of oviposi- 
tion in the new host. Although the adults show a decided predilection 
for a favored host in ovipositing and even, in certain species, a preference 
for the plants in which the larvae have fed for one or two generations, 
the instinct to oviposit seems to overbalance that of host selection, con- 
sequently new hosts are frequently selected — possibly more frequently 
in nature than is generally realized. As an example of this, take 
Cyllene pichis requiring hickory cut during the winter. This con- 
dition would be fully met in tops left during logging operations. When 
the timber cutting ceased, a concentration of adults would be left with 
none of the favored host plant available in the right condition. The 
grape, osage orange, and hackberry strains collected at Hummelstown, 
Pa., were in reality taken in a woods v/hich had been logged for hickory 
and in which operations had ceased three years prior to the finding of 
these strains. At Falls Church, Va., in June, 1920, adults of Neoclytus 
erythrocephalus were observed ovipositing on pine logs. Much infested 
ash, from the previous year, was lying about from which they had 
emerged in great numbers. 



NOTES ON THE ORGANIC ACIDS OF PYRUS CORONARIA, 
RHUS GLABRA, AND ACER SACCHARUM 

By Charles K. Sando, Junior Chemist, Office of Plant Physiological and Fermentation 
Investigations, Bureau of Plant Industry, United States Departjnent of Agriculture, 
and H. H. BarTLETT, Collaborator, Office of Plant Physiological and Fermentation 
Investigations, Bureau of Plant Industry, United States Department of Agriculture, 
and Professor of Botany, University of Michigan 

During the study of other compounds found in the plants in question, 
we have incidentally isolated and identified the organic acids of the 
fruits of the wild American crab apple {Pyrus coronaria L.) and the 
smooth sumac {Rhus glabra L-.). We have also made an examination of 
the product known as " maple sand " (found to be impure calcium malate) 
which is formed as a granular deposit in the pans during the process of 
boiling down sap of the sugar maple {Acer saccharum Marsh.) to make 
maple sirup. Every precise record of the distribution of plant products 
is distinctly worth while, and rather than hold our data on the acids of 
these three plants for incidental mention in papers dealing with other 
matters, we have thrown them together in the following notes. 

OCCURRENCE OF MALIC ACID IN PYRUS CORONARIA, AND ITS 
TRANSFORMATION INTO SUCCINIC ACID 

As might have been predicted from the botanical relationship of Pyrus 
coroiiaria to the common apple, the very sour fruit of the American crab 
apple was found to contain malic acid. It was also found that in water 
extracts of this fruit, made without heat, there is a transformation of 
malic into succinic acid, apparently through the action of enzyms of 
the fruit itself. This discovery will be of no little interest if further in- 
vestigations substantiate our belief that microorganisms were not con- 
cerned in the process. 

Cold water extractions of crab apples collected near Ann Arbor, Mich., 
were made in the presence of both chloroform and toluol, with the 
expectation of obtaining solutions of the fruit acids free from pectin 
and other colloidal substances. The extractions were made in large 
stone jars, tightly packed with sliced fruits and filled to the top with 
water saturated with chloroform and toluol. At the bottom there was 
an excess of chloroform and at the top an excess of toluol. The solution 
quickly became intensely sour. The extraction was allowed to take 
place for several weeks, at the end of which time the apple tissues were 
as green and hard as when collected, showing no change whatever in 
appearance. The infusion was brown at the surface layer, but clear and 
of a pale straw color below. Only the non-oxidized lower part of the 

Journal of Agricultural Research, Vol. XXJI, No. 4 

Washington, D. C. Oct. 22, 1921 

aad Key No. G-249 

(221) 



222 Journal of Agricultural Research voi. xxn,No.4 

solution was drawn ofif through a tubulature at the bottom of the jar. 
It was not until succinic acid instead of one of the usual fruit acids was 
isolated from this solution that suspicion arose as to the possibility of 
any fermentation, other than an autolytic one, having occurred, and it 
was then too late to examine the solutions for microorganisms. It can 
only be stated that there was no evidence that such were present. The 
high acidity would have prevented bacterial action, and the perfectly 
clear solution, saturated with chloroform and toluol, showed no evidence 
of the presence of yeasts. 

The infusion was neutralized by stirring with calcium carbonate. 
During this process rapid darkening took place, with the formation, 
presumably, of the same brown oxidation product that forms when a 
cut surface of apple is exposed to the air. A dark precipitate separated, 
which was not a salt of one of the fruit acids. It was filtered off. From 
the filtrate it was possible to get relatively pure succinic acid by acidi- 
fying with hydrochloric acid and shaking with ether, but the more econom- 
ical and easy procedure, by which a larger yield was obtained, is 
described below. 

The neutralized filtrate was evaporated to a small volume and pre- 
cipitated with several volumes of ethyl alcohol. An impure calcium 
salt separated as a sticky, molasses-like mass. It was dissolved in dilute 
hydrochloric acid, and the solution was concentrated until the succinic 
acid crystallized out. It was purified by crystallization from lo per 
cent nitric acid, and finally by repeated recrystallization from water. 

The pure acid thus obtained agreed in all properties with succinic acid. 
It melted at 184° to 185° C. (Rosenthaler (13) ^ gives 185° C). The 
reaction mixture obtained by heating with concentrated sulphuric acid, 
when diluted, boiled, and neutralized with ammonia, gave a red solution 
with a strong green fluorescence, a characteristic reaction of succinic acid. 

Combustions of the pure material dried at 110° C, resulted as follows: 

(I) Weight of sample, 0.2378 gm. ; H.^O, 0.1047 g"!-! CO2, 0.3556 gm. 

(II) Weight of sample, 0.3063 gm. ; HjO, 0.1356 gm. ; CO2, 0.4530 gm. 
Calculated for C4H6O4; C, 40.66 per cent; H, 5.12 per cent. 

Found: (I) C, 40.78 per cent; H, 4.94 per cent. (II) C, 40.33 per cent; H, 
4.96 per cent. 

Titration with sodium hydroxid (NaOH) gave the following results: 

(I) 0.1776 gm. acid required 30.037 cc. NJio NaOH. 

(II) 0.1453 g"^- ^cid required 24.424 cc. N/io NaOH. 
Calculated for C4Hg04; replaceable H, 1.707 per cent. 
Found: (I) 1.703 per cent; (II) 1.692 per cent. 

A silver salt was prepared and analyzed, giving the following data: 

(I) 0.4809 gm. salt gave 0.3119 gm. Ag. 

(II) 0.4818 gm. salt gave 0.3126 gm. Ag. 
Calculated for C4H404Ag2; Ag 65.02 per cent. 
Found: (I) 64.86 per cent; (II) 64.88 per cent. 

' Reference is made by number (italic) to "Literature cited," p. 22S. 



Oct. 22, 1921 Acids of Pyrus coronaria, Rhus glabra, Etc. 223 

It will be observed that if malic acid had been present in the cold water 
extract, it would have been discarded with the mother liquor from the 
first crop of succinic acid crystals, since malic acid is not only very deli- 
quescent but likewise difficult to crystallize from solutions containing 
sugars and other impurities. Some malic acid was doubtless lost at 
this point, but the large yield of succinic acid indicated that it was the 
chief acid derived from the fruits which had undergone autolysis. 

It was of course a matter of interest to find out whether or not succinic 
acid was present also in the living fruit. A new supply of crab apples was 
therefore heated with water in an autoclave at 20 pounds pressure, and 
the juice, after filtration through cloth, was evaporated to a small volume 
and treated with several volumes of alcohol, to throw out pectin and 
other colloids insoluble in alcohol. The alcohol was distilled from the 
filtrate, which was concentrated, in vacuo, to a sirup. From this sirup it 
was impossible to obtain even a trace of succinic acid, by either of the 
methods which had been successfully used with the cold water extracts. 
It contained, on the other hand, a large quantity of malic acid, identified 
by the preparation and analysis of its silver salt. The concentrated sirup 
mentioned above was diluted with water, which brought about a sepa- 
ration of a small precipitate of red pigment, which was filtered off. When 
lead acetate was added to the filtrate, the acidity of the solution was so 
great that the first increment caused no precipitation of lead malate but 
did throw out a small amount of dark precipitate, which was of course 
removed. Further addition of lead acetate gave a voluminous precipi- 
tate of lead malate (A), which was filtered off and washed. A second 
yield of lead malate (B) was obtained from the solution by the addition 
of alcohol. The two precipitates were separately decomposed with 
hydrogen sulphid, neutralized with sodium hydroxid, and silver nitrate 
solution was cautiously added. The first few drops of the silver nitrate 
produced a dark precipitate which was removed by filtration. Further 
addition of silver nitrate caused white silver malate to separate. The 
precipitates were dried at 105 C. and analyzed as follows (two samples 
each from A and B) : 

(I) 0-5317 gin- salt from A gave 0.3315 gm. Ag. 

(II) 0.7249 gm. salt from A gave 0.4518 gm. Ag. 

(III) 0.3067 gm. salt from B gave 0.1882 gm. Ag. 

(IV) 0.5374 gm. salt from B gave 0.3303 gm. Ag. 

Foimd: (I) 62.34 per cent; (II) 62.32 per cent; (III) 61.36 per cent; (IV) 61.46 
per cent. 

Pure silver malate would have given 62.00 per cent silver. In view of 
the fact that the acid itself was not purified before the silver salt was 
formed, the analytical results are sufficiently close. Doubtless other 
acids than malic are present in very small quantity in the crab apple. 
The significant fact is that the fresh fruit contains malic acid as the pre- 
dominant acid, and not enough succinic acid so that we were able to 
isolate it. 



224 Journal of Agricultural Research voi. xxn. no. 4 

Although the old observations and experiments of Dessaignes (7) and 
of Liebig ( //) showed that succinic acid was formed from malic acid when 
calcium malate was present in mixtures being fermented by yeast, never- 
theless succinic acid is generally found in plants in such small quantities 
and always so intimately associated with asparagin that it is now cus- 
tomary to look upon it as a degradation product of protein rather than 
as directly related in metabolism to the other plant acids. The possi- 
bility of amino compounds giving rise by enzym action in the plant to 
succinic acid is sufl&ciently indicated by such discoveries as that of 
Ehrlich (8), who has traced the production of succinic acid by yeast to 
the fermentation of glutamic acid, and of Harden (jo), who has shown 
that putrefactive bacteria {Bacillus coli communis), in the presence of 
glucose, will transform aspartic acid almost quantitatively into succinic 
acid. However, it must be stated that the whole subject of the place of 
succinic acid in metabolism is much in need of investigation. It would 
be a decided step forward to show that it is possible for malic acid to be 
transformed directly into succinic acid by enzym action, as appears to 
have taken place by autolysis in the crab apple. 

We wish to indicate the possibility that such a transformation takes 
place and to point out that green fruits containing malic acid afford ideal 
material for a study of the problem. We do not wish, however, to give 
the impression that the possibility of fermentation by microorganisms 
was absolutely excluded in our work. 

To anyone who may be inclined to take up the problem of acid trans- 
formations in green fruits, a word of caution may not be amiss with regard 
to the statements that have crept into general reference books such as 
those of Czapek {6, p. 434) and Wehmer (14) with regard to the distri- 
bution of succinic acid in plants. It is recorded from a number of unripe 
fruits but has actually been isolated or satisfactorily identified in very 
few cases. In 1876 Brunner and Brandenburg (2) isolated it from the 
juice of uiu-ipe grapes (Viiis vinifera L.) . The source of most subsequent 
reports is a paper published in 1886 by Brunner and Chuard (5). These 
authors called attention to the earlier observation of Buignet (4) that 
the juice of green fruits is capable of absorbing a large amount of iodin, 
which enters into chemical combination with some constituent of the juice. 
At the same time, a precipitate is formed, which Buignet erroneously 
supposed to be the iodin compound. Brunner and Chuard, taking up 
the problem at this point, showed that the iodin compound remained in 
solution; and they obtained evidence which satisfied them that it was a 
glucosid of monoiodosuccinic acid, derived from a naturally occurring 
glucosid of succinic acid. Their investigation covered a considerable 
number of green fruits and plant juices; and they actually isolated suc- 
cinic acid, as such, from unripe gooseberries and from the petioles of 
rhubarb. In the other instances it was merely inferred from analogy 
that the supposed succinic acid glucosid was present. The procedure 



Oct 23, 1921 Acids of Pyrus coronaria, Rhus glabra, Etc. 225 

was to treat the juice with lead acetate, which supposedly threw out all 
iodin-absorbing compounds except the succinic acid glucosid. Then the 
presence of the latter, which was never isolated at all, was inferred from 
two circumstances: (i) that the purified juice absorbed iodin, and (2) 
that, after the absorption of iodin, a precipitate could be obtained with 
basic lead acetate, supposed to be lead monoiodosuccinate, which when 
treated with a mineral acid to liberate the free monoiodosuccinic acid, 
and then with finely divided metallic silver, gave malic acid. The pro- 
duction, under these circumstances, of malic instead of tartaric acid was 
thought to indicate that iodosuccinic acid had been present rather than an 
iodin derivative of the widely distributed malic acid. 

The weakness of the whole argument is sufficiently obvious without 
going into detail, since neither the putative glucosid of succinic acid nor 
the iodosuccinic acid was isolated; and it was not shown that the basic 
lead acetate precipitate was free from lead malate, which one would 
naturally expect to be found there. To the physiologist who is interested 
in the ripening of fruits it will be clear that the whole problem of the dis- 
tribution and significance in metabolism of succinic acid is much in need 
of more study. Especially, there can be no doubt that Buignet's iodin- 
absorbing compound (4), whatever it may be, should be taken account 
of in studies of fruit ripening. It exists in large amount in the unripe 
fruit and disappears as ripening proceeds. As far as we are aware, it is 
not even referred to in the recent literature of the subject. 

ACIDS OF RHUS GLABRA 

The acid of the sour, red pericarp of the sumacs (several species related 
to Rhus glabra) has been variously reported by different investigators as 
citric, malic, and tartaric. Gallic acid has likewise been reported. 
The closely related species of true sumacs are doubtless alike as to acid 
content. Our work, confined to R. glabra, has verified the findings of 
Rogers {12) nearly a century ago, and Frankforter and Martin (9) that 
the fruit acid is malic, nearly all in the form of the acid calcium salt. 
We were also able to isolate free gallic acid, which seems not to have 
been reported from this particular species. There are statements in the 
older literature that free gallic acid occurs in the leaves of the European 
sumac, R. coriaria L. 

The berries of Rhus glabra were boiled with successive quantities of 
distilled water. The water solutions were clarified and largely freed 
from tannin by boiling with hide powder and o^gg albumen, and were then 
shaken with ether. The combined ether extracts were evaporated to a 
sirupy consistence and deposited gallic acid as a yellow powder. The 
latter was filtered off on a Buchner funnel and crystallized repeatedly 
from water. It was obtained in pure and almost colorless condition by 
precipitation from solution in absolute alcohol by chloroform, or by 
recrystallization from glacial acetic acid. As obtained by crystallization 



226 



Journal of Agricultural Research voi. xxn.No.4 



from water it formed brown aggregates of large crystals containing one 
molecule of water. (Calculated for CyHgOs. HjO, HjO, 9.57 per cent; 
found, 9.39 per cent.) It was identified by the usual tests. Mr. N. A. 
Lange made combustions of some of the purified acid and of its triacetyl 
derivative, the results of which he kindly permits us to publish as follows : 

I. The acid gave C, 50.19 per cent; H, 3.92 per cent. Calculated for gallic 

acid: C, 49.40 per cent, H, 3.56 per cent. 

II. The acetyl derivative gave C, 53.91 per cent; H, 4.13 per cent. Calculated 

for triacetyl gallic acid, C, 54.39 per cent; H, 4.06 per cent. 

The melting point of the triacetyl gallic acid, stated variously in the 
literature from 151° to 165° and 166° C, was 162° to 163° C. 

After the removal of tannin and gallic acid the aqueous extract from 
the berries was largely neutralized with calcium carbonate and filtered 
hot, after considerable concentration. Alcohol threw out a voluminous 
precipitate, the first fractions taffy-like, later ones solid. These fractions 
were treated with enough hydrochloric acid to form the acid calcium 
salt, and were repeatedly treated with animal charcoal and recrystallized 
from hot water. 

The pure crystals were dissolved in water, exactly neutralized with 
standard alkali; and normal silver malate was precipitated by the addi- 
tion of silver nitrate. The four successive fractions of the crude calcium 
salt were designated A, B, C, and D, and each was purified and converted 
into the silver salt. In addition, a portion of fraction A was purified 
by further recrystallization and was obtained in two portions called Aa 
and Ab, from which silver salts were also prepared. The duplicate 
analytical figures for all of the silver precipitates are given in Table I. 

Table I. — Duplicate analyses of silver salts prepared from a series of precipitates obtained 
by fractional separation with alcohol from an aqueous solution of calcium salts of the 
organic acid of the sumac fruit 



Fraction. 



A-I... 

A-II. 

Aa-I. 

Aa-II 

Ab-I. 

Ab-II 

B-I... 

B-II. 

C-I... 

C-II. . 

C-III . 

D-I. . 

D-II. 



Weight of 
silver salt. 



3.2895 
•2133 
•4385 

^•1358 
.6137 
.9674 
■4541 
•6359 
.4242 

•5015 
.3921 

•4195 
•5263 



Weight of 
silver. 



0.1794 
.1318 
.2721 
.7048 

•3794 
.5962 
.2806 

•3932 
.2624 

•3099 
.2420 
.2596 
•3259 



Percentage 
of silver. 



61.96 
61.79 
62.05 
62.05 
61.82 
61.63 
61.79 
61.83 
61.85 
61.79 
61.71 
61.88 
61.92 



Oct 22. 1921 Acids of Pyrus coronaria, Rhus glabra, Etc. 227 

The figures from all the fractions are in excellent agreement with each 
other and agree fairly well with malic acid. The results prove beyond 
much doubt that only one acid is present in any quantity. Rogers {12), 
the first to show the presence of calcium malate in berries of Rhus glabra, 
did not attempt to prove that malic acid was the only one present, and 
subsequent work was less careful than his. Although convinced by the 
identity of the silver salts that nothing but malic acid was present in our 
material, we felt that the determinations should be closer to the calcu- 
lated value. The average of 13 determinations makes the percentage of 
silver in the silver salt 61.85, whereas the theoretical value is 62.00 for 
pure malic acid. We, therefore, prepared silver malate, using a Kahl- 
baum preparation of the acid, and made four silver determinations in the 
same manner in which our other determinations were made. The four 
determinations gave us 61.94 P^^ cent, 61.92 per cent, 61.81 per cent, 
and 61.91 per cent, averaging 61.89 per cent silver in pure silver malate 
by our method of preparation and analysis. There can, therefore, 
remain no doubt that the acid of sumac berries is all malic. 

MALIC ACID IN SUGAR-MAPLE SAP 

It is no new observation that mafic acid is present in the sap of the 
sugar maple. Cowles (5), for example, has published methods for the 
estimation of malic acid in maple products. Although it might have 
been anticipated that the granular precipitate known as "maple sand" 
which is deposited in the pans during the concentration of the sap would 
prove to be calcium malate, no one, as far as we know, has previously 
reported an analysis. Bloor (j) used "sugar sand" as a source of acid 
in his work on the transformation of malic acid into sugar by the tissue 
of the maple ^ but gave no data to bear out the natural and perhaps 
quite justifiable inference that the acid was actually malic. Our sam- 
ple was kindly obtained for us from Ohio, by Dr. Clinton A. Ludwig, 
now of Clemson College, S. C. It was only necessary to add to the 
"maple sand " sufficient hydrochloric acid to transform the crude calcium 
malate into the acid calcium salt. The latter was obtained pure by 
repeated boiling with animal charcoal and recrystallization from hot 
water. It was neutralized with alkali, and silver nitrate was added to 
precipitate the insoluble silver malate. Three separate analyses for 
silver gave the following results : 

(i) 0.4129 gm. silver salt gave 0.2560 gm. Ag. 

(11) 0.1922 gm. silver salt gave 0.1190 gm. Ag. 

(ill) 0.2892 gm. silver salt gave 0.1796 gm. Ag. 

Calculated for C4H405Ag2; Ag, 62.00 per cent. 

Found: (i) 62.00 per cent; (11) 61.91 per cent; (iii) 62.10 per cent. 

1 It may be noted that Bloor used tissues of "Acer saccarinum " for his work. Since he gives no authority 
for the name, one is left in doubt as to whether he means the silver maple (A. saccharinum L.) or the 
sugar maple (,A. saccharum Marsh.; A. saccharinum Wang., not I,.). 

65583°— 21-^ — 4 



2 28 Journal of Agricidtural Research voi. xxn, no. 4 

SUMMARY 

(i) The acid of the sour fruit of the wild American crab apple, Pyrus 
coronaria, is malic acid. When the fruit undergoes autolysis under 
anaerobic conditions, in the presence of chloroform and toluol, this acid 
appears to be transformed largely into succinic acid. Further experi- 
ments, however, will have to be made in order to repeat the observations 
and to determine the exact process involved. 

(2) The acid of the outer part of the red fruit of the smooth sumac, 
Rhus glabra, is malic acid, occurring in the form of the acid calcium 
salt. With it is associated a considerable quantity of free gallic acid. 

(3) Malic acid is present in the form of calcium salts (both acid and 
normal) in maple sap. The product known as "maple sand" obtained 
from the evaporating pans is crude calcium malate. 

LITERATURE CITED 
(i) Bloor, W. R. 

1912. STUDIES ON MALIC ACID. I. THE TRANSFORMATION OP MALIC ACID TO 
SUGAR BY THE TISSUES OP THE MAPLE (aCER SACCHARINUm). In JoUt. 

Amer. Chetn. Soc, v. 34, no. 4, p. 534-539- 
(2) Brunner, Heinrich, and Brandenburg, Rudolph. 

1876. UEBER DAS VORKOMMEN DER BERNSTEINSAURE IN UNREIFEN TRAUBEN. 

In Ber. Deut. Chetn. Gesell., Jahrg. 9, p. 982-984. 

is) ^^d Chuard, Ernest. 

1886. PHYTOCHEMiscHE STUDiEN. 7» Bef. Deut. Chem. Gesell., Jahrg. 19, 
p. 595-622. 

(4) BlHGNET, H. 

1861. RECHERCHES SUR LA MATI^RE SUCR^E CONTENUE DANS LES FRUITS 
ACIDES; SON ORIGINS, SA NATURE ET SES TRANSFORMATIONS. In Ann. 

Chim. et Phys., ser. 3, t. 61, p. 233-308. 

(5) CowLES, H. W., Jr. 

1908. THE DETERMINATION OP MALIC ACID IN FOOD PRODUCTS. In Jour. Amef. 

Chem. Soc, v. 30, no. 8, p. 1285-1288. 

(6) CzAPEK, Friedrich. 

1905. BIOCHEMIE DER PPLANZEN. 2 vol. Jena. 

(7) DessaignEs, v. 

1849. NOTE SUR LA CONVERSION DU MALATE DE CHAUX EN ACIDE SUCaNIQUE. 

In Ann. Chim. et Phys., s6r. 3, t. 25, p. 253-255. 

(8) Erhlich, Felix. 

1909. UBER DIE ENTSTEHUNG DER BERNSTEINSAURE BEI DER ALKOHOLISCHEN 

garung. In Biochem. Ztschr., Bd. 18, Heft 3/5, p. 391-423. 

(9) Frankporter, G. B., and Martin, A. W. 

1904. A CHEMICAL STUDY OP THE SEED OP RHUS GLABRA. In Amer. JoUT. 

Pharm., v. 76, no. 4, p. 151-158. 

(10) Harden, Arthur. 

19OI. THE CHEMICAL ACTION OF BACILLUS COLI COMMUNIS AND SIMILAR ORGAN- 
ISMS ON CARBOHYDRATES AND ALLIED COMPOUNDS. In JoUt. Chem. 

Soc. [London] Trans., v. 79, pt. i, p. 610-628. 

(11) LiEBiG, Justus. 

1849. UEBER DIE DARSTELLUNG DER BERNSTEINSAI7RE AUS APPELSAUREM 
KALK. In Ann. Chem. u. Pharm., Bd. 70, Heft i, p. 104-107. 



Oct. 2J.I92I Acids of Pyrus coronarta, Rhus glabra, Etc. 229 

(12) Rogers, William B. 

1835. ON THE EXISTENCE OP THE BI-MALATE OP LIME IN THE BERRIES OP THE 
sumach; and the mode op procuring it prom them in the CRYSTAL- 
LINE PORM. In Amer. Jour. Sci., v. 27, no. 2, p. 294-299. 

(13) ROSENTHALER, L. 

1914. DER NACHWEIS ORGANISCHER VERBINDUNGEN . . . I070 p., 3 fig., I pi. 

Stuttgart. (Margosches, B. M. Die chemische Analyse, Bd. 19/20.) 

(14) Wehmer, Carl. 

191 1. DIE PFLANZENSTOFFE . . . xvi, 937 p. Jena. Literatur, p. x-xiv. 



FERTILITY IN SHROPSHIRE SHEEP ' 

By Elmer Roberts 
Illinois Agricultural Experiment Station 

Heape (4) ^mentions that in some breeds young ewes bear fewer twins than 
older ewes. Carlyle and McConnell {2) reported some observations which 
they had made on the efifect of age on fertihty in sheep from which they 
concluded that ewes from 3 to 6 years old averaged a larger percentage of 
lambs than younger or older ewes, and also that i -year-old rams were not 
so prolific as those 2 or 3 years old. The same conclusions were reached 
by Humphrey and KJeinheinz (<5) from a study of later records of the 
Wisconsin flock. Recently Jones and Rouse (7) showed that in sheep the 
percentage of twins increased with age until 5 years, when there was a 
decided drop. 

The present paper gives the results of a study of the influence of age 
and season upon fertility in American Shropshire sheep. 

The source of data is the American Shropshire Sheep Record (j). 
Individuals with registry numbers between 325502 and 344869 have 
been used, date of birth noted, whether bom as single, twin, or triplet, 
and age of dams and sires looked up. 

AGE OF EWE AND FERTILITY 

Table I shows the percentage of lambs born as singles, twins, and trip- 
lets from dams of various ages. Ewes under i year and 6 months are 
grouped in the i-year class, those i year and 7 months to 2 years and 6 
months in the 2 -year class, and so on. The percentage in multiple births 
increases to 4 years and remains fairly constant through 8 years. For 
the older groups the numbers are too small to draw conclusions. 

I Paper No. i6 from the Laboratory of Genetics, Agricultural Experiment Station, Urbana, III. 
* Reference is made by number (italic) to "Literature cited," p. 234. 

Journal of Agricultural Research, Vol. XXII, No. 4 

Washington, D. C. Oct. 22, 192 1 

aae Key No. Ill.-ii 

(231) 



232 



Journal of Agricultural Research voi. xxn. N0.4 



Table I. — Age of ewe and fertility 



Age of dam in years. 


Total 
number of 
offspring. 


Percentage 
of singles. 


Percentage 

of twins. 


Percentage 
of trip- 
lets. 


Percentage 

in multiple 

births. 


I 


379 

2,299 

2,025 

1,762 

1,256 

942 

506 

405 

157 

96 

23 
3 
4 
I 

4 
5 
I 


77.0 
66.4 
63.6 
57-6 
58.0 

53-7 
56-3 
54-8 
62.4 

38.5 
60.9 

100. 
50.0 

100. 
75-0 
20.0 

100. 


23.0 
33-2 
36.1 
41.4 
43 -o 
46.0 

43-1 
44-5 
37-0 
61.5 

39-1 




23.0 
33-6 
364 

42 .4 




0.4 

•3 

I.O 
I.O 

•3 
.6 

•7 
.6 


1 


A 


e 


44.0 


6 


46.3 
43-7 
45-2 
37-6 
61.5 
391 


7 


8 





10 


II 




12 




I J 


50.0 




50.0 


Z/i 




If 


25.0 
80.0 




25.0 
80.0 


16 




20 * 
















9,868 


60.8 


38.7 


.6 


39-2 



" This may be a mistake in the record. 

AGE OF RAM AND FERTILITY 

Table II gives the percentages of lambs bom as singles, twins, and 
triplets bom from sires of various ages. From these percentages one 
can not ascribe to the ram any influence on fertility. Carlyle and 
McConnell {2) thought that i -year-old rams were not so prolific as 
older rams, but this is not borne out by the figures in Table II. 

Table II. — Age of ram and fertility 



Age of ram in years. 


Total 
number of 
offspring. 


Percentage 
of singles. 


Percentage 
of twins. 


Percentage 
of triplets. 


Percentage 

in multiple 

births. 


I 


1,101 
3.265 
2.552 
1,460 
650 

434 

244 

118 

71 

47 

2 

3 


58.7 
60.6 

59-1 
65.8 

55-5 
66.1 

70-5 
74.6 

63-4 

68.1 

100. 

lOO.O 


40.4 
390 
39-9 
33-8 
43-8 
33-6 
295 
25-4 
36.6 

25-5 


0.9 

•5 

I.O 

.2 
.6 
.2 


41-3 
39-4 
40.9 
340 
44-5 
33-9 
295 
25.4 
36.6 

319 


3 


■I 


4. .... 


e 


6 

7 


8 







6^4' 


10 


II 


12 


















9.947 


61.2 


38.2 


.6 


38.8 



TIME OF BIRTH AND TWINNING 



Heape (5), who gathered information from flock masters, states that 
55 per cent of them reported that twins were usually born early in the 



Oct. 



Fertility in Shropshire Sheep 



233 



lambing season. To test this point Table III was made, showing the 
month of birth and the percentages of singles, twins, and triplets. It is 
readily seen that a larger percentage of twins is bom early in the season 
than is bom later. Of the 3,790 lambs bom in January, February, and 
March 42.3 per cent are twins, while of the 4,617 born in April, May, 
and June only 36.1 per cent are twins. If the triplets are added in with 
the twins the percentages are 43.1 in multiple births for January, 
February, and March, and 36.7 for April, May, and June. As Heape (5) 
points out, this may be due to the ewes with the most vigorous and active 
generative systems coming into heat earlier in the season. This may 
be also affected by the fact that early in the breeding season more 
green feed is available, a factor influencing the number of twins pro- 
duced. 

Table; lll.—Months of birth {Shropshires) 



Month. 



January. .. 
February. . 

March 

April 

May 

June 

August. .. . 
September 
December, 



Total 
number. 



33 

471 

3,286 

3.615 
966 

36 



Percentage 
of singles. 



75-8 
56.7 
56.7 
62.4 
66.3 

75-0 
100. o 
100. o 



Percentage 
of twins. 



24.2 

43-1 
42.4 

37-0 
32.9 
25.0 



Percentage 
of triplets. 



In the hope that additional information might be obtained, a study 
was made of the Dorset breed (j), which produces a large number of 
young in the fall. Table IV gives the month of lambing and the percent- 
ages of singles, twins, triplets, quadruplets, and of all multiple births. 

Table lY.— Months of birth (Dorsets) 



Month of birth. 



January. . 
February . 

March 

April 

May 

June 

July 

August. . . . 
September 
October. . . 
November , 
December . 



Total num- 
ber. 



Percentage 
of singles. 



1,818 
2,386 

3.919 
2,366 

857 

296 

90 

102 

925 
1.546 
1,088 
1,418 

16,634 



61.5 
54-3 
52.7 
51-7 
';4.8 

59-8 
65.6 
68.6 

73-7 
66.2 
67.7 
61.0 

57-8 



Percentage 
of twins. 



37-1 
41.9 

43-9 

45-4 
43-1 
38.5 
27.8 

30-4 
25-3 
32.9 
30-5 
36.8 

39-7 



Percentage 
of triplets. 



1-3 
3-8 
3-2 
2.6 
2.1 

1-7 
6.6 
i.o 
•9 
•9 
1.8 
2.2 

2.4 



Percentage 
of quad- 
ruplets. 



Percentage 

in multiple 

births. 



38.S 
45-7 
47-3 
48.3 
45-2 
40.2 

34-4 
31-4 
26.3 
33-8 
32.3 
39-0 

42.2 



234 Journal of Agricultural Research voi.xxn.No.4 



From Table IV it can be clearly seen that the percentage of multiple 
births is greater in the spring. If the births occurring from February 
to June, inclusive, are combined it is found that 48.2 per cent are in 
multiple births, while for the other months the percentage is 34.9. In 
Shropshires a larger percentage of twins or multiple births occurs in 
January, February, and March than later. This condition does not seem 
to hold for the Dorsets. Therefore, this condition in the Shropshires is 
not likely due to more green feed early in the mating season. The 
causes of these significant differences in multiple births at different sea- 
sons among sheep are yet to be discovered. 

SUMMARY 

(i) Multiple births increase with age up to 4 years. From this point 
they remain fairly constant until 8 years. Beyond this age the numbers 
are too small to draw conclusions. 

(2) The age of the ram has no influence on the percentage of multiple 
births. 

(3) Among Shropshire sheep more multiple births occur early in the 
lambing season than later. 

(4) Among Dorsets more multiple births occur in spring than in fall. 

LITERATURE CITED 
(i) American Shropshire Registry Association. 

1912. AMERICAN SHROPSHIRE SHEEP RECORD. V. 25. 

(2) CARtYLE, W. L., and McConnell, T. F. 

1902. SOME OBSERVATIONS ON SHEEP BREEDING FROM THE EXPERIMENT STA- 
TION FLOCK RECORDS. Wis. Agr. Exp. Sta. Bul. 95, 19 p. 

(3) Continental Dorset Club. 

1900-17. continental dorset club sheep record. v. i-16. 

(4) Heape, Walter. 

1899. abortion, barrenness, AND FERTILITY IN SHEEP. In JouT. Roy. Agr. 
Soc. England, v. [60] (ser. 3, v. 10), no. 38, p. 217-248. 



(5) 



1900. note on the fertility of DIFFERENT BREEDS OF SHEEP, WITH RE- 
MARKS ON THE PREVALENCE OF ABORTION AND BARRENNESS THEREIN, 

In Proc. Roy. Soc. [London], v. 65, 1899, p. 99-111. 

(6) Humphrey, George C, and Kleinheinz, Frank. 

1907. OBSERVATIONS ON SHEEP BREEDING FROM RECORDS OF THE UNIVER- 
SITY FLOCK. In Wis. Agr. Exp. Sta. 24tli Ann. Rpt. 1906-07, p. 25-40, 

(7) Jones, Sarah V. H., and RousE, James E. 

1920. THE RELATION OF AGE OF DAM TO OBSERVED FECUNDITY IN DOMESTI- 
CATED ANIMALS. I. MULTIPLE BIRTHS IN CATTLE AND SHEEP. /« 

Jour. Dairy Sci., v. 3, no. 4, p. 260-290. References, p. 288-290. 



Vol. XXII OOXOBE)R 2^, 1921 No. 5 

JOURNAL OF 

AGRICUI/TURAlv 

RESEARCH 



CONTKNXS 

Page 

Relation of Soil Temperature and Other Factors to Onion 
Smut Infection - ^ - - - - - - 235 

J. C. WALKER and L. R. JONES 

(Contribution from Bureau of Plant Industry and Wisconsin Agricultural Experiment Station) 

A Physiological Study of Grapefruit Ripening and Storage 263 

LON A. HAWKINS 

(Contribution (torn Bureau of Plant Industry) 

Absorption of Copper from the Soil by Potato Plants - 281 

F. C. COOK 

(Contribution from Bureau of Chemistry) 



PUBUSHED BY AUTHOMTY OF THE SECRETARY OF AGRICULTURE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



w:a.shingxon, d, c. 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 

J. G. LIPMAN 

Dean, State College of Agriculture', and 
Director, New Jersey Agricultural Expert- 
tnent Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 

of Minnesota 

R. L. WATTS 

Dean, School of Agriculture; and Dtrtctor; 
Agricultural Experiment Station; The 
Pennsylvania State College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultural Experiment Station, New 
Brunswick, N. J. 



JOIfflALOFAGltianmffiSEARCe 

c — %..JS * "I a t ."— f" i <M~i - — ■ - 

Vol. XXII Washington, D. C, October 29, i923e^ew YtpNo. 5 



RELATION OF SOIL TEMPERATURE AND OTHER FAC- 
TORS TO ONION SMUT INFECTION 

By J. C. Walker, Pathologist, Office of Cotton, Truck, and Forage Crop Disease Inves- 
tigations, Bureau of Plant Industry, United States Department of Agriculture, and 
Assistant Professor of Plant Pathology, University of Wisconsin, and L. R. JONES, 
Professor of Plant Pathology, University of Wisconsin 

OCCURRENCE OF ONION SMUT IN RELATION TO CLIMATE AND 
CULTURAL PRACTICES 

The onion smut fungus, Urocystis cepulae Frost, was first reported by 
Ware {11) ^ in the Connecticut River Valley in 1869. At this early date 
it was causing some injury to the onion crop, and in 1888 it was reported 
by Thaxter (jo) to be of much importance in the old onion soils of south- 
em New England. During the years which have since elapsed it has 
successively appeared and become an economic factor in nearly all the 
more westerly regions of intensive onion culture of the northern States, 
from New York to Oregon. It is possible that this fairly rapid distri- 
bution of the parasite has been occasioned to some extent by smut spores 
carried with the seed, as already noted by Chapman {2) and Munn (7, 
p. 412). It has, however, more probably been brought about by the 
increasingly widespread distribution of onion sets. Many of these sets 
are grown in the northern States on smut-infested soils, and since they 
are shipped in quantity to all parts of this country, and even exported, 
their role in the wide dissemination of smut spores is obvious. 

Chance introduction of the smut fungus in this way in the northern 
commercial onion-growing sections seems almost certain to lead to its 
permanent establishment. This evidently results from the fact that the 
common intensive practice of continuous cropping with onions for an 
indefinite term of years tends, when once the inoculum is introduced in 
the soil, to favor its increase and wider distribution season by season until 
it becomes a factor limiting further success with this crop. While this 
holds true for the northern States, it does not seem to be so in the 
southern sections. This is the more noteworthy since northern sets 
grown on smutty soil are each year shipped into the southern onion 
districts for propagation. This regional limitation of the smut fungus 
was impressed upon one of the authors (Walker) in connection with a 

1 Reference is made by number (italic) to *'I,iterature dted," p. 261, 

Journal of Agricultural Research, Vol. XXII, No. s 

Washington, D. C. , Oct. 29, 1921 

aaf Key No. 0-250 



236 Journal of Agricultural Research voi. xxii. no. s 

survey which he made some two years ago of the chief onion-growing 
centers of the entire country, as a representative of the Office of Cotton, 
Truck, and Forage Crop Disease Investigations of the United States 
Department of Agriculture. In connection with this, he personally 
inspected the leading onion-growing sections of Texas and Louisiana 
and conferred with the plant pathologists of these two States, Drs. C. W. 
Edgerton and J. J. Taubenhaus. No evidence of the disease was found, 
and it had not been reported to the Experiment Station of either State.^ 

In comparing the distribution and occurrence of onion smut in differ- 
ent sections of the country, it is necessary to keep in mind that two 
distinct types of onion culture are practiced in the United States. The 
first is followed in practically all of the northern sections, the second is 
the rule in the southern commercial growing regions, and in one or more 
sections in the Pacific coast States. In the first, or northern, type the 
seed is sown directly in the field as early in the spring as the soil can be 
properly prepared — that is, in March, April, or May, according to local 
climatic conditions. The bulk of the crop is harvested in these northern 
districts in August and September. The Globe varieties predominate, 
including the red, yellow, and some white. In the second, or southern, 
type of culture the seed is sown in special beds in late summer. The 
seedlings are tlien transplanted to the main field during the early winter 
months and the crop is harvested during the period from April to July. 
Here the Bermuda, Italian, and Spanish varieties predominate. The 
survey previously referred to brought out the fact that onion smut has 
become established in essentially all of the older onion-growing sections, 
which practice the first type — with spring sowing of seed — while smut 
is either entirely unknown or of no economic importance in those locali- 
ties where the seed is sown in summer followed by transplantation. 
Wherein lies the explanation? As already suggested, it can not be due 
to the matter of chance introduction of the organism. This is certainly 
being distributed frequently and widely throughout the South. It would 
seem rather to be associated with some of the factors incident to the 
southern type of culture. The conspicuously different factors as out- 
lined above are three: (i) The varieties used, (2) transplantation vs. 
direct seeding, (3) climatic differences associated with season of culture. 

Greenhouse experiments, in which we have tested the different types, 
have shown that the Bermuda and Spanish varieties which are used in 
the South are as susceptible to smut infection as are the Globe varieties 
of the northern sections. Hence varietal resistance does not furnish the 
explanation. Turning to cultural methods, we find that in the South the 
seed beds in which the onions are grown preliminary to transplantation 
are usually of considerable size and are located as a rule in a portion of 

1 The authors are indebted to Doctors Edgerton and Taubenhaus for continued cooperation in the search 
for the smut in their respective States. They each reported again early in the current year that not a 
single specimen had as yet been found. 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 237 

the main field. Therefore, if the organism were present and environing 
factors were favorable, it does not seem probable that this method of 
culture would completely inhibit the disease. Indeed, judging from our 
experience with cabbage transplantation in relation to clubroot and 
other soil- or seed-borne diseases, this method, instead of reducing the 
trouble, is likely to serve as a ready means of distributing the parasites 
with diseased seedlings from localized centers to wider areas. We are 
thus forced to turn for explanation of the absence of smut in the South to 
the third suggestion, that relating to climatic differences, bearing in mind 
the respective cultural seasons. The most evident environmental dif- 
ferences associated with the two types of culture relate to soil tempera- 
ture and moisture during the time of seed germination and early seedling 
development, which constitute the smut infection period. In the north- 
em type, the spring-sown seed develops in a soil which is comparatively 
cool and which has in general a relatively high and constant surface 
moisture content. In the southern type, the summer-sown seed must 
germinate and pass the early developmental stages in a soil of relatively 
high temperature and subject to superficial desiccation. Our problem 
has, therefore, necessitated an attempt to analyze and evaluate the pos- 
sible factors associated with variations in soil moisture and soil tempera- 
ture during the seedling stage. 

INFECTION PERIOD 

It has been of obvious importance in this study to know quite defi- 
nitely the period in the development of the host at which infection actu- 
ally occurs. Thaxter {10) gave critical attention to the time and manner 
of infection, concluding that the fungus always invaded the young seed- 
lings below the surface of the soil and that, by subsequent growth of the 
host, the infected cells were commonly carried above the ground before 
visible signs of the disease appeared. He also noted that onion sets and 
onion bulbs replanted for seed growing were not attacked and suggested 
that the seedling was probably subject to attack in only the early stages 
of its development. Sturgis (9) later found that seedlings half as thick 
as a lead pencil and about 5 inches high, transplanted into smutty soil 
did not contract the disease. Sirrine and Stewart {8) , in an experiment 
started at Jamaica, N. Y., on May 2, sowed eight rows of onion seed, 
each ID feet in length, in soil free from smut. Alternate rows were left 
as controls. Soil from a smut-infected field was introduced in three 
ways: (i) in the furrow with the seed in two rows; (2) on the surface of 
the soil after the furrow was closed in one row; and (3) in a fourth row, 
on the surface of the soil 1 1 days after planting, this being shortly after 
the seedlings appeared above ground. The control rows remained 
healthy. Heavy infection occurred with the first treatment, slight 
infection with the second, no infection with the third. It seems possible 



238 



Journal of Agrictdtural Research voi. xxii, no. 



that in the third treatment, where the inoculum was merely placed on 
the surface of the soil when the seedlings were well started, the method 
failed to insure a sufficiently intimate and immediate contact of germi- 
nating spores with embryonic tissue to justify definite conclusions. 
Reviewing the evidence as a whole, however, it is obvious that the 
smut fungus is capable of invading the onion seedling for only a short 
period after seed germination. 

In order to define more exactly the limits of this period of smut infec- 
tion, we carried through a series of greenhouse trials. In the first of 
these 17 pots of sterilized greenhouse soil were planted with Red Globe 
onion seed which had been treated with i to 25 formaldehyde solution 
for 15 minutes. At two-day intervals beginning the eleventh day after 
planting, two pots were inoculated by mixing smut-infested soil in the 
upper layers of the pot, so that the inoculum was brought into close con- 
tact with the embryonic region of the cotyledon. At the time of inocu- 
lation, all retarded seedlings were removed, so that only plants of uniform 
height were considered in each case. All plants were pulled and exam- 
ined for signs of the disease three to four weeks after inoculation. The 
results of this experiment, given in Table I, show that, under greenhouse 
conditions at least, infection may occur until the cotyledon is about 2 }4 
inches above ground, or for a period of two weeks or more after sowing. 
Thus, the infection period appears to be slightly longer than that re- 
ported by Sirrine and Stewart (8) . 

Table I. — Relation of the stage of development of the onion seedling to infection by Uro- 

cystis cepulae 



Soil treatment. 


Pot No. 


Length of 
period 

between 
sowing and 
inoculation. 


Height of 

cotyledons 

above 

ground at 
inoculation. 


Number of 
plants. 


Percentage 
infected. 




I 
2 


Days. 
II 
II 


Inches. 

i-iK 


28 
18 


89 
67 




3 


13 


iK-2 


24 


21 




4 


13 


lK-2 


19 


26 


Inoculated 


5 
6 


15 

IS 


lK-2 

iK-2 


17 
16 


53 




75 




7 
8 


17 
17 


2 -2>^ 
2 -2K 


19 
20 


16 

25 




9 


19 


2K-3 


II 


00 




10 


19 


2K-3 


II 


00 




f ^^ 






14 


00 




12 






37 
22 
21 
18 


00 




13 
14 
15 
16 






00 


Uninoculated 






00 








00 








25 
45 


00 




i 17 






00 











Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 239 

The foregoing experiment was repeated in a somewhat cooler house, 
in which the temperature remained close to 15° C. most of the time, 
rising to about 20° during the middle of the day. Under these conditions, 
the maximum length of the cotyledon was about 2)4 inches. The data 
from this experiment are reported in Table II. The plants became im- 
mune at approximately the same time as noted in the first experiment — 
between the nineteenth and twenty-fourth days after sowing, when the 
cotyledon had about attained its full growth and as the first leaf was 
emerging. It will be recalled that the basal portion of the cotyledon, as 
with each of the later leaves, forms a collar or sheath inclosing the lower 
parts of the younger leaves. The question arose as to whether or not 
immunity to smut infection is directly associated with maturity of the 
tissues. If so, it would seem that the explanation of this later escape 
of the onion plant from infection lies in the fact that the maturing basal 
sheath forms a thin but normally complete barrier of resistant tissue 
between the potentially infective soil and the deeper-lying embryonic 
tissues of the younger developing leaves. The removal of this mechan- 
ical barrier might, therefore, permit of infection at a later stage. In 
order to determine whether this is the case, the following experiments were 
undertaken. After the thirty-first day, when the onion seedlings had 
passed the so-called susceptible period, the cotyledons were carefully 
removed from the plants in one pot, and infected soil was placed around 
the base of the exposed first leaf. Sixty per cent of the plants thus 
treated became infected as shown in Table II, pot 9. This proves that 
the first leaves are susceptible even after the cotyledon becomes immune. 
On the fortieth day, a i-inch layer of infested soil was placed on top of 
pots 10 and 1 1 , so as to surround the first leaves in proximity to the axils. 
Pot 10 was left at the same temperature (15° to 20°) for 24 days and pot 
II was removed to a temperature of 25° to 28° for the same period. 
About 5 per cent of the plants in pot 10 showed infection of the first leaf 
as compared with 28 per cent in pot 11. The reason for the increased 
percentage of infection at the higher temperature has not been satisfac- 
torily explained. It may simply have been consequent upon the stimu- 
lated growth of the onion foliage. However this may be, it is evident 
that the basal portion of the first leaf remains susceptible to infection 
for a short time, at least, after it emerges from the cotyledon. 

From a summary of these results it appears that our own experimental 
data regarding the duration of tlie period of infection agree in the main 
with those of previous investigators. The conclusion seems justified 
that disease resistance is correlated with tissue maturity, and that the 
onion cotyledon becomes immune to smut infection at about the stage 
when growth ceases. The rate and nature of growth of the cotyledon 
will naturally vary with environmental conditions; hence variation in 
the actual length of the infection period is to be expected. The mature 



240 



Journal of Agricultural Research voi.xxii.no.s 



basal sheath of the cotyledon thus protects the embryonic region of the 
younger leaves from infection. That portion of the first leaf which 
emerges from the cotyledon is susceptible to infection for some little 
time after emergence, but since it ordinarily is not actually in contact 
with infested soil, this fact is probably not of practical significance. 



Table II. 



-Relation of stage of development of the option seedling to infection by Uro- 
cystis cepulae 













Inter- 










Inter- 






val be- 






Pot 

No. 


Method of exposure to 
inoculation. 


val be- 
tween 
sowing 
and in- 
ocula- 
tion. 


Length of 
cotyledon 
at time of 
inocula- 
tion. 


Condition of 
first leaf at 
time of inocu- 
lation. 


tween 
inocu- 
lation 
and 
final 
exami- 
nation. 


Total 
number 

of 
plants. 


Per- 
cent- 
age 
smut- 
ted. 






Days. 


Inches. 




Days. 






1-3 

4 


Uninoculated 










lOO-f 
24 




Infected soil around base 


13 


I 


Not out 


27 


75 




of cotyledon. 














5 
6 


Do 


16 


2 to 2X 

2 to 2X 
2 to 2% 


do 


24 
24 
24 
24 


34 
41 
19 
36 


35 
8 


Do 


19 

24 
31 


... do 


7 
8 


Do 


Just out. .. . 
K to iK 
inches 




Do 


















above axis. 








9 


Cotyledons removed and 
infected soil placed 
around base of first leaf. 


31 


2 to 2^ 


. ..do 


24 


15 


60 


10 


One-inch layer of infected 
soil placed on surface of 
old soil so as to cover 
lower inch of aerial por- 
tions of plants. 


40 


2 to 2K 


yi to 4 inch- 
es above 
axis. 


24 


36 


6 


II 


Same as No. 10, except that 
pot was transferred to a 
temperature of 25° to 
28° C. 


40 


2 to 2>2 


...do 


24 


46 


28 



RELATION OF SOIL MOISTURE TO INFECTION 

Since there are these well-defined limits to the time of smut infection, 
the possibility becomes clearly evident that variable environmental 
factors during this limited period may exercise a controlling influence on 
the occurrence of the disease. As already stated, the problem seems to 
resolve itself primarily into the question of the relations of soil tempera- 
ture and soil moisture to infection. The results of several workers 
upon the grain smuts, as summarized by Jones (5), have shown that soil 
temperature may influence infection. Hungerford and Wade (4) have 
published evidence that high soil moisture, also, may favor infection of 
wheat by the smut fimgus Tilleiia tritici (Beij.) Wint. Variations in 
the moisture content of the surface layers of soil are likely to be wide, 
especially during the late summer planting season in the southern States 
when high temperatures and low humidity may cause rapid desiccation. 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 241 

An experiment was therefore carried out in which onion seedlings 
were grown in smut-infested soil in pots which were held at different 
degrees of soil moisture. Galvanized iron pots 5 inches in diameter and 
4 inches deep were used for these trials. Greenhouse sandy loam soil 
was used and its water-holding capacity was determined in advance by 
the two standard methods recommended by soil physicists ^ — that is, 
by means of the lo-inch cylinder and the i-cm. cup. The soils were 
brought to the desired low and medium water contents before they were 
placed in the culture pots ; and in those cases where the desired moisture 
approached the water-retaining capacity, the water content was finally 
adjusted after the soil was potted. Although these methods failed to 
secure exact uniformity in the physical compactness of the soils in the 
several series, they were considered satisfactory as to initial moisture 
conditions. The pots were weighed daily during the progress of the 
experiment, and water was added to replace the losses. Since it was 
realized that the surface layer of soil would change in moisture content 
through evaporation more rapidly than the lower layers, an effort was 
made to reduce this surface evaporation so far as practicable. To this 
end, tar paper covers were used until the seedlings came above ground, 
when glass covers were substituted for a few days, and finally mineral 
wool was packed on the surface about the seedlings to reduce evaporation. 
Absolutely uniform moisture throughout the pot could not be maintained 
even by this method, and the upper layers of soil unavoidably assumed 
a somewhat lower water content than the average for the pot. There- 
fore, at the end of the experiment, moisture determinations were made 
of the upper inch of soil, since this was the important part from the 
standpoint of smut infection. 

The soil was inoculated at the outset by the introduction of spores 
from diseased leaves and scales. The data from this experiment are 
given in Table III. Good germination took place within the range of 
10 to 15 per cent moisture content (45 to 70 per cent of the moisture- 
holding capacity). A high percentage of infection also occurred within 
this range. Above 15 per cent there was some decrease in germination 
together with a gradual reduction in infection. At the extreme, how- 
ever, where germination of seed was practically eliminated (pot i), one 
of the two plants surviving became infected. It is evident from these 
data that a good percentage of infection may be expected at a soil mois- 
ture content up to the limit where good germination and growth of the 
host plant occur. There was a gradual reduction of infection below 5 
per cent (23 per cent of the moisture-holding capacity), but this was not 
sufficient to insure good germination and support good growth. It 
may be concluded, therefore, that soil moisture does not function as a 
factor limiting infection with onion smut within the limits at either 
extreme where good germination and growth, of the host occur. 

1 We are especially indebted to Prof. E. Truog, of the Department of Soils of the University of Wiscon- 
sin for advice in connection with this work. 



242 



journal of Agricultural Research voi. xxii, ^fo. s 



Table III. — Relation of soil -moisture to infection by Urocystis cepulae 



Pot No. 



2. 
3- 
4- 
S- 
6. 

7- 
8. 

9- 
10 
II 
12 

13 
14 



Original 
moisture 
content 

(percentage 
of dry 

weight)." 



Moisture 
content of 
surface lay- 
er at end of 
experiment. 



19.6 
17. 1 
16.3 
18.2 
20.0 
15.0 
15.0 
10. o 
10. o 
10.7 

5-0 
8. I 

7-5 
6.6 



27. 1 
24.7 
23-7 
23-4 
14.0 
12.4 
II. 6 

9-3 
9.0 
2.4 

2-5 

1.4 
3-4 
1-3 



Number of 

seeds 

planted. 



100 
100 
100 
100 
100 
100 
100 
100 
100 
100 
100 
100 
100 
100 



Total num- 
ber of 
plants. 



52 
48 
50 
60 
66 
66 

71 
66 

32 
66 

63 

37 



Percentage 
of plants 
smutted. 



50 

O 

42 

71 
90 

97 



85 
94 
59 
50 
19 
35 



« Water-holding capacity, as determined by lo-inch cylinders, was 22.3 per cent; as determined by the 
i-cm. cup, it was 27.8 per cent. 
The calculated wilting coefficient of the plants was 2.3 per cent. 

TEMPERATURE RELATIONS 

In calling attention to the importance of soil temperature as a factor 
in the development of certain plant diseases, Jones (5) points out that 
several investigators have stressed its bearing upon infection in the case 
of stinking smut of wheat, Tilletia triiici (Beij.) Wint., and of the oat 
smuts, Ustilago avenae (Pers.) Jens, and Ustilago Levis (K. and S.) Mag. 
Heald and Woolman (j) showed that the amount of infection with the 
stinking smut of wheat was reduced as the mean soil temperature rose 
above 65° F. (18.3° C.) or fell below 40° F. (4.4° C). Humphrey (4) 
states, for the same disease, that soil temperatues of 0° to 5° C. and 
above 22° C. are decidedly unfavorable to infection. 

In studying the relations of soil temperature to the developemnt of a 
parasitic disease, consideration must be given to the possible influence 
of such temperature upon the host and the parasite independently. This 
may enable one to analyze with more confidence the effects when host 
and parasite are subjected simultaneously to the experimental condition. 
This has been done as far as practicable in connection with the present 
work. 

Unfortunately, germination of the fungus spores under artificial con- 
ditions has been so scanty that the effect of temperature upon the fungus 
has been necessarily limited to inoculation experiments with infested 
soil. However, the disease is produced so readily and consistently by 
the latter method that a very accurate index to the development of the 
fungus can be secured by varying the condition of the infested soil dur- 
ing the germination and early growth of the onion seedling. 

The soil-temperature experiments were all carried out in the green- 
house at Madison, Wis., during the winter months. The apparatus in use 



Oct 29, 1921 Relation of Soil Temperature to Onion Smut 243 

in the Department of Plant Pathology, University of Wisconsin, for the 
control of soil temperatures has been described by Jones (5) . Briefly, it 
consists of a series of water baths held at constant or nearly constant 
temperatures in which the glass or galvanized-iron culture pots are 
inserted.! s ^^-f?-' .'i.suuu-.. 

For these experiments galvanized-iron cylindrical pots 5 inches in di- 
ameter and 8 inches in depth were used. In order to overcome the in- 
fluence of the air temperature upon the upper layer of soil, the surface 
of the latter was kept at >2 to i inch below the level of the water. Tar- 
paper covers were placed over the pots until the seedlings came above 
ground; these covers were then removed, and a layer of mineral wool 
was placed on the surface of the soil. By this procedure the tempera- 
ture of the upper inch of soil was kept reasonably close to that of the 
deeper portions — that is, approximately that of the water in the tank. 
In order to follow any minor variations, readings were taken three times 
daily from thermometers inserted i inch below the surface of the soil. 
At the beginning of the experiments the moisture content of the soil was 
adjusted to two- thirds of the water-holding capacity. The pots were 
thereafter weighed at intervals of one to three days, depending upon the 
rate of water loss, and the moisture content was readjusted accordingly, 
either by adding water directly to the surface or by introducing it at the 
bottom of the pot through a glass tube. Obviously this method did not 
secure uniform distribution of moisture throughout the pot, and una- 
voidably the content of the upper layer of soil was somewhat lower than 
the average for the whole pot. It is believed that this variation had little 
if any influence, however, since other experiments, described earlier in 
this paper, showed that infection is quite uniform over a much wider 
range of soil moisture than here occurred. The seed was planted at a 
depth of I inch. Since the chlorophyll in the tops in some cases ob- 
scured the smut lesions, the plants were placed in alcohol acidified with 
acetic acid until thoroughly bleached before final examination for the 
disease was made. 

EFFECT OF SOIIv TEMPERATURE UPON THE DEVEI.OPMENT OF THE HOST 

Experimental studies to determine the relation of soil temperature to 
the rate and character of seed germination and seedling development 
were carried on in conjunction with those relating to infection, of which 
the results will be presented in the next section. It will be simpler, 
however, to discuss these two aspects of the problem separately, taking 
up first the relations of temperature to host development. 

Experiment I. — Seven pots of sterilized greenhouse loam soil were 
uniformly planted with 50 Red Globe onion seeds in each pot. One pot 
was then held at each of the following temperatures: 10° to 14°, 16.5° 

1 since this description was published, numerous improvements have been made from time to time by 
members of the Department. As now in use these are termed the "Wisconsin soil temperature tanks." 



244 



Journal of Agricultural Research 



Vol. XXII, No. s 



to 1 8°, 19° to 22°, 24° to 26°, 27° to 29°, 30° to 31°, 35° C. The moisture 
content of the soil was held at two- thirds the water-holding capacity (22 
per cent of dry weight). The air temperature of the greenhouse was 
kept at about 15° with a rise to 20° during the middle of the day. These 
conditions as to soil moisture and air temperature were such as had 
proved favorable for both host and parasite development in the earlier 
trials. The first seedlings to appear above ground were those at 27° 
to 29°, those at 24° to 26° came up shortly afterward, then those at 
19° to 22°. Good growth took place at these three temperatures, but 
germination was very slow at lower temperatures. At the highest tem- 
perature, 35°, a few seeds germinated, but growth was very slight. The 
plants were all removed and the roots washed out on January 7, 
1920, 29 days after the seed was sown. The data given in Table IV 
summarize the condition of tlie plants at this date. It will be seen that 
at this early stage in the development of the plants there was a tendency 
for best root development at about 21° or below, while the best develop- 
ment of tops took place at this point or above. 

Table IV. — Development of onion seedlings in sterilized greenhotise soil held at 22 per 
cent oj the dry weight or two-thirds the moisture-holding capacity, and at different soil 
temperatures. Data on January 7, IQ20, 2Q days after sowing 



Soil temperature. 



"C. 
10 to 14 . . . 
16.5 to 18. . 
19 to 22 . . . 
24 to 26. . . 
27 to 29 . . . 
30 to 31. . . 



Num- 


Total 


Per- 


Per- 


Per- 
centage 
with 
first 
leaf. 


Average 


Average 


ber of 


num- 


centage 


centage 


dry weight 


dry weight 


seeds 


ber of 


with 2 


with 3 


of tops per 


of roots per 


planted. 


plants. 


roots. 


roots. 


plant. 


plant. 












Gm. 


Gm. 


SO 


41 


22 





17 


0.00224 


. 00046 


50 


30 


17 


3 


70 


.00293 


. 00023 


SO 


32 


Sb 


25 


100 


.00396 


. 00043 


SO 


28 


32 





82 


•00307 


.00028 


50 


14 


SO 





86 


.00285 


.00014 


50 


"8 


13 





63 


.00212 







Average 

total dry 

weight per 

plan I. . 



Gm. 

0.0027 
.0031 
.0044 
•0033 
.0030 
.002 1 



» The reduced stand at 27° to 29° and 30° to 31° C. was due to damping-off fungi. 

Experiment II. — ^The experiment was repeated, with some modifica- 
tions, starting April lo, 1920. The Red Globe and Yellow Bermuda 
varieties were used. Two pots of each variety were kept at each of the 
following soil temperatures: 14°, 20°, 25°, 28°, 30° C. The air temper- 
ature ran sHghtly higher (20° to 30°) during the middle of the day and 
dropped to about 15° for the most of the night. Both the rate and the 
percentage of germination were noted, and the data are recorded in 
Table V. In both varieties the most rapid germination took place at 25°, 
although the rate was only slightly less at 20°, 28°, and 30°. At 14° the 
seedlings were distinctly slower in starting ofif. The plants from one pot 
of each series were removed on the thirteenth day. The dry weights of 
the tops and roots as given in Table V were so small at this age that com- 
parison on this basis does not have any great value. The tendency for 
rapid development of tops as compared with roots at 20° or above is, 
however, shown very strikingly in Plate 25. The plants in the remaining 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 



245 



pots were removed on the thirtieth day. The relative dry weights of 
roots and tops then secured are shown in Table V and those of the Red 
Globe are graphed in figure i. 

While the temperature relations of the two varieties were alike in their 
main features, there was an interesting minor difference, possibly indica- 
tive of the better adaptation of the Red Globe for northern culture and of 
the Yellow Bermuda for southern. In both cases with these onions, as 
indeed holds generally in our experiments with other plants, the best 

.0020 



.008 



I 




.00^4 



\ 
I 

\.ooz 

\ 



.000 



^a^zr 



^r^o^^,^. 



'^''■?'^rM>s\ 



I 

I 

.00/0 "1 

I 

.000s \ 



Z'^" 20'' 2S'=' 23*^ 



30'' 



.pooo 



Fig. I. — Relative developments of dry weight in tops and in roots of Red Globe onion as shown at end of 
30 days' growth in a series of culture pots kept at the several soil temperatures indicated, with all other 
factors, including air temperature, alike for all. Note that the best root development occurs at the low 
temperatures (12° to 15° C.) whereas the tops are forced more strongly at higher temperatures (20° to 25°). 

root development occurs at relatively lower temperatures (12° to 20° C.) 
while best top development occurs at higher temperatures (20° to 25°). 
When the varieties are compared, it is seen that with the Globe both 
roots and tops grew relatively better at somewhat lower temperatures 
than did those of the Bermuda. Thus the data at 30 days show the 
maximum root development of the Globe at 14° while that of the Ber- 
muda was at 20° ; with tops the maximum was at 20° for the Globe and at 
25° for the Bermuda. 



246 Journal of Agricultural Research Voi.xxn. no. 5 














•ia . 




t \ 




■-H 




' ^i2 <a 




4J 

1 


rt 000 o.i:T3 3 > , 




■3^ 


0^ 1- V- ^- 25^ ^ >- s 




5 a 


•- S S S? S ?!•-"• " " ? S 

cS o*^^^^" ra T^ — •-t'O 

Ji-a '- -0 -d T) ^ ta S-d-a 9 . 
|8g8882 8^8S^s 




4) 
W 

2 





> 


"a 


<; 


2 


















<? 


2 MO 


00 o 


•5 00 " >^oo r 


0\ ^ m 


b 


.Ot^ vOCOCOO<N MOOr^ 

Soo ooooo 008 
;joo ooooo 000 

6 


6 




K,w • 


!> C 


h Tj- M M r^so vo :> t^ 






.O>0 COI^O^MO 0\00> 1 




s; -t r^ C NO vO t^ I 






V- M 


(3 8 c 


OOOOO 000 




►^ fc- i<-r 


OOOOO 000 1 




6 




... 


, ^ • 


r^ Tf ov u^nO 00 M (»^ »o 1 




n 2 


\0 nC 


■t <t ■* TfNO •* «5 -"t 1 




g w. a 










^ 01 d 
































■3 






















•a 
























"o-ti 


« 




















fl 


l« 




















M 41 


« 




















i« a 


13" 






















se. ° 


'0 




















M 


V 


ooooo 


000 1 


i 




•>.-o 


■0 -a -a /q -a 


t3-OT3 


a 


> 





















^ 


< 





















a 










2 


2 -So 


« t> 


t^ 0^ 


5^^ 






r*^ (N r^ ^ t 




•d 


S 88 
!0 8 8 






888 
000 




9 -M . 


-0 >/■ 


00 t^ 


OWN 




2-Sa 


^!88 


00 CO -0 rO « 

88888 


On r^oo 

888 




6 






1 >*-. m 


00 


<^ ■* I' 


Ov N 






fO r^ 


m-O 10 10 "J 


t fO ^ 




a >- g 










1^^ 










' J3 . 


<> fi 


Tf OiO CO 


10>0 M 




t- *j ► 


10 f^ 


t^ I^ 1/^ 


•0 »/io 


d 


2gS 











{h S-a 








•a 

CIS 

.g 










, 


t^ fO 


•^00 t^ M 


wio On 


i 




r^ 


t^M3 M NO 


-0 >« 10 


u 


Wfcx) 








"0 


J3 . 


"1 


e< \n rr, 


^ W500 



to 

C9 


.a& 





t^ NO 


NO lO 10 




^■o 
















1 




m 


rO D CO t^ 


** fO On 


^•s>' 




NO i/i nr to 


lo ■>}• ■* 




(^ i-i m 

wg-S 










>t5 


•a- 


»nco -Nt 


1/100 




^ " " 


N M W5 « M 


M M CO 




V CS 










•*-! 
















rt 






>. 






■o 










3 

a 




rt 






t-i 




►^^ 






01 




K" 




J2 


5 


PQ 


"3 








rt 








>< 






1 



Oct 29, 1921 Relation of Soil Temperature to Onion Smut 247 

EFFECT OF soil. TEMPERATURE UPON INFECTION 

At present the chief interest in these data focuses upon the question of 
any possible bearing of the rate of development of the host plant at 
different temperattires upon predisposition to, or escape from, smut 
infection, recalling that such infection is practically limited to the seed- 
Hng stage before the maturity of the cotyledon. It has just been noted 
that the promptest seed germination and most rapid growth of tops 
during this early seedling stage occur at fairly high temperature, 20° to 
25° C, with a rather pronounced drop in rate of aerial growth at tempera- 
tures below 20°. It is to be expected, therefore, that in the northern 
onion-growing sections where the seed is planted in early spring there 
will be a rather tardy germination and slow early development of tops, 
the growth energies during the seedling stage being directed under this 
climatic environment to a relatively stronger development of the root 
system. In the South where the seed is planted in the comparatively 
warm period of early autumn, we should expect a more rapid top growth 
at the outset, with correlated strength of root development coming later 
in the autumn as the soil becomes gradually cooler. 

The naturally infested soil was secured near Racine, Wis., from a badly 
diseased field of sandy loam rich in organic matter. The soil which was 
artifically inoculated consisted of a greenhouse mixture of loam and sand 
to which were added fresh spores from smutty onion leaves. In order to 
test the efficacy of this method of soil inoculation a preliminary planting 
of onion seed was made in advance of the final experiments. This gave 
a high percentage of smut infection, showing that the method of introduc- 
ing the inoculum was satisfactory. Several early trials indicated that 
below 25° C. soil temperature variations have little effect on the relative 
amount of infection. The results of two such experiments, nearly cover- 
ing the range of onion seed germination, are given in Table VI (experi- 
ments I and 2). It is evident from these figures that abundant infection 
occurred between 10° and 25°, both with naturally infested and with 
artificially inoculated soil. The number of pustules per plant as shown 
in Plate 25 proves that the fungus was very active even at low tempera- 
tures. Above 25° infection is reduced very rapidly, as indicated by 
both the percentage of infected plants and the number of pustules per 

plant. 

In order to determine more closely the point at which infection is 
inhibited four more serial experiments were conducted (experiments 3 
to 6) in which the temperature was kept as constant as possible at 2- 
degree intervals between 25° and 31° C. The results given in Tables 
VII and VIII, and illustrated in Plates 26 and 27, show that abundant 
infection took place at 25° to 26°, while it was greatly reduced at 27° to 
28° and completely inhibited at 29° or above. The infected seedlings 
from experiment 5 show the great reduction in the amount of disease per 



248 



Journal of Agricultural Research vo1.xxii,no.s 



plant at 27° to 28° as compared with 25° to 26°. It is interesting to note 
that infection was reduced more abruptly at 27° to 28° in the artificially- 
inoculated soil than in that naturally infested. This may be due to the 
age of the inoculum, a smaller percentage of the spores being functional 
in the former soil, or perhaps to the presence of a smaller amount of 
inoculum. 

Table VI. — Relation of soil temperature to infection of onion seedlings by Urocystis 

cepu lae 



Experiment i. 


Experiment 2. 


Naturally infested soil. Begim May 3, 1919; com- 
pleted May 30, 1919. Records not kept as to soil 
moisture nor as to air conditions. 


Artificially inoculated soil. Begim Dec. 10, 1919; 
completed Jan. 6, 1920. Soil moisture held at 
22 per cent or two-thirds the water-liolding capac- 
ity. Air temperature 13° to 23° C, relative hu- 
midity 45 to 75 per cent. 


Soil temperature. 


Total ntmi- 
ber of 
plants. 


Smutted 
plants. 


Soil temperature. 


Total num- 
ber of 
plants. 


Smutted 
plants. 


"C. 
10 to 13 


25 
5" 

47 
25 


Per cent. 

72 

80 

100 

15 



10 to 14 


64 
49 
63 
56 
52 
36 


Per cent. 
98 

98 


18 to 20 


16 5 to 18 


22 to 24 




25 to 30 


2 2 to 26 




28 to 34 


27 to 29 


93 
8 





20 to •?! 







" stand reduced by damping-off fungi. 

Table VII. — Relation of soil temperature to infection of onion seedlings by Urocysti^ 

cepulae 



Experiment 3. 



Naturally infested soil 
pleted Jan. 10, 1920 



Begun Dec. 20, 1919; com- 
Soil moisture held at 25 per 



cent or two-thirds the water-holding capacity. 
Air temperature, 13° to 2?° C: relative humiditv. 



Air temperature, 13 
4S to 75 per cent 



Soil temperature. 



"C 

19 to 22 

23 to 26 

26 to 27 

27 to 28 



Total num- 
ber of 
plants. 



44 

100 
86 



Smutted 
plants. 



Per cent. 

93 
96 

S7 
12 



Experiment 4. 



Artificially inoculated soil. Begun Dec. 18, 1919; 
completed Jan. 12, 1920. Soil moisture held at 13 
per cent or two-thirds the water-holding capacity. 
Air temperature, 13° to 25° C; relative humidity, 
AS to 75 per cent. 



Soil temperature. 



C. 



23 to 26 
26 to 28 
29 to3i , 



Total num- 
ber of 
plants. 



50 
40 



Smutted 
plants. 



Per cent. 



98 

12 
O 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 


249 


Table VIII. — Relation of soil temperature to infection of onion seedlings by Urocystis 

cepulae 


Experiment 5. 


Experiment 6. 


Naturally infested soil. Begrun Jan. 16, 1920; com- 
pleted Feb. 10, 1920. Soil moisture held at 25 per 
cent or two-thirds the water-holding capacity. Air 
temperature, 13° to 25° C; relative humidity, 40 
to 80 per cent.. 


Artificially inoculated soil. Begun Jan. 16, 1920; 
completed Feb. 10, 1920. Soil moisture held at 
15 per cent or two-thirds the water-holding capac- 
ity. Air temperature, ij° to 25° C; relative hu- 
midity, 40 to 80 per cent. 


Soil temperatiu-e. 


Total num- 
ber of 
plants. 


Smutted 
plants. 


Soil temperature. 


Total num- 
ber of 
plants. 


Smutted 
plants. 


T. 
25 to 26 


89 
103 

77 
43 


Per cent. 
100 

47 




°C. 
25 to 26 


106 
98 
47 
30 


Per cent. 


27 to 28 


27 to 28 


29to3o.s 


2Q to ^0. < 





30 to •?2 












It may be concluded from the foregoing experiments that a high per- 
centage of infection may be expected up to 25° C, above which there is 
a rather abrupt reduction, leading to complete inhibition at 29°. There 
appears to be no lower limit of temperature for infection within the range 
where onion seeds will germinate and normal growth occur. 

After it was clearly shown that no infection v/ould take place at 29° C, 
the question arose as to how long seedlings must grow at this temperature 
to become entirely immune. It has been shown that at moderate tem- 
peratures the plant becomes immune in about 20 days, or at about the 
time when the cotyledon has reached its maximum growth. To deter- 
mine whether or not this condition is altered when the plants are grown 
at 29°., several pots each of the naturally and the artificially inoculated 
soil were started off at this high soil temperature. Pots were then 
transferred from time to time to a lower temperature favorable for 
infection (15° to 20°) where they were held for about three weeks before 
they were examined for signs of the disease. The results of these experi- 
ments are summarized in Table IX. 

It is quite evident that the amount of infection was markedly reduced 
by an exposure of 15 to 18 days at 29° C. Complete inhibition of infection 
by even more protracted exposure to this high temperature was not 
attained. However, where infection did occur there was usually not 
more than one lesion per plant, which in the majority of cases was so 
situated that subsequent infection of newly forming leaves would be 
impossible. It is thus quite certain that where seedlings develop at 
about 29° for the first 20 days the amount of damage from smut will 
be negligible, especially in an area where the amount of inoculum is 
slight. 



250 



Journal of Agricultural Research voi.xxn.Ncs 



Table IX. — Effect oj different soil temperatures upon onion smut infection. In all 
cases except the fifth, tenth, and eleventh, the pots -were held for the stated period at 20° 
C, where infection was inhibited, then transferred to 15° to 20°, a temperature favor- 
able for infection. In the fifth and tenth, where the continuous temperature was high, 
note that practically no smut developed; in the eleventh, where the soil temperature was 
continuously low, note that practically all the plants were smutted; in the remaining 
series, where the plants were transferred from the higher temperature (29°) to the lower 
(75° to 20°) after varying periods, note that long exposures at the higher temperature 
tended to reduce the amount of infection. 











Extent of infection at end of 












experunent. 




Type of soil inocu- 
lation. 


Length 
of ex- 
posure 

t0 29''C 


Size of plants at time 
of removal to low 
temperature. 






Pot 

No. 


Total 

number 

of 


Per- 
centage 
dis- 


Extent of infection. 










plants. 


eased. 








Days. 










I 


8 


II 


Cotyledons i inch 
long. 


16 


94 


63 per cent dis- 
eased first leaf. 


2 




15 


Cotyledons 2 to 
2>^ inches long. 


17 


41 


12 per cent dis- 
eased first leaf. 


3 


) ^T3 


18 


do 


12 


25 


25 per cent dis- 
eased first leaf. 




.2^ 








4 


t-fl "^ 


28 


First leaves i-f 
inches high. 


9 


II 


II per cent dis- 
eased first leaf. 


5 




, 35 


Continuous expo- 
sure at 29°. 


6 







6 


1 


12 


Cotyledons iK to 
2 inches long. 


29 


48 


14 per cent dis- 
eased first leaf. 


7 


1 6 Coty le dons 2 ^to 


40 


13 


3 per cent dis- 




Si 


3K inches long. 






eased first leaf. 


8 


1 


23 


First leaves out 
in about one- 


24 


8 


8 per cent dis- 
eased first leaf. 






half plants. 








9 




27 


First leaves out 
in most plants. 


16 


13 


6 per cent dis- 
eased first leaf. 


10 


38 


Continuous at 


34 


3 


per cent dis- 




ai 




29°. 






eased first leaf. 


II 





Continuous at 


76 


99 


Most of these 








15° to 20°. 






plants died in 
cotyledon stage. 



It is interesting to note also in this connection that continued expo- 
sure of onion roots to a temperature of 29° C. led to the gradual slowing 
up of groM^th. With the transfer of the pots to the lower temperatui-e 
(15° to 20°) both root and top development were greatly stimulated. 
In attempting to correlate these results one must keep in mind the 
fact that in nattire the temperature conditions under which the onions 
develop are much different from those in the experimental pots. 
Whereas in the pots the soil temperature is uniform throughout their 
depth, in the natural soil there is a gradual decrease in temperature 
at progressively greater depths. The temperature of the upper inch runs 
extremely high during the summer months because of direct exposure 
to the sun's rays, and this is the area critical for infection by onion smut. 
The young roots, on the other hand, as they develop progressively 
reach strata of lower temperature, which are more favorable for their 
growth. 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 251 

EFFECT OF HIGH AIR TEMPERATURE UPON THE DEVEI^OPMENT OF SMUT 

The experiments reported above in which the soil temperature was 
varied were carried on at an air temperature of 15° to 20° C. The latter 
is considerably lower than the air temperature which prevails in southern 
onion sections at the time when young seedlings are starting off. This 
is shown for one section (Laredo, Tex.) in Table XIII, where the mean 
air temperature is about 30° during most of the onion-planting time. 
The question arose as to what efifect these high air temperatures might 
have upon the development of smut in the aerial portions of the plant. 
Five clay pots of onions in naturally infested soil were started off in a 
greenhouse running at about 25°. Previous observations had shown 
that the pustules become evident in the cotyledons on about the twefth 
day under these conditions. Accordingly the plants were allowed to 
grow at this temperature for nine days, at which time a few lesions were 
barely visible. In order to prove that good infection had aheady taken 
place, 10 plants were removed from infested soil and washed thoroughly 
in running water to remove any external inoculum, after which they were 
transplanted to clean soil. Within two days lesions were distinctly 
visible in these plants, and smut developed in 8 out of the 10. 

As a control on this method of removing the external inoculum 11 
plants grown in clean soil were moistened and covered thoroughly with 
infested soil. They were then washed in running water and trans- 
planted to clean soil. No smut developed. On the ninth day after sow- 
ing, 4 of the 5 pots were removed to a greenhouse running at 30° to 33° C, 
one being allowed to remain at 25°. One pot was then transferred from 
the higher temperature back to 25° at the end of each the second, fourth, 
ninth, and fourteenth days. This exposure to the higher temperature 
resulted in a stimulation of host plant growth for a few days. When 
the plants were allowed to remain at this high temperature, however, 
for three weeks distinct stunting became evident, while more prolonged 
exposure resulted in death. The plants so transferred were allowed to 
continue growth at 25° for three weeks or more, when they were examined 
for the presence of smut. The final results are given in Table X. It 
was evident that the gradual elimination of smut which took place was 
proportional to the length of exposure to the higher temperature (29°). 
After 14 days of exposure only small lesions developed on 16 per cent 
of the onion plants, although presumably 80 per cent or more of these 
plants were originally infected while they were growing at the lower 
temperature. This experiment was repeated several times with prac- 
tically the same results, namely, that exposure of plants bearing in- 
cipient infections to a temperature of 30° to 33° for 12 to 15 days almost 
entirely checked further development of the parasite. 
65768°— 21 2 



252 



Journal of Agricultural Research voi. xxii. No. 5 



Table X. — Effect of high temperature, following infection, in inhibiting the further 
development of smut. The plants were from a series of pot cultures started at 25° C. 
and held there until incipient infection had occurred, then transferred for the period 
indicated to a high temperature, J0° to 33°, and finally brought back to the original 25°. 



Length of exposure to temperature of 30° to 33°. 



None (25° throughout) 

2 days 

4 days 

9 days 

14 days 




Percentage 
smutted. 



94 
45 
37 
33 
17 



It is to be noted that in the experiments just reviewed potted plants 
were used. The entire pot, thus including roots and tops of the experi- 
mental plants, was exposed to the stated temperature condition. 

The question then arose whether the results secured were due entirely 
to the effect of high air temperature upon the fungus or to an indirect 
effect of the changed conditions upon the metabolism of the host. In 
order to throw some light upon this point two experiments were con- 
ducted in which seedlings were grown in infected soil at three constant 
soil temperatures and each of two air temperatures, 25° and 30° to 33° C. 
The results (Table XI) at the lower air temperature (25°) coincided 
closely with those previously secured at air temperatures of 13° to 25°, 
inasmuch as abundant infection occmred when the soil temperature was 
25°, while complete inhibition was attained at 30°. It is, however, 
significant to note, in comparison with the results in Table X, that, with 
the soil temperature held at 20° or 24°, the 30° to 33° air temperatm-e 
did not greatly check the development of the disease. It appears, then, 
that roots as well as tops must be exposed to the inhibitive higher tem- 
perature, 30° to 33°, in order fully to check the parasite after incipient 
infection has taken place. This suggests that the inhibitory effect may 
be due in part at least to the influence of the environmental conditions 
upon the metabolism of the host and not entirely to a direct effect upon 
the fungus itself. 

Table XI. — Effect of different combinations of soil and air temperature upon onion smut 

infection 



Air temperature. 



2S 

30 to 33 . 



Soil 
tempera- 
ture. 



20 
25 
30 
20 
24 
30 



Experiment i. 



Number 
of plants. 



23 
68 

42 

Op 

73 
24 



Percentage 
smutted. 



100 
97 



86 
o 



Experiment 2. 



Number 
of plants. 



8S 
49 
22 

41 
.■50 
30 



Percentage 
smutted. 



95 
92 

o 
46 
60 

o 



1 stand reduced by damping-o£f fimgi. 



Oct. 29, 1931 Relation of Soil Temperature to Onion Smut 253 

^FFBCT OF MODERATEI/Y HIGH TEMPERATURES UPON SYSTEMIC INVASION 

OF THE PI^ANT 

Thaxter (jo, p. 134) observed that in some instances the smut fungus 
may infect and develop in the cotyledon without invading the first leaf, 
with the result that the plant eventually outgrows the disease. Obser- 
vations lead us to believe that this may vary with different tempera- 
tures. It has been pointed out that at a temperature of about 25° C. 
the most rapid top growth of the onion seedling occurs, while at tem- 
peratures below 20° the top growth is much retarded. Two pots of 
infested soil were sown with onion seed and placed in greenhouses, 
one at 24° to 28°, with a maximum of about 36°, for one or two hours 
on sunny days, the other at 15° to 20°. A high percentage of cotyledon 
infection occurred in both pots. After 31 days 24 out of 29 plants at 
the high temperature were infected, but the pustules were all confined 
to the cotyledons and no infection of first leaves had developed, although 
the plants were now in the second leaf stage. At the low temperature, 
on the other hand, of approximately the same number of plants, only 
9 had survived on the thirty-seventh day, and 8 of these showed infec- 
tion in the second leaves. It appears, then, that rapid growth of tops 
at about 25° may result in a large percentage of plants outgrowing the 
disease after the cotyledons become infected. The results of successive 
field plantings, discussed in the next paragraph, seem to confirm this 
judgment. The importance of the practical bearings of this matter are 
such as to justify further critical attention. 

EFFECT OF SUCCESSIVE PLANTINGS THROUGHOUT THE GROWING SEASON 

UPON INFECTION 

The laboratory experiments described early in this paper have shown 
that onion smut infection is greatly reduced where a constant soil tem- 
perature of 27.5° C. is maintained during the susceptible period of the 
plant's growth, while a temperature of 29° thus applied completely 
inhibits infection. Moreover, as explained in the last paragraph, when 
plants are growing in infested soil with temperature of air and soil 
held at about 25°, although a high percentage of cotyledon infection 
may occur, there is a greater tendency than at lower temperatures for 
the plants to outgrow the disease, owing to the rapid growth of tops. 
These results combined to justify the expectation that successive field 
plantings of onion seed throughout the growing season might show 
considerable variations in the percentage of smut infection. In the 
onion field the soil temperature usually varies widely during 24 hours, 
often reaching a maximum considerably above 29° during the day and 
descending to a minimum much below this at night. Under Wisconsin 
conditions the daily mean temperature gradually rises during the spring 
and early summer months and falls during the latter part of the growing 



254 



Journal of Agricultural Research 



Vol. XXII, No. s 



season. It seemed possible, therefore, that by making successive plant- 
ings a period might be found for this latitude when the mean soil tem- 
perature is sufficiently high to materially check or completely inhibit 
onion smut infection. 

In order to test this out, a series of plantings at intervals of from 8 to 
14 days was begun on June 18, 1920, at Madison, Wis. Onion seed 
was sown in smut-free soil in an open trench about i inch deep and was 
then covered with naturally infested soil similar to that used in certain 
of the laboratory experiments. Two varieties. Red Globe and Yellow 
Bermuda, were used, one lo-foot row of each variety being put in at 




S/ 5 

Fig. 2. — Graph showing the daily mean soil temperature at a depth of i to 2 indies as it occurred in the 
"successive planting" plots. Since the weather continued rather cool, one bed was covered with glass 
to insure a higher temperature. The temperature of the uncovered bed is shown by the solid line, 
the temperature of the glass-covered bed by the broken line. For further details see Table XII and 
and accompanying text. 

each planting. Temperatures of the soil at a depth of i to 2 inches were 
obtained by means of a self-recording thermograph. The hourly mean 
temperature for each day was then secured by adding temperatures as 
recorded for each hour and dividing the sum by 24. These computa- 
tions are represented graphically in figure 2. Since the weather in 
July was unusually cool, a higher mean soil temperature was secured for 
some of the plots by covering them with an ordinary glass cold frame. 
Inasmuch as the dry weather and high temperature would cause a rapid 
desiccation of the surface layer of soil, the plots were watered thoroughly 
on alternate days or oftener during the early growth of the plants. The 



Oct. 29. 1921 Relation of Soil Temperature to Onion Smut 



255 



data collected from this field plot are tabulated in Table XII. Obser- 
vations were made by pulling plants at several points in each plot, and 
examining for smut lesions after the chlorophyll had been removed by 
means of alcohol and acetic acid. The first observation was made on 
the twenty-first to the twenty-third day after planting. Subsequent 
observations were made as indicated in the table. 

TabIvE XII. — Development of onion smut in successive plantings in the field at Madison, 

Wis., ig20 



Date of 


Variety. 


Treatment. 


First observa- 
tion, 21 to 23 
days after 
planting. 


Second observa- 
tion, 29 to 31 
days after 
planting. 


Third observation, 44 
to 65 days after planting. 


planting, 
1920. 


Ntim- 
ber of 
plants 
exam- 
ined. 


Per- 
cent- 
age 
smut- 
ted. 


Num- 
ber of 
plants 
exam- 
ined. 


Per- 
cent- 
age 
smut- 
ted. 


Time 

after 
plant- 
ing 
(days). 


Num- 
ber of 
plants 
exam- 
ined. 


Per- 
cent- 
age 
smut- 
ted.ff 


June 18 


Red Globe... 
[Red Globe. . . 


Uncovered 
. ..do 


108 
62 
45 

41 
20 

52 

40 

35 

46 
26 


S3 
73 
49 

10 
10 


13 
14 





& SO 

42 

47 
28 

2>7> 


C40 
«86 

/;3 




52 
44 


31 
SI 


3 
39 


June 26 


Yellow Ber- 
muda. 
[Red Globe . . . 


...do 

. . .do 




65 
65 

65 
53 
53 

53 
53 


61 
24 

9 

135 

39 

91 

45 




July 10 


Yellow Ber- 
muda. 

Red Globe . . . 

Red Globe... 

Yellow Ber- 
muda. 

Red Globe... 

Yellow Ber- 
muda. 


...do 

Covered . . 
Uncovered 
...do 

Covered « . 
Covered a . 





II 








13 

9 

24 


July 19 



















" Covered for 15 days only. 
•6 Observation 39 days after planting. 

«^ Extent of infection: Systemic, 26 per cent; confined to dead cotyledon, 14 per cent. 
<* Extent of infection: Systemic, 10 per cent; confined to dead cotyledon, 66 per cent. 
« Extent of infection: Systemic, 5 per cent; confined to dead cotyledon, 81 per cent. 
/ Extent of infection: All cotyledon infections. 
Extent of infection: All systemic infections. 

An analysis of the data secured can be made by referring to Table XII 
and figure 2. It will be seen that the soil temperature mean gradually 
rose until July 23 to 29, after which there was a gradual drop. At no 
time did the mean in the uncov red plot reach the inhibiting temperature 
(29° C), but it closely approached this point during the warmest portion 
of the season. In the covered plot, however, the mean remained above 
29° continuously until the cover was removed on August 3. The two 
important points to be considered in the respective plantings were (i) 
the amount of original infection which was determined three to four 
weeks after planting (see first and second observations in Table XII) 
and (2) the extent to which the disease either became systemic or was 
entirely outgrown by the plants during the following four or five weeks 
(see third observation in Table XII). 



256 Journal of Agricultural Research voi. xxii. no. s 

Considering first the amount of original infection, it will be seen that a 
high percentage of disease resulted in all the plantings of June 18 and 
June 26. The somewhat lower infection in that of June 18 may be ex- 
plained in part at least by the fact that a smaller quantity of inoculum 
was used than in subsequent plantings. The next two plantings (July 
10 and 19) were so made that the resulting seedlings were exposed during 
early growth to the maximum soil temperature of the season. By refer- 
ring to Table XII it will be seen that associated with this higher tem- 
perature there was a decided reduction in the amount of infection in 
even the uncovered plots, while in the covered plots, where the mean 
temperature remained continuously above 29° C, no infection whatever 
occurred. 

Considering, secondly, the extent to which the disease became sys- 
temic or was outgrown, it will be seen that in the planting of June 18 a 
majority of the infected plants showed systemic invasion at the second 
observation (thirty-ninth day). In the next planting (June 26), how- 
ever, by the time of the second observation most of the external signs of 
the disease were confined to the dead cotyledons. The amount of 
systemic infection increased somewhat, however, at the third observation 
(39 per cent). 

In the third planting (July 10) it is interesting to note first that the 
plants in the covered plot remained entirely free from infection. In the 
uncovered plot, although some cotyledon infection was noted at the sec- 
ond observation, no disease whatever was found at the third observa- 
tion. This indicates that the time when the temperature was at its highest 
point the infected plants succeeded best in outgrowing the disease. 

The field data secured in the foregoing experiments at Madison are 
thus in general accord with the experiments performed under controlled 
conditions. In such controlled experiments the amount of smut infec- 
tion falls as the soil temperature rises toward 29° C. and is totally inhib- 
ited above this temperature. Likewise in the field trials with successive 
plantings there was a gradual reduction in the amount of infection fol- 
lowing the rise in the mean soil temperature, with omplete inhibition of 
infection where the mean was kept above 29° for two or three weeks after 
planting. Complete freedom from infection under these Wisconsin field 
conditions was secured only by growing the plants under artificial condi- 
tions in which by covering the plants with glass the temperature was 
raised several degrees above the normal. It is, however, to be noted that 
the summer of 1920, when the foregoing results were secured, was some- 
what cooler than the average. The weather records of other years indi- 
cate that in a hot summer complete inhibition of smut infection would 
be secured by such summer plantings. 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 



257 



CORRELATION OF EXPERIMENTAL RESULTS WITH FIELD CONDI- 
TIONS OF THE SOUTHERN STATES 

These results obtained in both greenhouse and field experiments justify 
the question as to the part played by soil temperature in determining 
smut infection in the onion fields of the more southern States. As 
noted at the beginning of this article, a recent survey of southern onion 
sections indicates that smut is not prevalent in the southern fields — for 
example, in Texas — as it is in the northern onion sections. As was 
earlier explained, it is the practice in these southern fields to plant the 
onion seed in late summer or early autumn. It is thus quite possible 
that the mean temperature for the surface inch of soil in southern onion 
sections is considerably above the maximum for onion smut infection 
during and immediately following the sowing of seed. According to 
Mally (6), onion seed is sown in the Laredo district of southern Texas 
as early as August i , while most of the seed is planted about September 
10 to 25. The mean air temperature as recorded at Laredo, Tex., by 
the United States Weather Bureau for August, September, and October, 
1917, is given in Table XIIL 

Table XIII. — Mean air temperatures for August, September, and October, iQiy, at 

Laredo, Tex.''' 



Day of 
month. 


August. 


September. 


October. 


Day of 
month. 


August. 


September. 


October- 




"F. 


" F. 


" F. 




°F. 


"F. 


°F. 


I 


89-5 


85.5 


71.0 


17 


90. 


79.0 


80.5 


2 


88.5 


86.0 


80.0 


18 


90-5 


83.0 


81. s 


3 


88. s 


86.5 


80.5 


19 


91. 


7!?-5 


76. 


4 


88.5 


87.0 


80.5 


20 


92. 


79-5 


67..-; 


S 


89.0 


88.0 


80. s 


21 


90. 


79-5 


62. 


6 


88. <; 


87-5 


78.0 


22 


90. 


80.0 


63-5 


7 


89. i 


88.0 


81.0 


23 


90. 


79-5 


68. s 


8 


88.5 


88.5 


82.5 


24 


91. 


79.0 


61. 


9 


89.0 


89-.=; 


64-5 


25 


89-5 


80. ■; 


66.5 


10 


90-5 


88.0 


64-5 


36 


87.0 


81. 5 


76.5 


II 


90- .■; 


83-5 


70-5 


27 


89-5 


81.5 


74-5 


12 


88.0 


85.0 


73-.'; 


28 


86.5 


71- S 


78. s 


13 


89-5 


86. =; 


7S-0 


29 


89.5 


71- 5 


75- 


14 


88.5 


88.5 


77.0 


30 


84.5 


71- S 


SO-."; 


I c 


90- S 

88.5 


83-5 
83-5 


76. 

80.0 


31 


86. 







16 









° Obtained by averaging the daily maximum and minimum temperatures. 

Table XIII shows that the air temperature ranged very high during 
August and September, the onion-planting period. In this connection 
it is to be noted, moreover, that the records of Bouyoucos (i) in Michigan 
indicate that surface soil temperatures may considerably exceed air 
temperatures. Thus, his observations showed that the maximum tem- 
perature for the upper quarter inch of all the soils he studied was about 
16° C. higher during hot, clear days than that of the air at an elevation 



258 Journal of Agricultural Research Voi. xxu. No. s 

of 4 feet, while the minimum temperatm-e of all the soils used, except 
peat, was 0.5° to 1.0° C. higher, as a monthly average, than that of the 
air. Our own observations in Wisconsin are in general accord with 
these Michigan records. Assuming that the temperature of the surface 
layer of soil under Texas conditions likewise averages several degrees 
higher than the air, it is evident that the mean never went below the 
point where infection is entirely inhibited (29° C, or 84° F.) during 
August and seldom below it during September. Continuing up to 
October 8 there were only a few days when the air temperature fell 
below 27.5° C. (81.5° F.), the point at which our experiments have shown 
smut infection to be markedly reduced. It seems probable, therefore, 
that even if onion smut were introduced into this Laredo soil, it would stand 
small chance of infecting onion seedlings to the extent of establishing 
the disease as a permanent factor. The data available are not sufficient 
to justify the attempt at more detailed geographic correlation of onion 
smut occurrence with the temperature factor. We believe, however, 
that the conclusion is justified that soil temperature during the early 
seedling stage must be considered as a limiting factor in determining the 
occurrence of the disease in any locality. It must be left with local 
observers to make use of this fact in interpreting conditions as they 
occur in any particular region. 

SUMMARY 

Onion smut was first noted in the Connecticut River Valley in 1869. 
Since then it has successively appeared and become an economic factor 
in nearly all of the northern onion-growing sections from New York to 
Oregon. This has probably resulted from chance introduction of the 
organism with seed or bottom sets, followed by its accumulation in the 
soil where continuous cropping with onions is practiced. The disease 
has not appeared in the southern onion-growing sections of Texas 
and Louisiana, although they are exposed to similar chance introduction 
of the parasite and the continuous cropping method is common. 

These facts have raised the question as to wherein lies the explanation 
of the regional limitation of the disease. The southern method of cul- 
ture, characterized by special seed bed and transplantation of seedlings, 
does not offer sufficient explanation for the absence of smut. No differ- 
ence in susceptibility between northern and southern varieties has been 
found. Is regional limitation explained by differences in environ- 
mental factors in the North and the South at the time when the seedling 
is susceptible to infection, that is, during the first two or three weeks after 
germination? An analysis of certain of these factors in relation to 
infection has been the object of the present investigation. 

The cotyledon of the onion is susceptible to attack by the smut organ- 
ism up to the time it attains full growth, a period of about three weeks. 



Oct. 29, 1921 Relation of Soil Temperature to Onion Smut 259 

varying somewhat with environment. Cotyledons remaining free from 
infection during this period become resistant and serve as a barrier to 
subsequent invasion of the embryonic region of the true leaves. Conse- 
quently, if infection is prevented by environing conditions during this 
period of susceptibility, the plant will remain free during the remainder 
of its growth. 

Experiments were conducted in which seedlings were grown on smut- 
infested soil held at various soil moisture contents. A high percentage 
of infected plants resulted over the entire range in which good germina- 
tion and growth of the host occurred. At either extreme, very high or 
very low moisture, there was some reduction in amount of infection, but 
with it occurred a corresponding decrease in seed germination and rate 
of growth of the plants. Soil moisture, therefore, does not appear as a 
serious limiting factor in onion smut infection. .^rj v/ot; 

The relation of soil temperature to the development of the host and 
the parasite was studied by growing plants in pots held experimentally at 
a series of constant soil temperatures in the special apparatus known as 
the "Wisconsin soil temperature tank." 

Seed germination and growth took place over a range of soil tempera- 
ture from 10° to 31° C. Most rapid seed germination and development 
of tops occurred at soil temperatures of 20° to 25°, while as a rule the 
best development of roots occurred below 20°. 

A high percentage of plants grown on smutted soil were infected at 
soil temperatures ranging from 10° to 25° C. A decided reduction in 
infection was noted at about 27°, and complete freedom from the disease 
resulted at 29°. In these experiments all plants were under uniform 
conditions of air temperature, which ranged from 15° to 20°. 

The relation of variations in air temperature to the development of 
the disease was then studied. 

Exposure of plants bearing incipient infections of the fungus in the 
aerial parts to an air and soil temperature of 30° to 33° C. so disturbed 
the relations between parasite and host as to preclude any further develop- 
ment of the disease. This was shown by growing plants at a tempera- 
ture favorable for infection (15° to 20°). Then, just as the pustules of 
the disease were beginning to appear (tenth to twelfth day), the plants 
were removed to a room held at 30° to 33°. This stimulated top growth 
for a few days, which was followed by a decided checking of the plants 
and death after three or four weeks. However, if after 12 to 15 days 
at the high temperature the plants were returned to the original tem- 
perature (15° to 20°), they grew normally, but the fungus in nearly all 
cases failed to produce spores, and the plants remained free from further 
invasion. 

Experiments were then performed in which seedlings were grown on 
infested soil held at 20°, 25°, and 30° C. with a uniform air temperature 
of 30° to 33°. A high percentage of infection resulted at soil temperatures 



26o Journal of Agricultural Research voi. xxii, no. $ 

of 20° and 25°, but none at 30°, showing that high air temperature 
alone is insufficient to check the development of the disease. It appears 
probable that the failure of the fungus to complete its development in 
the case described above (where the plants after infection were exposed 
to an air and soil temperature of 30° to 33°) was brought about at least 
in part by some marked disturbance of the metabolism of the host and 
not simply by the direct effect of the high air temperature upon the 
fungus in the aerial parts of the seedling. 

Comparison between the development of the disease in plants grown 
at 15° to 20° and at 24° to 28° C. (air and soil) was made. A high per- 
centage of cotyledon infection occurred in both cases. At the lower 
temperature the disease proceeded as usual to the infection of the true 
leaves. At the higher temperature, however, the plants tended to out- 
grow the disease, this being associated with a more rapid rate of top 
development which apparently enabled the plants to slough off the 
smutted cotyledons before infection of the first true leaf occurred. 

The foregoing conclusions as to the dominant influence of soil tem- 
perature upon onion smut infection, while primarily based on greenhouse 
experiments, have been found to accord well with field developments. 

Successive out-of-door plantings at Madison, Wis., made in inoculated 
soil during the growing season, resulted in a gradual reduction of infection 
as the season advanced and the soil temperature rose. Complete freedom 
from smut was attained when the daily mean soil temperature at i to 2 
inches depth remained at or slightly above 29° C. for two to three weeks. 
There was also a tendency, as the temperature rose, for the seedlings to 
outgrow the disease by the sloughing off of the diseased cotyledons before 
infection of the first leaf occurred. 

An examination of records from one of the southern onion sections 
(Laredo, Tex.) shows that during a good share of the critical period for 
onion smut infection (August and September) the mean air temperature 
is above that at which complete inhibition of infection was attained in 
our experiments (29° C. or about 84° F.). If we assume, as observed 
in northern sections, that the mean temperature for the upper layer of 
soil is several degrees higher than that of the air at this time of the year, 
it is reasonable to conclude that even though the smut organism were 
introduced into southern onion sections, its development would be pre- 
vented or greatly minimized, first, by the prevention of infection due to 
high temperatures, and, secondly, by the rapidly developing tops out- 
growing the disease, should occasional infections occur. 

In general we believe, therefore, that the regional distribution of onion 
smut in the United States is conditioned upon the soil temperature 
during the seedling stage of the plant's growth, the infection and develop- 
ment of smut being favored by the relatively low temperatures and 
inhibited by the high temperatures, with approximately 29° C. as the 
critical point. 



Oct. 29, 19" Relation of Soil Temperature to Onion Smut 261 

It is hoped that the evidence here recorded may lead to the accumula- 
tion of further field data bearing upon this particular problem by inves- 
tigators in various places, especially in the southern States. It is also 
believed that these results illustrate well the importance of more per- 
sistent inquiry by the experimental method into the relation of environ- 
mental factors to the occurrence of disease of plants in general. 



(I 
(2 

(3 

(4 

(S 

(6 

(7 

(8 

(9 

(10 



LITERATURE CITED 
BouYOUCOS, George J. 

1916. SOIL TEMPERATURES. Mich. Agr. Exp. Sta. Tech. Bui. 26, 133 p. 
Chapman, George H. 

I910. NOTES ON THE OCCURRENCE OF FUNGOUS SPORES ON ONION SEED. MaSS. 

Agr. Exp. Sta. 22 Ann. Rpt., 1909, pt. i, p. 164-167. 
Heald, Frederick D., and Woolman, H. M. 
1915. BUNT OR STINKING SMUT OF WHEAT. Wash. Agr. Exp. Sta. Bui. 126, 

24 p., 5 fig. (in text and on pi.). 
HuNGERFORD, Charles W., and Wade, A. E. 

1920. RELATION BETWEEN SOIL MOISTURE AND BUNT INFECTION IN WHEAT. 

(Abstract.) In Phytopathology, v. 10, no. i, p. 53. 
Jones, L. R. 

1917. SOIL TEMPERATURES AS A FACTOR IN PHYTOPATHOLOGY. In Plant World, 

V. 20, no. 8, p. 229-237, 2 fig. Literature cited, p. 236-237. 
Mally, F. W. 

1915. THE BERMUDA ONION. Tex. Dept. Agr. Bui. 46, 56 p. 
MuNN, M. T. 
1917. NECK-ROT DISEASE OF ONiONG. N. Y. State Agr. Exp. Sta. Bui. 437, p. 
361-455, II pi. Bibliography, p. 4SO-455- 
SiRRiNE, F. A., and Stewart, F. C. 

1900. EXPERIMENTS ON THE SULPHUR-LIME TREATMENT FOR ONION SMUT. N. Y. 

State Agr. Exp. Sta. Bui. 182, p. 145-172, i pi. 
Sturgis, William C. 

1896. TRANSPLANTING, AS A PREVENTIVE OF SMUT UPON ONIONS. In Conn. 

Agr. Exp. Sta. 19th Ann. Rpt. 1895, p. 176-182, pi. i. 
Thaxter, Roland. 
1890. the "smut" OF ONIONS, (UROCYSTis cEPUOrAE FROST). In Conn. Agr. 
Exp. Sta. Ann. Rpt. 1889, p. 129-154, pi. 1-2. 
Ware, Benjamin P. 
1870. EXPERIENCE OF A PRACTICAL FARMER. In 17th Ann. Rpt. Mass. Bd. 
Agr. 1869, Appx., p. 1-16, 



PLATE 25 
Relation of soil temperature to the development of onion seedlings. 

Upper row. — Seedlings of Red Globe variety 13 days old. Each cluster was the 
entire crop from one experimental culture pot. All were grown in like virgin soil and 
at the same air temperature (i5°to2o°C.) but with gradation in the soil temperature of 
the respective pots as follows (left to right): 12° to 14°, 20°, 25°, 28°, 30°. Note that 
there is a tendency for greater root development in relation to top growth at the lower 
temperatures. This was especially marked at the lowest temperature, 12° to 14°. 
For further details see Table V and the accompanying text. 

Lower row. — Seedlings of Yellow Bermuda variety grown imder same conditions as 
those in upper row. 

(262) 



Relation of Soil Temperature to Onion Smut 



Plate 25 




L. 



Journal of Agricultural Research 



Vol. XXII, No. 5 



Relation of Soil Temperature to Onion Smut 



Plate 26 




Journal of Agricultural Researcli 



Vol. XXII, No. 5 



PLATE 26 

Relation of soil temperature to the infection of onion seedlings by the smut fungus, 

Urocystis cepulae. 

Representative seedlings taken from the experimental culture pots, showing the 
influence of soil temperature upon the amount of smut. All the pots alike contained 
smut-infested soil. The air temperature and other aerial factors were the same for all. 
Soil temperature was the only factor varied experimentally, the temperature grada- 
tions extending from about 10° to 30° C. 

Note the abundance of smut at 10° to 22°, as shown in the upper row. A slight re- 
duction occurred at 23° to 26°. At 27° to 29° the reduction is sharply marked. At 
29° to 3 1 ° inhibition is complete. For the percentages of seedling infection and other 
details see Table VI and the accompanying text. 



PLATE 27 
Relation of soil temperature to onion smut infection. 

This shows the results from a series of experiments in which the methods described 
for Plate 26 were repeated with the soil temperatures restricted to the critical limits 
between 25° and 29° C. and controlled more exactly. Note the marked reduction in 
infection at 27.5° and complete inhibition at 29°, thus establishing, but with more 
exactness, the conclusions illustrated in Plate 26. 

For the percentage of infected seedlings at these temperatures and other details, see 
Table VIII and the accompanying text. 



Relation of Soil Temperature to Onion Smut 



Plate 27 



\ V 



( \ 



ZS^'C 



1 



27.5°C 




Journal of AgriculturalResearch 



Vul. XXII. No. 5 



A PHYSIOLOGICAL STUDY OF GRAPEFRUIT RIPENING 

AND STORAGE ' 

By IvON A. Hawkins 

Plant Physiologist, Office of Horticultural and Pomological Investigations, Bureau of 
Plant Industry, United States Department of Agriculture 

In an earlier investigation (7) ^ the changes in Florida-grown grape- 
fruit during storage were studied, particular attention being paid to the 
sugar and acid content of the pulp or edible portion of the fruit as influ- 
enced by some six different storage temperatures. It was found that the 
acid content decreased in cold storage while the total sugar content re- 
mained about the same. The percentage of cane sugar decreased and 
the reducing sugar content increased. At the higher temperatures, 
common storage (about 55° to 60°, 70°, and 86° F.) there was in some 
cases apparently an increase in acidity and a reduction in the amount of 
sugar, especially in fruit stored for long periods. The shrinkage, which 
was very marked in the ventilated packages at these high temperatures, 
made the obtaining of definite evidence on this point impossible. 

The investigation described in the present paper is concerned with the 
acid and sugar changes in the fruit on the tree as well as with the changes 
which take place, both in warm storage and in cold storage, in fruit picked 
at monthly intervals. The control of the pitting which occurs commonly 
on grapefruit in cold storage is given some attention. 

PLAN OF THE EXPERIMENTS 

" Common Florida " ^ fruit from two trees was picked at monthly inter- 
vals for four months, beginnmg July 27, making five different picks. At 
the last three pickings fruit was also harvested from two additional trees 
in the same grove. The fruit was expressed to Washington and sampled 
on arrival. Part of the fruit was placed in warm storage at about 
70° F. and part in cold storage at 32°. Analyses were made of the fruit 
stored in the warm storage each month for two months and on fruit from 
the last three pickings stored in cold storage, after it had been in storage 
four months. By this plan it was possible to obtain data on the changes 
in the fruit on the tree from a month or so before the fruit was in con- 
dition to pick for market until the last of the season, and to compare the 
changes which took place in cold storage in fruit picked from the same 
trees at different times of the season. 

1 This paper gives the results of a portion of the work carried on under the project " Factors affecting the 
storage life of fruits." 

2 Reference is made by number (italic) to " Literature cited," p. 278-279. 

' The writer's thanks are due Mr. W. J. Krome for the picking and shipping of all the " Common Florida" 
fruits used in these exijeriments. 

Joimial of Agricultural Research, Vol. XXII, No. s 

Washington, D. C. Oct. 29, 192 1 

aag Key No. 0-251 

(263) 



264 Journal of Agricultural Research voi. xxii. no. s 

METHODS OF ANALYSES 

The fruit was prepared for sampling and sampled as in the previous 
work. Analyses were made for acids, sugars, both reducing sugars and 
total sugars, dry weight, shrinkage, thickness of peel, and percentage of 
peel. In addition the acidity and specific gravity of the expressed juice 
of the fruits were determined and the solids-acid ratio calculated after the 
usual method. The acidity determinations were carried out as in the 
previous work, as were practically all the other determinations with the 
exception of the extraction of the sugar from the pulp. In the sugar 
extraction a method was followed similar to that described in work on 
potatoes (6). The weighed pulp was placed in a liter volumetric flask which 
was then filled to volume with 85 per cent alcohol. It was allowed to 
stand with frequent shakings for about three weeks, the losses from 
evaporation, of course, being made up by adding alcohol. The alcoholic 
solution of sugar was then separated from the residue by filtration, and 
the sugars were determined in aliquots of the filtrate. 

The first lot of grapefruit was of small size, green in color, with very 
little juice in the pulp. No solids-acid determinations were made on 
this lot. They were, however, maintained in warm storage for two 
months. At the end of this period many of them had assumed the char- 
acteristic yellow color of the ripe grapefruit. 

The second pick, received August 29, was much further advanced, 
being about 50 per cent colored and of good size. The third and fourth 
picks, those of October 25 and November 28, respectively, were in fine 
condition for shipping and are what would be considered midseason 
fruit. The November 28 fruit was fair, possibly a little coarser than 
the two picks immediately preceding. No sprouted seeds were found 
in any of the fruits, however. The date of picking might be considered 
as in the latter part of the grapefruit season for this locality and for this 
variety. 

The analyses of grapefruit picked from trees i and 2, from warm 
storage at about 70° F. for one and two months, are shown in Tables I 
and II. In the analytical work the analyses were usually made in 
duplicate, and both analyses are given in the tables, as this furnishes 
evidence on the experimental error in the method of sampling. The 
tables are self-explanatory. 

An inspection of Tables I and II shows that in the first four pickings 
there is in all cases an increase in the acid content of the pulp, while in 
the last picking from both trees there is no decided increase. A com- 
parison of the acid content, as determined in the analyses of the 
pulp and the acid content of juice, shows a similar behavior. The acid 
content of the juice is, as a rule, markedly higher than that of the pulp, 
due, of course, to the fact that in the last-mentioned case the weight of 
fibrous material is taken into consideration in calculating the percentage 



Oct. 29, 1921 Physiological Study of Grapefruit Ripening 265 

of acid. In the fifth, pick from both trees there is no decided variation 
in the acid content of the pulp during storage, and the percentage of 
acid in the juice does not change as much as in fruit from any of the other 
four picks. With the sugars, the percentage of reducing sugars and of 
total sugars is always greater at the end of two months in warm storage, 
except in the fifth pick. The reducing sugar increases most, due prob- 
ably to the inversion of some of the cane sugar which is less in all cases 
after two months in storage. 

It was brought out in the earlier publication on grapefruit storage 
that there was an indication that the acid content of the fruit was slightly 
increased during a long period of warm storage. It was pointed out 
also that definite evidence on this point was difficult to obtain because 
the structure of the fruit prevented accurate calculation of the shrinkage 
of the various portions. Further evidence, mostly of an indirect nature, 
may be derived from the data on sugar and acid content of the fruit, 
found in Tables I and II. As was mentioned above, there is in all cases 
an apparent increase in the acid content and the total sugar content of 
the pulp, due for the most part undoubtedly to loss of water during 
storage. In the tables it is noticeable that the solids-acid ratio is usually 
less after two months in storage. This indicates, of course, that the 
increase in soluble solids is not proportional to the increase in acidity 
and that some soluble substance or substances other than titratable 
acids decreased in the storage period. This occurs in five cases out of 
seven on which data were obtained. The other two cases, tree 2, third 
pick and fifth pick, show slight increases, 0.07 and o. 1 1 , respectively. 

These data are corroborated in the total sugars-acid ratios, which are 
calculated by dividing the percentage of total sugar as dextrose by the 
percentage of acid as citric. In the 10 cases the ratio of sugar to acid 
is less in 7, practically the same in 2, and greater in i. Indications 
are, then, that there is usually an increase in the ratio of sugar to acid 
under the conditions of the experiment. This could be brought about 
by either decreasing the sugar content of the fruit or by increasing the 
acid content or by a combination of these two factors. It is notice- 
able that in 6 cases out of 10 the acid-sugar ratio is greater after one 
month in storage than it is after two months at the same temperature. 
The acid and sugar in the fruit from warm storage will be considered 
later in comparison with the changes taking place in cold storage. 

There is in most cases not much variation in the percentage of dry 
matter during storage, though there seems to be a tendency, more 
marked in some cases than in others, toward an increase. This seems 
probable, as the shrinkage where determined is from 14.4 to 23.3 per 
cent for the full two months in storage. The percentage of peel always 
decreases during storage at this temperature, due to the loss of water and 
wilting. This is evident in the decrease in thickness of the peel, which is 
very marked, especially in the earlier picks. 
65768°— 21 3 



266 



Journal of Agricultural Research voi. xxii, no. s 



CHANGES IN FRUIT ON TREES 

The analyses of fruit from the various pickings at the time it was placed 
in storage (Tables I and II) show marked differences in composition. 



PfRl 
con 

ri 


, 


\ 






























X 


Mf, 


























10 
fts 

9 






^1 


^ 


















.'^ 


^-^ 










\^ 


:^ 


■==r 










^^'^ 


:^ 


?^^ 
















































^,_ 


>^ 










■=^ 


^^ 




7 






-^ — -| 




















Z 
























y 


/^ 


^ — 




6 

55 

5 




















y 


^ 


























y 


^ 




















•\ 


^ 


o»&-- 


y- 


/ 




















/ 


■^o-V^^ 


,o>-. 


•" 


















4 

as 

3 
Z5 

Z 
\3 

1 






y 


























' 


y 


/ 




















^^ 


:^ 






















^i^ 


■■—::. 


^ 














^<- 




Sucsc 


SE 




j^^,.^ 


"^^^ 












2_S}i 


CPOSt. 




z^^ 






,,/ 


y 


jjcSfiS- 




&• 


^ 


















^ 






-^t^ 





























"^^^ 


^^^^^ 


Acid 

















— 








.Tree 


1 


























— 


.Tree 


2 




























AU&i 






SEPT. 


\ 


( 


XT 1 






NOV. 1 






DEC. 1 





Fig. I. — Graphs showing changes in percentage of dry matter, acid, and sugars calculated as dextrose in 
fruit on two Common Florida trees from August to December. 

For convenience in comparison the data for sugars, acids, and dry matter 
are shown graphically (fig. i). In these curves the percentage is plotted 
on the ordinates and the time interval between pickings on the abscissae. 



Physiological Study of Grapefruit Ripening 



267 







t~. 


00 


r^ 00 


t^ H 


C) 


M 








t^ a 


[ 




■2 A B^ 


M 


■* 1^ Oi 


OfOHtstooOOO tO-^OO 


M 


J^ 


4 4 N 


N t^ t^vd d d »0 n' to M H 00 















5 8^ 


TO (^ 


10 o\vo 


o> c^ t^ • 




H IT) 


'S 


" " 


^ to "1 


(N 0\ to f^ . 


• to N 


•*00 ^ 


X 


sii^g 


M M 


4 4 to 


to i^ t-*'d ■ 


• ■■)• tH 


M M 03 


^ 


W "i-> 


.— ,^ 


•— »—■*—.- 


_^_. : 






fe 














•d w .A 


^ 


ro r^ N 10 












1; CJ 




r^ 1/ 


1 to tt to to N 10 














M 


.^ ,. 


) t4 s6 d d " 4 












p^.s P 


_.>_ 


—^^.-^ 


-„-«^ --V— 












•i=i a J 


00 M 


O^t^O\O00f^^^0 -Ot^t^ 






t~ 


lOOO lOvO HI t^vO -^ -^ ^00 






4 4 N 


ci^O'vdtO d OC^-^j-MMt^ 



a 


jH S a«, 




— ,^^-- 


^w^ H M H M 


oS.H 


CO t^ 


M M 


N Tf 


>0 « CO 




l^ fO c< 


^ f 10 o\ 


t 
3 



H H 


4 w 


ti vd d 


t^ 4 M M .i 


"^ t- 9, 


M 


t to 


N OvvO >o 


Tt M M 




u 1; '^ 





0. M 


ej <^ " t 00 


H «00 




S.2 




« r^ to 


to>d v3'>d d. M 10 


M d 00 








M 


















•a 4 


00 fO 


00 CO 


oi-o a too tH 


to M to 




ai-sji 




^ -T Tf 


'J- o\oo 00 vo 00 o\ 


tooo CI 




H-aj^ 






*"* 




M 












lsa>^ 


M -0 


W-O \0 00 M -^00 00 to 


M M 


'a 


" " 


00 00 »o a u^^O t^ M OiOO 




•a 


rt;5 




to N N 


N lnv^»}■o^0^c^to 


M di 


1 


t^""^^ 










"^ U) ^ 


00 


M 00 H 


to to M t^ 01 to 


00 tooo 




?> OJ 


M 


lovo 


OS lo 0\ tf-O to 10 


Ct Tf to 




^^«s 


H M 


c^ « « 


c^ lnir)4dd\t^i/S 


M d» t^ 




Ph.s 












•2 , M'^ 

■- a 1> 


■*CC 


t^CO 00 


>o 10 to to^O to 


»0 OS »o 






ts to 


« too tOvO 1^ 

«io>o4didvts4 


o\ 


1 
a 

g 


h-q| 






>_^" 






1000 


S §, S 


00 CI 


OS M 


■* N 




^J- rf to 0. t~ ti 00 
c! w uS t^ d ,^ ri- 


M d -d 


m"" ^0 






" 


" 


m 












•a il. CO 


fO 


r^ONOO CAt^O »oi?> 


tj- M VO 




H (U . 




MMvooot^O»-<r^ i>^0 w 


T^ c* to 




Ah .2 m 
















« •* m tt d\ cj\ 100 


M cjs>d 




t3 I Mod 
•a cu— 


00 t^ 


01 Q 

o\ 6 N 

M to M 


»n 0\ »o 10 r^ 

H tf tj- t^ M vd 10 










•a 
'a 


H w Qq 




-V---V 


__^_ _!!_" 










•a J, 

5 a a-y 

aa Q 


ro I^ 

-0 


f^ to 00 Tt- 

o\ ov ao'^totoct 










■(-> 
















Wi 


01 m Q 1" 
















1-t 


m "w 




--'— .--^ 


^-— -— , 




























"o i 2' 


w 0. 


C>'0 CO 00 t^ ^ >/^ 00 












(U D N 


>0 TT 


l^ 1 


M -0 -0 to <N to tooo 












<J t;; M >. 
























Ci t^ to C< M « fosd 












^.b"!, 


>_^^ 


^-y— — V 




































■0 
















V 














•s 





























c« 











































1 i 


§ 










•c 


























■3 
"3 










J 






•0 




















'S 
"3 


3 J, 


J3 






^ 


S 










1 ^ 


3 S ?•«: 
-i: -3 -a c 
"o 7 '0 ■€ 


J 


It-? 








a 


ho 61 


' ^ 1 ^ |i|?l2 

b S S3fe-3-s? 






C8 Cl 








-U -1. 














■0 


a> 1 






< 




p. 


Ai 


ft 


(- 


ft 


ft 


i" 


ft 


< 


<7) 


tfi 


1 



268 



Journal of Agricultural Research voi. xxii, no. s 



J. So 















lo c^-O 


t 


>-» I^ M 


fO 










fO 













N « 






O 


M CO t<; 


to 




o 


00 ro 


^^ 


O 


0\ 
























^ I woo 









a 2^ 
^ -so 









o rt g +; 









l-ll 



) 0\ N O 00 vO 



^OOOH M OCOW O t^O fD 
HI CO 04 ro fOO 'O ^ 0\ 0\ H -^ 



*ivO 00 O ^O fn Tj- 



fn ^ fO I 



I -^ 0\ O rt 



■* 00 O 



Ov (^ 00 00 w) 
't ^ 0\ CO VI 



■ -^ f^ in M o\ t- lo 
W W W lO ■^ 4- o\ 



lO »o Tf M O CO "^ lOO iri ^ H O 



O O ^0 NvO 

lO 4 o\ -^ -^ 



vO 00 N M -tj- O O 



■ t^ t^ lo w OS 

^ rr f*j o» c> 6i*o 






*0 "(too lO -^ w ^O 



00 ^ 01 "* Q O 



w H ro CO < 



a d 



.9 3 



00 ro (O 
M fo t^ 

H 6 00 



0\ Oi 



i 6 >o 



<J fu, fl, a, H 



2-g 



tj (J tJ.ii "JIJ s^^ 



Oct. 29, 1921 Physiological Study of Grapefruit Ripening 269 

From an inspection of the curves it is evident that there is a decrease 
in acidity as the season advances, the acid being highest in both cases at 
the beginning of the season. The acid content is lowest at the fourth 
pick and rises shghtly at the fifth pick. ColHnson ( j) shows a somewhat 
similar decrease in acidity. This writer analyzed the fruit at more 
frequent intervals but apparently did not begin sampling so early in 
the season. According to his work there is a general tendency toward 
lower acidity, though in a few of his series of analyses there is a higher 
acidity toward the end of the season than in the midseason fruit. 

There is a rise in the percentage of total sugars during the season, the 
total sugar content of pulp of the fifth pick being about double that of 
the first pick. Collinson shows an increase in the total sugar content, 
but it is not so marked, due probably to the fact that his series begins 
later in the season. As showm in the curves (fig. i), the rise in total sugar 
content during the first month is very sharp. The increase in percentage 
of reducing sugar during the season is much more gradual and regular 
than that of the total sugars. The percentage of this sugar in tlie pulp 
a little more than doubles in the four months of the experiment. Much 
the same ratio of increase is found in the total sugars. The cane sugar 
curves are not so regular. There is, however, a marked increase in the 
percentage of cane sugar. The mean of the two sucrose curves is always 
higher than that of the reducing sugars except at the last sampling. 
Collinson records a series of analyses in which the reducing sugar is 
markedly higher than the sucrose during the latter part of the season. 
The irregularities in the total sugar curves in the present work are due 
to the variation in sucrose content. 

The percentage of dry matter, as determined in this work, is highest 
at the first of the season between 11 and 11.5 per cent but drops in the 
first month to between 9.5 and 10 per cent, the third and fourth analyses 
giving about the same results. There seems to be, however, an increase 
in the dry weights in the last month. 

A comparison of the percentage of peel (Tables I and II) at the time 
the five different lots of fruit were placed in storage shows there is a de- 
crease from 2)Z-Z to 21.2 and 45.6 to 18.2 per cent of peel by weight for 
trees No. i and 2, respectively. The percentage of peel decreases much 
more rapidly in the first month than in the succeeding month. In fact, 
in the case of tree i there is only a slight decrease in the proportion of 
peel to pulp in the last three pickings. At the time these last three 
pickings were made, the fruit was ready for market. 

As might be expected, the decrease in thickness of the peel, as meas- 
ured in these experiments, parallels the decrease in percentage of peel. 
The peel was found to be 6.8 mm. and 8.1 mm. thick, respectively, for 
trees i and 2 when the first pick was placed in storage, while it measured 
4.5 mm. and 3.2 mm. at the first samphng of the fifth pick. This is a 
reduction of 34 per cent and 60 per cent in the thickness of the skin fca: 



270 Journal of Agricultural Research voi. xxii, no. s 



the season. It is evident that the proportion of peel to pulp and thick- 
ness of the peel decrease as the fruit matures. 

A comparison of the acid and sugar changes in grapefruit in growth 
and ripening with the acid and sugar changes of other fruits in the same 
period of their life history brings out some interesting correlations and 
differences. The total sugar content of deciduous fruits usually increases 
during the growing and ripening period. This has been shown for 
apples by Lindet, (8) Bigelow, Gore, and Howard (2), and others; for 
pears by Ritter (12), Riviere and Bailhache (11), Magness (10), and by 
Bigelow and Gore (i), for peaches. Numerous other investigations cor- 
roborating this point might be mentioned. The literature pertinent to 
this subject may be found in the works referred to here. With certain 
vegetables a somewhat similar increase in total sugars is found. This 
was brought out by Hasselbring (5), working with sweet potatoes, and 
Sando (14) with tomatoes. This evidence would seem to indicate that 
in fruits or vegetables where sugar is stored the percentage of sugar 
calculated on a wet-weight basis increases during the growing season — 
that is, there is not only an absolute increase but an increase in proportion 
of sugar present as compared to the sum of the other constituents. This 
increase in some cases is due to an increased content of reducing sugars, as 
in the tomato, or may be due to an increase in both reducing sugars and 
cane sugar, as in the apple, pear, and peach, or for the most part to an 
increase in cane sugar, as in the sweet potato. 

In regard to the acid content of fruits which contain both sugar and 
acids in appreciable quantities, there is sometimes an increase and some- 
times a decrease in acidity as the season advances. In pears there is 
generally a decrease. Magness (10), however, found that pears from the 
Yakima district, Washington, and Medford district, Oregon, showed an 
increased acidity as the season advanced. Apples, according to the 
analyses of Lindet (8), and Bigelow, Gore, and Howard (2), exhibit a 
decrease in acidity as the growing season advances. Peaches, on the other 
hand, increase in acid content as they approach maturity. The decrease 
in acidity of grapefruit during the growing season is comparable to the 
usual behavior of the acidity in pears and in apples. 

COLD STORAGE EXPERIMENTS 

As mentioned earlier in this article, experiments on the cold storing of 
grapefruit were carried out during the 1920-21 season. Fruit from four 
trees was used. These trees included the two from which fruit was 
obtained for the warm storage work, the fruit being from the lots desig- 
nated third, fourth, and fifth picks in the experiments aheady described. 
Not sufficient fruit was available for this work from the fifth pick from 
tree i , so only two experiments were possible with fruit from this tree. 

Table III gives the results of analyses made at the time the fruit was 
placed in storage and four months later. As was pointed out, because 
of the structure of the fruit it is very difficult to obtain definite evidence 
on the changes of the various constituents of the pulp. While the 



Oct. 39, 1921 



Physiological Study of Grapefruit Ripenivg 



271 



fruit for each experiment was carefully selected from a lot of fruit all 
harvested at the same time from a single tree, the variation in composi- 
tion of the fruit on this tree introduces a possible error which it is hardly 
possible to calculate. It is only by obtaining a large amount of evidence 
that a clear indication of the direction of the change can be obtained. 
It was accordingly deemed advisable to give in this table all the data 
obtained in the analyses in the 1 1 different storage experiments carried 
out in this portion of the investigation. The table is self-explanatory. 

Table III. — Percentage of acid, sugar, and dry matter in pulp, thickness of peel and per- 
centage of peel, acid and soluble solids in juice, and solids-acid ratio of " Common Flor- 
ida" grapefruit before and after storing four months at J2°F. 

TREE I 





First lot. 


Second lot. 


Third lot. 




When 
placed in 
storage. 


After 4 

months in 

storage. 


When 
placed in 
storage. 


After 4 

months in 

storage. 


When 
placed in 
storage. 


After 4 

months in 

storage. 




f T rn 


0.88 
.90 

2-55 
2. 74 

3- II 
2. 72 
=;. 66 
5-46 
9-43 
9-37 
26. 9 

5-5 
5 

I. 02 
8.87 
8.66 


I. 00 

1. 02 

2. 96 
3- 04 
3-13 

3. 22 
6. 09 
6.26 
9-45 
9-54 

21.8 

S 


0.88 
.92 
3-48 
3-09 
3- II 
3- SI 
6. ."59 
6.6 

9-74 
9-74 
21. 6 

4-3 
4.9 

I. 01 
II. 25 
II. 16 






citric 1 I. 08 

Percentage of reducing f 2. 62 














Percentage of cane sugar 


1 2.61 
I 2. 93 
















as dextrose \ 5. 51 

Percentage of dry mat- 1 f 9. 49 
ter il 9. 63 


















Thickness of peel (in 


5-S 






Percentage of shrinkage 
Acidity of juice as per- 
centage of citric 

Soluble solids (Brix) . . . 






1.27 
9-45 

7.38 


I. 14 

ID. II 
8.81 





















TREE 2 



Acid as percentage of 
citric 

Percentage of reducing 
sugar as dextrose 

Percentage of cane sugar 
as dextrose 

Percentage of total 
sugar as dextrose 

Percentage of dry mat- 
ter 

Percentage of peel 

Thickness of peel (in 
mm.) 

Percentage of shrinkage 

Acidity of juice as per- 
centage of citric 

Soluble solids (Brix) . . 

Solids-acid ratio 



1. 06 
2.47 

2.44 

2. 74 
2. 52 

21 
96 



54 
S 



5-6 



1.28 
9.40 
7-31 



I 2.61 

}2.6s 

} 5-26 

9.46 

23-3 

4.4 
3 

I. 19 

10. 04 

8.44 



I. 06 

3. 01 
. 2.92 

3-3° 

3-38 

6.31 

. 6.30 

' 9.66 

. 9-43 
21. 2 

4. 2 



I. 18 

IO-33 

8.73 



I. 00 
3-37 
3-54 
3-03 
2-75 
6. 40 
6. 29 
9.86 
10. 16 
21. 7 

4-3 
4 

I. 09 
10. 16 
10. 29 



I- 15 
3-73 
3.66 

3-24 
3-25 
6.97 
6. 91 

10.30 

18.2 



I. 19 

10.79 

9- 03 



I. 10 

3-29 

3-59 

6.88 

9-77 
9.82 
22. 4 

4.8 
3-9 

I. OS 

11-35 
10. 84 



272 



Journal of Agricultural Research voi. xxii. no. s 



Table III. —Percentage of acid, sugar, and dry matter in pulp, thickness of peel and per- 
centage of peel, acid and soluble solids in juice, and solids-acid ratio of "Common Flor- 
ida" grapefruit before and after storing four months at 32° F. — Continued 

TREE 3 



First lot. 



When 
placed in 
storage. 



After 4 

months in 

storage. 



Second lot. 



When 
placed in 
storage. 



After 4 

months in 

storage. 



Third lot. 



When 
placed in 
storage. 



percentage of 



Acid as 
citric . 

Percentage of reducing 
sugar as dextrose 

Percentage of cane 
sugar as dextrose 

Percentage of total sugar 
as dextrose 

Percentage of dry mat- 
ter 

Percentage of peel 

Thickness of peel (in 
mm.) 

Percentage of shrinkage . 

Acidity of juice as per- 
centage of citric 

Soluble solids (Brix) . . . 

Solids-acid ratio 



|- 1.06 

\ 2. 46 
I 2.58 
I 2.83 
f 3.01 
L 5-29 
I .?• 59 
L 8.77 
[ 8.78 
24 

5-3 



I. 26 
8.6s 
6. 92 



2-55 
2-35 
2.58 
2.77 

S-I3 
5.12 
8. 10 

8-53 
24. 2 

4-7 
5 

I. 17 

II. 19 

9- SI 



I. 10 

1. 09 
2.77 

2. 50 
2.81 
2.81 
5.58 
5-31 
9-56 
9. 16 



4-9 



03 
07 
22 

50 
16 

75 
38 
25 
19 
67 



21.8 



I. 29 
I. 18 
3-36 
3- 51 
3.01 

3-2? 

6-37 
6.76 
9.82 
9.68 

22. 2 
4-3 



1.25 
9.66 
7.71 



08 
46 

73 



I. 29 

10. 29 

7-94 



TREE 4 



percentage of 



Acid as 
citric . 

Percentage of reducing 
sugar as dextrose 

Percentage of cane 
sugar as dextrose . . . . 

Percentage of total sugar 
as dextrose 

Percentage of dry mat- 
ter 

Percentage of peel 

Thickness of peel (in 
mm.) 

Percentage of shrinkage . 

Acidity of juice as per- 
centage of citric 

Soluble solids (Brix) . . . 

Solids-acid ratio 



I I. 18 
[ I. 10 
f 2. 70 
1 2.66 
f 2. 62 
^ 2.56 
^ 5-32 

I 5-22 
9. 28 

I 9-53 
22 

5-4 



1.36 

9-35 
6.86 



94 
03 
,18 
80 
31 
69 
49 
49 

91 

4 



I. 17 
ro. 01 
8.55 



1.24 

3-04 
1.78 
.^.82 



58 
57 
3 



4-S 



I- 31 
10.23 

7-79 



I. 04 



4.9 
2,- 2, 

I. 04 

10. 16 

9-73 



I. 21 
I. 22 
3-75 
3-85 
3. 10 
3.02 
6.85 
6.87 
9.87 
9.90 



I. 26 

10.86 

8.60 



From Table III it is evident that in every case there is a lower acidity 
in the fruit after it has been held in storage four months than in fruits 
from the same tree and picking when placed in storage. This is in accord- 
ance with the findings reported in the previous publication and would 
seem to establish this point definitely. The fact that in the present 
experiments fruit was picked at three dififerent times during the growing 
and ripening season strengthens the evidence. 



Oct. 29, 1921 Physiological Study of Grapefruit Ripening 273 

The total sugar content is usually slightly higher at the end of the four 
months' storage period, though there are several instances in which it is 
lower. These cases are mostly in well-matured fruit of the last pick. 
The increase in total sugars is due for the most part to an increase in 
the reducing-sugar content, as there is usually a marked decrease in the 
percentage of cane sugar during storage. There is never more than 5 per 
cent shrinkage during these four months. This shrinkage is doubtless 
partly from the peel and partly from the interior portion or pulp. 

The fact that in most cases there is an apparent increase in total sugars 
can be accounted for by the loss of water and consequent shrinkage. It 
is very evident from these data that there is no appreciable diminution 
in the amount of sugar in the grapefruits in four months at 32° F. On 
the other hand, there is without doubt no considerable increase. It is, 
of course, probable that some of the pectins and other hemicelluloses or 
the glucosid in the fruit break down slowly, and it is possible that some 
reducing substance is formed from these decomposition products. 

A comparison of the behavior of the acids and sugars in grapefruits 
stored in warm storage (Tables I and II) with the results obtained in the 
cold storage experiments just considered brings out some rather striking 
differences. In the data obtained from the warm storage experiments 
there is evidence of an increase in acidity or a decrease in total sugars or 
both — that is, in most cases the ratio of total sugar to acid decreases, while 
in the cold storage the reverse is true. This is corroborated by the acidity 
and soluble solids of the juice. In the warm storage experiments the 
solids-acid ratio is in most cases less after two months in storage, while in 
the cold storage there is always a decrease in acidity and an increase in 
solids-acid ratio. It is evident that there is an increase, or at least not 
a decrease, in acidity in warm storage and a decided decrease in cold stor- 
age. It would, therefore, seem probable that some of the processes which 
go on in the fruit stored in the warm are modified when the fruit is placed 
in cold storage. It is possible, of course, that in respiration the acid is 
used up in cold storage while the sugars are used in warm storage. There 
is an indication that the sugar content may decrease slightly in the fruits 
held in warm storage, while there is no evidence of change in the per- 
centage of sugar in the cold-stored fruits. Magness (9) has shown that the 
composition of the gases in the interior of apples and potatoes varies 
with the temperature at which they are held. For example, he found 
that the gas from the interior of Yellow Newtown apples stored at 2° C. 
(about 35° F.) analyzed 14.2 per cent O2 and 6.7 per cent CO2, while at 
30° C. (86° F.) the extracted gas was 3.2 per cent O2 and 21.4 per cent 
COj. The air surrounding the fruits used in these experiments was practi- 
cally the same in both cases. The oxygen content was low and the carbon- 
dioxid content high in the fruit at high temperatures because the oxygen 
was used up in respiration faster than it could diffuse in from the outside. 
While no such determinations have been made on grapefruits, it seems 



274 Journal of Agricultural Research voi. xxii. no. s 

probable from the size of the fruit and the thickness and structure of the peel 
that in fruit held at high temperatures for any considerable period there 
would be a low oxygen pressure. This might result in some intermolecular 
respiration and the formation of acid. At low temperatures the respira- 
tion rate would be considerably decreased, while the rate of diffusion of 
O2 through the tissues would not be so greatly reduced, and sufficient 
oxygen might be present for the breaking down of the compounds used 
in respiration of COj and HgO. A careful investigation of this point is 
needed. The work of Gerber {4) is of interest in this connection. 

The dry weights are about the same at the conclusion of the experi- 
ments as at the beginning. There may be a slight diminution in the 
percentage of dry matter, but this apparently lies within the experi- 
mental error of the determinations. The variation in thickness of peel 
and percentage of peel is so great that thera is frequently a higher per- 
centage of peel after the fruit has been stored four months than when it 
was placed in storage. This is undoubtedly due to the lack of uniformity 
in the fruits and the low percentage of shrinkage. 

The loss in weight during the four months' storage is from 3 to 4.9 per 
cent, averaging around 4 per cent. The relative humidity of the storage 
rooms was around 75 per cent. The fruit was not in the best condition 
for merchandising at the end of this storage period, as it was in many 
cases badly pitted. It is doubtful whether this method of storage would 
be applicable to commercial conditions if the fruit were placed directly 
in cold storage. 

EXPERIMENTS IN THE CONTROL OF PITTING 

As was mentioned in the earlier paper (7) , grapefruit tends to pit in 
cold storage. This pitting begins as a small indentation of the skin in 
practically any region of the surface. The sunken area gradually in- 
creases in size, frequently becoming as much as i cm. in diameter. They 
are usually, in the type of fruit used in these experiments, about i mm. 
in depth. In time they may take on a brown color. This coloring occurs 
more quickly if the fruit is removed to a warm room. These pits may be 
very numerous on the surface of the fruit, in many cases coalescing in 
irregular shaped patches. 

Cross sections of these pits show that they are formed by a breaking 
down of the layer of tissue containing the oil vesicles. There is appar- 
ently no disintegration of the tissue. The cells and vesicles simply flatten 
out as if subjected to local pressure, the layer of tissue becoming brown. 
The injury apparently does not extend to any distance in the spongy 
tissue ben-ath this oil-bearing layer, and it is only after a long period that 
any evidence of the discoloration appears on the inside of the peel. The 
pulp of the fruit is apparently uninjured. The affected fruit, however, 
is very unsightly, and badly pitted fruit would hardly be salable in a 
normal market. It was evident that unless some method of preventing 



Oct. 29, 1921 Physiological Study of Grapefruit Ripening 275 



this pitting was worked out the storing of grapefruit for any considerable 
period would hardly be commercially practicable. 

Experiments were, therefore, undertaken to see if it were possible to 
treat or handle the fruit so that it could be cold-stored without this danger 
of pitting. As was mentioned earlier, fruit stored in warm storage, 70° 
to 86° F., or in common storage (about 55° to 60°) apparently does not 
pit. It was considered possible that if fruit were cured for a time in warm 
storage before being placed in cold storage the injury from this blotching 
and breaking down of the surface of the peel might be obviated. Ac- 
cordingly a lot of I dozen fruits from tree i , third pick, was maintained at 
a temperature of 70° and a humidity of about 65 per cent for one month, 
then removed to cold storage (32°) and xamined at intervals. At the 
end of three months in cold storage none of these fruits were pitted, while 
about 60 per cent of the fruit from the same lot olaced directly in cold 
storage at 32° were badly pitted. 

The experiment was repeated with grapefruits of the Duncan, Marsh 
Seedless, and Silver Cluster varieties from Polk County, Fla., which were 
placed in storage February 12, 1920. Part of the fruit of each lot was 
placed directly in 32° F., and the rest of the three lots were placed in the 
curing room and maintained at a temperature of about 70° with a relative 
humidity around 60 per cent. Portions of the lots from the curing room 
were removed to 32° cold storage at intervals. The entire storage period 
was three months for all lots. The results of the experiment are shown 
in Table IV, in which are given the length of time in curing, the time in 
cold storage, and the percentage of pitting of the different lots. In 
these experiments the pitting is given as slight and bad pitting. Bad 
pitting is applied to pitting that would markedly injure the sale of the 
fruit. Slight pitting refers to pitting that while noticeable does not par- 
ticularly injure the fruit for sale. It is at most a few spots usually small. 

It is noticeable in Table IV that most of the control fruit that was 
placed directly in cold storage without curing is pitted and that there is a 
high percentage of bad pitting. In the Duncan, 6 per cent was good, 
while the poorest lot of cured fruit of this variety was about 90 per cent 
good. There was more pitting in the cured Silver Cluster than in the 
Duncan and somewhat more in the Marsh Seedless than in the Silver Clus- 
ter. The data obtained in this one storage experiment are hardly suffi- 
cient, however, to justify the conclusion that Duncan grapefruit store 
better than Silver Cluster and Marsh Seedless. The experiments, how- 
ever, seem to show that the pitting can be controlled by proper curing 
before the fruit is placed at the low temperatures. The specific effect of 
this curing, by exposure to warm temperatures from one to six weeks, 
on the tissue of the peel so that the pitting is prevented has, of course, 
received little attention. Pitting has all the external appearance of 
injury considered to be due to Colletotrichum gloeosporioides (Penz.) by 
Rolfs, Fawcett, and Floyd (zj) and figured by them. This fungus, 



276 



Journal of Agricultural Research voi. xxii, No. 



however, has a high optimum temperature, and it seemed highly improba- 
ble that its growth could be controlled by exposing to high temperatures 
and that it caused most damage at temperatures around 32° to 40° F. 
It was possible, however. This point was investigated by Dr. F. V. 
Rand, of the Laboratory of Plant Pathology, Bureau of Plant Industry. 
The results of this work are as yet unpublished. The following account, 
however, is based on Dr. Rand's work. 

Table IV. — Results of storage experiments with Duncan, Marsh Seedless, and Silver 

Cluster grapefruit 



Num- 
ber of 
fruits. 


Num- 
ber of 
days in 
curing 
room. 


Num- 
ber of 

days in 
cold 

storage. 


Tem- 
pera- 
ture of 

cold 
storage. 


Num- 
ber of 
good 
fruits. 


Per 
cent- 
age of 
good 

fruit. 


Num- 
ber 
with 
slight 
pitting. 


Per- 
centage 
with 
slight 
pitting. 


Num- 
ber 
with 
bad 

pitting. 


Per- 
centage 

with 

bad 
pitting. 


5° 
27 

23 
14 
10 


a 00 

12 

19 
25 

41 


90 
78 
71 
65 
49 


°F. 
32 

32 
32 
32 
32 


3 
24 

23 
13 
10 


6 

88.9 
100 

92.8 
100 


10 
3 


20 
II. I 


37 


74 






I 


7-1 



















MARSH SEEDLESS 



82 
42 
30 
19 

12 


a 00 
10 
18 
24 
30 


90 
80 
72 
66 
60 


32 
32 
32 
32 
32 


3 
II 
12 
16 
10 


3-6 
26.2 
40 
84.2 
83-3 


6 


7-3 


73 
31 
12 


89 
73-8 
40 


6 

3 
2 


20 

15.8 

16.7 











Sn,VER CLUSTER 



46 
16 
17 

20 


a 00 

9 

18 

25 


90 
81 
72 
65 


32 
32 
32 
32 


23 
14 
15 
17 


50 

87-5 

88.2 

85 






23 


5° 


2 
2 
3 


12.5 
II. 8 
15 















" Controls placed directly in cold storage. 

Cultures were made from the pits and from the tissue of the peel 
between the pits. Cultures were also made from the peel of cured fruits 
which had been in cold storage for three months after curing. The 
results are shown in Table V. 

It is evident from Table V that Colletotrichum was almost univer- 
sally present in the peel of these Florida grapefruits and that while it 
is usually to be found in the pit it is just as common in the normal peel 
of the pitted fruit or the cured fruit. It is, of course, impossible to 
assert from the evidence at hand that the fungus does not cause the 
breaking down of the peel. The cold storage might so affect the physi- 
ology of the peel as to make it susceptible to fungus attacks, while curing 



Oct. 29. 192 1 Physiological Study of Grapefruit Ripening 



277 



and warm storage render it resistant. This, of course, is somewhat 
doubtful. The case is somewhat analagous to that cited by Winston 
(15) in regard to tear stain, which has up to now been considered to be 
due to Colletotrichum gloeosporioides, mainly because this fungus was 
usually found in cultures from the diseased areas. In the present work 
it seems fair to conclude that whether or not the fungus causes the pit- 
ting it is controlled at least to a large extent by curing before placing the 
fruit in cold storage. 



Table V. 



-Results of cultural experiments with pitted grapefruit and with fruit from 
same lot which was unpitted « 



Date. 


Source. 


Num- 
ber of 
fruits. 


Num- 
ber of 
pieces 

of 
tissue. 


Col- 
leto- 

tri- 
chum. 


Clado- 
spo- 
rium. 


Alter- 
naria. 


Pusa- 
rium. 


Peni- 

cil- 

lium. 


Ster- 
ile. 


Mis- 
cel- 
lane- 
ous 
fungi. 




Pits 

Between 
pits 

Pits 

Between 
pits 

Pits 

JBetween 
pits 

Control . . . 

[Pits 

JBetween 

1 pits 

(Control . . . 
[Pits 

Between 
1 pits 

Control . . . 

Pits 

JB etween 
1 pits 

Control . . . 

[Pits 

JBetween 
1 pits 

Control . . . 


6 

I 
2 

2 

7 

4 
5 
3 

2 
I 
3 

3 
3 

4 

2 
2 
5 

2 
4 


96 

7 
3i 

17 
53 

29- 
70 
32 

27 
14 
38 

17 
23 
38 

24 
20 
93 

38 
84 


33 

I 
17 

10 
30 

15 
42 
21 

12 
14 


31 

5 


I 






31 

I 
16 

7 












































3 
3 






I 

3 
4 










1 












22 i 4 

1 TI 










Feb. I, i92i(not sterilized) 
































38 

6 




Feb. 3. 1921 (sterilized 3 


3 
16 
38 

15 
17 
68 

27 
63 






I 
I 
2 






















Feb. 3, 1921 (not sterilized) 






2 

I 

I 


10 

4 
10 


7 












Feb. 8, 1921 




6 

3 

S 


7 
S 


10 




2 

3 



o Unless otherwise stated, all pieces of tissue were sterilized two minutes in i to 1,000 bichlorid solution 
and were washed three times in sterile tap water before pouring plates. The control fruits were without 
signs of the pitted spots under investigation. ' ' Between pits ' ' refers to sound tissue between the spots. 

GENERAL DISCUSSION AND CONCLUSION 

In the investigation of grapefruit storage described in the foregoing 
pages it has been brought out that in warm storage the percentage of 
acid calculated to the wet weight of the pulp increases markedly in two 
months' storage. There is evidence that this increase is not due entirely 
to loss of water from the pulp, but that there is an increase in the amount 
of acid present. There is evidence indicating that there may be a slight 
decrease in the sugar content in warm storage. In cold storage there is a 
decrease in the acidity very marked after four months in storage, while 
there is little change in the amount of total sugars present. A possible 
explanation of this difference in the behavior of the sugars and acids in 
warm and cold storage was pointed out. This phase of the problem 
deserves further attention. The investigations on the changes in the 



278 Journal of Agricultural Research voi. xxii, no. s 

fruit during development on the tree showed that the total sugar con- 
tent increased while the acidity decreased, the increase in sugar content 
being very marked. 

Fruit on the tree increases in palatability and food value. There is, 
of course, always danger that the seeds will sprout in the varieties 
containing seeds if the fruit remains on the tree too long. There is also 
danger that the fruit will drop or be shaken from the tree by high winds. 

It is of interest to note that the behavior of the acids and sugars 
during growth and in cold storage is similar to the behavior of these 
constituents of some of the deciduous fruits — that is, it is apparently 
possible to remove the fruit from the tree after it is well along toward 
maturity and to ripen it in storage. The result will be an apparently 
sweeter fruit, due to loss of acidity and a reduced bitterness, the naringin 
or bitter principle breaking down in storage. A period in cold storage, 
then, renders the fruit more palatable. From the experiments detailed 
above it seems probable that the pitting of grapefruit can be controlled 
by curing at 70° F. before they are placed in cold storage. Investiga- 
tions are in progress at the present time on this last-mentioned phase of 

the Avork. 

LITERATURE CITED 

i) BiGELOW, W. D., and GorE, H. C. 

1905. STUDIES ON PEACHES... U. S. Dept. Agr. Bur. Chem. Bui. 97, 32 p. 
2) and Howard, B. J. 

1905. STUDIES ON APPLES... U. S. Dept. Agr. Bur. Chem. Bui. 94, 100 p., 
30 fig., 5 pi. 

3) COLLINSON, S. E. 
I913. SUGAR AND ACID IN ORANGES AND GRAPEFRUIT. Fla. Agr. Exp. Sta. 

Bul. 115, p. 1-23. 

4) Gerber, Charles. 

1896. RECHERCHES SUR LA MATURATION DES FRUITS CHARNUS. In Ann. Sci. 

Nat. Bot., ser. 8, t. 4, no. 1/6, p. 1-280, pi. 1-2. 

5) Hasselbring, Heinrich. 
1918. BEHAVIOR OF SWEET POTATOES IN THE GROUND, /n Jour. Agr. Research, 

V. 12, no. I, p. 9-17, I fig. 

6) Hawkins, Lon A. 

1916. EFFECT OF CERTAIN SPECIES OP FUSARIUM ON THE COMPOSITION OP THE 

POTATO TUBER. In Jour. Agr. Research, v. 6, no. 5, p. 183-196. 
Literature cited, p. 196. 

7) and MagnESS, J. R. 

1920. SOME CHANGES IN FLORIDA GRAPEFRUIT IN STORAGE. In JoUr. Agr. 

Research, v. 20, no. 5, p. 357-373. Literature cited, p. 372-373. 

8) LiNDET, L. 
1894. RECHERCHES SUR LE D^VELOPPEMENT ET LA MATURATION DE LA POMME 

A cidrE. In Ann. Agron., t. 20, p. 5-20. 

9) Magness, J. R. 

1920. COMPOSITION OF GASES IN INTERCELLULAR SPACES OP APPLES AND POTA- 
TOES. In Bot. Gaz., v. 70, no. 4, p. 308-316, i fig. Literature cited, 
p. 316. 



Oct. 29, 1921 Physiological Study of Grapefruit Ripening 279 

(10) Magness, J. R. 

1920. INVESTIGATIONS IN THE RIPENING AND STORAGE OF BARTLETT PEARS. 

In Jour. Agr. Research, v. 19, no. 10, p. 473-500, 8 fig. Literature 
cited, p. 499-500. 

(11) Rivi^rE, Gustave, and Bailhache, Gabriel. 

1908. iSTUDE relative a la PROGRESSION ASCENDANTE DU SUCRE ET A LA 
PROGRESSION DESCENDANTS DE l'ACIDITE;, DANS LES FRUITS DU 
POIRIER, DEPUIS LEUR FORMATION JUSQU'A LEUR MATURITY. In 

Jour. Soc. Nat. Hort. France, ser. 4, t. 9, p. 284-289. 

(12) RiTTER, Georg. 

1910. UEBER DEN CHEMISCHEN REIFUNGSPROZESS DER FRUCHTE, MIT BESON- 

DERER BERUCKSiCHTiGUNG DES OBSTES. In Deut. Obstbauztg., 
Jahrg. 56, Heft 31, p. 429-435- 

(13) Rolfs, P. H., Fawcett, H. S., and Floyd, B. F. 

1911. diseases of CITRUS FRUITS. Fla. Agr. Exp. Sta. Bui. 108, p. 25-47, 

fig. 10-23. 

(14) Sando, Charles B- 

1920. THE PROCESS OF RIPENING IN THE TOMATO, CONSIDERED ESPECIALLY 

FROM THE COMMERCIAL STANDPOINT. U. S. Dept. Agr. Bul. 859, 38 

p., 3 fig., 4 pi. (1-2 col.). I,iterature cited, p. 32-35. 

(15) Winston, John R. 

1921. TEAR-STAIN OF CITRUS FRUITS. U. S. Dept. Agr. Bul. 924, 12 p., 2 pi. 

Literature cited, p. 12. 



ABSORPTION OF COPPER FROM THE SOIE BY POTATO 

PLANTS 

By F. C. Cook 

Physiological Chemist, Miscellaneous Division, Insecticide and Fungicide Laboratory, 
Bureau of Chemistry, United States Departm.ent of Agriculture 

RESULTS OF PREVIOUS INVESTIGATIONS 

Some of the results obtained by a few investigators on the absorption 
of copper by plants and cells may be summarized as follows : 

Schander^ found that copper in a soluble form is a poison for plant 
cells of both high and low order. 

Tschirch - believes that living plants are able to absorb copper through 
their roots and also through the epidermis of the leaves, the amount of 
copper absorbed being very small, however. 

Haselhoff ^ stated that soluble copper salts are injurious to plants at 
a concentration of lo mgm. of cupric oxid per liter. When soluble cop- 
per salts are added to the soil the plant materials, especially the potash 
and the lime, are dissolved and washed away, as a consequence of which 
the fertility of the soil is decreased. The action of copper sulphate is 
more severe on some crops than on others. The presence of calcium car- 
bonate in the soil prevents or decreases the toxicity of solutions of copper 
sulphate. 

True and Gies ^ have shown that when lime is used with copper sul- 
phate solutions the toxicity of the copper is decreased. They state that 
when there is lime in the soil four times the amount of copper that can be 
allowed when no lime is found may be present in a soil without exerting 
a toxic action. 

Forbes ^ found that com grown in soil containing copper held most of 
the copper in the roots rather than in the tops. He states also that the 
toxicity of copper depends on the combination in which it exists in the 
soil, the physical characteristics of the soil, and the chemical composition 
of the soil, and on climatic and moisture conditions, as well as on the crop 
grown. 

iScHANDER, Richard, uber die PHYSIOWJCISCHE WIRKTTNG DER KT7PFERV1TR10UCAI.KBRUHE. In 
Landw. Jakrb., Bd. 33, Heft 4/5, p. 517-584. 1904. 

2 Tschirch, A. das kxtpfer vom standpunkte der gerichtuchen chemie, toxicologie und hy- 
giene. 138 p., 2 fig. Stuttgart. 1893. Bibliographical footnotes. 

'HaSELHOFF, Emil. UEBER DIB SCHADIGENDB WIREtTNG VON KUPFERSULFAT UND KUPFERNITRAT- 

MALTlGEM WASSER AUF BODEN UND PFLANZEN. In Laudw. Jahrb., Bd. 21, p. 263-276, 2 pi. 1892. 

* True, Rodney H., and GiEs, William J. on the physiologicai, action of some op the heavy 
METALS IN mixed solutions. In Bvd. Torrey Bot. Club, v. 30, no. 7, p. 390-402. 1903. 

^ FoBLBES, R. H. certain effects under irrigation of copper compounds upon crops. Ariz. Agr. 
Exp. Sta. Bui. 80, p. 145-238, 16 fig., 4 pi. (i col.). Bibliography, p. 236-238. 

Journal of Agricultural Research, Vol. XXII, No. s 

Washington, D. C. Oct. 29, 1921 

aah Key No. E-17 

(281) 
65768°— 21 4 



282 Journal of Agricultural Research voi. xxii, No. s 

OBJECT OF PRESENT EXPERIMENTS 

The experiments discussed in this paper were undertaken to determine 
what proportion of the copper present in standard Bordeaux spray, in 
Pickering's limewater Bordeaux spray, and in a solution of copper sul- 
phate, of equal copper content, is absorbed by potato plants when the 
sprays or solution are applied directly to the soil in which the vines are 
growing. The comparative distribution of the absorbed copper in dif- 
ferent parts of the potato plants was also studied. 

The copper in the Pickering spray was in an insoluble form, basic cop- 
per sulphate, with no excess of lime present. The copper of the Bordeaux 
spray was in an insoluble form, with a large excess of lime present. The 
copper of the solution of copper sulphate v/as soluble. It was believed 
that a comparative study of these three sprays, containing copper in equal 
amounts, would show the extent to which the excess lime of Bordeaux 
spray is instrumental in preventing the absorption of copper by the roots 
of the potato plants, as well as the relation of the absorption of copper 
from a soluble copper compound to that from an insoluble copper com- 
pound when applied to the soil. 

EXPERIMENTAL WORK 

The tests were conducted on the Aroostook Farm of the Maine Agri- 
cultural Experiment Sta.tion, at Presque Isle, Me., on Caribou type soil. 
A single row, 8 feet long, of Norcross strain of the Green Mountain variety 
of Irish potato plants was used for each of four plots which were treated 
in the following manner : Plot i , sprayed with standard Bordeaux, 3-3-50 
formula, containing 0.75 per cent of copper sulphate; plot 2, sprayed 
with an "A ' formula Pickering limewater Bordeaux spray, containing 
0.70 per cent of copper sulphate; plot 3, sprayed with a solution con- 
taining 0.75 per cent of copper sulphate; and plot 4, a control plot, un- 
sprayed. 

The vines were 20- inches above ground when the first applications 
were made. At each application i gallon of the spray or solution was 
applied directly to the ground within 6 inches of the stems of eight potato 
plants in each plot, each vine thus receiving i pint of the solution to each 
treatment. An equal amount of water was applied to the roots of eight 
control plants at the time the other applications were made. Applica- 
tions were made on July 27, August 8, August 17, August 24, and August 
30, 1917. 

PREPARATION OF SAMPLES 

Vines and tubers from each of the four plots were taken for analysis 
at frequent intervals. 

The vines from the various plots were dried in the air, then washed in 
running water and held for 30 seconds in a 4 per cent solution of hydro- 
chloric acid, after which they were immediately washed in water and 



Oct. 29, 1921 A bsorption of Copper from the Soil by Potato Plants 283 

finally in distilled water. The vines were next dried for 16 hours in an 
oven at 110° C. Separate analyses of leaves, stems, roots, and tubers 
were made. 

Five or six tubers from each plot were thoroughly washed, rinsed in 
distilled water, and dried with a towel. The tubers were pared, passed 
through a grinder, well mixed, and transferred to a Mason jar with rubber 
and top. Care is necessary in securing a uniform sample of the ground 
tubers for analyses, as the water and solids separate very rapidly. 

Samples of soil were taken 6 inches deep, near the roots of the treated 
plants, from the various plots at the time the plants were sampled. The 
soil samples were held in Mason jars with rubbers and tops until analyzed. 
Before analysis the stones and other foreign matter were removed from 
the samples. 

DETERMINATION OF COPPER IN VINES AND TUBERS 

From 5 to 10 gm, of the dried leaves and stems, and from i to 5 gm. 
of the roots were taken for copper analyses. The samples were ashed 
in 4-inch porcelain dishes, 30 cc. of 5 per cent nitri acid were added, and 
the whole was allowed to remain overnight. The solutions were filtered 
and washed, after which ammonia was added to faint alkalinity. They 
were brought to a boil, cooled, and made to volume, usually 150 cc. The 
precipitated iron and alumina were removed by filtration, and an aliquot 
of the filtrate was taken for the determination of copper. 

Table I. — Copper found in potato vines and tubers'^ 



Plants taken for analysis. 



Aug. 8.. 
17- 

24. 

Sept. 3. 
Average 



Parts analyzed. 



Leaves. 
vStem.. . 
Root... 
Leaves . 
Stem . . 
Root... 
Tubers. 
Leaves. 
Stem.. . 
Root... 
Tubers. 
Leaves. 
Stem . . 
Root... 
Leaves. 
Stem . . . 
Root.. . 
Tubers. 



Soil treated 


Soil treated 


Soil treated 


with Pick- 


with Bor- 


with CuSOi 


ering spray 


deaux spray 


solution 


(0.7s per 


(0.75 per 


(0.75 per 


cent 


cent 


cent 


CUSO4). 


CuSOi). 


CUSO4). 


Per ceiit. 


Per cent. 


Per cent. 


0. 004 s 
. 001 15 




0. 006=; 


0. 0070 


.0079 





.0136 


. 0052 


. 0097 


. 0100 


. 0069 


.0023 


. 0042 


.0047 


.0036 


.0030 


. OIOI 


. 0004 
.0179 




. 0001 


.0258 


. 0109 


. 0029 


. 0048 


. 0067 


.0128 


. 0104 


.0130 


. 0001 


. 0001 


. 0001 


.00-53 


. 0069 


.0225 


. 0160 


. 0069 


. 0300 


. 0107 


.0179 


. 0081 


.0030 


• OO.S7 


. 0104 


. 0081 


. G085 


. 0146 


. C002 


. 0001 


. 0001 



Control 
plot. 




0017 
o 
0069 
0027 
0012 
0002 



<» Analyses made on dry basis. 



284 Journal of Agricultural Research voi. xxu.no.s 



As a rule, 25-cc. aliquots were evaporated to dryness in 50-cc. por- 
celain dishes on the steam bath, and the residue was taken up in 5 cc. 
of distilled water. Two drops of acetic acid and 3 drops of i per cent 
solution of potassium ferrocyanid were added, and the color was im- 
mediately compared with that of standard solutions of copper sulphate 
which had been evaporated with ammonium nitrate and taken up in 
5 cc. of distilled water. 

Copper in the ground tubers was determined by the same procedure, 
using 50 gm. of the moist sample. The analytical data are recorded in 
Table I. 

DieTERMINATlON OF COPPER IN SOEUS 

One hundred gm. of the well-mixed soil samples were treated with a 
mixture of 80 cc. of nitric acid and 20 cc. of sulphuric acid in large por- 
celain casseroles. The mixtures were heated on the steam bath and 
til en on the hot plate until the nitric acid fumes were removed. The 
residues were extracted with 200 cc. of water, filtered, washed, and made 
to 500 cc. volume. After evaporation to 200 cc, the iron was precipi- 
tated with ammonia and the solutions were made to volume. They were 
next filtered and aHquots were made acid with hydrochloric acid, through 
which hydrogen sulphid was passed for 20 minutes, or until all the copper 
was precipitated. The precipitated copper after settling was filtered 
and dissolved in 10 cc. of nitric acid, the filter paper and precipitate being 
transferred together. Ammonia was added to faint alkalinity, and the 
solutions were evaporated to dryness in small porcelain dishes. The resi- 
dues were taken up in 5 cc. of distilled water, two drops of acetic acid 
and three drops of i per cent potassium ferrocyanid were added, and the 
copper was estimated by colorimetric comparisons. In some cases after 
evaporation to dryness it was necessary to take up in water, filter, wash, 
and repeat the evaporation to remove precipitated material. 

If present in large enough amounts copper may be determined electro- 
lytically, by a method based on the procedure given by Forbes, Free, and 
Ross.^ 

The results of the analyses of the first and last samples of soil taken 
appear in Table II. This table gives also the results of a series of tests 
on the soil around the roots of potato plants which had been commer- 
cially sprayed with Bordeaux, with Pickering spray, and with a solution 
of copper sulphate, to determine whether any appreciable amounts of the 
copper occur in the soil beneath the sprayed vines. 



> Forbes, R. H. certain effects under irrigation of copper compounds xtpon crops. Arlr. Agr. 
Exp. Sta. Bui. 80, p. 145-238. 16 fig., 4 pi. (i col.). 1916. Bibliography, p. 236-238. Part 3, Appendix: 
Methods of analysis, with the collaboration of E. E. Free and W. H. Ross, p. 229-233. 



Oct. 29, 1921 Absorption of Copper from the Soil by Potato Plants 285 



Tabl^ II. — Copper found in soil 

SPRAYS APPLrED TO SOIL NEAR PLANT ROOTS 



Sample 
No. 


Date of 

sampling. 


Description of samples. 


Description of plots. 


Total 
copper 

found 
in soil. 


I 


July 26 
Aug. 24 
Sept. 3 


Samples taken before any 
copper was added to the 
soils. 

Samples taken just before 
last application of copper 
to soil. 

Samples taken after the last 
application of copper to 
the soil. 


Control 


P. p.m. 
2 


2 

3 
4 
5 
6 

7 
8 

9 
10 


Pickering "A" formula 

Bordeaux ^-^-c;o 


I 
2 


CUSO4 solution 

Control 


2 

S 
211 
256 
250 

2 


Pickering "A" formula 

Bordeaux •?— •?— t;o 


CUSO4 solution 

Control 


Pickering "A" formula 

Bordeaux 3—3—50 


225 
243 

449 




CUSO4 solution 







SPRAYS APPLIED TO VINES IN COMMERCIAL PRACTICE 



July 16 



Aug. 31 



Samples taken before any 
sprayings were made in 
1917. 

Samples taken after last 
sprayings were made in 
1917. 



Control 

Pickering "C" formula. 

Bordeaux 5-5-50 

Pickering "A" formula 

Control 

Pickering "C" formula 

Bordeaux 5-5-50 

Pickering "A" formula 



DISCUSSION OF RESULTS 



VINES AND TUBBRS 



The leaves, stems, and roots of the plants from the soil receiving 
the Pickering spray showed an increased copper content with each 
successive analysis (Table I). The largest percentage of the copper was 
held by the leaves. The roots held an appreciable part of the copper, 
the amount increasing from o in the first sample to 0.0160 per cent in 
the sample taken on September 3. The tubers contained only minute 
amounts of copper. 

The plants from the Bordeaux treated soil showed irregularities, 
particularly with respect to the copper content of the roots and stems. 
The leaves and stems contained more copper than those of the plants 
from the Pickering treated soil, while the roots contained less copper 
than the roots of the plants from the Pickering treated soil. The 
amounts of copper found in the tubers were small. 

The vines grown in the soil treated with a solution of copper sulphate 
showed a marked progressive increase in copper content of the roots 
with each succeeding analysis. The leaves contained somewhat larger 



\ 

286 Journal of Agricultural Research voi. xxn.No. s 

amounts of copper than the stems, but not as much as the roots. The 
leaves contained less copper than the leaves of the plants grown on the 
Bordeaux or Pickering treated soils. The tubers from the plot treated 
with copper sulphate solution were as low in copper as those from the 
other plots. 

The analyses of the various portions of the conti'ol plants showed the 
presence of copper, but in smaller amounts than in the plants grown on 
soil treated with the copper sprays. 

The results of the copper absorption experiments indicate that the 
potato plants, with the exception of those grown in the soil receiving 
the solution of copper sulphate where the roots were distinctly injured, 
distributed the largest part of the absorbed copper to the leaves, while 
the roots and stems contained appreciable amounts of copper. In all 
normally sprayed potato plants the largest proportion of the copper is 
said to be found in the leaves. 

The plants grown on the soil treated with a solution of copper sulphate 
were small and lacking in vigor. The roots had but few hairs, and 
showed other signs of injury. The large percentage of copper found in 
the roots, together with the small size of the roots, indicated some inter- 
ference with the normal metabolism of the vines. The toxic effect of 
the soluble copper salt was exerted primarily on the roots of the plants. 
It was apparent that the soluble copper sulphate had injured the potato 
plants, while the insoluble copper compounds had not. 

The vines from the Bordeaux plot contained a little more copper than 
the vines from the Pickering plot, indicating that the extra lime of the 
Bordeaux spray did not aid in preventing the absorption of copper by 

the plants. 

soil, 

The results of the analyses of the first and last samples of soil taken 
show that no water-soluble copper was found in any of the samples 
examined. The amount of copper in the first set of samples (Table II, 
No. 1, 2, 3, and 4) which were taken before any copper had been added 
to the soils, is practically the same in all cases. The sets of samples 
taken before and after the final treatment of the plots show the pres- 
ence of a large amount of copper in the samples receiving the copper 
treatments. This means that copper in an insoluble form may be pres- 
ent in the soil in marked amounts without exerting any apparent toxic 
action on the growth of potato plants. 

But little copper was found in the soil as a result of spraying with 
copper sprays according to commercial practice. 

On September 5, shortly after the last treatment of the soil, a few hills 
of potatoes were dug. The weights and number of the tubers, the per- 
centage of decayed tubers, as well as the notes taken on the size and 
appearance of the vines are given in Table III. 



Oct.29, I92I Absorption of Copper jrofyi the Soil hy Potato Plants 287 



Table III. — Effect on potato tubers and vines of applications of sprays to soil 





Tubers." 




Spray used. 




Weight. 


stand and condition 
vines. 






Total. 


Average. 








Ounces. 


Ounces. 




Bordeaux (0.75 per 


3 (i large and 2 


8 


2^ 


Normal. 


cent CUSO4). 


small) in ihill. 








Pickering (0.7 per 


8 (5 large and 3 


29 


zH 


Do. 


cent CuSOJ. 


small) in 2 hills. 








Copper-sulphate solu- 


17 (all small) in 4 


23 


iK 


Small and stunted. 


tion (0.75 per cent 


hills. 








CUSO4). 










Unsprayed (control) . 


8 (3 large, 3 medium, 
and 2 rotten) in 2 
hills. 


24 


3 


Normal stand ; 
blight. 



a Rot found only on control tubers. 

These data show tliat the solution of copper sulphate had a very dis- 
astrous effect on the growth and yield of the tubers. The only decayed 
tubers found were obtained from the unsprayed plot. These results are 
so few that they can be considered only as suggestive. 

vSUMMARY 

Potato plants grown in soil treated with insoluble copper compounds 
contained more copper in the leaves than in the stems, while but little 
copper was found in the roots. The tubers showed only traces of copper. 

When the soil was treated with the copper sulphate solution, the roots 
were injured and the normal metabolism of the vines was disturbed. 
The tubers from these vines were small and the vines stunted. The roots 
of these plants held more copper than the leaves. 

The soluble copper sulphate added directly to the soil caused injury 
to the plants, while the insoluble copper compounds of the sprays did 
not. The excess lime of the Bordeaux spray did not reduce the amount 
of copper absorbed by the plants compared with the plants grown on 
the Pickering plot. 

Practically the same amounts of copper were found in all the soil 
samples tested. Samples of soil from sprayed potato fields showed but 
minute amounts of copper. 



ADDITIONAL COPIES 

OF THIS PUBLICATION MAY BE PROCURED FROM 

THE SUPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTrNG OFFICE 

WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 
Subscription Price, S3.00 Per Year 

V 



Vol, XXII NOVKIvIBE^R 5, 1921 No. 6 



JOURNAL OF 




RE SEAJR.CH 



CONXKNXS 

Page 

Pale Western Cutworm (Porosagrotis orthogonia Morr.) - 289 

J. R. PARKER, A. L. STRAND, and H. L. SEAMANS 

( Contribution from Montana Agricultural Experiment Station) 



Biology of Embaphion muricatum - - - - - 323 
J. S. WADE and ADAM H. BOVING 

( Contribution from Bureau of Entomology ) 



PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



WASHINGTON, D. C. 

GOVERNMENT PRINTING OFFICE 

1921 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARL F. KEIyI<ERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 

J. G. LIPMAN 

Dean, State College of Agriculture, and 
Director, New Jersey Agricultural Experi- 
ment Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of EnU^ 
mology and Economic Zoology, Agrieui- 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS 

Dean, School of Agriculture, and Director, 
Agricultural Exptrimenf Station; Tks 
Pennsylvania State College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultural Experiment Station, Nev7 
Brunswick, N. J. 



JOURNAL OF AGMCETIMLEESEARI 

Vol. XXII Washington, D. C, NovEm^'^*^?2i No. 6 

PALE WESTERN CUTWORM (POROSAGROTIS 
ORTHOGONIA MORR.) 

By J. R. Parker, Assistant Entomologist, Montana Agricultural Experiment Station; 
A. L. Strand, Assistant State Entomologist of Montana; and H. L. Seamans, 
Special Assistant in Cutworm Investigations, Montana Agricultural Experiment 
Station ' 

INTRODUCTION 

Extensive injury to grain crops by cutworms was reported from north 
central Montana during the period from 1915 to 1920. It was at first 
supposed that well-known species were responsible for the damage and 
the usual method of control, poisoned bran mash, was recommended. 
The repeated failure of this method led to a study of the situation, and 
from the results of numerous rearing records and the personal investiga- 
tion of many infested fields it was found that the greater part of the losses 
was caused by the pale western cutworm {Porosagrotis orthogonia Morr.), 
a species previously not considered of economic importance in Montana. 
The enormous damage which it has done during the last six years, the 
rapidity with which it has extended its range, its unusually long period 
of larval feeding, its comparative freedom from parasites, and the fact 
that it works underground and can not be controlled by poisoned bran 
mash, stamp P. orthogonia as the most dangerous of all our western 
grain cutworms, not excepting even the army cutworm {Chorizagrotis 
auxiliaris Grote). 

HISTORY OF THE SPECIES 

The species was given its specific name in 1876 by Morrison (15),^ 
who described it as A gratis orthogonia from specimens collected at 
Glencoe, Nebr. In 1890 the species was placed under the genus Porosa- 
grotis by Smith {16, p. 129), who also gave a description of the adult 
and recorded its occurrence in the following new localities: Colorado, 
New Mexico, Arizona, and Utah. Dyar {j, p. 139) lists the species and 
gives its range as the Rocky Mountain region. In 1905 it was reported 

1 The color plate and drawings for this article were done by Miss Helen Lund, with the remarkable 
accuracy characteristic of her work. Mr. K. M. King, an undergraduate assistant in 1919, conducted the 
rearing experiments during that year, and many of the observations recorded are based on his very complete 
insectary notes. 

^ Reference is made by nxunber (italic) to "Literature cited," p. 320-321. 

Journal of Agricultural Research, Vol. XXII, No. 6 

Washington, D. C. Nov. 5, 1921 

aai Key No. Mont.-8 

(289) 



290 Journal of Agricultural Research voi. xxii, no. 6 

(doubtfully) by Dod {2, v. 57, p. 53) from Calgary, and in 1908 by 
Hampson (9, p. 102) from Prairie, Alberta. 

The species was looked upon as a rare insect until 191 1, when Gibson 
(5, 6) reported it under the name Porosagrotis delorata Smith as destroy- 
ing large areas of wheat in southern Alberta, where one correspondent 
claimed to have lost 320 acres before June 21. Hewitt {10, p. ijj) also 
refers to this outbreak in his annual report for 19 12, and in his report 
(jj, p. 506) for the following year the species is recorded along with 
Euxoa ochragasier Gn. as having destroyed between 30,000 and 35,000 
acres of crop in 19 12 in about the same territory where damage occurred 
the previous year. 

An account of the insect's first appearance, its depredations during 
191 1 and 19 1 2, and a brief review of the control experiments carried on 
the following year was given by Gibson (7) in 19 14. In this article he 
states that Porosagrotis delorata Smith and P. orthogonia Morr. are the 
same and records an adult of the species as having been taken at Regina, 
Saskatchewan, on August 10, 1904. Hewitt {12, p. 861-862) states 
that in 19 13 this cutworm caused much less damage to crops in southern 
Alberta than in the preceding year. 

The most complete published account of the species was written by 
Gibson (<?, p. 30-31) in 19 15. A brief description of the larva and adult 
is given, together with notes on its life history and habits. In this 
article the common name "pale western cutworm" is used, apparently 
for the first time. 

During the season of 19 14 Porosagrotis orthogonia was again present 
in Alberta, and experiments in its control were conducted by Strickland 
(17), who found that surface applications of the bran mash were wasted 
but that gratifying results were secured when a molasses-and-shorts 
mixture was harrowed into the soil. The next account is by the same 
author {18), who gives a brief statement of the life history and makes 
recommendations for the control of the species by cultivation methods 
and the modified use of a poisoned-shorts mixture. 

According to Hewitt's 191 6 report {13), the pale western cutworm was 
seldom seen in 19 15. Experiments, however, were conducted by Strick- 
land which confirmed his earlier conclusions that shorts is preferable to 
bran and that when the soil is moist harrowing in the poison is not so 
advantageous as it is on dry soil. 

In his 19 19 annual report (j, p. 8) Cooley points out the habits of the 
species which make it such an important pest and places its control as 
one of the most important entomological problems in Montana. 

A review of the life history of the species, descriptions of the various 
stages, and colored drawings of ^gg, larva, and adult, are given by Maxson 
{14, p. 45-46) in his work on sugar-beet insects published in 1920. 



Nov. 5>i92i Pale Western Cutworm 291 

DISTRIBUTION 

Published records of the occurrence of Porosagrotis orthogonia are as 
follows: Glencoe, Nebr., by Morrison (15, p. 2jg) ; Colorado, New Mexico, 
Arizona, and Utah by Smith {16, p. i2g) ; Rocky Mountains by Dyar 
(j. P- 139) ; Calgary, Alberta, (doubtfully) by Dod {2, 37, p. 53) ; Prairie, 
Alberta, by Hampson (9, p. 102) ; and southern Alberta by Hewitt 
{10, p. 177), Gibson (7) and Strickland {17). 

In Montana Porosagrotis orthogonia now occurs throughout the State 
east of the continental divide. It has been most abundant in the tier 
of counties which lies just east of the foothills of the main range of the 
Rocky Mountains and which extends from the Canadian border to 
within 100 miles of the southern border of the State. 

Mr. E. H. Strickland has kindly given the following information on 
the present distribution of Porosagrotis orthogonia in Canada : 

Our records indicate that it is practically confined to southern Alberta, extending 
as far north as latitude 51° and east to longitude 108°, although it has been recorded 
as far as Regina, Saskatchewan. The maximum intensity, however, is confined to 
an area that does not extend more than 100 miles east of the Rocky Mountains. 

Dr. William Barnes, of Decatur, 111., who has an extensive collection 
of western noctuids, has generously furnished the following records of 
Porosagrotis orthogonia specimens in his collections: Denver, Oak Creek 
Canyon, Lavetta, and Alamosa, Colo.; Deming and Fort Wingate, N. 
Mex. ; Provo, Vineyard, and Eureka, Utah; Yellowstone National Park, 
Wyo.; Reno, Nev. ; Redington, Ariz.; and Kern County, Calif. 

Mr. George M. List states that Porosagrotis orthogonia is fairly com- 
mon at Fort Collins, Colo., 430 moths having been taken at a trap dur- 
ing the season of 1920. )<;■ ;0 I; 

To Mr. C. N. Ainslie, of the Bureau of Entomology, United States 
Department of Agriculture, we are indebted for a record of 63 Porosagrotis 
orthogonia moths reared from a shipment of larvae from Dickinson, 
N. Dak., on June 10, 1920. 

From the foregoing records it is evident that Porosagrotis orthogonia 
occiu-s at least in scattering numbers throughout the southwestern and 
northwestern States with the possible exceptions of Oregon, Washing- 
ton, and Idaho, where as yet it has not been collected. Correspondence 
with entomologists throughout the territory where P. orthogonia has 
been recorded indicates that it has never been of economic importance 
outside of the heavily infested areas in Montana and Canada. 

METHODS OF STUDYING 
IN THE INSECTARY 

Larvae were reared in individual tin boxes 1.5 inches in diameter. 
The bottom of each box was covered with filter paper which was slightly 
moistened at each feeding. This prolonged the freshness of the wheat 



292 Journal of Agricultural Research voi. xxn.No. 6 

and dandelion which were used as food, and the cans could be easily 
cleaned by replacing the filter paper whenever it became soiled. 

Pupae were placed in moist, well-pulverized soil in individual glass 
vials I inch in diameter and 4 inches deep. Each vial was filled to a 
depth of 2 inches with well-firmed soil in which a round hole i inch 
deep was punched to receive the pupa, which was placed in it with the 
anterior end uppermost. A small twig was placed in each vial so that 
the moth upon emerging could hang from it and allow the wings to expand. 
The vials were closed by cheesecloth held in place by rubber bands. 
The best results were obtained by keeping the pupae in uniformly moist 
and mellow dirt. Extreme dryness or excessive moisture often resulted 
in the death of the pupse. Uniform moisture conditions were more 
easily obtained by allowing water to run slowly down the side of the 
tube instead of flooding it over the surface of the soil. 

After the moths emerged they were placed in wire s,creen covered 
tin cans 3.5 inches in diameter and 2.5 inches deep. An inch of moist 
soil was kept in the bottom of the cans, and alfalfa or clover blossoms 
were added each day for the moths to feed upon (PI. 30, A) and to 
hide under. Cutworm moths of all kinds seem very contented in these 
cans, and with Porosagrotis orthogonia no difficulty was encountered in 
getting the females to mate and lay eggs. 

Eggs were placed on filter paper in pint Mason jars with the caps 
lightly screwed down. A few drops of water were placed on the filter 
paper from time to time to provide the proper amount of humidity. 

IN THE FIELD 

A very good opportunity to watch the development of this insect 
under natural conditions was afforded in a heavily infested field at Wil- 
sall, Mont., in 19 19. This field was first examined on May i and was 
visited several times a month all summer. Many fields in other parts of 
the State were also visited, but that at Wilsall was the only one where 
Porosagrotis orthogonia was followed through all stages of its develop- 
ment. During the summer of 1920 a temporary field station was estab- 
lished at Willow Creek, Mont., in a district where thousands of acres of 
wheat had been destroyed during May and June. Moths appeared in 
large numbers during August and September and were under observa- 
tion at all hours, both day and night. Special attention was given to 
the egg-laying habits, and for this purpose two observation cages were 
set up. The cages were 2 -foot cubes with screen wire sides and solid 
metal tops. They were placed over sunflower plants, and the ground 
inside the cages was covered with soft dirt, stubble, clods, baked earth, 
and green plants, thus offering the moths nearly all the natural condi- 
tions of the neighborhood. Experiments with trap lights were also 
carried on at the Willow Creek field station. 



Nov. s. I92I Pale Western Cutworm 293 

SEASONAL HISTORY AND HABITS 
DURATION OF EGG-IvAYING PERIOD 

During the season of 19 19 the eggs were found to be well developed 
in the ovaries when the female moth first emerged, and egg laying began 
and was completed within a short time under insectary conditions. 
The first eggs were obtained from reared moths on August 17, the aver- 
age period between emergence and the beginning of egg laying being 
four days. Moths collected in the field on August 26, which appeared 
to have just emerged, laid numerous eggs the following day and con- 
tinued to lay until September 9. 

In 1920 no records were available from reared moths, but a study of 
moths in the field seemed to indicate that the eggs that season were 
not fully developed when the moths first emerged. Thus out of 35 
moths examined on August 24 only one had well-developed eggs in the 
ovaries. On September i many moths were found with the ovaries 
filled with well-developed eggs. These moths were mostly badly rubbed 
specimens, indicating that they had probably emerged some little time 
before. 

Our field observations show that the height of the egg-laying period 
is during the last week in August and the first week in September. 
Eggs in smaller numbers may be laid during the first three weeks in 
August and as late as October i . 

WHERE EGGS ARE LAID 

The first eggs obtained were from moths confined in tin rearing cans. 
When the soil in the cans was dry and light most of the eggs were placed 
from X iiich to i inch below the surface and could be found only by careful 
searching. When the soil was hard and lumpy eggs were scattered 
about on the surface and could be easily seen. In the rearing cans the 
eggs were not always laid in the soil. Many of them were placed on 
the stems, leaves, and flowers of alfalfa, and frequently scattering eggs 
were found on the sides of the cans or on the screen covers. Thus out 
of 243 eggs found in one can, 180 were found in the soil, 62 were found 
on the stems, leaves, and flowers of alfalfa, and i was found on the 
side of the can. Some were laid singly and others were in clumps of 
2, 3, or 4, and sometimes as many as 40. 

In outdoor cages, where a variety of soil conditions and various kinds 
of vegetation were available, eggs were laid only in loose, dry dirt. 

Under field conditions the eggs are very difficult to find, and the only 
ones we have ever found were secured by carefully examining the soil 
at the exact spots where females were seen in the act of egg laying. 
Eggs are found most frequently in loose, mellow dirt from X to >^ inch 
below the surface. This is an important point, and will be discussed 
later in connection v^^ith the habits of the moths and from the standpoint 
of control methods. 



294 



Journal of Agricultural Research voi. xxii. No. 6 



NUMBER OF EGGS 

The only females upon which we have complete recor ds are five moths 
reared and mated in the insectary. They averaged 315 eggs, the lowest 
number per moth being 248 and the highest 453. Ten moths brought 
in from the field averaged 132 eggs, but these had probably laid numerous 
eggs before they were caught. Under normal field conditions the aver- 
age number per female is probably between 300 and 400. The records 
of the individual moths are shown in Tables I and II. 



Table I. — Time between emergence and egg laying, length of egg-laying period, and 
number of eggs laid by reared specimens of Porosagrotis orthogonia 



Moth 
No. 


August, 19 19. 


Eggs 

in 

ovaries 

at 
death. 


Total 


13 


14 


IS 


16 


17 


iS 


19 


20 


21 


22 


23 


24 


2S 


26 


27 


28 29 


30 


31 


ber 
eggs. 




E 
E 










243 

D 
E 
E 
E 


Ds 

























249 


3 


248 










9 
























258 












88 


4S 


96 
66 

30s 


S4 
2SI 

2 


D 

2S 
12 














283 














26 


75 


D 

12 










453 































































E= emerged. D=dead. 

Average period between emergence and egg laying 4 days. 

Average period of egg laying 3 to 4 days. 

Average number of eggs i'^s- 

Table II. — Period of egg laying and number of eggs laid by Porosagrotis orthogonia 
moths brought in from the field and kept in rearing boxes 



Moth 
No. 


August. 


September. 


Eggs 

in 

ovaries 

at 
death. 


Total 


26 


27 


28 


29 


30 


31 


I 


2 


3 


4 


5 


6 


7 


ber 
eggs. 




Caught 










63{ 
1 


D 

100 
D 
23 

45 

2 

6 
IS 

6 
16 


] 












4 

91 

88 

103 

30 

I 



152 


167 




do 






















116 




do 


22 

2 


74 
5 


43 
3 




I 


1 
D 
fD 

\ 3 

m 

\ 3 

m 

\i4 
/D 
I 8 

36 












184 




. .do 


4 




1 










107 




do 


f 










112 


6 


do 




4 
35 
26 


56 

IS 
20 


25 
30 
32 


12 
22 


1 










134 




do 


20 


\ 










127 


8 .. 


...do 


( 

D 

44 










116 




. ..do 




\26 


1 






106 




do 














/ 
D 





































D=dead. 

Average period of egg laying 3 to 4 days. 

Average number of eggs 132. 

DURATION OF EGG STAGE 

The length of the egg stage is exceedingly variable, depending largely 
upon moisture conditions. It may be as short as 1 1 days or may extend 
over several months. Eggs laid in the insectary August 19 hatched 
August 3c. On September 30 several newly hatched larvae were found 



Nov. s, 1921 Pale Western Cutworm 295 

at Wilsall. Hundreds of eggs were laid in the breeding cans during the 
last week in August and the first week in September, but with the ex- 
ception of three larvae which hatched August 30 no eggs hatched unless 
they were placed in a very humid atmosphere. The larvae mature 
within the eggs in from 10 to 20 days and may remain for months in this 
condition, waiting merely for proper moisture conditions to allow them 
to break through the eggshell. Examples of this may be of interest. 
Moth number 1427-G laid 160 eggs on August 24 and 25. These were 
placed on a piece of filter paper in a Mason jar. They were allowed to 
stand in the laboratory for 10 days and were then placed in an incubator 
and held at day temperature of 80° F. and night temperature of 60°. 
At the end of three weeks practically all the eggs had darkened, and the 
black heads of the young cutworms could be seen through the eggshells, 
but none had hatched. On October 4 a few drops of water were added 
to the filter paper, and when the jar was opened 24 hours later 70 larvae 
were found to have hatched. On October 1 1 water was again added to 
this jar, and 40 more larvae hatched out. 

On October 27 a few drops of water were added to another jar of 62 
eggs which were laid August 29. In two hours 8 larvae had hatched, 
but no more hatched during the next six hours. Twenty-four hours 
later all of the eggs had hatched. On October 27 moisture was added to 
eggs that had been kept in the greenhouse since August 30, and in two 
hours many of them had hatched. On this same date one of these eggs 
was placed on a block of plaster of Paris, and water was slowly dropped 
upon it from a medicine dropper. At the second drop the larva began 
to move within the egg. Soon it began to move its mandibles and after 
several attempts the eggshell was punctured, and within 30 minutes 
after the first drop of water was added the larva was free from the shell 
and actively moving about. On November 20 eggs that were laid 
August 30 and had been kept for a month in a small tin can on a shelf 
directly over a radiator were examined, and living larvae were removed 
from them by carefully breaking the eggshells with fine needles. 

On November i twenty eggs which were laid on August 29 and had 
been kept indoors were placed in two small wooden boxes and buried in 
a pail of damp sand. The pail was set on the ground, outdoors, where 
it was covered with snow practically all winter. The eggs were exam- 
ined once a month, but none hatched until the second week in April 
when all the eggs were found to have hatched and the larvae were alive 
and vigorous. 

From our studies of the egg it would appear that if there is sufficient 
moisture and proper temperature condition the majority of the eggs 
will hatch in the fall, while if it is unusually dry or cold weather starts 
early the eggs will not hatch until the following spring. Strickland {18) 
found eggs on frozen ground December 3, which would indicate that in 
Canada some of the eggs at least do not hatch until spring. 



296 Journal of Agricultural Research voi. xxii. no. 6 

FIRST APPEARANCE OF I^ARV^E 

Larvae may appear in the fall. This is proved by the fact that three 
larvae were found at Wilsall on September 30 and that larvae hatched in 
the rearing boxes during October and November whenever sufficient 
moisture was added. Large numbers of larvae must have hatched at 
Wilsall during the fall of 1918, for as soon as the snow left the ground 
the following spring fourth- and fifth- instar larvse were found in large 
numbers. We have no records of injury to wheat during the fall months, 
but it seems quite probable that in years when there is considerable 
moisture and mild weather during October and November great damage 
may be done. 

The larvae begin to feed shortly after the wheat begins to grow in the 
spring. In 191 9 at Wilsall 80 acres of winter wheat were completely 
destroyed by May i, which indicates that the worms must have been 
active 10 days or 2 weeks previous to that date. 

In 192 1 at Willow Creek first-instar larvae were found on March 3. 
The weather had been warm for about a week, and winter wheat was 
starting to grow again. No larger larvae could be found, and it seemed 
as though the very small first-instar larvae must have just hatched. 

PERIOD OF LARVAIv FEEDING 

One of the reasons why Porosagrotis orthogonia is such a dangerous 
insect is the unusually long period of heavy larval feeding which extends 
until the middle of June or even to July 10 in the case of late-hatched 
specimens. Judging from the reports of injury received, the larvae 
attract most attention during the month of May and the first two weeks 
in June, differing decidedly in this respect from Chorizagrotis aiixiliaris, 
which generally has reached the height of its destructiveness by April 
15 and has practically disappeared by May i. 

The length of the larval stage as determined for 20 larvae, 5 from each 
of 4 parent moths, under insectary conditions varied from 62 to 151 
days and averaged 118 days, as shown in Table III. 

All of the larvae were kept under very similar conditions, and no 
reason has been found for the wide variation. They were always fed at 
the same time and were kept on one tray in individual rearing boxes, 
thus giving practically identical conditions of moisture, temperature, and 
food. In spite of this similarity of conditions we find that larva 2ie 
pupated 62 days after hatching from the egg, while larva 2 id from the 
same parent and from the same egg cluster took 124 days, or just twice 
as long, to reach the same stage of development. The number of instars 
was also found to vary. Thus the number of larval instars for the five 
larvae from each of the moths was as follows: Moth 21 — four had 7 
instars and one had 8; moth D — all five passed through 8 instars; moth 
42 — two had 7 instars and three had 8; moth 24 — one had 7 instars and 



Nov. 5, 192 1 



Pale Western Cutworm 



297 



four had 8. According to the rearing records of other larvae than those 
shown in the table, several individuals passed through 9 larval instars 
and one passed through 10. Eight instars, however, is the usual number 
and the minimum is 7. 

Table III. — Duration of larval instars of Porosagroiis orthogonia under insectary 

conditions 



Larva record No. 



Number of days in each instar. 



IV. 



VII. VIII. 



Num- 
ber of 
days in 
larval 
stage. 



2ia 

2lb 

SIC 

2 id 

2ie 

Da 

Db 

Dc 

Dd 

De 

42a 

42b 

42c 

42d 

426 

Ma. 

Mb 

Mc 

Md 

Me 

Average 



12 
12 

9 
12 

13 
12 
12 

13 
II 
10 

14 
12 

13 
9 

7 
13 
13 

7 



5 
6 
6 
9 
7 
9 
9 
10 

7 
9 
8 

5 

10 
8 
II 
6 
7 
9 
12 

7 



II 

8 

10 

10 

7 

9 

12 

12 

9 
10 
II 

7 

8 

10 

9 
12 
10 



9 
7 
9 
7 
13 
10 

14 
10 
10 

9 
10 
10 
10 

15 

9 

10 

10 



12 
10 
14 
9 
II 
II 
II 

9 

10 

17 
II 

9 
18 
12 
12 

10 

13 

9 

9 

10 



13 
13 
18 

13 
17 
13 
9 
16 
16 
15 
14 
20 

17 
12 

13 
9 
18 
16 
14 
13 



9.4 



9.8 



II- 3 



14.4 



44 
29 

32 
16 

14 
14 
14 
21 
16 
19 
25 
41 
34 
14 
20 



18 
18 
26 



45 



37 
48 

63 

45 
51 
32 



2>Z 
33 
75 
56 
65 
43 



99 

85 

98 

124 

62 

125 
121 

151 
125 
150 
123 
104 
112 
112 
119 
148 

144 
148 

131 
83 



22. 6 



29. 6 



118 



FEEDING HABITS OF LARVA 

The larva differs from that of most cutworms in its feeding habit in 
that it almost invariably attacks the plant below the surface of the 
ground. The most frequent type of injury is the eating away of the 
central stem and its surrounding sheaths from }i inch to i inch below the 
surface of the ground. In many instances the stem is severed and the 
lower part not eaten, and frequently it is only slightly chewed into before 
the worm moves on to another plant. Even slight injury to the under- 
ground portion of the central stem usually results in the death of the 
plant. A very small portion of each plant is actually eaten by this 
cutworm, and its capacity for destruction is thus greatly increased. 
The first indication of injury is the presence of wilting or dried plants 
which can be easily lifted out of the soil without bringing the roots. 
The destruction of grain that is just pushing through the ground is par- 
ticularly rapid, one worm being able to cut off plant after plant in quick 
succession. The worm usually moves along the drill row, taking each 
plant as it goes (PI. 30, D). Where one crop has been destroyed and 



298 Journal of Agricultural Research voi. xxn.No. 6 

the land has been reseeded the worms often attack sprouting grain and 
sometimes even gnaw into the kernels before they germinate. 

The larvae feed during both day and night. Freshly cut plants 
have been found repeatedly during the day and cutworms have been 
found with their heads inside the sheath of the plant in the act of feeding 
on the central stem. In the rearing boxes no difference could be noticed 
between the amount of day and of night feeding. 

According to observations made by Strickland and reported by Gibson 
(7) it is the habit of the larva — 

to travel over the surface of the soil and when a suitable plant for attack is found 
it immediately burrows and feeds just below the surface. 

In our observations we have never vdtnessed this habit. Considerable 
time has been spent during the late afternoon and evening in heavily 
infested fields, and except in rare instances we have never seen Porosa- 
grotis orthogonia larvse above ground. On the other hand, we have 
closely examined hundreds of newly attacked plants where there was no 
sign that the soil had been disturbed at the surface by the burrowing of 
the larva. In order to find out whether the worms would work from 
one plant to another without coming to the surface, wheat seedlings 
were started 3 inches apart in a flat in the greenhouse, and when the 
wheat was up about i inch 12 half-grown cutworms were placed in one 
end of the flat. At the same time a line of plaster of Paris was placed 
across the center of the flat so that any traveling of the worms over the 
surface at the center would be indicated by lines through the plaster of 
Paris. The seedlings at the end of the flat where the worms were intro- 
duced were immediately attacked, and within a week all of the seedlings 
in the flat had been cut off below the surface of the soil and without any 
indication that any of the larvae had crossed the plaster of Paris line. 

If it were the natural habit of Porosagrotis orthogonia to travel over 
the surface of the ground in getting from plant to plant, it would come 
in contact with poisoned bran mash scattered on the ground, and it 
seems as if there would be no difficulty in controlling it by the ordinary 
methods. The fact that poisoned bran mash is useless against Porosa- 
grotis orthogonia, together with our field and insectary observations, leads 
us to believe that it very rarely comes to the surface in getting from one 
plant to another but instead moves underground, generally along the 
drill row. 

We have received occasional reports from farmers stating that pale 
western cutworms had been seen feeding above ground during and im- 
mediately following rains, but only one instance of this kind has come 
under our own observation. This was at Willow Creek on June 27, 
1920. A light thunder shower at 6.30 p. m. cooled the air and wet the 
ground to the depth of i inch. As soon as the storm was over, numer- 
ous cutworms were seen crawling over the surface in a wheat field that 



Nov. 5. 1921 Pale Western Cutworm 299 

was known to be badly infested with Porosagrotis orthogonia. It was 
estimated that about 50 per cent of the total number of cutworms in 
the soil were on the surface at any one time. They appeared in greatest 
numbers in the spots where the grain had previously been cut off. At 
first they merely wandered about over the surface, but later on, as it 
became dark, they started feeding. Observations were continued until 
II p. m., at which time the worms were still moving about and feeding 
upon the leaves of wheat and grass. At 5 a. m. the next morning a 
few worms still remained on the surface, but all disappeared as the 
sun came up. The soil around stray wheat plants was noticeably 
stirred up where the worms had come up and gone down. That this 
habit of feeding above ground is not a common one is shown by the 
fact that only in rare instances have we ever found any injury to that 
part of the grain plant which is above the ground, and in such cases 
there was always some doubt as to whether the injury might not have 
been done by some other species. 

LARV^ DO NOT LEAVE FlElvDS AFTER GRAIN IS DESTROYED 

Another unusual habit of Porosagrotis orthogonia is that it seldom 
migrates even though its food supply becomes exhausted. If, in fol- 
lowing along the drill row, it fails to find a plant within a few feet, it 
simply remains where it is, perhaps for several weeks, without feeding 
or growing to any extent. In fields that are only partially infested the 
injury shows up as scattered bare spots, and in such places the larvae 
do concentrate along the edges of the standing grain, but we have never 
known them to migrate more than a few rods. This habit of remain- 
ing in the fields where grain has been destroyed has a very important 
bearing on farm practice, as will be shown by the following example: 
Eighty acres of winter wheat at Wilsall in 19 19 were completely de- 
stroyed by May i, the ground being left entirely bare. The field was 
reseeded to spring wheat the second week in May. On May 24 the 
grain was just coming through the ground and was being rapidly cut 
off by cutworms which had remained in the field since the winter wheat 
had been destroyed, some three weeks before. The worms continued 
to feed for several weeks and destroyed all the spring wheat. 

The ability to go for a long time without feeding was well shown by 
a half-grown larva which remained in a rearing can for 12 weeks with- 
out food and was then fed and reared to maturity. 

FOOD PLANTS 

In Montana this cutworm has been most commonly found feeding 
upon winter and spring wheat. Oats, barley, rye, flax, and alfalfa 
have also been attacked. In the insectary, larvae have fed readily and 
grown rapidly upon dandelion. In Canada, Gibson reports Porosagrotis 



300 Journal of Agricultural Research voi. xxii no. e, 

orthogonia larvae as feeding upon fall and spring wheat, oats, barley, flax, 
beets, onions, cabbages and carrots. 

PERIOD OF INACTIVITY BEFORE PUPATION 

Although Porosagrotis orthogonia larvae are mature and have prac- 
tically ceased feeding by the middle of June they do not pupate until 
nearly a month later. During this period they occasionally feed 
slightly, but for the most part they remain in a semidormant condition, 
gradually turning whitish in color and shrinking in size just previous 
to pupation. This was noticed both in the field and under insectary 
conditions. Notes taken at Wilsall June 20, 1919, state that on that 
date cutworms were decreasing in numbers and were nearly all full 
grown. This field was visited again on July 4, when many whitish larvae 
were found, some of which had formed earthen cells, but no pupae were 
found in a two-hour search. 

Records kept on 75 larvae in the insectary showed an average period 
of 20 days of complete inactivity previous to pupation and a period 
of 26 days in which only very slight feeding took place. 

PUPAI^ PERIOD 

Pupation generally takes place about the middle of July. Out of 
80 specimens collected as larvae at Wilsall in May, 1919, and reared to 
adults in the insectary, the average date of pupation was July 19, the 
earliest date July 2 and the latest August 1 1 . This checked out almost 
exactly with conditions in the field at Wilsall. 

About a month is spent in the pupal stage. The average length of 
pupal period of 80 specimens was 29^^ days, the shortest 2 1 days and 
the longest 40 days. 

The pupae are protected by a cemented earthen cell and are usually 
found at a depth of 3 to 4 inches in the soil beneath the plants where 
they last fed. 

SEASONAL ABUNDANCE OF ADULTS 

The earliest emergence of Porosagrotis orthogonia moths which we have 
on record is July 31, although Gibson {8, p. 30-ji) reports the emergence 
of a moth of this species on July 19, In general, the period of greatest 
abundance is during the last two weeks in August and the first week in 
September. 

The field at Wilsall where Porosagrotis orthogonia larvae destroyed two 
seedings of wheat during May and June, 19 19, was searched for moths 
on August 7, but none could be found. A trap light was run until mid- 
night on this date, and not a moth of this species was taken. On August 
26 the field was again visited, and numerous moths were found during 
the day, and at night they came to trap lights by the thousands. The 
majority of the moths taken at this time were in prime condition and 



Nov. s, 1921 



Pale Western Cutworm 



301 



looked as though they had just emerged. One week later the number 
of moths was greatly reduced ; the)'' were difficult to find during the day, 
and very few came to lights at night. On September 30 an entire day 
was spent in searching the same field, but not a moth could be found. 
The ovvTuer of the field had been disking and drilling throughout the 
month of September and during the first half of the month had frequently 
seen moths fly up as the ground was disturbed but had seen none after 
September 15. 

Seventy-five larvae collected in this field May i and reared in the 
insectary at Bozeman emerged as adults on the dates shown in Table IV. 
In 1920 at Willow Creek the first moth was caught on August 9, the 
heaviest flight was from August 19 to 24, and several moths were seen as 
late as October 8. 



Table IV. — Dates of emergence of Porosagrofis orthogonia moths in igiQ 



July 31. 
Aug. 4 

5 
6 

7 
8 

9 

10 
12 
13 
14 
IS 
16 



Number 
of moths. 



Aug. 17 

18 
19 

20 

22 

23 
24 

25 
26 
27 
28 
31 

Sept. I. 



Number 
of moths. 



4 

13 

3 

2 

5 
S 
4 
4 
2 
I 
I 
I 
2 



EGG-LAYING HABITS 

Egg laying was first witnessed at Willow Creek in 1920. Several 
gravid females were placed in outdoor observation cages in which a 
variety of soil conditions and vegetation was offered and were closely 
watched for several days. On the afternoon of September 4, at 4.45, 
one of these females was seen laying eggs. She crawled over clods, 
stubble, and plants, constantly feeling with the ovipositor the objects 
which she walked upon. On reaching a patch of soft earth she stopped 
and carefully worked the abdomen into the soil until the wings were flat 
on the ground. After remaining quiet for a short time she moved and 
repeated her actions in another spot. Three ovipositions were made in 
15 minutes, and after each one the dirt was stirred as the abdomen was 
withdra-'vn and the hole left covered with dirt. The dirt around these 
holes was carefully removed with a teaspoon and eggs were found in 
clusters of 3 or 4 about % inch below the surface of the ground. A total 
of II eggs were recovered from the three ovipositions. 



302 Journal of Agricultural Research voi. xxn, no. t 

Porosagrotis orthogonia moths were seen laying eggs in the open at 
Willow Creek on September 5, 1920. Just before dark moths were seen 
flying over a freshly worked, summer-fallowed field, being most abun- 
dant on the higher knolls and along the ridges where the soil was soft 
and loose. One moth was followed for some distance. She would fly a 
few feet, never getting over 10 inches above the ground, and would then 
crawl a short distance, continuously feeling the surface with her ovi- 
positor. On reaching soft dirt she stopped and laid eggs for six minutes, 
going through the same actions as the moth observed in the cage. When 
she left the ground she flew straight away for at least X ^^^ ^t a height 
of 20 to 30 feet above the ground and was finally lost to view. Five eggs 
were recovered from this oviposition. Other moths were seen flying to 
the ridges and knolls, but it was too dark for further observations on this 
date. A few days later another moth was observed laying eggs on a 
knoll in the same field. One oviposition was made which lasted 23 min- 
utes, during which time the moth was not in the least disturbed by any 
movements of the observer. When the ovipositor was finally with- 
drawn the moth swung around X iiich and started in again, this time 
remaining quiet for 1 7 minutes, after which she crawled under a clod to 
hide. This moth had oviposited for 40 minutes, and 12 eggs were re- 
covered from the two holes. Moths were seen ovipositing along the 
knolls and ridges in this field for several days. 

Moths in egg-laying show their preference for spots in the field where 
the soil is softest and also indicate a preference for freshly worked fields 
over those which have become caked and hard. Across the road from 
the freshly worked, summer-fallowed fields in which egg laying was 
observed was another summer-fallowed field which was spotted with 
Russian thistles and in which the soil was caked on the surface, due to a 
rain some two weeks earlier. Moths were continually observed flying 
into this field, but they usually flew on across it to the knolls in the 
freshly worked field, even though it was ]4 mile farther. Very few 
moths flew to similar knolls in the caked field, and those that alighted hid 
under the thistles or clods of dirt and made no attempt to lay eggs. 
Further evidence of this preference for mellow fields will be brought out 
later in this paper. 

ATTRACTION OF THE MOTHS TO LIGHTS 

Our first experiments in attracting the moths to lights were conducted 
at Wilsall on the evening of August 26, 19 19. A large Coleman gas lamp 
was placed on the ground in the field where the grain had been destroyed 
the previous spring. As soon as it grew dark Porosagrotis orthogonia 
moths began to come to the light at the rate of one every two or three 
minutes. The lamp was placed upon bare sandy soil and the ground was 
well lighted for several feet on all sides. The moths usually struck the 



Nov. 5, 1921 Pale Western Cutworm 303 

ground from 2 to 15 feet from the light and then crawled toward it, 
where they could be easily picked up. As soon as it became totally dark 
the moths came to the light so rapidly that two men could not keep them 
picked up, and from 9 o'clock until midnight 282 females and 164 males 
of P. orthogonia were placed in rearing cans or killing bottles. This by 
no means represented the total number that came to the light, for hun- 
dreds escaped. Many different species of noctuids were attracted to the 
light, but fully 95 per cent were P. orthogonia. Ten of the females thus 
captured averaged 132 eggs. (Table II.) 

In another part of the same field a Duro moth trap was run through- 
out the nights of August 26 and 27, 19 19, and each morning the pan was 
well filled with Porosagrotis orthogonia moths. During the two nights 
4,900 moths were caught, of which 4,200 were males. It is difficult to 
understand the preponderance of females caught at the larger light 
between 9 and 12 p. m. and the very small percentage of females caught 
at the smaller light during the entire night. 

Experiments with trap lights were conducted on a somewhat larger 
scale at Willow Creek. A trap was designed which was made up of 
utensils commonly found on every farm and which would serve other 
purposes when not in use as a trap light (PI. 30, B). It consists of a 
No. 2 galvanized-iron washtub and a No. 2 barn lantern. In addition, 
a galvanized-iron arch is made which fits across the tub and serves the 
dual purpose of deflecting the moths and holding the lantern. When 
the arch is wired firmly and the lantern swung in place the flame of 
the lantern is just above the edge of the tub. "W^Tien set in place, the 
tub is staked down to prevent its being blown over, and about 4 inches 
of water are poured into it. About }i to }{ inch of kerosene is floated 
on the water to kill the moths which fall into it. 

Eleven such traps were put out at Willow Creek, and observations 
were made during the flight period of Porosagrotis orthogonia. Two 
traps for catching moths alive were also used, and when these showed 
that P. orthogonia was beginning to fly the tub traps were put out on 
fields that had been heavily infested with worms. 

During the first few nights the catches were small and the moths 
were counted. The numbers increased nightly until the height of 
flight, which was from August 19 to 24, inclusive. The night flight 
gradually decreased after the latter date. When the numbers became 
too large to count they were estimated, and during the height of flight 
they were measured in pints. As a pint measure holds from 962 to 1,000 
moths, the measuring of moths gave a fairly accurate count. During 
the heavy flight several of the traps ran as high as 4,000 moths in a 
single night. The entire season 's catch of Porosagrotis orthogonia moths 
in the 11 traps was 82,488. The catch on individual nights is shown in 
Table V. 



304 



Journal of Agricultural Research 



Vol. XXII, No. 6 



Table V. — Number of Porosagrotis orthogonia moths caught at trap lights at Willow 
Creek, Mont., during the season oj igzo 



Aug. 



Sept. 



13 
14 
IS 
16 

17 
18 

19 



22 

23 
24 

25 

26 

27 
28 
29 
30 
31 



Weather conditions and remarks. 



Number 
of moths 
caught. 



127 

342 

1,528 

1,566 

449 



68 



11,720 



Rain in afternoon; night warm, cooler toward midnight 

Early part of night warmer than usual; warm after midnight. . . 

Warm all night 

Night moderately warm 

Cooler, especially so after 9.30 p. m. ; not as many moths flying . 
High wind in afternoon blew over traps; cold and windy after 

dark ; no moths out 

Windy; cold soon after dark, almost frost; moths flew only a 

few minutes 

Warm at 8 p. m.; moths flying heavily; windy and cooler after 

ir.30; fewer moths out 

Warm west wind most of night; heavy flight of moths 13 , 990 

Same as night before; height of flight 9.15 p. m 14.950 

Warm, light west wind 13 , 650 

Warm, west wind; wind strong and cold after 12.30 a. m 16,250 

Cold and cloudy after 9 p. m. ; southwest wind ! 7 , 490 

Windy and cold ; moonlight; no moths flying ' 

Windy until 8 p. m.; clear, cold; full moon; no moths flying. 

Wind and heavy rain ; cold ; no moths 

Rain all morning; cold, windy night; traps not lighted ! 

Cool; little wind; few moths flying early; too cold after 9 p.m. . . 68 

Clear, cool; bright moon; no moths flying 

Cloudy to 9.30; clear, cold; bright moon 

Moths flying in daylight after noon; few flying after dark 

No moths flying after dark; a few found feeding 

Few moths out after dark ; not attracted to lights 

Three Nociua c-nigrum caught; no others flying to light 

No moths flying to traps; traps taken up 



119 

171 



Total. 



82,488 



In all observations made at Willow Creek no moth was ever seen to 
land on the ground on its way to a light trap as did the moths at Wilsall 
the previous year. This may have been due to the fact that the tub 
hid the light so it would not strike the ground, and in order to keep in 
the path of light the moth had to fly straight to the trap. This was 
usually the case, and for the most part moths flying to the traps came 
on a straight line from 4 to 15 feet above the ground. They either struck 
the arch or lantern or went straight on over the trap. 

On a still, dark, fairly warm night the moths would come to the traps 
in varying waves of abimdance for which there was no apparent reason. 
There would be a cloud of moths for a few minutes and then they would 
come in scattering two's or three's. If the wind was strong no moths 
were caught in the traps and no moths could be found moving about on 
the ground. No moths were caught during a rain or ever after a rain, as 
long as the ground and vegetation remained wet. When the moon 
was bright, moths were not caught nor were any seen flying, though they 
would start the minute the moon went behind a cloud. Practically no 
moths were caught after the temperature had dropped below 58° F. 



Nov. 5, 1921 Pale Western Cutworm 305 

MOTHS BOTH NOCTURNAL/ AND DIURNAL 

During the last half of August when the nights were warm and night 
flying was at its height the moths remained inactive during the day, 
hiding under clods and weeds. As the nights grew colder, the moths 
flew only an hour or two after dark, and on September i they were seen 
flying during the day. On this date they began flying about 4.30 p. m. 
and were seen in abundance feeding upon sunflowers, golden rod, 
tumbling mustard, yellow greasewood and lamb's quarter. All but one 
of the moths seen at flowers at this time were males, a search under 
weeds and clods at the same time revealing only females. At 8 p. m. 
when the flower patches were visited moths were still feeding in large 
numbers, practically all of them being females. They paid no attention 
to lights, and none were caught in a trap light set close by. As the night 
grew colder all of the moths disappeared and could be found hiding under 
clods or weeds. On the following day moths were found feeding at 
flowers at i p. m., and at 3.30 p. m. a patch of yellow greasewood 
{Chrysanthus jrigidus), which seemed to be the favorite flower, had 
attracted dozens of Porosagrotis orthogonia, nearly all of which were 
males. At 5.30 p. m. the patch was again visited, and it was found that 
the males were then leaving and that females were flying to the flowers 
from a nearby summer-fallowed field. On September 3 moths of various 
species were found feeding during the morning, and at noon the flowers 
of the yellow greasewood were covered with moths, none of which were 
P. orthogonia. At 3 p. m. about 10 per cent of the moths present at 
flowers were P. orthogonia. These gradually increased in numbers until 
6.15 p. m., at which time practically all other species had disappeared. 
After the moths had finished feeding they invariably flew toward the 
higher ridges and knolls in neighboring cultivated fields. Many of the 
moths in coming to the flowers were seen to fly from 200 to 500 yards 
directly against a stiff breeze. Moths were seen flying to flowers in large 
numbers until September 8, when a cold rain and wind occurred. The 
males were always found feeding earlier in the day and the females later, 
although neither was ever found before noon. 

ECONOMIC IMPORTANCE 

The record of this cutworm during the last 10 years has demonstrated 
its capacity for doing enormous damage to grain crops. When in 191 1 
Porosagrotis orthogonia, then an obscure insect, suddenly increased in 
numbers and did considerable damage to grain in southern Alberta {10, p. 
iji) little importance was attached to it. In the following year, however, 
when 33 per cent of all the grain sown in the Lethbridge land district 
was destroyed and an accurate estimate by the superintendent of the 
experiment station at Lethbridge placed the actual loss from this insect 
at from 30,000 to 35,000 acres (jj, p. 506) it was looked upon as a pest of 

; 65769°— 21 2 



3o6 Journal of Agricultural Research voi. xxn, no. 6 

major importance. During the last two years P. orthogonia has been re- 
sponsible for losses in central Canada amounting to several million dollars. 

In Montana the pale western cutworm has been on the increase since 
it was first noticed in 19 15 and is now the most destructive insect pest 
with which the grain grower has to contend. In 19 15 at Conrad 80 acres 
of wheat were destroyed and were reseeded to oats, which was also taken. 
Flax was then seeded, but this also was so badly injured that the owner 
plowed the field and summer fallowed it. This instance was typical of 
scores of losses in the district now composed of the counties of Chouteau, 
Teton, and Pondera. 

During the next two years, especially in 19 17, great losses were sus- 
tained throughout the north central portion of the State, due to cut- 
worms which worked entirely beneath the surface of the soil and which 
were doubtless no other than Porosagrotis orthogonia. In 1919 the pale 
western cutworm appeared in destructive numbers farther south and 
caused severe losses in Park and Jefferson Counties as well as in the 
previously infested area. A conservative estimate of the losses for the 
year based on the reports of county agents, hundreds of questionnaires 
returned by farmers, and the personal investigation of many fields in 
different parts of the State, is at least 200,000 acres. In 1920 the injury 
in Jefferson and Park Counties was more widely extended, and there was 
a decided increase in the damage done in many of the districts previously 
infested. The loss over the entire State for the year is placed at 250,000 
acres, valued at $3,000,000. In the Willow Creek district in Jefferson 
County a careful survey conducted in 1920 showed that 29 per cent of 
the total seeded area had been destroyed by this cutworm, and a similar 
survey in several of the northern counties showed a loss of 35 per cent of 
the grain crops planted. 

To show perhaps a little more clearly what this cutworm has been 
doing it may be stated that 100 fields personally inspected during the 
summer showed a loss of 2,437 acres out of a total of 6,844 ^^ 1919, and 
in 1920 a loss of 3,382 acres out of 6,844, or 35.7 per cent in 1919 and 
49.4 per cent in 1920. Mr. George O. Sanford, manager of the Sun River 
irrigation project, has stated to us that of the 15,300 acres seeded to crop 
on the Greenfield Bench in 1920, 7,345 acres was a total loss and that 
some damage was done to the remainder. Using the figures he has given 
for the average yields on the undestroyed acreage — wheat 11.5 bushels, 
oats 20.86 bushels, and flax 6.31 bushels — the average value of tlie prin- 
cipal farm crops of that section was at least $15 per acre. Accordingly, 
using that as a fair valuation per acre of the crops destroyed, the pale 
western cutworm inflicted a loss of $110,175 in this one comparatively 
small territory. Although these losses took place on irrigable land, no 
water was available until after the first of June. Were it not that irri- 
gation made it possible in some cases to reseed and grow a late crop on 
part of the originally destroyed area, the loss would have been 55 per 
cent instead of 48 per cent of the acreage in that district. 



Nov. 5. 1921 Pale Western Cutworm 307 

EXPERIMENTS IN CONTROL 
ORDINARY CUTWORM CONTROI, METHODS NOT EFI^ECTIVE 

Early in our study of Porosagrotis orthogonia it became apparent that 
the ordinary method of scattering poisoned bran mash over an infested 
field was not effective in controlling this species. On May 2, 19 19, pois- 
oned bran mash was scattered over a heavily infested field in southern 
Montana at the rate of 20 pounds to the acre and was followed by three 
other applications on successive days. The field was exammed each 
day by the owner, who reported that he could not find a single dead worm. 
On May 7 the treated area was carefully examined by one of the writers 
but no dead worms could be found, nor could any decrease in the number 
of live worms be noted. On May 7, 19 19, poisoned bran mash was scat- 
tered over 2 acres of heavily infested wheat in northern Montana. During 
the next 10 days no results whatever were secured from this treatment. 
County agents and numerous farmers have reported that attempts to 
poison this species by the ordinary method of scattering poisoned bran 
mash over the surface have always resulted in failure. 

At Willow Creek in 1920 pale western cutworms were noticed crawling 
over the surface of the groimd in the evening after a rain, and an attempt 
was made to kill them by scattering poisoned bran mash during the 
night. The bait was scattered soon after dark over an area which in- 
cluded bare ground, scattering wheat, and a good stand of wheat, all 
heavily infested. Observations were made during the night by the aid 
of automobile headlights, and many of the worms were seen feeding upon 
the bait. Two days later a search was made for dead cutworms. In 
the area where there was no vegetation it was estimated that 60 per 
cent of the worms were killed; where there was a scattering of wheat 
the percentage of dead worms was 50; and where there was a good stand 
of wheat 43 per cent were killed. It is possible that several night appli- 
cations of poisoned bran mash during rainy weather might bring about a 
satisfactory control, but as yet we have not had the opportunity to 
try it. 

POISONED BRAN MASH HARROWED INTO THE SOU. 

Strickland reports {17) that poisoned bran mash harrowed into the 
soil gave gratifying results. This method was tried out at Wilsall in 
May, 19 19. Poisoned bran mash was scattered over 3^ acre of heavily 
infested wheat at the rate of 25 pounds to the acre. On several square 
rods where the worms were thickest the mash was worked well into the 
soil with a hand rake, and the remainder of the treated area was thor- 
oughly worked with a spike-toothed harrow. The plot was carefully 
examined three days after the poisoned bran mash was applied, and it 
was estimated that the treatment was not more than i per cent effective. 
Very few dead cutworms could be found, and eventually all of the wheat 
was destroyed. 



3o8 Journal of Agricultural Research voi. xxn.No.6 

POISONED BRAN MASH DRII,IvED INTO THE SOII# 

Since Porosagrotis orthogonia very rarely comes to the surface to feed, 
placing the poisoned bran mash beneath the soil was tried in the hope 
that the cutworms would thus come in contact with it and feed upon it. 
The most promising way of doing this seemed to be with a seed drill. 
This method was tried out in northern Montana at Havre and in southern 
Montana at Wilsall. 

TESTS AT HAVRE 

At Havre two formulae were used. 

FORMULA NO. I 

Shorts pounds . . 25 

Paris green do ... . i 

Oranges 4 

Molasses quarts . . 2 

Water gallon. . i 

FORMULA NO. 3 

Shorts pounds. . 25 

Paris green do ... . i 

Molasses gallon . . i 

Water quarts . . 2 

These mixtures after being prepared were spread out and allowed to 
dry for 24 hours. When dry, No. 2 was distinctly stronger smelling, 
although both had a good molasses odor. The reason for using the 
large amount of molasses in these formulae w^as to secm-e a distinct odor 
in the dried material which we hoped might attract cutworms in the 
soil for some little distance. 

The dried material was seeded into the ground with a Van Brunt 
drill at the rate of 16 pounds to the acre and at a depth of about 1.5 
inches. Six acres were ti"eated. The drill was run at right angles to the 
rows of grain so that the worms in working from plant to plant would 
only move a few inches before coming in contact with the bran. The 
greatest difficulty encountered was in getting the bran to feed evenly 
through the drill. When it was sufficiently dr}^ to be well divided it 
was too light to force its way through and it was necessary to agitate 
the mixture continuously in the seeder box to get any-where near an 
even distribution. 

The field was examined two days after the poisoned bran mash was 
drilled in, and it was found that formula No. i had killed approximately 
50 per cent of the worms while formula No. 2 gave slightly better results 
with a kill of about 55 per cent, which was not enough to prevent the 
destruction of the crop. 



Nov. 5. 1921 Pale Western Cutworm 309 

TESTS AT WILSALt 

On June 8, 19 19, a similar test was conducted at Wilsall. The follow- 
ing formula was used : 

Shorts pounds. . 25 

Paris green do .... ^ 

Salt :':v^...v;?..';:.:;v.r. : do % 

Molasses quarts . . 2 

Water. gallon . . i 

After mixing, the mash was spread out to dry, which with a hot sun 
and a fair breeze was accomplished in half a day. The mixture was 
distributed over 25 acres of infested wheat at the rate of 12 pounds of 
the dry mash to the acre. Sixteen acres were sown with all the spouts 
of the drill working and 9 acres with every other one closed. The drill 
was run across the old grain rows. The greatest difficulty encountered 
was the same as in the test at Havre — the mixture was too light to feed 
evenly through the drill. This was overcome by using two men on 
the drill, one to drive and one to keep the bran shaken down where it 
would come in contact with the disks of the drill. This was done by 
frequently pounding the seeder box with a padded hammer and punch- 
ing out packed masses with a small stick. Dead and dying worms were 
found the second day after the poisoned bran was drilled in, and on the 
third day a careful examination was made and it was estimated that 
from 50 to 60 per cent of the worms had been killed. The field was 
examined two weeks later, and there was a very noticeable difference in 
the number of worms found in the treated and untreated areas, but this 
did not prevent total destruction of the crop. 

TEST AT WIL1.0W CREEK 

During May, 1920, poisoned bran mash was distributed with a grain 
drill over a very badly infested field at Willow Creek. Cutworms were 
uniformly scattered over a 40-acre field of spring wheat, and at the time 
the poisoned bran mash was applied had destroyed about half the plants. 
Conditions were ideal for a good test of control methods. The following 
mixtures were used : 

FORMUI,A NO. I 

Shorts pounds. . 25 

Paris green do ... . i 

Molasses quarts.. 2 

Salt pound. . i 

Water gallon . . i 

After mixing, this was thoroughly dried out and was then seeded 2 
inches deep through a Van Brunt drill at the rate of 20 pounds to the 
acre. Two acres were sown. This mixture did not feed uniformly 
through the drill unless constantly agitated. 



310 journal of Agricultural Research voi. xxu. no.6 

FORMULA NO. 2 

Shorts pounds . . 25 

White arsenic ■ do ... . i>^ 

Molasses quarts. . 2 

Salt pound . . i 

This was prepared and distributed in the same manner as formula 
No. I. Two acres were sown. 

FORMULA NO. 3 

Shorts pounds . . 25 

Paris green do .... i 

Salt do .... I 

This was mixed dry and seeded at the rate of 12^^ pounds to the 
acre. It ran through the drill about the same as the mixtures which 
were mixed wet and then dried. Two acres were sown. 

FORMULA NO. 4 

Shorts pounds . . 25 

White arsenic ; .* ! .'.'.. 1 do ... . i}4 

Salt do ... . I 

This was prepared dry and then thoroughly mixed with an equal 
bulk of wheat. This combination ran through the drill very evenly, 
the wheat being heavy enough to carry the bran through the drill without 
clogging. Two acres were seeded at the rate oi \2}i pounds of bran to 
the acre. Three days after the poisoned bran was put out the field was 
examined and it was estimated that about 10 per cent of the worms 
in the treated areas had been killed. No difference could be seen in the 
effectiveness of the various formulae, and numerous living cutworms 
remained in all the plots. One week later the plots were again examined 
and the number of dead cutworms had not materially increased. A 
final examination of the field was made on June 14, three weeks after the 
poisoned bran was put out. Cutworms were found in abundance on all 
plots, and in plots 1,2, and 3 practically every spear of wheat had dis- 
appeared. In plot 4, which was seeded with a mixture of wheat and 
poisoned bran, the wheat was about 3 inches in height and was being 
rapidly cut off, 50 per cent of the new stand being already destroyed. 
From a practical standpoint the control on all plots was a complete 
failure and an absolute waste of materials. 

POISONED BAIT SPRAY FOR ADUI,TS 

The presence of large numbers of Porosagrotis orthogonia moths at 
flowers led us to try out the following poisoned bait spray: 

Water gallon . . i 

Molasses pint . . ^ 

White arsenic ounce . . }4 

Amyl acetate do % 



Nov. 5, 1921 



Pale Western Cutworm 



311 



This was scattered in coarse droplets over flowers and vegetation 
where moths were abundant. Many flies and bees were killed, but no 
moths were observed feeding upon the bait, and dead moths were never 
found in the vicinity of the sprayed vegetation. 

CUI^TURAI, METHODS AS A MEANS OF CONTROI, 

In our study of Porosagrotis orthogonia under field conditions we have 
repeatedly noticed instances where crops in one field were completely 
destroyed, while in an adjacent field the grain escaped unharmed. This 
suggested that the manner in which the ground was worked before the 
crop was put in might have been responsible for the great difference in 
the amount of damage done in the two fields, and in 1920 a survey was 
conducted with the object of determining the relation of cultural methods 
to cutworm abundance. This survey was conducted in two ways : (1) By 
an auto trip through the districts most heavily infested by means of which 
hundreds of farmers were personally interviewed and the histories of 
their fields obtained for the period 1919-20; (2) by questionnaires sent 
to all farm bureau members in counties where Porosagrotis orthogonia was 
known to be present. 

The percentage of cutworm losses under various cultural methods as 
shown by a study of fields, the owners of which were personally inter- 
viewed, is shown in Table VI. 

Table Vl. ^Percentage of Porosagrotis orthogonia injury in IQ20 under various methods 
of cultivation in preparation for seeding 



Cultivation between previous crop and 1920 crop. 



Number of 


Total 


Acres 


fields. 


acres. 


lost. 


8 


465 


200 


39 


2,250 


1,301 


36 


1.536 


661 


13 


643 


138 


51 


2,465 


666 


18 


1.332 


526 


39 


3. "4 


267 



Percentage 
lost. 



Fall double disked 

Spring double disked 

Spring single disked 

Fall-plowed ; disked or harrowed be 'ore seed- 
ing 

Spring-plowed" disked or harrowed before 
seeding 

Spring-harrowed 

Summer-fallowed 



43- o 
57- o 
43- o 

21. o 

27. o 
40. o 

8.5 



A study of the results shows a high percentage of cutworm injury in 
all cases where the stubble was only disked or harrowed before seeding. 
Fields which were plowed either in the fall or spring showed a somewhat 
lower percentage, while summer-fallowed fields showed only the very 
small loss of 8.5 per cent. 

While the average cutworm loss in summer-fallowed fields was low, 
yet several individual fields suffered severe losses. It was therefore 
decided to make a study of the histories of summer-fallowed fields during 
the two seasons of 19 19 and 1920. Since the moths were known to 



312 



Journal of Agricultural Research voi. xxii, no. 6 



prefer loose mellow soil for egg-laying, it was thought that the condition 
of the ground in summer-fallowed fields during the egg-laying period 
might have considerable rafluence on the number of eggs deposited in 
the field and on the percentage of loss the following spring. Since egg 
laying does not begin until about August 15, fields which are not culti- 
vated or disturbed in any way after July 1 5 become more or less crusted 
and caked. Fields which are cultivated in any way during the last part 
of July or during August, on the other hand, are very likely to be soft 
and mellow during the egg-laying period, thus offering tlie very con- 
ditions which the moths are seeking. Forty-eight fields, for which we 
had data for both 19 19 and 1920, were therefore classified as crusted, if 
they were worked only before July 15, or as mellow, if they had been 
worked after that date. The percentage of loss for the variously worked 
fields is shown in Table VII. 

Table VII. — Percentage of Porosagrotis orthogonia injury during IQIQ and IQ20 in 
"crusted" and "mellow" sumtTier-f allowed fields. Data secured by personal interview 
with grower 



Condition of field and time of cultivation. 


Number of 
fields. 


Total 
acres. 


Number of 

fields 
infested. 


Acres 
lost. 


Percentage 
lost. 


"Crusted" — worked only before 
July 15 


27 
21 


1,828 
1,562 


3 
14 


14 

425 


00. 7 
27. 2 


"Mellow" — worked after July 15. . 



Farmers were asked in questionnaires sent to farm bureau members 
in counties infested with Porosagrotis orthogonia whether they had 
noticed any relation between the condition of the soil in summer-fallowed 
fields during August and the amount of pale western cutworm injury 
the following spring. Sixty-eight grain growers answered this question. 
Fifty-three said that injury was most severe in fields where the surface 
soil was well pulverized, or, as one farmer stated it, "The more mulch 
the more worms." Seven reported that the greatest injury had oc- 
curred in fields that had been crusted during August, and five stated 
that they could see no relation between soil conditions and cutworm 
injury. 

The foregoing data, together with the fact that we have seen ovi- 
positing females show a distinct preference for mellow fields, leads us to 
the conclusion that the physical condition of the soil during the egg- 
laying period has a very important bearing upon the amount of Poro- 
sagrotis orthogonia injury that may occur the following spring. Accord- 
ing to the data at hand greatest injury may be expected in fields in 
which the surface soil is loose and well pulverized during the egg-laying 
period. This loose, mellow condition may have been brought about in 
summer-fallowed fields by tillage during late July and August or it may 



Nov. 5. I92I Pale Western Cutworm 313 

be a natural condition such as is found on knolls and ridges where the 
soil is generally light and easily drifted. In fields where a crop is re- 
moved during July or August the surface crust may become broken and 
pulverized in numerous places by the disturbance of the soil in connec- 
tion with harvesting, thus offering the moths many desirable spots for 
egg laying. Injury may be least expected to occur in fields in which the 
surface soil is hard or crusted during the egg-laying period. In most 
instances this condition can be brought about by not disturbing the 
ground in any way between July 15 and September 15. If farmers in 
preparing their grain fields for seeding will be governed by these prin- 
ciples it is believed that Porosagroiis orthogonia injury can be greatly 
reduced. Fortunately this method of handling summer-fallowed fields 
does not interfere with approved farm practices, and in fact agrees very 
closely with the recommendations of the agronomists. 

NATURAL ENEMIES 

Unlike most of our common cutworms, Porosagroiis orthogonia suffers 
comparatively little from attack by natural enemies. Much difficulty 
has been encountered in rearing various other species taken in the field 
as larvae, particularly the army cutworm, Chorizagrotis auxiliaris, on 
account of the high percentage that developed disease or parasites. 
This has not been the case with the present species. Our records for 
19 15 show that out of a large number of army cutworms reared indi- 
vidually only 35 per cent were brought through to the moth stage, 
parasites emerged from 24 per cent, 2 1 per cent died of disease, and the 
remaining 20 per cent died in the pupa stage, mostly on account of in- 
sect parasites. In 1919, 55 per cent of P. orthogonia larvse handled 
in the same way were reared to adults. Of the 45 per cent that died, 
very few seemed to die of any disease, and parasites emerged from only 
two larvae. 

In 1920, out of 960 Porosagroiis orthogonia larvae collected in the 
field, 13.7 per cent were parasitized, 12.2 per cent by Diptera and 1.5 
per cent by Hymenoptera. The parasites which emerged were 14 Bon- 
netia compta Fall and i Peleteria robusta Wied. 

The common wild birds of the prairie are the most beneficial natural 
check that we have observed. The western grasshopper sparrow, Am- 
modramus savannarum bimaculatus Swainson, particularly, has been 
watched while digging out the larvae and carrying them away to its 
young. In many parts of the cutworm-infested regions it has been a 
common sight toward the last of June to see thousands of small excava- 
tions made by the western grasshopper sparrow, horned larks Otocoris 
alpestris leiicolaema Coues (PI. 30, C), and possibly other wild birds in 
their search for the larvae. 

Although the common ground squirrel, Citellus richardsoni Sabine, has 
at times been known to seek out and devour large numbers of cutworm 




314 Journal of Agricultural Research voi. xxii, no. 6 

larvae, we do not believe that ground squirrels are of much importance 
as a natural check. 

In some instances both larvae and adults of Calosoma teptdum Lee. 
have been observed to be especially numerous about cutworm-infested 
fields and are, we believe, one of the lesser important predators. 

DESCRIPTION OF STAGES 
EGG 

Spheroidal, flattened dorso-ventrally, glistening milk-white when first 

laid (PI. C, i), later becoming dull gray: i mm. in diameter, 8 mm. in 

height. Around the micropyle is the 

■. . v.. .i. ..•,•■■•:••., .' usual rosette which lies in the cen- 

'/C ^:--^r'-^'''Vi'.'''f ■■(■''''■■.•' ter of a finely reticulated area about 

. •. .-• 0.3 mm. in diameter. The pattern of 

■•, '• the reticulation is shown in figure i. 

From the edge of the reticulated area 

]!...;.... about 30 slightly raised longitudinal 

■/••.'.?■• ribs radiate toward the base, extend- 

fi. •' " ■ . 

■.■."'^:;{.. ing approximately four-fifths the dis- 

'"'/""■■/•■■■■■■} •! V '■^..•"■•. tance from the apex to the base. 

"■•■■■'••■.••....• ■;'■"■ The ribs are sometimes irregularly 

branched or connected by cross 

^^u'Z?'"'''''"^"''", "'■'A'^^'?!?'"-" ""'^^f'"'' ridges. The shallow channels be- 
about the micropyie. Highly magnified. " 

tween the ribs are transversely 
striated and lightly pitted. The chorion on the basal fifth of the egg 
is smooth and glistening without ridges or definite reticulation. 

I^ARVA ^ 
FIRST INSTAR 

Head width, 0.4 to 0.43 mm.; average 0.41 mm. 

Total length of body, 2.4 to 3 mm.; average 2.8 mm. 

The head (PI. 28, A) is a very dark glistening brown, almost black; 
clypeus and front same color as epicranium; adfrontals indistinct, but 
apparently extending almost to deep indentations at summit; mandibles, 
labrum, ring-shaped sclerite resting on mentum, and cardo dark brown; 
submentum, margins of stipes next the mentum, the antennae, and labial 
palpae brownish; ©cellar region blackish. 

The posterior portion of the thoracic shield dark brown; anterior 
margin very light brown. The thoracic legs are brownish with dark 
brown transverse lines immediately anterior and partially surrounding 
coxae. 

■ In the description of the larva the naming of the various parts follows very largely the system of William 
T. M. Forbes U). 



Nov. 5, 1931 Pale Western Cutworm 315 

Prolegs on segments 9, 10, and 13 (counting the head as the first seg- 
ment); crotchets brown; rudimentary prolegs slightly visible as small 
tubercles on segment 8. 

There is a reddish brown mottling over the lateral and dorsal regions, 
becoming more distinct along the posterior segments. Before the larva 
has taken food this coloration appears much darker. The general color 
of the newly hatched larva is brown. The spiracles dark brown, setae 
single and arising from minute brownish tubercles. 

SECOND INSTAR 

Head width, 0.47 to 0.576 mm.; average 0.53 mm. 

Total length of body, 2.8 to 3.77 mm.; average 3.3 mm. 

The head (PL 28, B) remains a dark, shiny brown; mandibles and 
labrum dark brown, mandibles at teeth and labrum at notch blackish; 
ring-shaped sclerite resting on mentum blackened ventrally and with its 
setae within the sclerite; cardo dark brown; submentum, margins of 
stipes next the mentum, antennae, and labial palpse brownish; ocellar 
region blackish. 

Thoracic shield dark brown posteriorly, but not as dark as head; 
anterior margin light brown. On each side of the light brown dorsal 
stripe is a dark spot on the anterior portion of the shield with several 
dark spots laterad. 

Prolegs on segments 8, 9, 10, and 13; crotchets on eighth segment con- 
sisting of only two or three hooks; rudimentary prolegs on segment 7 
beginning to show; crotchets reddish brown. 

The dorsal stripe is gray, bordered with broken lines of brown; sub- 
dorsal and lateral stripes brownish; spiracles dark brown; setae single, 
tubercles brownish. 

THIRD INSTAR 

Head width, 0.68 to 0.786 mm.; average 0.75 mm. 

Total length of body, 4.19 to 5.5 mm.; average 4.29 mm. 

Head brown (PI. 28, C) with upper parts of lobes of epicranium dark 
brown; front brown but with its lower margin together with clypeus 
dark brown; mandibles reddish brown to black at teeth; labrum dark 
brown with blackened notch, ring-shaped sclerite resting on mentum 
blackened ventrally and with its setae within sclerite ; cardo dark brown ; 
submentum, margins of stipes next the mentum, antennae and labial 
palpae brownish; ocellar region blackish. 

Thoracic shield dark brown, almost black at posterior region; an- 
teriorly it is a lighter brownish gray with a small black spot on either 
side of the light dorsal stripe and with several dark spots laterad. 

Thoracic legs tinged with brown, claws and markings anterior to and 
partially surrounding basal joints reddish brown. 



31 6 Journal of Agricultural Research voi. xxii, no. 6 

The dorsal stripe along the body is made of broken gray which is 
bordered irregularly with brown. Subdorsal and lateral stripes brownish; 
general color same as dorsal stripe, or lighter, with a greenish tinge. 
Spiracles dark brown; setae single, tubercles greenish brown. 

FOtTRTH INSTAR 

Head width, 0.84 to 1.14 mm.; average 1.02 mm. 

Total length of body, 6.5 to 12.5 mm.; average 8.9 mm. 

Coloration of head slightly modified from preceding instar; front not 
as dark, clypeus a lighter brown, and cranial lobes considerably darker 
at top (PI. 28, D) ; mandibles are black at teeth and fade to reddish 
brown to dark brown near articulations; labrum dark brown with 
blackened notch; ring-shaped sclerite resting on mentum blackened 
ventrally and with its setae within the sclerite; cardo dark brown; 
submentum, margins of stipes next the mentum, antennae and labial 
palpae brownish; ocellar region blackish. 

Posterior part of thoracic shield dark brown; anteriorly it is lighter, 
and in this lighter area are three distinct dark spots on either side of the 
dorsal line and also larger dark spots toward the lateral ends of the 
shield. 

Thoracic legs tinged with brown, claws and markings anterior to and 
partially surrounding basal joints, reddish brown. 

Prolegs on segments 8, 9, 10, and 13 and rudimentary prolegs on 
segment 7; crotchets reddish brown. 

The dorsal line is a greenish gray partially broken and bordered with 
brown; subdorsal and lateral lines brownish; general color green to 
gray; spiracles dark brown; tubercles are greenish, setae single and 
ringed at base with a light color. 

FIFTH INSTAR 

Head width, 1.38 mm. to 1.98 mm.; average 1.83 mm. 

Total length of body, 11. 5 mm. to 18.0 mm.; average 16.1 mm. 

General color of head much lighter (PI. 28, E) ; ocellar region very 
dark brown; ocelli i, 2, and 6 colorless, other three dark; mandibles 
reddish brown to black at teeth; lower margin of labrum reddish, 
blackened at notch; cardo and submentum bro^vn with the margins of 
stipes next the mentum same color and with the stripe becoming wider 
about the base of the palpifer; sclerite resting on mentum blackened. 

The two bands of dark brown on the epicranium and bordering the 
adfrontals become prominent for the first time in this instar (PI. C, 2). 
The rest of the head seems to have lost color, leaving these two stripes 
which run from points even with the base of the clypeus to the second 
epicranial setae, above which they gradually fade out about the first 
epicranial setae. 



Nov. 5,1921 Pale Western Cutworm 317 

The thoracic shield is brownish with a light dorsal stripe ; in the lighter 
area on the anterior margin of the shield and on either side of the dorsal 
stripe are distinct blackened spots, with other dark spots toward the 
lateral margins of the shield. 

Thoracic legs tinged with brown, claws and markings anterior to and 
partially surrounding basal joints reddish brown. 

Fully developed prolegs on segments 7, 8, 9, 10, and 13 are concolorous 
with body; crotchets brownish. The anal plate is marked with a trans- 
verse row of small brownish spots anterior to the setae. 

The dorsal line is a greenish gray and bordered with brown; the sub- 
dorsal and lateral stripes brownish; the general color is about the same 
or perhaps a trifle lighter than the dorsal stripe. 

SIXTH INSTAR 

Head width, 1.98 to 2.64 mm.; average 2.41 mm. 

Total length of body, 2.0 to 2.5 cm.; average 2.2 cm. 

Head light brown; ocellar region very dark brown; ocelli i, 2, and 6 
transparent, other three brownish; mandibles reddish brown to black 
at teeth; lower margin of labrum reddish, blackened at notch; cardo and 
submentum brown with the margins of stipes next the mentum same 
color with the stripe broadening apically about the base of the palpifer. 

The two conspicuous bands of dark brown persist on the epicranium 
bordering the adfrontals (PI. 28, F); the stripes become a lighter color 
at top, ending near the first epicranial setae; a denser colored portion of 
each band follows the adfrontals almost to their apex. 

The thoracic shield is brownish with a light dorsal stripe ; in the lighter 
area on the anterior margin of the shield and on either side of the dorsal 
stripe are distinct blackened spots with other dark spots toward the 
lateral margins of the shield. 

The thoracic legs tinged with brown, claws and markings anterior to 
and partially surrounding basal joints reddish brown. 

Fully developed prolegs are found on segments 7, 8, 9, 10, and 13, con- 
colorous with body; crotchets brownish. The anal plate on the thirteenth 
segment, a pale green in color, possesses a transverse row of brownish 
spots anterior to the setae; a light-colored dorsal stripe runs through the 
plate. 

The dorsal stripe is a broken gray green bordered with light brown; 
the subdorsal is brownish, but the lateral has become lighter in color, 
brownish gray. Sets single, ringed at base with a light-colored area; 
tubercles greenish brown; spiracles black. 

SEVENTH INSTAR 

Head width, 2.70 to 3.18 mm.; average 2.93 mm. 
Total length of body, 2.9 to 3.2 cm.; average 3.02 cm. 
I Head light brown; ocellar region very dark brown, ocelli 1,2, and 6 
transparent, other three brownish; mandibles reddish brown to black at 



3i8 Journal of Agricultural Research voi. xxn.No. 6 

teeth; lower margin of labrum reddish; ringed-shaped sclerite resting on 
mentum blackened ; cardo and submentum brown with margins of stipes, 
next the mentum, same color with the stripe broadening about the base 
of the palpifer. 

The bands of dark brown bordering the adfrontals have the same 
appearance as in the previous instar (PI. 29, A). 

The thoracic shield is dark brown with a distinct light-colored dorsal 
stripe on either side of which toward the anterior margin of the shield 
is a small blackened area with other dark spots toward the lateral margins 
of the shield. 

The thoracic legs are tinged slightly with brown toward the apical 
joints; claws reddish brown, and color markings remain the same about 
the basal joints. Prolegs on segments 7, 8, 9, 10, and 13 are concolorous, 
crotchets brownish; the anal plate is a gray green with dark spots in a 
transverse row anterior to the setae; a light-colored dorsal stripe runs 
through the plate. 

The dorsal stripe is gray-green, and the brown borders in the previous 
instars appear greenish in this one; the subdorsal and lateral stripes are 
greenish with a slight tinge of brown ; the general color is a greenish gray ; 
setae single and ringed at base with light area; spiracles black, tubercles 
green; the pulsating dorsal vessel can be easily seen through the epi- 
dermis. 

Eighth instar 

Head width, 3.18 to 3.42 mm.; average 3.36 mm. 

Total length of body, 3.1 to 3.6 cm.; average 3.34 cm. 

The general color of the head is a light brown with a slight yellowish 
tinge; ocellar region dark brown, ocelli i, 2, and 6 colorless, other three 
dark brown to black; the front is a trifle lighter in color than the clypeus; 
the adfrontals, which are made very distinct by a darker brown coloration 
following the frontal sutures and by the dark bands following the epi- 
cranial sutures, extend to the bottom of the deep indentation separating 
the epicranial lobes at the summit; the mandibles are black at teeth and 
at points of articulation, and between lies an area which is a very pale 
brown in color; submentum and cardo brown; chitinized brownish stripe 
on margin of stipes, next the mentum, broadened about base of palpifer; 
antennae, labial and maxillary palpse brownish. 

The two bands of dark brown which border the adfrontals and which 
are very conspicuous in the fifth, sixth, and seventh instars, are some- 
what reduced, especially the upper parts of the bands in the regions of 
the first and second setae, in the advanced stage of the last instar (PL 
29, B) ; during the first days of the instar the bands extend from points 
even with the base of the clypeus to the second epicranial setae, running 
between the setae and the adfrontals; here they become slightly less 
dense in color and divided, parts of the stripes continuing along the 



Nov. s, 192 1 Pale Western Cutworm 319 

adfrontals and the other, lighter but wider parts, extending back to the 
regions of the first epicranial setae. 

The thoracic shield is pale brown with a prominent whitish dorsal 
stripe; there are small spotted dark brown areas toward the lateral 
margins of the shield and several smaller brown spots on either side of 
the dorsal stripe. 

The thoracic legs are tinged with brown, especially laterad; about the 
base of the coxae and femora anteriorly are reddish brown stripes. 

The prolegs on segments 7, 8, 9, 10, and 13 are concolorous with body 
and possess reddish brown crotchets. 

In the first part of this instar (PI. C, 3) there are distinct dorsal and 
subdorsal stripes, the dorsal appearing as dark green and produced by 
the pulsating dorsal vessel beneath the epidermis, and subdorsal as 
brownish ; a broken whitish lateral stripe is quite distinct; in the advanced 
stage of the instar with the exception of the tubercles there are no mark- 
ings on the body, which becomes a bleached out yellowish color. 

The spiracles are black; setae, which are a reddish brown especially 
in the head region, are single; tubercles a greenish brown but imme- 
diately about the base of each seta there is a ring of lighter color; the 
anal plate is marked by a transverse row of dark brown spots anterior 
to the setae. 

PUPA 

Length 17.5 mm., width 5.7 mm. 

Typical noctuid pupa; labial palpa? exposed for entire length; maxillae, 
mesothoracic legs and antennae of practically same length and extend- 
ing almost to caudal margin of wings ; prothoracic femora exposed ; tips 
of metathoracic legs visible and mesothoracic legs not extending to eye 
pieces; dorsal cephalic margins of abdominal segments 5, 6, and 7 marked 
with many small chitinized circular pits which extend to ventral surfaces 
of segments but where they are fewer in number and less prominent. 
The slightly bifurcate, blackish, rough cremaster ends in two stout often 
incurved spines set far apart. The color of pupa varies from a light 
straw color to a dark brown, according to age (PI. C, 4). 

ADULT ' 
"Agrotis orthogonia" nov. sp. 

All the tibiae spinose. Antennae of the male strongly serrate. Middle of the 
second joint of palpi black, its outer edge and tip, as well as the third joint, light. 
Head and thorax gray. Anterior wings dark gray; all the markings well expressed; 
half-line followed by a white shade line; basal space lighter than the other portions 
of the wing; interior line forming a very long outward projection below the submedian 
vein, and another shorter one on the costa, the line is white and distinct, bordered 
with black on each side, between the submedian and subcostal veins it is straight, 
except one lobe below the median vein, to which the concolorous, black edged clavi- 
form spot is attached; subcostal median and submedian veins white, and contrasting 

1 The description of the adtilt is quoted from Morrison (z^). 



320 Journal of Agricultural Research voi. xxn. no. 6 

(PI. C, 5); orbicular spot elliptical, with an outer black ring, within which appears a 
white annulus, inclosing the gray center; reniform spot large and of the usual shape, 
the portion of its black annulus, beneath the median vein, separated and very dis- 
tinct; exterior line rounded, formed of interspaced luniform marks, followed by a white 
shade line; subterminal space rather lighter than the median space, terminal space 
again dark; a series of partially effaced cuneiform marks, before the white subterminal 
line, which forms two short teeth on the second and third median branches. Pos- 
terior wings whitish at the base, with a black terminal band and contrasting white 
fringes. Beneath whitish, the center of the median space dark, and the neighbor- 
hood of the median vein, on the anterior wings, clothed with long soft hair. 

Expanse, 34 mm. 

Hab. Glencoe, Nebraska. Received from Mr. G. M. Dodge. (No. 66). 

The nearest ally of this fine species is the European Agrotis vestigiales Rett. 

LITERATURE CITED 

(1) COOLEY, R. A. 

1919. SEVENTEENTH ANNUAL REPORT OF THE STATE ENTOMOLOGIST OF MONTANA. 

Mont. Agr. Exp. Sta. Bui. 133, 15 p. 

(2) DoD, F. H. Wolley. 

1901-06, PRELIMINARY LIST OP THE MACRO-LEPIDOPTERA OF ALBERTA, N.-W. T. 

In Canad. Ent., v. 33, p. 40-42, 157-172, 1901; v. 36, p. 345-355. 
1904; V. 37, p. 17-28, 49-60, 145-156, 173-184, 221-230, 241-252, 
1905; V. 38, p. 45-54. 89-94, 263-267, 1906. 

(3) Dyar, Harrison G. 

1902. A LIST OP NORTH AMERICAN LEPIDOPTERA... U. S. Nat. Mus. Bill. 52, xix, 

723 p. List of works quoted, p. ix-xix. 

(4) Forbes, William T. M. 

1910. A structural study op some caterpillars. In Ann. Ent. Soc. Amer- 
ica, V. 3, no. 2, p. 94-132, pi. 10-20. Bibliography, p. 125-127. 

(5) Gibson, Arthur. 

1912. THE entomological RECORD, 1911. In 42d Ann. Rpt. Ent. Soc. On- 
tario, 1911, p. 89-112. 
(6) 

(7) 



1912. cutworms and army-worms. Canada Dept. Agr. Div. Ent. Bui. 3 (70, 
Exp. Farm ser.), 29 p., 10 fig. 



1914. A NEW destructive CUTWORM OP THE GENUS POROSAGROTIS, OCCURRING 

IN WESTERN CANADA. In Jour. Econ. Ent., v. 7, no. 2, p. 201-203. 

(8) 

191 5. CUTWORMS AND THEIR CONTROL. Canada Dept. Agr. Div. Ent. Bui. 10. 

31 p., 20 fig. 

(9) Hampson, G. F. 

1908. notes on noctuidaE... In Canad. Ent., v. 40, no. 3, p. 102-107. 

(10) Hewitt, C. Gordon. 

1912. report op the dominion entomologist, /w Canada Exp. Farms Rpts. 
[i9ii]/i2, p. 173-189, pi. 5. 

(11) 

1914. report prom the division of entomology... [i9i2]/i3. In Canada 

Exp. Farms Rpts. [i9i2]/i3, p. 499-518, pi. iq. 

(12) 

1915. REPORT FROM THE DIVISION OP ENTOMOLOGY... [i9i3]/i4. In Canada 

Exp. Farms Rpts. [i9i3]/i4, p. 851-876. 



Nov. s. 192 1 Pale Western Cutworm 321 

(13) Canada Department op Agriculture. 

1917. REPORT OF THE DOMINION ENTOMOLOGIST, [i9i5]/i6. 73 p., 4 pi. Ottawa. 

(14) Maxson, Asa C. 

1920. PRINCIPAL INSECT ENEMIES OK THE SUGAR BEET IN THE TERRITORIES 
SERVED BY THE GREAT WESTERN SUGAR COMPANY. vii, 157 p., 30 fig. 

9 col. pi. Denver, Colo. Bibliography, p. 147-148. 

(15) Morrison, H. K. 

1877. descriptions OP NEW north AMERICAN NOCTUIDAE. In PfOC. BoStOH 

Soc. Nat. Hist., v. 18, 1875/76, p. 237-241. 

(16) Smith, John B. 

1890. contribution toward a MONOGRAPH OF THE INSECTS OF THE LEPI- 
DOPTEROUS FAMILY NOCTUIDAE OF TEMPERATE NORTH AMERICA. RE- 
VISION OF THE SPECIES OF THE GENUS AGROTIS. U. S. Nat. MuS. Bul. 
38, 237 p., 5 pi. 

(17) Strickland, E. H. 

1915. POISONED BAIT FOR CUTWORMS. In Canad. Ent., V. 47, no. 7, p. 201-204. 

(18) 

1916. THE CONTROL OF CUTWORMS IN THE PRAIRIE PROVINCES. Canada Dept. 

Agr. Ent. Branch Circ. 6, 8 p., 5 fig. 

65769°— 21 3 



PLATE C 
Porosagroiis orthogonia: 



—Eggs. 

— Cast head of fifth-instar larva. 

— Eighth-instar larva. 

— Pupa in earthen cell. 

— Adult, male. 

(322) 



Pale Western Cutworm 



Plate C 



1 ^^i3^ 






Hit- i^i/^'<MA i\ I 







Journal of Agricultural Research 



Vol. XXII, No. 6 



PLATE 28 
PoTOsagrotis orthogonia: 
A, B, C, D, E, F. — Cast heads of first- to sixth-instar larva?. 



Pale Western Cutworm 



Plate 28 






Journal of Agricultural Research 




Vol. XXII, No. 6 



Pale Western Cutworm 



Plate 29 



O 



O 



^ 



: • ' ' /; 



(■ -/t ■ I 




^ 







\ 



ep/. 



/ 



III 



m 



m 



IV 



U 



M 



)M 



^ 



hi, 



vadfii 
fr. 



,di 



o. 



^ 



a 



B 



/b 



K 



Journal of Agricultural Research 



Vol. XXII, No. 6 



PLATE 29 

Porosagrotis orthogonia: 

A. — Cast head of seventh-instar larva. 

B. — Cast head of eighth-instar larva with setae numbered. Adf., adfrontal sclerite, 
adf. i, adf. it, its setae; fr., frontal sclerite; fr. i, frontal setae; fr. 0, frontal puncture; 
cL, clypeus; cl. i, cl. ii, its setse; Ihr., labrum; ant., antennae; md., mandible; md. i, 
md. ii, its setae; i to xi, setae of epicranium, / to IV , first four ocelli. 



PLATE 30 

Porosagroiis orthogonia: 

A. — Moth feeding on clover blossom. 

B. — Light trap. 

C. — Excavation made by horned lark in digging out cutworm. 

D. — ^Wheat field attacked by the larv'se. 



Pale Western Cutworm 



Plate 30 




£\ ^^mtft/0^' 






7 ' ' \ ^ 







V4 t' 






^1 -1^^ ^ 



''• " fi 






M.r^'M :. 



1. 

- \1 



rT 



fo" 









'%V - - ■ 



>.i 



Journal of Agriculturjil Researcli 



Vol. XXII, No. 6 



BIOIvOGY OF EMBAPHION MURICATUM 

By J. S. Wade, Scientific Assistant, Cereal and Forage Insect Investigations, Bureau 
of Entomology; systematic description of the larva by Adam G. Boving, Bureau 
of Entomology, United States Department of Agriculture 

INTRODUCTION 

Considerable damage has been wrought during the past six or seven 
years by the larvae of Embaphion fmiricatum Say and related species of 
false wireworms to growing wheat and other field crops throughout the 
semiarid and middle western United States. The area of greatest 
injury embraces approximately the western half of Nebraska, Kansas, 
and Oklahoma, and the eastern third of Colorado and New Mexico, 
although losses of varying magnitude have been reported in various 
localities over the greater part of all these States. In view of the obscure 
character of the injury, it seems quite probable that much crop damage 
commonly charged to other causes in reality has been brought about by 
this pest. The steady transformation in recent years of grassy prairies 
into cultivated fields has been an important factor, because the removal of 
native food plants causes this and related species to feed more and more 
upon cultivated grains. The hardiness of the insect and the rapidity of 
its adaptation to changed conditions and to new host plants indicate 
that this species is potentially a serious menace to grain production 
within the infested region. 

EARLY RECORDS 

The species under discussion was originally described in 1824 as Akis 
muricata by Thomas Say (/) ,^ who stated that it — 
inhabits Arkansa at the Rocky Mountains 

and that — 

as it does not entirely agree witli any genus the characters of which Latreille has 
noted, it may be proper to remove it to the Blapsidae, under a separate genus, which 
may be named Embaphion. 

This description was reprinted in 1859 in the LeConte edition of 
Say's (x) works, with a brief supplementary editorial note indicating 
relationship of the genus Embaphion to the genus Eleodes. A single 
specimen from Texas was described as Eleodes contusum by LeConte (2) 
in 1853, who stated that it — 

resembles E. muricatum Say but is longer and narrower with the broad margin of 
the elytra more suddenly reflexed and almost perpendicular. Although so different 
in form, this genus is only distinguished from Eleodes by the inferior plane of the 

' Reference is made by number (italic) to "I<iterature cited," p. 334. 

Journal of Agriculture Research, Vol. XXII, No. 6 

Washington, D. C. Nov. s. 192 1 

aaj , Key No. K-103 

(323) 



324 Journal of Agricultural Research voi. xxii. no. 6 

mentum being more rounded and more deeply impressed; its anterior margin is 
slightly incised; the lateral angles are so much reflexed as to be invisible; the tarsi 
are silicate beneath and fringed at the apex and sides with short spines; the middle 
joints of the posterior tarsi appear more elongated than in Eleodes. I have grave 
doubt of the generic value of any of these differences, and several nondescript species 
from New Mexico seem to be intermediate both by the form of the body and by the 
dififerences in the mentum. 

The characters distinguishing Embaphion contusum from Embaphion 
muricatum were discussed by LeConte (j) in 1859. A brief resume of the 
previous history of the genus was made by Lacordaire {4, v. 5, p. 152; 
atlas, pi. 50, fig. 2) in 1859, in which attention was called to Say's inade- 
quate designation of the genus Embaphion and to variation of the species 
with its geographical distribution. Horn (5, p. 320-322) in 1870, in his 
discussion of the genus, indicated the feeble taxonomic characters which 
separate this genus from Eleodes. In referring to the species Embaphion 
muricatum he states : 

This species may be readily distinguished from the others of the genus by the very 
broad foliaceous margin of the thorax and elytra, very strongly reflexed. The elytral 
margin extends beyond the apex and the two meet on a line with the suture. The 
thoracic margin is broad and widens behind, so that the hind angles are prominent, 
sub-acute, and project backwards over the basal angles of the elytra. The thorax 
itself (less the margins) is narrow, longer than broad, and about equal to half the width 
of the elytra (without margin). The disc of elytra (without margin) is elongate oval, 
the humeral angles not prominent and are rounded. The angles formed by the margin 
are nearly right. The base of the thorax is strongly trisinuate; the base of the thorax 
proper being rounded, that of the margin on each side emarginate. The base of elytra 
is emarginate at middle, and on each side broadly rounded. 

He stated further that Embaphion concavum, described by LeSonte 
(2) in 1853, is 

merely a large form with more strongly reflexed margins. The elytra of both forms are 
sculptured with approximate series of fine punctures, each bearing a short hair. 

Blaisdell's (//, p. 473-477) very full discussion (1909) of the adult 
forms of the species and their taxonomic relationships leaves little to be 
desired. He especially emphasizes the salient type characters. 

Margins of the thorax and elytra broadly foliaceous and strongly reflexed, basal 
angles of the prothoracic margins projecting strongly backward over the basal angles 
of the elytra. 

DISTRIBUTION 

Nebraska: Alliance, altitude 3,971 feet, August, H. F. Wickham; Beaver City, alti- 
tude 2,150 feet, M. H. Swenk {12), September, J. S. Wade; "Nebraska," May to 
August, H. F. Wickham (<S). 

New Mexico: Chico, altitude 6,882 feet, September, D. J. Caffrey; Clovis, August, 
H. F. Wickham; Koehler, altitude 5,500 feet, June, V. L. Wildemiuth, August, 
W. R. Walton; Vaughn, September, H. F. Wickham; Maxwell, altitude 5,894 feet, 
D. J. Caffrey; Willard, altitude 6,091 feet, H. F. Wickham. 

Kansas: Clark County, altitude 1,962 feet, June, F. H. Snow {10); Colby, altitude 
3,150 feet, August, J. S. Wade; Dodge City, altitude 2,509 feet, August, J. S. Wade; 
Hamilton County, altitude 3,000 feet, F. H. Snow {10); Liberal, altitude 2,839 feet, 






Nov. 5, 1921 Biology of Embaphion muricatum 325 



July, J. S. Wade; Meade, altitude 2,503 feet, July, J. S. Wade; Morton County, alti- 
tude 3,000 feet, F. H. Snow {10); Norton, altitude 2,284 feet, August, J. S. Wade; 
Rice Coimty, June, H. F, Wickham; Scott City, altitude 2,971 feet, August, J. S. 
Wade; Wallace County, altitude 3,000 feet, F. H. Snow (7); Wellington, altitude 
1,205 feet, July, J. S- Wade; "Kansas to Texas," G. H. Horn (5), "Western Kansas: 
In Arkansas and Smoky Hill Valleys," E. A. Popenoe ((5). 

North Dakota: Dickinson, altitude 2,411 feet, August, H. F. Wickham; "Dakota," 
W. G. Dietz; "Dakota," May to August, H. F. Wickham {8). 

Colorado: Bellevue, altitude 8,993, H. F. Wickham (p); Colorado Springs, altitude 
6,072 feet, H. F. Wickham {g) ; Denver, altitude 5,279 feet, April, H. Soltau; Greeley, 
altitude 4,652 feet, June, H. F. Wickham; Fort Collins, altitude 4,994 feet, H. F. 
Wickham (p); LaSalle, altitude 4,676 feet, September, H. F. Wickham; Pueblo, 
altitude 4,685 feet, October, H. Soltau; West Las Animas, H. F. Wickham (p); 
"Colorado," May to August, H. F. Wickham (8). 

Texas: Amarillo, altitude 3,676 feet, August, H. F. Wickham; Canadian, altitude 
2,340 feet, August, H. F. Wickham ; Mobeete, July, H. S. Barber; Texline, altitude 
4,694 feet, September, I. R. Crawford. 

Montana: Assinniboine Mountains, Hubbard and Schwarz; "Montana," May to 
August, H. F. Wickham (8). 

South Dakota: Alexandria, altitude 1,354 feet. 

Mexico: Nuevo Laredo, Tamaulipas, Hoge. 

INJURY 

The principal damage caused by these insects is that wrought by the 
larvae during the fall in devouring recently sown or newly sprouted wheat 
grains shortly after the seed wheat has been drilled. These larvee often 
may be found in large numbers in infested fields at such periods working 
steadily along through the soft soil of the drill rows, either wholly devouring 
or destroying for germination purposes every wheat grain within a drill row 
for many yards. Within the region of greatest infestation the principal 
injury is done between September 20 and October 15. The injury to the 
grain is characteristic of this family. Sometimes the entire contents of 
the grain are removed, lea\'ing all or part of the shriveled outer husk; in 
some cases the ends of the grain are nibbled away or portions of the ven- 
tral crease are neatly fmrowed out. The adults also are known to feed 
upon wheat grains and other seeds, being present around the bases of 
wheat stacks in July, where they may be found tearing away the spikelets 
of grain in newly cut wheat heads to devour the kernel within, or they 
may be found feeding upon the scattered grains. The extent of the injury 
varies annually in accordance with seasonal conditions, little or no damage 
being done in localities where an abundance of rainfall occurs, and where 
temperature and other factors are favorable to growing crops, whereas 
at the same time considerable loss may be experienced in other localities, 
varying from 10 to 50 per cent or more of the wheat of an entire neighbor- 
hood, where weather and other conditions render normal development 
of this crop impossible. In view of the fact that the larvae of this pest 
usually may be found working with those of other nearly related species 
of true and false wireworms, it becomes increasingly difficult to isolate and 
estimate singly the exact amount of injury wrought by this particular pest. 



326 Journal of Agricultural Research voi. xxu, no. 6 

HABITS 

The larvae are exceedingly active and quick and, if exposed to light by 
the plow or otherwise disturbed, have the power of ^vriggling very 
quickly down out of sight into the soil. They are also occasionally 
found upon the surface of the ground feeding upon seeds of weeds and of 
other plants, in spots where the soil may be slightly moist and where 
they are covered by wheat shocks or by matted masses of dried Russian 
thistles or other weeds. While they appear to prefer habitats where 
there may be a slight degree of moisture, such as moist, poorly drained 
spots in fields, and cool, damp cellars, yet they do not live long in thor- 
oughly wet soil. Both larvae and adults often may be found in numbers 
beneath dried weeds along irrigation canals. The larvae habitually feed 
during warm weather at a depth varying from 2 to 5 inches, according to 
condition of the soil. As they burrow from place to place, they feed upon 
the roots and seeds of plants, and possibly to a certain extent upon 
organic matter where this is sufficiently decayed. When placed under 
artificial conditions the larvae feed readily not only upon germinating 
wheat, but upon com and roots of grasses. They are cannibalistic in that 
they feed upon other larvae of the same species which die or become 
weakened because of injury or disease. They also feed upon their own 
exuviae. 

The adults, in common with those of other nearly related species, are 
very hardy and active and appear to be able to withstand considerable 
variations of temperature. While they, like the larvae, appear to prefer 
cool, moist spots, they do not survive temperatures as low as —9° F. 
They have been collected in August beneath wheat shocks in fields where 
the temperature was as high as 100°. The adults easily climb all over 
wheat where standing or in the stack or shock, and they burrow with 
apparent ease far into the piles of unthrashed grain. They are also fre- 
quently found in the burrows of small mammals. During periods of pro- 
longed drought the beetles may seem to have entirely disappeared, yet 
immediately following a shower or rainstorm, curiously enough they 
reappear in large numbers, where previously none could be found. 

DESCRIPTIONS 

EGG 

Size slightly variable, being i.i to 1.3 mm. in length and 0.60 to 0.62 mm. in width; 
shape circular in cross section and oval in longitudinal section; without sculpturing; 
color pure white when first deposited, changing to yellowing brown before hatching. 

MATURE LARVA * 

Length 27 mm.; color testaceous with head and legs somewhat darker; anterior 
and posterior margins of prothorax and posterior margins of the following segments 
castaneous-testaceous. Surface corneous. Form elongately cylindrical, more than 

1 Description and Plates 31 and 32 by Adam G. Bdving. 



I 



Nov. 5. 1921 Biology of Embaphion muricatum 327 

10 times longer than wide; dorsally verj'^ convex, ventrally flattened; pygidium 
movable in the directions up and down, subconical, obtusely pointed. Head, ven- 
tral sides of the thoracic segments and of the first abdominal segment, legs, and 
pygidium (PI. 32, C) clothed with rigid or soft setae; rest of body glabrous with very 
few and small ventral hairs. 

Cranium (PI. 31, B) rounded, nutant, exserted, one-third broader than long (from 
epistomal margin (epi) to foramen occipitale), broadest medianly, dorsally somewhat 
convex. Anterior frontal angle (fa) low and rounded. Frons (/) three-fourths the 
length of cranium, about as long as wide with extreme width anteriorly, side margin 
convex. Epicranial halves (epc) meeting dorsally; epicranial sutiue one-fourth the 
length of cranium; ventrally (PI. 31, E) the halves are separated by gula (gu); dor- 
sally with a few, laterally and ventrally with many hairs. Gula and submentum 
(sm) both distinct, coriaceous. Gula almost square, with tentorial pits (ip) at the 
middle of the side margins. Submentum trapezoidal, broadest posteriorly; side 
margins slightly concave and adjacent to maxillary articulating area. Clypeus 
(cl, PI. 31, B) trapezoidal, widest behind, length to extreme width as one to four, 
medianly with slight transverse deepening, set on each side with one minute seta 
near the middle line and two well-developed setae near the lateral margin. Labrum 
well-developed, movable, transversely rectangular, length to width as one to three, 
anterior margin almost straight, anterior comers rounded; disk on each half with a 
median transverse series of five large setae, and an anterior series of three long, thin, 
and straight setae; right behind those but on the ventral side of labrum another series 
of four shorter, stronger, and ciurved setae. Epipharynx (eph, PI. 31, A) forming the 
buccal surface of labrum, soft-skinned with posterior transverse, broad, sinuous, chitin- 
ous band, that carries one pair of stublike sharp teeth; on the soft-skinned part 
anteriorly to these teeth a pair of tiny hooks; near anterior margin scattered minute 
setae and ring-shaped punctitfes. Just behind antenna two ophthalmic spots, both 
transverse, slightly posteriorly convex, the anterior a little more external and about 
three times longer than the posterior; immediately in front of the anterior are numer- 
ous setae; the ophthalmic spots are likely to disappear in full-grown larvae. Antenna 
(PI. 31, B) closely behind the mandible, attached in articular cavity with distinct 
border; basal antennal membrane well developed; basal article cylindrical, about as 
long as epicranial suture, second article as long as basal article, more clavate, apical 
article very small, conical, papilliform, carrying one seta; no supplementary appendix 
besides the apical article. Mandibles (Pi. 31, F) of right and left side differing in 
shape; both apically bifid (a^, a^); both with one tooth (/) between apex and molar 
part (m); tooth of right mandible, however, prominent and placed near apex, that of 
left mandible less developed and placed closer to molar part; molar part of right mandi- 
ble with bituberculate crown, that of left mandible with hollow crown; ventrally (PI. 
31, D) with cutting part deeply excavated; exterior stirface ("the back of the mandi- 
ble") distally with a slightly carinate margin (PI. 31, F, c), proximally with a soft- 
skinned, whitish swelling (s) from an excavation (e) opposite the molar part; three 
to four strong setae from the anterior portion of the swelling, two from the posterior, 
several small, soft setae near dorsal mandibular articulation. Maxilla dorsally com- 
pletely covered by mandible; palp (PI. 31, E) siu-mouiiting mala (ma) (maxillary 
lobe) with one-third of its own length; palpiger (pag) small, ring-shaped; basal article 
about as wide as long, second article cylindrical, somewhat narrower and more than 
twice as long as basal article, apical article two-thirds as long and half as thick as the 
second, conical, with soft tip; each article with one or two thin setae; mala (ma) on 
buccal stuface (PI. 32, F) with two series of well-developed, somewhat curved setae; 
base of stipes (PI. 31, E, bs) (that is, region where stipes and cardo meet) rather short; 
proximal half of inner margin of stipes (is ,) connected with exterior half of maxillary 
articulating area (ar^), distal half (js- 2) right behind mala, free; ventral stipital 
surface with several strong setae; other setae on the exterior surface; cardo as long as 



328 Journal of Agricultural Research voi. xxii, no. e 

exterior margin of stipes, adjacent to slightly curved hypostomal thickening {hyp) 
between fossa for ventral mandibular condyle {fm) and fossa for tip of cardo (/c); inner 
margin of cardo connected like stipes with exterior half {ar ^) of maxillary articulating 
area. Maxillary articulating area protuberant, soft, divided into two halves; exterior 
half (ar I) connected with maxilla, subdivided into an upper and low^er portion; 
interior half (ar^) connected with submentum, entire; no seta. Mentum {me) 
almost square, side margins free; on each side about five setse of different length. 
The two stipites labii {stla) fused into a slightly chitinized unit, carrying on each 
side two setae; labial palp about half as long and half as thick as maxillary palp; 
basal and apical articles slightly different in length, basal article somewhat clavate, 
apical article conical and half as thick as basal article; ligula {li) small, narrow conical, 
with one terminal pair of setae. Hypopharyngeal sclcrite (PI. 32, A, G, H, hsc) 
elongate rectangular, projecting, strong; anteriorly tricuspidate with median cusp 
largest; disk somewhat excavate with a posterior semiglobular tubercle; molar part 
of mandible and hypopharyngeal sclerite grinding together (PI. 31, D, F, G). The 
hypophar>'ngeal bracon (PI. 32, A, G, H, hbr) is well developed as a chitinous rod in 
the buccal membrane between the ventral mandibular articulation and the hypo- 
pharyngeal region. Prothoracic legs (PI. 31, C, H, I; 32, B) considerably stronger 
than the mesothoracic and metathoracic ones and with coxae attached so closely 
together that they almost touch each other at base. Coxa of first pair about as long 
as wide ; many fine , scattered hairs on exterior and interior sm"f aces ; trochanter about 
as long as coxa, on the inner side (PI. 31, H) distally with two spinelike setae and also 
with a few other thin hairs; femur {fe) about as long and wide as trochanter, armed 
with five spinelike setae, also with many thin, scattered hairs; tibia {ti) about twice as 
long as thick, almost same length as femur but not fully as wide, armed with five 
spinelike setse and also with fine, scattered hairs; tarsus {ta) of almost same length 
as tibia, claw-shaped, strong, but rather slender, with backward-facing surface distally 
excavate and proximally carrying a round soft-skinned region around a short but strong 
seta; another and similar seta set close to it at the end of the excavation; otherwise 
no setae or hairs on tarsus. Second and third pairs of legs inserted farther apart than 
the first pair; the arrangement of their setae very similar to that of the first pair, but 
the proportional sizes between the articles somewhat different from those of the 
first pair. Prothoracic eusternum (PI. 32, B, eu) large, rhomboidal, anteriorly almost 
reaching the front margin of the segment, only separated from this margin by a small 
presternal area (y); the hypopletu-al chitinization {hi and h-^), and especially its 
prehypopleural part (A i), large and strong; prothoracic tergal shield siibquadrate, 
slightly wider than long, with anterior and posterior margins, as mentioned above, 
darker than the rest of the shield and finely longitudinally striated. Mesothorax 
and metathorax with transverse, sub triangular, narrow presternum {y), laterally 
adjacent to poststernellura (2) of the preceding segment; hypopleiiral chitinizations 
{h I and h n) well developed, but considerably smaller than those of prothorax; 
poststemellum of metathorax not present, preepipleurum of mesothorax {e i) 
subtriangular, carrying first thoracic spiracle; preepipleurum of metathorax not 
distinctly limited, carry'ing the rudimentary second thoracic spiracle; postepipleiuiim 
(e u) of both segments well developed, more or less fused with the corresponding 
preepipleura ; mesothoracic and metathoracic tergal shields transversal, subrectangu- 
lar, about twice as wide as long, right behind anterior margin with a dark transverse 
line; posterior margin darker than rest of segment, finely longitudinally striated. 
Typical abdominal segment (that is, one of the eight anterior abdominal segments) 
with fused sternal and hypopleural areas {ster), covered by a single, longitudinally rec- 
tangular shield, which posteriorly has a rather dark, transverse, longitudinally finely 
striated margin; one seta present near the anterior and one seta near the posterior 
margin; additionally the sternum of first abdominal segment is anteriorly densely 



Nov. 5, 1921 



Biology of Embaphion muricatum 



329 



set with small, soft sets; similar outfit lacking on the other abdominal segments. 
Epipleural region narrow. Tergal region with a dark line above the spiracle. Tergal 
shield (ter) single, posteriorly with a dark, longitudinally striated margin. Anterior 
abdominal segments transverse, slightly wider than long; sixth, seventh, and eighth 
abdominal segments subquadrate. Ninth abdominal segment smaller than the 
preceding segment ; dorsal part or pygidium pointing upwards, subconically produced, 
above somewhat flattened, below broadly convex, apex obtuse, laterally with margin 
set with a single series of strong, short setse, whole siuface with scattered, fine setae; 
ventral part of ninth segment small, transverse, soft. Tenth abdominal segment 
(or "anal segment") small, with trilobate upper transverse anal lip, with a pair of 
conical and, except at the tip, setose ambulatory warts, laterally to anus a small 
triangular lower lip. Spiracles (Pi. 32, I) annular, shortly oval, transversely placed; 
opening at the bottom of cup-shaped peritrema, linear, unprotected by hairs. The 
number and development of setae on the first pair of legs vary according to species 
and do not offer any generic character. The same is the case with the setal arrange- 
ment of pygidium. 





Fig. I. — Pupa of Embaphion mu- 
ricatum, dorsal view. 



Fig. 2. — Pupa of Embaphion mw 
ricaium., ventral view. 



PUPA (FIG. I, 2) 



Length 11 mm. Width 5.6 mm. Free. Arcuate. Color pinkish white, with 
ferruginous tinge on pronotiun. Femora and tarsi fuscate, other appendages partly 
translucent. Pupa becomes more strongly colored immediately before issuance of 
adult. Head pressed to prostemum. Pronotum broad and projecting somewhat 
anterior to head, making the head nearly invisible from above. Frons impressed. 
Vertex prominent. Antennae placed backward near sides of prothorax. Mesonotum 
narrow. Legs not pressed against body. Tips of wing cases extending to the an- 
terior margin of metanotum. Second to fifth abdominal segments bearing on each 
side of tergites flat, lacerated protuberances, obtuse, pointed and directed posteriorly. 
Eighth segment ending in two slightly divergent, acute processes. 



330 



Journal of Agricultural Research voi. xxn.N>. e 



ADULT (fig. 3) * 

Oval to oblong-oval, brownish to piceous black, thoracic and elytral margins very- 
broad and foliaceous, strongly reflexed. 

Head small, less than twice as wide as long, plane, sides of the frons slightly promi- 
nent, punctate, punctures very feebly subasperate, fine, not dense, each with a small 
curved and short seta, frontal suture usually not visible. Antennce rather long, quite 
slender, outer four joints very slightly compressed and scarcely widened, tliird joint 
shorter than the next two taken together, fourth scarcely longer than the fiftli, the 
latter and sixth subequal, seventh shorter, eighth feebly shorter than the seventh 
and slightly triangular, ninth and tenth suborbicular, eleventh subovate. 

Pronotum with margins very broadly foliaceous, the margin more than one-half 
wider than the disc, the latter comparatively narrow, longer than wide at middle, very 

feebly convex, usually with irregular de- 
pressed areas; finely, more or less subas- 
perately and sparsely punctate; reflexed 
margins wider posteriorly and more or less 
concave, a little more distinctly punc- 
tate, punctures less sparse, each with a 
short curved seta; o/>ex deeply and feebly 
subquadrately emarginate, the emargina- 
tion about one-half wider than deep, sides 
almost parallel, and scarcely margined; 
sides evenly but not strongly arcuate, 
moderately converging from base to apex; 
base proper feebly arcuate, not margined 
and about equal to the length, 1^.' ,rally 
sinuate; apical angles rather narrowly 
rounded and formed by the advanced 
foliaceous margins and nearly as long as 
the head; basal angles are posteriorly 
prominent, subacute, and projecting back- 
ward over the basal angles of the elytra. 
Propleurcc smooth and impunctate; in- 
ferior surface of the foliaceous margins 
obsoletely punctate. 

Elytra oval to elongate oval; margins 
broad andreflexed, angles at humeri nearly 
rectangular and more or less truncate at base, posteriorly extending beyond the apex, 
the two meeting on a line with the suture above the true elytral apex, and defined 
from the same by a slight groove, borders evenly arcuate from base to apex or more 
or less parallel basally ; base evenly but not strongly emarginate; humeri proper broadly 
rounded and not prominent ; sides proper more or less evenly arcuate ; apex proper not 
produced and narrowly rounded; disc plane, feebly convex, at times slightly concave, 
the inflexed sides nearly straight and oblique, gradually and not strongly arcuately 
declivous behind; surface sculptured with approximate series of fine asperate punc- 
tures, which become more irregular and slightly denser laterally. Each puncture 
bears a short and ratlier robust cur\'ed seta. 

Epipleurce narrow, not attaining the humeral margin and not dilated, but gradually 
narrowing to apex, not defined from the inflexed sides of the elytra, and on the same 
plane; superior margin obsolete, except near apex; elsewhere represented by a line 
of punctures or a faint groove. 




Fig. 3. — Adult of Embapkion tnurkalum, dorsal view. 



^Reprinted from F. E. Blaisdell (//, p. 473-476). 



Nov. 5, 1931 Biology of Embaphion muricatum 331 

Sterna more or less dull, finely and not distinctly sculptured. 

Parapletirse smooth, rather sparsely but not very distinctly punctate. 

Abdomen horizontal, very finely and sparsely punctulate, obsoletely rugulose and 
quite evenly convex. 

Legs rather slender, moderate in length. Anterior femora mutic, pro tibial spurs 
similar in the sexes, the anterior slightly longer than the posterior. Protarsi simple. 

LIFE HISTORY AND DEVELOPMENT 

The principal observations on the hfe history and development of 
Embaphion muricatum, as given below, were made under laboratory and 
field conditions in south-central Kansas, at an altitude of approximately 
1,200 feet. Under different conditions of latitude, altitude, and humidity 
there would doubtless be found more or less marked variations. The rec- 
ords are unfortunately based on incomplete studies for, owing to working 
conditions and to pressure of other duties, there was no opportunity to 
conduct a sufficiently extensive series of experiments to render all obser- 
vations conclusive. 

The eggs are deposited in loose, dry, or slightly moist soil at a depth 
of ^ to I inch, sometimes singly, but more often in clusters of two or 
three to a dozen or more eggs at one place. At temperatures ranging 
from 80° to 90° F. the average period of incubation is approximately 
10 days, whereas at temperature's of 68° to 70° F. the egg stage is approxi- 
mately 13 days. Undoubtedly weather conditions and the time of year 
have a direct bearing on the duration of the egg period. 

During the later stages of development and shortly before hatching, 
the surface of the egg becomes light brown in color, and the shell appears 
to expand slightly and to become more flexible, while the movements of 
the young larva can be noted within. During the process of hatching, 
the struggles and the lifting pressure of the young larva burst the shell 
and the larva emerges by rather slow periodic movements, as its integu- 
ment is very soft and fragile. Though the young larva often remains 
for some time near the place of hatching, yet it is capable of locomotion 
soon after emergence. All normal eggs of the same egg cluster usually 
hatch within a short period, generally a few hom-s. While abnormal 
weather conditions may prolong the period of hatching, no injurious 
effects of such retardation are noted in the eggs. No infertile eggs were 
ever collected under field conditions. Soon after the emergence of the 
larvae the empty eggshells become more and more contracted and dried 
up, until eventually only tiny, shrivelled fragments remain. 

Upon hatching the larva averages 3.5 to 3.75 mm. in length and is 
yellowish white. The color changes sHghtly after each molt until at 
maturity the larva becomes a deep yellow. 

The length of the larva stage, according to an experiment consisting of 
3 1 larvae hatched in June and kept in a cool cellar at an average tempera- 
ture of 68° F., averaged for the siu-vivors 79 days, while in an experiment 
consisting of 49 larvse, under similar conditions, it varied from 76 to 



332 Journal of Agricultural Research voi. xxii, no. 6 



96 days, though the average duration was 85 days. The larvae as hatched 
were placed in small tin salve boxes containing about X inch of slightly 
moist soil and split wheat grains. As the larA^se became large, whole 
wheat grains were used as food. Under field conditions many of the 
larvse appear to become nearly mature during late fall and overwinter 
in this condition. From about November i to March 15 in the latitude 
of southern Kansas they are exceedingly inactive and feed but little. 
The rapidity of growth of the larvffi undoubtedly depends to a large 
degree upon weather and seasonal conditions and the quantity and 
quality of food available. Shortly before the period of pupation the 
larva does not feed and assumes a semidormant stage of approximately 
7 to 9 days' duration. 

The pupa stage, when rearings were conducted under laboratory 
conditions, comprised 18 to 20 days. The pupse are pinkish white im- 
mediately after transformation, and as development proceeds the color 
changes to light yellow. Shortly before the adults emerge the append- 
ages take on a yellowish brown tint. 

The newly issued adults are of a brighter color, and the chitinous 
portions of the body are soft. Within a few days, however, the color 
darkens and the integument hardens so that the newly emerged adults 
are not distinguishable. Under artificial conditions mating does not 
become general for a week or more after emergence. Oviposition and 
feeding appear to occur usually at night. The adults are crepuscular. 
They may be found abroad in greatest numbers on cloudy days or in 
early morning or late evening. On clear days, during the middle or 
warmer portion of the day, they remain under shelter. While usually 
inactive at such periods, if disturbed they will run with great rapidity. 
The insect may overwinter both in the adult and in the larva stages. In 
the latitude of southern Kansas, however, the mortality of such over- 
wintering adults is great. 

ENEMIES 

While the incomplete character of the life-history work performed with 
Embaphion muricahtm afforded comparatively little opportunity for 
obtaining parasites under artificial conditions, or for obtaining data on 
other enemies for possible use in control work, yet some noteworthy 
information was obtained. From adults of E. imiricaium collected by 
the MTiter from barley at Colby, Kans., on August 25 there were reared 
on October 23 adults of a parasite determined by A. B. Gahan of the 
Bureau of Entomology as Perilihis eleodis Viereck {13). No life-history 
work on these parasites was attempted. 

Considerable difficulty was experienced in rearing larvae owing to the 
presence in the cages of a fungus, Metarrhizium anisoplae Metsch. Soc. ; 
and though the apparatus and soil were sterilized, yet the mortality was 
sufficient at times to interfere to a marked extent with the rearing. A 



Nov. 5, 1921 Biology of Embaphion muricatum 333 

number of the larvae in the cages were also attacked by an obscure bac- 
terial disease. This appeared to be identical with that described by 
Prof. Swenk {12). There would appear somewhere upon the body 
sutures small circular or irregularly shaped dark brown spots, and these, 
after a few days, would become larger, until in some instances they 
would cover one-third to one-half of the body surface. This disease 
usually caused the death of the larvae within varying periods of time. 
Larvas found under normal field conditions are sometimes found to be 
affected both with Metarrhizium and with the disease. 

CONTROL 

While no extended series of experiments relative to control of the 
insect thus far has been found possible, yet the information secured on 
the subject has been sufficient to assure the practical value of the measures 
here recommended in reducing or preventing damage. 

A systematic rotation of crops is one of the most effective procedures 
in cutting down dam.age. The maximum injury always may be found 
upon those areas where the ground has been cropped to wheat continu- 
ously for several years, whereas the minimum injury is found where 
com, kafir, milo, and other crops are grown which require some degree 
of cultivation during the growing season. An important factor in migra- 
tion and infestation lies in the fact that the beetles are wingless and 
therefore become dispersed much more slowly than do winged forms. 

A number of fields within infested areas, which were also infested by 
the corn earworm {Chloridea obsoleta Fab.) and other insects of somewhat 
similar habits, were plowed by farmers during late fall or early spring to 
destroy the pupae, and it was found that such measures were of consid- 
erable value in control of the false wireworms in the soil. The pupal 
cells were crushed and the pupae buried or thrown out upon the surface, 
where they were exposed to the elements and to predatory enemies. 

It is not only good farm practice but also advantageous as a control 
measure to destroy and remove from infested fields and adjacent fence 
rows all clumps of Russian thistles and other weeds or heavy growths of 
grasses likely to shelter these beetles. 

While adults in small numbers are known to feed upon poisoned bran 
mash and similar preparations, experiments in poisoning the larvae were 
not satisfactory. It appears doubtful that such poisoning will ever 
prove of practical value in dealing with this pest. 

Late sowing of wheat in the fall also has been tried as a possible control 
measure, but does not appear to be successful unless the season is a very 
dry one, and even then if the seed has to lie in the ground any appreciable 
length of time before rain and germination much injury is likely to result, 
for the larvae are most active in the dry, loose soil under such conditions. 
65769°— 21 4 



PLATE 31 
Embaphion muricatum: 

A. — Epipharynx (eph) and anterior margin of labrum. 

B. — Head: cl., clypeus; fa, anterior angle of front; epi, epistoma; /, frons; epc, 
epicranium. 

C. — Lateral view of larva. 

D. — ^Mandibles and hypopharyngeal sclerite from below. Concavity of molar part 
of left mandible grinding against the sclerite. 

E. — Gula, labiiun, and right maxilla from ventral side: gu, gula; tp, tentorial pit; 
sm, submentum; me, mentum; stla, stipes labii; li, ligula; hyp, hypostoma; fm, 
fossa for mandible; fc, fossa for cardo; ar, maxillary articulating area; ca, cardo; sti, 
stipes maxillaris; &.y, basis of stipes; z>i and Uj' i^mer margin of stipes; jmo, mala maxil- 
laris (probably lacinia); pag, basal membrane of maxillary palp. 

F. — Dorsal side of right and left mandible, hypophar>Tigeal sclerite between them: 
a^ and a^, the bicuspidate mandibular apex; t, tooth of cutting edge; m, molar part; c, 
carinate edge on exterior side of cutting part of mandible ; s, soft-skinned, seta-bearing 
elevation below the carinate edge; e, margin of chitin framing the soft elevation. 

G. — Mandibles and hypopharyngeal sclerite from below; no grinding in this position. 

H. — Left anterior leg showing the anterior face of the leg hanging perpendicularly 
down from a horizontally placed larva. 

I. — Left anterior leg, exliibiting its posterior face: cox, coxa; ir., trochanter; /e, 
femur; ti, tibia; ia, claw-shaped tarsus, shortly but not correctly designated as "claw." 



Biology of Embapliion muricatum 



Plate 31 




Journal of Agricultural Research 



Vol. XXII, No. 6 



Biology of Embaphicn muricatum 



Plate 32 




/^c^^^ 



Journal of Agricultural Research 



Vol. XXII, No. 6 



PLATE 32 
Embaphion muricatum: 

A. — Lateral view of buccal cavity with mouthparts removed: eph, epipharynx; 
hsc, hypopharyngeal sclerite; hbr, hypopharyngeal bracon;/jw, fossa of ventral con- 
dyle of mandible; oes, oesophagus (note the distance between sclerite and entrance 
to oesophagus). 

B. — Ventral view of head, the thoracic segments, and the anterior portion of first 
abdominal segment: y, presternum; eu, eustemum (Snodgrass) or basistemum 
(Crampton); hi, prehypopleurum ; hn, posthypopleurum ; 2, poststernellum (prester- 
num and poststernellum constitute together the ventral intersegmental region); ej, 
preepipleurum ; %, postepipleiunim ; te, tergite; ster, sternal shield of abdominal 
segments; ep, abdominal epipleimim ; ter, abdominal tergite. 

C. — Pygidium, lateral view. 

D. — Pygidium, dorsal view. 

E. — Pygidium, ventral view; IX, ninth abdominal ("pygidial") segment; X, 
tenth abdominal ("anal") segment. 

F. — Maxillse, ligula, labial palpi seen from the buccal cavity. (Hypopharyngeal 
region removed.) 

G. — Hypopharyngeal region, oesophagus, and hypopharyngeal bracon which all 
were removed from figure F: hsc, hypopharyngeal sclerite; hbr, hypopharyngeal 
bracon; fm, mandibular ventral xossa; oes, oesophagus. 

H. — Hypopharyngeal region, same piece as figure G, reversed: hsc, base from which 
hypopharyngeal sclerite originates; hbr, hypopharyngeal bracon; /w, mandibular 
ventral fossa; oes, oesophagus. 

I. — First thoracic spiracle. 



ADDITIONAL COPIES 

OF Xms PUBLICATION MAY BE PROCURED FEOM 

THE SUPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTING OFnCE 

WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 

Subscription Price, $3 Per Year 



iToL. XXII NOVEMBER 12, 1921 No. 7 

JOURNAL OF 

AGRICULTURAL 
RESEARCH 



CONTENTS 

Page 

Genetic Behavior of the Spelt Form in Crosses between 
Triticum spelta and Triticum sativum- - - - 335 

CLYDE E. LEIGHTY and SARKIS BOSHNAKIAN 

( Contribution from Bureau of Plant Industry ) 

Plum Blotch, a Disease of the Japanese Plum Caused 
by Phyllosticta congesta Heald and Wolf - - - 365 

JOHN W. ROBERTS 

( Contribution from Bureau of Plant Industry ) 

A Comparison of the Pectinase Produced by Different 
Species of Rhizopus - - "" ~ " -371 
L. L. HARTER and J. L. WEIMER 

(Contribution from Bureau of Plant Industry) 



PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE, 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



WASHINGTON, D. C. 

GOVERNMENT PRINTING OFFICE 

1921 



EDITORIAL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARL F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALLEN 

Chief, Office of Experiment Stations 

CHARLES L. MARLATT 

Entomologist and A ssisfant Chief, Bureau 
of Entomology 



FOR THE ASSOCIATION 
J. G. LIPMAN 

Dean, State College of Agriculture, and 
Director, New Jersey Agricultural Experi- 
ment Station, Rutgers College 

W. A. RILEY 

EntoTnologist and Chief, Dnision of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 
of Minnesota 

R. L. WATTS 

Dean, School of Agriculture, and Director, 
Agricultural Experiment Station, The 
Pennsylvania State College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultural Experiment Station, New 
Brunswick, N. J. 



JOlNiaOFAGiaaiTiAllSEMCH 

Vol.. XXII Washington, D. C, November 12, 1921 ,^ No. 7 



GENETIC BEHAVIOR OF THE SPELT FORM IN CROSSES 
BETWEEN TRITICUM SPELTA AND TRITICUM SATI- 
VUM ' 

By Clyde E. LEighTY, Agronomist, Office of Cereal Investigations, Bureau of Plant 
Industry, United States Department of Agriculture, and Sarkis Boshnakian, Depart- 
ment of Plant Breeding, College of Agriculture, Cornell University 

INTRODUCTION 

In connection with genetic studies of density in the wheat spike, it 
was noted that the appearance of dififerent specific forms in certain wheat 
crosses introduced marked irregularities in density curves of the second 
and following generations, and that the densities of Triticum sativum 
Lam., T. polonicum Linn., T. spelta Linn., etc., were affected in different 
degrees when a certain known density factor was introduced through 
hybridization. In some instances there were partial and sometimes 
total inhibitory effects in regard to density, depending upon the sub- 
species and also the kind of density factor involved in the cross. 

The occurrence of these irregularities which appeared to be caused 
by the spelt character in some crosses led to the study of the nature and 
genetics of the species T. spelta. Although studies have been made of 
the mode of inheritance of the spelt form in a large number of inter- 
specific crosses, only the different modes of inheritance in crosses where 
the parents are spelta and sativum are presented in this paper. In other 
crosses, such as turgidum Linn. X sativum, durum. X sativum, dicoccum 
Schr. X .yafwwm, etc., spelts invariably appear in the F2 generation. The 
mode of inheritance of these spelt forms is complex and variable, so their 
discussion here has been omitted. 

The plants on which these studies were made were grown, with a few 
exceptions, on Arlington Farm, near Washington, D. C, or on the Plant 
Introduction Station, Chico, Calif., both operated by the United States 
Department of Agriculture. The crosses were made at the former place 
in 1913. 

> The specific name T. sativum as used in this paper refers only to the forms T. vulgare Vill., T. compactum 
Host., and T. capitatum Schlz. These three forms are essentially the same species, their differences being 
merely a question of intemode length. The word wheat is frequently used as an English designation for 
these forms, and when so used does not include such other forms as T. durum Desf . , T. polonicum, etc. 

Journal of Agriailtural Research, Vol. XXII, No. 7 

Washington, D. C. Nov. 12, 1921 

aak Key No. G-isa 



336 



Journal of Agricultural Research voi. xxn, no. 7 



SPECIFIC DIFFERENCES BETWEEN TRITICUM SATIVUM AND TRITICUM 

SPELTA 

The shape of the outer or sterile glume is an important character in the 
dififerentiation of wheat species. The glume of the true T. sativum form 
(fig. I , B) is, as a rule, soft, with a somewhat pointed apex. It is rarely 
and very weakly keeled along the entire length. About 0.5 to i mm. 





A 



B 



Fig. I. — Characteristic spikelets of Triticum spelta (A) and Triticum sativum (B). Note upright and 
tight position of the glumes of spelta and the loose, spreading habit of satiiiim. Spelta has a flat shoulder 
(a) with two or three indentations; the shoulder of sativum (d), though varying widely, is generally not so 
prominent. It is often rather tapering with no indentations. The base of the sterile glume of spelta (b) is 
broad, showing firm attachment to the rachis. The glumes can not be opened without breaking them at the 
base. In sativum the base (/) is narrow and is weakly attached to the rachis, and the glumes can be opened 
easily. There are one or more depressions («) at the base of the glume in sativum which are not present in 
spelta. The glume of spelta is more or less uniform in width. It is stiff and has prominent corrugations 
(c); that of jaijVwwj is narrow at base, widens, then tapers again gradually. It is very thin and soft, and the 
veins on the glumes are not so prominent. Sativum has a weak keel while spelta has a very strong one . 
The Spelta spikelet usually develops only two kernels; sativum, often develops three or more. 

above the point of attachment there is a wrinkle or depression. The 
base is rather narrow and is very weakly attached to the rachis. The 
glume characters of the typical spelt, on the other hand, are quite 
different (fig. i, A). The glume is stiff and thick, with a very blunt 
apex. It is strongly keeled and has no depression above the base, 
which is wide and firmly attached to the rachis. 



\ 



Nov. 12, 1921 Genetic Behavior of Spelt Form in Crosses 337 

Some wheats exhibit some of the spelt characters in a very weak form. 
For instance, some are more or less strongly keeled or have a flat shoulder 
at the distal part of the glume. While such wheats are classified as 
T. sativum, they are not true sativum in the sense used here. They do 
not represent the type. 

Aside from the glumes, these species have other distinguishing char- 
acters. Of these, the brittleness of the rachis and nonshattering qualities 
of the spikelet in spelt may be mentioned as contrasted with the sativum 
types, the latter being nonbrittle and shattering easily. 

Although the heads of the spelt varieties commonly grown at experi- 
ment stations in this country usually are lax, laxness of the head is not 
necessarily a characteristic of the spelt. Compactness may easily be 
introduced into spelts when crossed with T. compactum, which is really 
only a wheat carrying a genetic factor or factors for compactness. 

The characters differentiating these two species are recapitulated 
below. This list includes also some other minor characters. 

T. sativum. T. spclta. 

Outer glumes — Outer glumes — 

Weakly attached by narrow base. Firmly attached by wide base. 

Weakly keeled. Strongly keeled. 

Apex tapering. Apex blunt. 

Depression or wrinkle near point of at- No wrinkle near base. 

tachment. 

Veins not prominent. Prominent parallel longitudinal 

veins. 

Shoulder narrow to broad, with no Shoulder broad, with two or three 

dentation. dentations. 

Glume soft. Glume, lemma, and palea firm and 

thick. 

Spikelets — Spikelets — 

Kernel loosely held between glumes Kernel tightly held (nonshattering). 

(shattering). 

Spikelets spreading somewhat from Spikelets oppressed tightly against 

rachis. the rachis. 

Usually three or more kernels per Usually two and rarely three kernels 

spikelet. per spikelet. 

Rachis — Rachis — 

Tenacious. Fragile. 

There also are differences in the shape of the kernels of these two 
species (i).^ 

LINKAGE OF SPECIFIC CHARACTERS 

In order to understand the manner in which these two forms are here 
classified, it is necessary to refer to the linkage of their specific characters, 
without discussing the details. 

Most of the characters tabulated above show two limits of contrasts. 
When a wheat is crossed with a spelt, a gradation of forms naturally 

' Reference is made by rmmber (italic) to " Literature cited," p. 363-364. 



338 Journal of Agricultural Research voi. xxii.no. 7 

appears in the Fj generation, ranging from the true spelt to the typical 
wheat form. In the segregates, a plant which has the glume form of 
the spelt invariably has a brittle rachis and a nonshattering habit. If 
the glume form is intermediate, the brittleness and shattering qualities 
also are intermediate These three characters do not segregate inde- 
pendently. A plant with sativum glumes, for instance, has not been 
found whose rachis is brittle (a characteristic of the spelt only), nor 
have we obtained a spelt-glumed plant which has as tenacious a rachis 
as that of T. sativum. These species may present other characters, 
such as pubescence, beardedness, glume color, etc. These are common 
to both and segregate independently irrespective of the species. The 
presence of correlation between some of these specific characters in 
sativum X spelta crosses also has been noted by von Tschermak (12). 

Unless the linkage is absolute it does not exclude the possibility of 
the occurrence of crossovers, but if crossing over ever occurs with respect 
to these characters it must be very rare. 

The linkage of specific characters of spelt is very similar to the linkage 
of a number of glume characters of the wild oats in crosses between wild 
and cultivated forms described by Surface (u) and by Love and Craig (5), 

It is absolutely necessary to bear in mind that the determination of the 
species is based only upon the presence, intensity of development, or 
absence of the specific characters (glume characters, brittleness of rachis, 
and seed-holding habit) , all of which are linked to a very large extent. 
These characters show no independent segregation. It is obvious, 
therefore, that they do not mendelize independently. The characters 
which do segregate independently, such as pubescence, beardedness, 
glume color, etc., are not taken into consideration.^ 

METHOD OF CLASSIFYING THE FORMS 

The second and subsequent filial generations of spelta X sativum crosses 
show numerous gradations between the two parental forms. Some of 
the F2 plants produced a progeny consisting of individuals which were 
decidedly spelts, while others produced a progeny which, while all spelts, 
yet resembled T. sativum to some extent and were distinctly different 
from the former group of plants where all the individuals were markedly 
spelt-like. 

Ten arbitrary classes were made in order to record the degree of in- 
heritance of spelt characters. The typical or intense spelt which ex- 
hibited all the specific spelt characters in their extreme forms was graded 
I. As the exhibition of the intensity of these characters diminished — 
that is, as they tended to approach those of the wheat — ^the heads were 

' Even these independently segregating characters vary in intensity and quality in these two species 
when segregating in crosses. 



Nov. 13, 1921 Genetic Behavior of Spelt Form in Crosses 339 

classed 2, 3, and so on to 9. Class 10 includes only true wheats, the 
forms which show absolutely no trace of the spelt characters. It must 
be emphasized that these 10 classes are purely arbitrary, based upon 
the appearances of the heads, and are not intended to represent any geno- 
typic classes. The types representing these 10 classes are presented in 
Plate 33, A. 

When the degree of inheritance of the spelt characters by a heterozy- 
gous population is recorded on two different occasions, a variation in 
the class frequencies naturally may be expected, but the general form 
of the curve remains practically the same. The error which may affect 
the conclusions rests in the decision as to whether a particular plant 
belongs to class 9 or 10. The degree of error depends upon the cross 
examined. 

Of the plants discussed herein, which were tested in the Fg generation, 
there were five cases where individuals belonging in class 9 had been 
placed in class 10 and only one case where a class 10 individual was 
erroneously recorded as belonging to class 9. In the tables these six 
corrections have been made. As the conclusions are based not merely 
on the F2 individuals but upon the progeny of these, it does not seem 
likely that this source of error could have affected the results to any 
extent. 

The spelt parents used for making the crosses graded from i to 4 ; the 
variety of wheat known as Gatineau ^ and herein considered as speltoid 
in form graded from 4 to 7; and all the T. sativum parents of course 
graded 10. As the discussions in this paper are confined only to the 
spelt character it does not seem necessary to describe further the agrono- 
mic and botanical characteristics of the plants used in making the crosses, 
as these have no direct bearing on the subject. 

FAMILIES SHOWING THE PRESENCE OF ONLY ONE FACTOR FOR 

SPELTING 2 

The hybrids of the first filial generation (FJ of a spelta X sativum cross 
are slightly intermediate in form, resembling the spelt more than the 
sativum parent. They grade from 4 to 7, depending upon the cross. 
They possess all the characteristics of a spelt, but the spikelets may be 
somewhat more open and the grains may not be so firmly held within 
the glumes. The spelt characters are so nearly completely dominant 
that they inhibit all wheat characters. 

In the second generation a segregation is obtained where the indi- 
viduals vary, producing forms ranging from the typical wheat form to 

' This variety originated as a cross of Red Fife 9 (T'. sativum) and Goose S {T. durum) (lo, p. 239). It 
is a striking illustration of the spelt-like segregates which appear when these two wheat species are crossed. 
The heads are sc much like spelt that at blooming time especially it is easily mistaken for spelt, but it 
thrashes free and is otherwise like the common wheats. 

' For brevity, "spelting" is used throughout this paper in place of "inheritance of the spelt characters." 



340 



Journal of Agricultural Research vou xxii, no. ? 



the spelt. The curve of these gradations is not a binomial frequency 
curve, but, on the contrary, more individuals are found, as a rule, at the 
extremities of the range than near the center. This fact may be ob- 
served in Table I, which shows the forms of the Fj segregates. 

Table I. — Number of plants of the F^ generation falling into each of lo classes, based on 
presence or absence of spelt characters, with total number of spelts arid wheats actually 
obtained, calculated numbers on basis of j to i ratio, deviation, probable error, and ratio 
between deviation and probable error 



Series. 



Crosses. 



Number falling in class — 



To- 
tal. 



13255a 

13260a 

13263a 
3094a 

3085a 

13124a 

13125a 



Spelt sel. i344oXDale Gloria set. 

13401 

Spelt sel. 13438 X Turkey (C. I. 
3375) sel. 13389 

Dale Gloria sel. i340iXspelt 

Black Bearded spelt X Early Red 
Chief 

Black Bearded spelt X Jones Long- 
berry 

Vulgare (C. I. 3338)Xspelt sel. 

13437 

Crimean (C. I. 3340) sel. 13351X 
spelt sel. 13437 



63 

25 

81 

50 
76 

5° 



Total of crosses . 



418 



Series. 



13255a 
13260a 
13263a 
3049a 
3085a 
13124a 
13125a 



Crosses. 



Spelt sel. i344oXDale 

Gloria sel. 13401 

Spelt sel. 13438 XTurkey 

(C.I- 3375) sel. 13389- 
Dale Gloria sel. 13401X 

spelt 

Black Bearded spelt X 

Early Red Chief 

Black Bearded spelt X 

Jones Longbcrry 

Vulgare (C. I. 3338) X 

spelt sel. 13437 

Crimean (C. I. 3340) sel. 

i335iXspeltsel. 13437 

Total of crosses . . . 



Number of plants. 



Obtained. 



Spelts. Wheats. 



52 
49 
18 

52 
37 
56 
41 



305 



113 



Calctilated. 



Spelts. Wheats. 



54-8 

47-3 
18.8 
60.8 

37-5 
57- o 
37-5 



313-5 



IS- 7 
6.2 
20.3 
12.5 
19. o 
12.5 



104.5 



Devia- 
tion. 



8.5 



Prob- 
able 
error. 



2-5 
2-3 

1-4 
2.6 
2. I 

2.5 
2. I 



6.0 



Ratio 

Dev . 
P.B. 



•74 
•57 

3-38 
.24 
.40 

I. 67 



I. 42 



As it is impossible to distinguish the homozygous spelts from the 
heterozygous forms, in determining the ratios all the spelt and speltoid 
forms (classes i to 9) have been grouped together and compared with 



Nov. 13, 1921 Genetic Behavior of Spelt Form in Crosses 



341 



the wheat forms (class 10), which show no trace of spelt characters. 
The proportions between spelta and sativum forms of each of the crosses 
taken separately and that of the totals of these two groups approximated 
the monohybrid ratio of 3 to i. The obtained ratio of the totals of the 
crosses was 305 speltoid and spelt forms to 113 wheats, the expections 
being 313.5 to 104.5, respectively, showing a deviation of 8.5 with a 
probable error of ± 6.0. 

Two of the Fj families, 13260a and 13255a, gave the results shown in 
Table II when the F3 generation was grown. 

Table II shows that the Fj population of the families tested consisted 
of individuals in the proportion of i homozygous spelt to 2 heterozygous 
forms and i homozygous wheat. 

Table II. — Number of F2 plants from series ij26oa and Ij2^ja which proved to be homo- 
zygous spelts, heterozygous forms, and homozygous wheats {i : 2 : i) when tested in the 
/• 3 generation 





13260a, Spelt X Turkey. 


132553, Dale Gloria X Turkey. 


Nature of data. 


Homozy- 
gous 
spelts. 


Heterozy- 
gous 
forms. 


Homozy- 
gous 
wheats. 


Total. 


Homozy- 
gous 
spelts. 


Heterozy- 
gous 
forms. 


Homozy- 
gous 
wheats. 


Total 


Obtained 

Calculated 

Deviation 


8 

7-5 
•5 


13 
15.0 

2. 


9 

7-5 
•5 


30 


7 

8.2 

1.2 


14 

16.5 
2-5 


12 
8.3 

3-8 


33 



The ratios of the totals of the forms produced in the F3 generation by 
the heterozygous Fg individuals of these two families (Table III) seem 
to conform to the foregoing assumption, although the ratios of the 
forms produced by each of the F2 individuals sometimes are not so 
close to the 3 to i expectation. Of the total individuals produced by 
the Fj heterozygous plants of series 13260a (spelt X Turkey) , 212 were 
spelts and 71 wheats. These results were surprisingly close to the 
expectation, the deviation from the calculated numbers being but 
0.3 with a probable error of ± 4.9. In series 13255a (Dale Gloria X spelt), 
the numbers obtained from the heterozygous individuals were 365 
spelts and speltoids and 156 wheats; the deviation here was 25.8 with a 
probable error of ±6.7. This apparent dominance of the spelt character 
over that of the wheat and its segregation into the 3 to i ratio are in 
accord with the observations of Pitsch, as cited by von Tschermak 
{12, p. ijg) and of Kajanus (j). 



342 



Journal of Agricultural Research voi. xxii. no. 



Table III. — Numbers of spelts and wheats produced in the F3 generation from the F^ 
heterozygous plants, and comparison of these with theoretical expectations, calculated 
on the J to I basis 





Total F3 
plants. 


Number of plants. 


Devia- 
tion. 


Probable 
error. 


Ratio 


Pedigree No. 


Obtained. 


Calculated. 


Dev. 




Spelts. 


Wheats. 


Spelts. 


Wheats. 


P.E. 


132603-3 . . . . 

6 

7. ... 

9. ... 

10. . . 
13- •• 

14. .. 

15. .. 
19. .. 
25. .. 

28. .. 

29. . . 

30. .. 


23 
20 
28 
17 
23 
24 
17 
20 
29 
17 
17 
18 

30 


13 
16 
20 
10 
18 
17 
13 
16 
26 
13 

12 
16 
22 


10 

4 
8 

7 
5 
7 
4 
4 
3 
4 
5 
2 
8 


17-3 
15.0 

21. 
12.8 

17-3 
18.0 
12.8 
15.0 
21.8 
12.8 
12.8 

13-5 

22. 5 


5-7 
5-0 
7.0 
4. 2 

5-7 
6.0 
4.2 

5-0 
7.2 
4.2 
4.2 
4-5 
7-5 


4-3 
I. 
I. 
2.8 

•7 

I. 

. 2 

I.O 

4.2 
. 2 
.8 

2-5 

•5 


1.4 
1-3 

I- 5 
I. 2 
1.4 
1.4 
1.2 

1-3 
1.6 
I. 2 
I. 2 
1.2 
1.6 


3-1 
.8 

•7 

2-3 

•5 
•7 
. 2 
.8 
2.6 
. 2 

•7 
2. 1 

•3 


Total.. 


283 


212 


71 


212.3 


70.7 


•3 


4.9 


. I 


i325Sa-7 •• • ■ 
8. ... 
10. . . 
II . . 

15. .. 

16. .. 

17. .. 
21 . . . 

23 •• ■ 

24. . . 

25. . . 


36 

19 

24 

35 
59 
33 
13 
52 
49 
35 
44 
37 
50 
35 


34 
15 
20 

27 

44 

23 

4 

28 

29 
18 

33 
25 
33 
32 


2 
4 
4 
8 

15 
10 

9 
24 
20 

17 
II 
12 

17 
3 


27. 

14-3 
18.0 
26.3 

44-3 

24.8 

9.8 

39- 
36.8 
26.3 

33- 
27.8 

37-5 
26.3 


9.0 

4-7 
6.0 
8.7 

14.7 
8.2 
3-2 

13.0 

12. 2 
8.7 

II. 
9.2 

12.5 
8.7 


7.0 

•7 
2. 

•7 

•3 

1.8 

5-8 

II. 

7.8 

S.3 



2.8 

4-5 
5-7 


1-7 
1-3 
1.4 

1-7 
2. 2 

1-7 

1. 

2. I 
2. 
1-7 


4.1 

•5 
1.4 
4.1 

. I 
1. 1 
5-8 
5-2 

3-9 
4.9 


26. .. 
28. .. 
3°- •• 


1.8 
2. 1 

1-7 


1.6 
2. I 
3-4 


Total.. 


521 


365 


156 


390.8 


130. 2 


25.8 


6.7 


3-9 



FAMILIES SHOWING THE PRESENCE OF TWO SPELT FACTORS 

Of the crosses studied, two families, 13126a (Giant Squarehead X spelt) and 3019a 
(spelt X Salt Lake Club) produced a very low proportion of wheat types. Not much 
importance would have been attached to the irregular behavior of these families if 
an apparently similar behavior had not been observed in another sativum X spelta 
cross. The manner of segregation of the progeny of these two crosses is given in 
Table IV. 

Table IV. — Degree of spelling and proportions of spelts and wheats obtained in the 
F2 generations of spelta X sativum crosses which did not segregate in the 3 to T ratio 





Degree of spelting in class— 


Total. 


Niunber of plants. 


Devi- 
ation. 




Series. 


I 

12 

27 


2 

7 
15 


3 

7 
9 


4 

6 

7 


s 

9 
3 


6 

8 

5 


7 

3 
2 


8 

6 

5 


9 

12 
8 


10 

4 
2 


Obtained. 


Calculated. 


bable 
error. 




Spelt. 


Wheat. 


Spelt. 


Wheat. 


13126a '. . . . 
3019a ^ 


74 
83 


70 
81 


4 
2 


69.4 

77.8 


4.6 
5-2 


0.6 
3-2 


I. 40 
1.49 



1 Series 13126a, Giant Squarehead (C. I. No. 3351, selection 13366) X spelt (selection 13437). 
' Series 3019a, white spelt X Salt ijake Club. 



Nov. 12. 1921 Genetic Behavior of Spelt Form in Crosses 343 

In series 13126a (Table IV), only 4 wheats were produced in a popula- 
tion of 74 F2 plants, while in series 3019a, 2 wheats were produced in an 
F2 population of 83 individuals. The deviation from the 3 to i ratio is 
so great that even by grouping class 9 with class 10 — that is, by making 
generous allowances for observational error — the proportion approached 
more nearly the 15 to i ratio. On the basis of the 15 to i ratio, the 
expectation in series 13126a, is 69.4 to 4.6, in series 3019a, 77.8 to 5.2. 
The deviations are 0.6 and 3.2, and the probable errors ±1.40 and 
± 1.49, respectively. 

On examining the Fg generation produced from 27 plants of series 
13126a, it was found that 12 of these had produced only spelts (Table V), 
two plants yielded only sativum types, and the remaining 13 F2 plants 
yielded progeny of mixed forms. Assuming that the spelt parent in 
this particular cross carried two spelt factors, Sj and Sj,^ the first two 
generations will consist of the following genotypic forms: 



Pi 


(Giant Squarehead) 


S1S1S2S2 X wibib2>-'2 


(Winter spelt) 




T. sativum 




T. spelta 


Fi 




(Speltoid) 




F2 


I SjSjS2S2 


I ^1^1^2^2 4 W1S1O2S2 


2 S1S1S2S2 






2 Oj^Oj02S2 


2 SiSiS2S2 






I D]^OiS2S2 








2 OjSj02W2 








I SjSj^0202 




Tot 


al I T. sativum. 


I q spelt. 





If these genotypes were carried through the F3 generation the theo. 
retical behavior of each of the Fg plants would be as follows: 

GROUPS. F2 GENOTYPES. PHENOTYPES OF THE Fs AND THEIR BEHAVIOR IN THE F3 GENERATION. 

A I SjSjSjSj I wheat will yield wheat only. 

B I SjSiS2S2 

2 t3iOi02S2 

1 S1S1S2S2 

2 DiSib202 
I SiSjv!)2v32 

C 4 SiSiS2S2 4 Spelts will segregate 15:1 

15 2 S1S1S2S2 1 ^ cr.«ife will oo^.-o„o+» ... \^ unstable forms. 



>7 spelts will yield spelts only. 



4 SiSiS2S2 4 spelts will segregate 15:1 | 

2SisaS }4 spelts will segregate 3:1 f 



Total 16 

When the performances of the Fj plants were examined, a close 
approximation was found to the above-mentioned theoretical ratios. 
The numbers of constant wheats, constant spelts, and unstable spelt 
forms obtained are shown in Table V, together with the theoretical 
expectations. 

' In the factorial explanations given in this paper the spelt factors are assumed to be Si and S2. Although 
the assumption of the factors si and sz to stand for the wheat (T. sativum) character will fully agree with the 
results obtained, so far as the ratios go, there is no evidence as yet to warrant the assumption that the wheat 
and spelt characters are allelomorphic to each other. In fact, results with other specific crosses show the 
possibility that these are caused by two sets of independent factors. The behavior of the sativum X spelta 
crosses may be compared with the behavior of a maize cross where one parent has yellow endosperm and 
purple aleurone color (YYCCPP), and the other differs from this by its lack of purple color (YYCCpp). 
In such a cross, where the F2 shows segregation into 3 purple to i yellow, the assumption that yellow and 
purple are allelomorphic may be used as a working hypothesis for crosses of this type, although it is not the 
correct explanation, as endosperm and aleurone color are two different characters altogether. 



344 



Journal of Agricultural Research voi. xxii.no. 7 



Table V. — Frequencies of spelt and wheal classes in F3 progeny of F2 individuals of 
family Iji26a, Giant Squarehead X spelt 



FAMILIES CONSISTING OP WHEATS ONLY 





Class 
of F2 
parent 
plant. 


Classes of spelt inheritance Fa — 


Number of plants. 


Pedigree numbers of Fs. 


I 


2 


3 


4 


5 


6 


7 


8 


9 


10 

4 
10 


Spelts 
and 
spel- 

toids. 


Wheats. 


Total. 


7 


10 
10 




4 
10 


4 
10 


2? 







FAMILIES CONSISTING 


OF SPELTS ONLY 








2 


5 
3 
9 
7 
3 
6 

4 
5 
6 


6 

5 

9 

4 

3 
I 
2 
3 


6 

4 

2 
2 

I 
2 

I 

3 
4 


4 

I 

6 

I 
2 
3 
3 
I 
2 
2 


I 

I 

I 

2 

I 
I 

3 


1 

2 

2 
2 

I 
I 


I 
I 

I 

I 


I 

I 
3 


I 

I 
I 


13 

2 

3 
2 

I 

I 




18 
10 
13 
13 
13 
5 
18 

14 
5 

14 
10 

5 




18 


A, 


10 


c 


13 
13 
13 
5 
18 


7 


8 


12 


IC 


16 


14 
5 

14 
10 


20 


22 


24. 


I 


21; 


5 
















Total 
























138 




138 



























FAMILIES CONSISTING OF 


SPELTS 


AND 


WHEATS 


(heterozygous f 


2 plants) 


\ 
I 


5 
6 
8 

9 

8 
8 

7 
5 
8 
I 
I 
9 
5 


I 

2 

2 
I 
I 
2 


2 

3 
2 

I 

3 

5 
2 

3 

5 

I 


5 

I 

I 
3 

4 

5 
I 


3 

I 

I 

I 
2 

2 


I 


I 
I 

I 

I 


I 

I 
I 

I 


I 

2 

I 

I 
I 

I 
I 


I 
I 
I 

3 
I 

13 

3 

3 

2 

3 

I 


I 
2 
2 

I 
I 
I 

4 
I 
I 

I 

4 
2 
I 


16 
9 

5 
3 
7 

14 
7 

14 
16 

5 
18 

4 

5 


I 

2 
2 

I 
I 
1 

4 

I 
I 
I 

4 
2 
I 


17 
II 


6 





7 
4 
8 


10 


II 


i-j 


15 
II 


14. 


17 


15 

17 

6 


18 


IQ 


21 


22 


26 


6 


27 


6 






Total progeny of he- 
terozygous F2 plants. . 
Expectations 
























123 
108. 75 


22 
36.25 


145 





























SUMMARY AND GROUPING 





Types. 


Number of plants. 




Groups. 


Obtained. 


Calculated 

7:8:1. 


Deviation. 


A 


Wheats producing wheats only 


2 
12 
13 


1-7 
II. 8 

13-5 


0.3 
. 2 


B 


Spelts producing only spelts 


Cand D... 


Spelts producing both spelts and wheats. . 


•5 



Nov. 12. 1921 Genetic Behavior of Spelt Form in Crosses 



345 



The agreement between the proportions expected and those obtained 
is very close indeed to the 7 to 8 to i ratio, and perhaps too close to be 
ordinarily expected from such a small population. 

The analysis may be carried a step further. As shown above, the 
plants which would show unstability in the F3 (groups C and D), were 
expected to be of two different genotypes. One of them, containing the 
S1S1S2S2 forms (group C), was expected to segregate in the 15 to i ratio, 
while the other (group D), containing the S1S1S2S2 and S1S1S2S2 genotypes, 
should segregate in the simple 3 to i monohybrid ratio. Apparently 
the individuals belonging to each of these two groups are those analyzed 
in Table VI. 



Table VI. — Analyses of unstable spelts of the F2 generation, series 12126a. Number of 
individuals of the F^ generation produced from Fn plants of groups C and D, compared 
with the theoretical expectation 

■p2 INDIVIDUALS APPARENTLY SEGREGATING IN THE 1 5 TO I RATIO (GROUP C, 81818252) 





Total. 


Number of plants. 


Devia- 
tion. 


Probable 
error. 


Ratio 

Dev. 
P.E. 


Pedigree. 


F3 obtained. 


F3 calculated. 




Spelts. 


Wheats. 


Spelts. 


Wheats. 


I3i26a-i . . . . 
II . . . 

13. . . 

17. . . 

18. . . 


17 
8 

IS 
15 
17 


16 
7 
14 
14 
16 




IS- 9 
7-S 
14. I 
14. I 
IS- 9 


I. I 

■ s 

-9 

-9 

I. I 


0. I 

• s 

. I 
. I 
. I 


0.68 
.46 
.62 
.62 
.68 


0. I 

1. I 
. I 
. I 

. I 


Total.. 


72 


67 


5 


67- S 


4-5 


• s 


1.38 


•3 



F2 INDIVIDUALS APPARENTLY SEGREGATING IN THE 3 TO I RATIO (GROUP D, S1S1S2S2 

AND S1S182S2) 



i3i26a-6. . . . 


II 


9 


2 


8.3 


2.7 


0.7 


0. 96 


0.7 


9 


7 


5 


2 


5-3 


1-7 


•3 


•77 


•4 




4 
II 


3 

7 


I 


3-0 
8.3 


I. 









14. . . 


4 


2.7 


1-3 


.96 


1.4 


19. . . 


6 


5 


I 


4-S 


i-S 


•S 


.72 


•7 


21 . . . 


22 


18 


4 


16.5 


s-s 


I- 5 


1-37 


. I 


26. . . 


6 


4 


2 


4-S 


1-5 


•5 


.72 


•7 


27. . . 


6 


S 


I 


4-S 


I- 5 


•S 


.72 


•7 


Total.. 


73 


56 


17 


S4-8 


18.2 


I. 2 


2. 50 


-5 



Total Fa plants segregating in 15 to i ratio, obtained, 5. 
Total F2 plants segregating in 15 to i ratio, calculated, 6.7. 
Deviation, 1.7. 

Total F2 plants segregating in 3 to i ratio, obtained, 8. 
Total F2 plants segregating in 3 to i ratio, calculated, 6.7. 
Deviation, 1.3. 



The data in Table VI show that forms were obtained in the F2 some 
of which segregated in the 15 to i and others in the 3 to i ratio as ex- 
pected. The agreement to the theoretical numbers of the progeny of 
each F2 plant is as close as can be expected with such small numbers, even 



34^ Journal of Agricultural Research voi. xxn, no. 7 



though it is borne in mind that the values of the probable error are likely 
to be too high in data of this kind. 

Summing up the types of groups C and D, the former yielded 67 spelts 
to 5 wheats, and the latter group yielded 56 spelts to 17 wheats, the 
deviation being 0.5 and 1.2 and the probable errors ±1.38 and ±2.50, 
respectively. 

Of the 2 7 plants tested (Tables V and VI) , 5 (group C) showed an ap- 
proximation to the 15 to I ratio and 8 (group D) to the 3 to i ratio. The 
theoretical number of plants belonging to each of the two unstable groups 
was 6.7 — that is, one-fourth of the total Fj plants tested. Comparing the 
results obtained with those expected, it will be noted that there were in 
the F, generation two (1.7 actual) SjSiSaSg individuals less and one (1.3 
actual) SjSiSjS, or s^SiSaS, more than expected (Table VI) . 

In considering the 15 spelt to i wheat segregation, it should be borne 
in mind that if the progeny of a heterozygous Fj plant is less than about 
10 individuals, the chances are that the wheat form, which is expected 
to appear but once in a population of 16 individuals, will not be obtained. 
Such heterozygous plants producing only spelts and no wheats, on account 
of their small F3 population, would be classified under group B. Had Fj 
plants 13126a — 12, — 20, and — 25 produced more than 4 or 5 individuals, 
I or 2 of them might have produced a wheat form which would have placed 
them in group C. The experimental ratios then would almost coincide 
with the theoretical. 

Considering the closeness of agreement even in the details of the 
analysis, with such small numbers, there seems to be no question that we 
are dealing here with two spelt factors and that the ratio observed is the 
ratio of 15 to I. 

In the absence of more experimental evidence, the simplest hypotheses 
were given to account for the 3 to i and 15 to i segregations. Notwith- 
standing the surprisingly close agreements between the experimental and 
theoretical ratios, however, the real explanation concerning the produc- 
tion of the spelt character is still a matter of speculation. 

The same spelt parent plant was used in crosses 13124a, 13125a, and 
13126a. The wheat parents were of different varieties. No satisfactory 
explanation can be offered as to why the same spelt parent should pro- 
duce a 15 to I ratio in cross 13126a and a ratio of 3 to i in crosses 13124a 
and 13125a. Three possibilities, however, may be mentioned. 

1. The spelt plant used in the above-mentioned three crosses might 
have been heterozygous for one of the spelt factors. Such a heterozygous 
SjSiSjSj plant bears gametes producing 3 to i and 15 to i ratios in the 
F2 generation when crossed with a double recessive (SjSjSjSj) wheat. 

2. The spelt parent may be assumed to have carried two spelt factors 
Sj and S2 and some of the wheats might have carried an inhibiting factor 
I. If the wheat carried the I factor the ratio of spelt to nonspelt would 
be about 3 to i and if it did not the ratio would be 15 to i. 



Nov. ij. I93I Genetic Behavior of Spelt Form in Crosses 



347 



3. The spelt might have carried a spelt factor Sj and in addition 
another factor 83, which would produce the spelt character if there were 
present its complementary factor C, which might have been supplied by 
the wheat parent. In this case if C were present the ratio would be 
about 8 spelts to i wheat; and if C were not present, it would be 3 to i. 
The ratios to be expected on the basis of this last possibility, however, 
are not in accord with the experimental results. 

The second explanation seems to be the most plausible of the three. 
The assumption that some wheats carry the factor for the inhibition of 
the spelt character is not a mere speculation but a fact, as will be seen 
later when the question of the production of synthetic spelts by crossing 
two wheats is taken up. As commercial strains of T. sativum are not 
purified with respect to inhibitors, there are undoubtedly some strains 
which contain individuals heterozygous for this factor. If such a plant 
is used, for instance, as the female parent and is crossed with a spelt 
carrying the factor for spelting, according to the hypothesis some of the 
seeds will produce Fj progeny where in some instances the ratio will sug- 
gest a 15 to I, and in others a 3 to i segregation. As will be seen later, 
other modifications of these ratios may also be expected to arise. 

PROGENY OF STABILIZED SPELTOID X SATIVUM CROSSES 

The speltoid form used in the crosses which will be considered now 
is commercially known as "Gatineau." The variety originated from a 
cross between T. durum and T. sativum. It grades usually from 4 to 7 
in the classification used in this study for spike form and so resembles 
the commercial spelts in this respect. It does not have the brittle rachis 
of spelt, and the grain thrashes from the glumes more easily than the 
grain of spelt, being like some of the tight-glumed wheats. It is neither 
a typical spelta nor a typical sativum. 

The F^ plants of the crosses between Gatineau and T. sativum were 
almost like Gatineau. The Fj generation consisted of forms which were 
intermediate; typical spelts of classes i to 2 were not found. The classi- 
fication of the Fj individuals is reproduced in Table VII. 

Table VII. — Classes of F2 generation plants of two speltoid X wlieat crosses. Numbers 
of individuals obtained compared with the theoretical expectations 





Degree of spelting. 


Number of plants. 


De- 
via- 
tion. 




Series. 


I 


2 


3 

2 

I 


4 

2 



5 

4 
5 


6 

9 
9 


7 

10 

7 


8 

II 
9 


9 

15 
15 


10 

15 
17 


Total. 


Obtained. 


Calculated. 


Prob- 
able 




Spel- 
toid. 


Wheat 


Spel- 
toid. 


Wheat 




13228a ^ 

13229a 2 










68 
63 


Si 
46 


15 
17 


51.0 
47.2 


17.0 
IS- 8 


2. 
I. 2 


2.4 

2-3 



> Series 13228a Turkey (C. I. 3375. Sel. 13389) X Gatineau (C. I. 2959. Sel. 13403). 
'Series 13339a Seneca Chief (C. I. 3372, Sel. 1338S) XGatineau(C. I. 3959, Sel. 13403). 



348 Journal of Aqriculiural Research voi. xxn.No. 7 

So far as the ratios of speltoid to wheat forms are concerned, these 
crosses segregated in the simple mendeUan fashion. In series 13228a 
(Turkey X Gatineau) there were 53 speltoids and 15 wheats, showing 
a deviation from the theoretical numbers of 2.0 with a probable error of 
±2.4 In series 13229a (Seneca Chief X Gatineau) there were 46 speltoids 
and 17 wheats, the deviation here being 1.2 with a probable error of ±2.3. 
In both cases the approximations of the figures obtained to those 
expected are within the range of their probable erros. Therefore, it can 
safely be concluded that the ratio is 3 to i and that there very likely is 
but one spelt factor difference. 

The question of interest in the inheritance of this speltoid form 
(Gatineau) is not so much in its 3 to i ratio as in the way it differs in 
details from the spelt X wheat crosses first discussed, which segregated in 
the ratio of 3 to i. 

In order to compare the Fj curves of these two groups of spelt X wheat 
crosses, the Fj frequencies in Table VII (series 13228a and 13229a) and 
the first two series in Table I (13255a and 13260a) are represented graphi- 
cally (fig. 2). 

The comparative characteristics of the curves of these two sets of 
crosses are as follows : 

The curves of the true spelt X wheat crosses (13255a and 13260a) begin 
at class I , where they have their highest spelt frequencies. They gradu- 
ally drop until they reach classes 5 to 9, inclusive, where there seems to 
be an indefinite fluctuation of frequencies. Then the curves suddenly 
rise again at class 10, which contains the spelt-free populations. 

As to the curves of the wheat X speltoid crosses, the Fj spelt popula- 
tions begin at about class 3, where but a very few individuals are found. 
Beginning at class 5, the curves steadily rise until they reach their maxi- 
mum height at class 10. The curves produced by the true spelt crosses, 
it will be recalled, continuously dropped instead of ascending. 

When these two sets of crosses are compared it will be observed that, 
although there is but one spelt factor difference in each, the spelt factor 
present in crosses 13255a and 13260a is entirely different from the spelt 
factor present in crosses 13228a and 13229a. The wheat parent has 
had no influence is producing this variation in distribution, the same 
wheat parent plant, Turkey, C. I. No. 3375, selection 13389, having been 
used in crosses 13260a and 13228a. 

GENERAL DISCUSSION OF THE GROUPINGS WITHIN THE SPELT AND 

SPELTOID CLASSES 

Having discussed the question of ratios, let us turn our attention to 
the analyses of the details of the variations within the spelt and speltoid 
classes. 

The possibility of distinguishing the homozygous from the heterozyg- 
ous spelts of the Fj generation is of primary interest. Table VIII has 



Genetic Behavior of Spelt Form in Crosses 



349 











_ . 






1 






SER/ES /326ba ^P£^T/9/V^^ TUf^KEY 

SERI^ /325Sa fSPEET/^ND p/?L£ ^LORf/^ 

SERIES i3228a TURKEY /f/yPO^XIf^ ^^f- 
SERIES /3229<Z^£N£C/9 CM/£F^NOGffn 


1 




-♦••^•^ 


V£/fu\ 


















\ 




































/, 
















j 


















/ 


\\ 


















\\ 




\ 












t 


If 




\ 










K 


1 


If 




VN 








y 




1 

1 


II 




\ 


\ 




/ 


\ 


/ 




If 




\ 


\ 




/ 




/ 




1 




\ 


\ 




/ 








1 




^ 


', j^ 




/ 








f 








^sJ 


f 








1 








r 










f 








4\ 




V 




J 




/ 




s 

1 

I 




/ 




^ 






/-" 


\ 


1 

f 








/ 





OSOPSEOF SP£LT/Ne 

Fig. 2.— Degrees of spelting of F2 population true spelt X sativum crosses (13255a and 13260a) as compared 
with the curves of sativum X speltoid crosses (13228a and 13229a) of the same generation. 



350 



Journal of Agricultural Research voi. xxn. no. ? 



been prepared to show these differences. Two sets of frequencies are 
represented. One set represents the classes of spelt inheritance of the 
Fj plants of series 13255a and 13260a, which produced nothing but 
spelt in the F3 generation. The other set shows the plants of the same 
generation and series which proved to be heterozygous for the spelt 
character and produced spelts as well as wheats in ratios approximating 
3 to I. These figures are taken from the data in Tables IX and X. 

Table VIII. — Comparison of classes of spelt inheritance of tested homozygous and 
heterozygous plants of the F^ generation 





Grades of spelting. 


Totals. 






I 


2 


3 


4 s 


6 


7 


8 


9 




Homozygous spelts: 

I'ja'iica 


3 

I 


3 

5 


I 
I 












7 
8 


1. 71 

2. 25 


13260a 


I 
























Average of means 


1.98 
























Heterozygous spelts: 

i?2cca 


I 

4 


2 
3 


2 

I 


2 


2 
2 


I 


3 

I 




I 
2 


14 
13 


4.64 
3-69 


13260a 




Average of means 


4. 16 

























From Table VIII it is seen that the plants which proved to be homozy- 
gous for the spelt character occur from class i to class 4, the average of 
their means being 1.98. The Fj plants which proved to be heterozygous, 
on the other hand, came from practically all classes, the average of the 
mean classes being 4.16. 

Table VIII shows that, although there is no sharp difference in the 
phenotypic appearances of the homozygous and heterozygous individuals, 
yet as a rule the Fj spelts of the speltoid classes (classes 5 to 9) are far 
more likely to be heterozygous for the spelt character than those of the 
true spelt classes (classes i to 4). 

The comparison of the distributions of the F3 populations of the Fj 
homozygous spelts of the 3 to i and 15 to i segregating families is of 
special interest, as, among other things, it supports the two-spelt-factor 
explanations already given. The distributions of the F3 populations of 
constant spelt-producing Fj individuals of each of the three series 13126a, 
13255a, and 13260a are recorded in Tables V, IX, and X, respectively, 
and are represented graphically in figure 3. 



Nov. 12, 1921 Genetic Behavior of Spelt Form in Crosses 



351 




CI/9SSES OFSP£L TEDA/ESS 



JO 



§<50 




•3 <!f- ,S 6 7 8 
CL/fSSES OFSPEl TEDNE6S. 



10 

































— 


— 




— 


— 




— 








— 


1 




90 


















































/ 


I 






















































1 


1 




CO 80 
^ 70. 


















































1 


1 
















































~ 


~ 


1 


1 




6' 


fA 


/£ 


|?y 


'S/£ea 


c 


?//5 


WT^Q(//9/?1 


-» 


p 


mm 


V7 


•5V 


''iS 


z- 


r 


1 


1 






L. 






















































V 












































1 




1 




^ 




A 


































J 


l" 








( 




1 




P 


Ji 




I— 
































/ 


\ 


~ 






1 


~ 










\ 










































1 








^ 'f-O 






\ 


-., 






/ 


^ 




















1 




h\ 








v 








k ^^ 






) 


r^' 


s 




^/ 


^\ 


























\ 






f 


~; 


JT 


1 


1^0- 


-a 


^ 


■^- 


t-- 


V 


■sj 




V 

\ 


\ 
















i 
















^ 






-15 

■18 






v 


-w 


'^ 


k- 




^\ 
















/ 








\ 




1 




/ 




1 


^ 




3 


? 




^ 




x*^ 












• 


^5 










\ 




«/ 


il 








■/o' 




^ 


^ 


2 


^. 


^ 






?s 


I 


V 










/ 












^ 




/ 






r- 






3 




!? 


K^ 


' '\ 




'^ 


n 






^ 




V 














1 


; 


u 


J^ 


X I 








/ 


^ 




X 


/ 


S 






J^ 


»*j 




> 


-V 


A 


^ 


V 


--. 


.^^ 


r-' 


— 


k 




/" 






^i 


O' 


■i? 


^ 






z 










V 








/ 


^ 


^ 


«< 

<' 


^ 


*^ 


(£* 


^ 


>»: 


s 






le 












_ 









_ 










~ 












— 


^ 













is t3 ^- ^5" /9 7 A 

CLffSSES 0FSP£LTEDN£6a 



/O 



Fig. 3. — Comparison of Fj generation curves of the progeny of F2 spelts which yielded only spelts 
(series 13260a, 13255a, and 13126a). 

70494°— 21 2 



352 



Journal of Agricultural Research voi. xxn. no. 7 



Table IX. — Frequencies of spelt and wheat classes in F3 progeny of Fj individuals in 
family Ij2^§a, Dale Gloria X spelt 

FAMILIES CONSISTING OF SPELTS ONLY 



Pedigree 

number 

Fj. 


Class 
of F2 
parent 
plant. 


Classes of spelt inheritance. 


Number 
of spelts 

and 
speltoids. 


Num- 
ber of 
wheats. 


Totals. 


I 


2 


3 


4 


s 


6 


7 


8 


9 


10 


I 


2 

2 

I 
I 
2 

I 

3 


4 
5 
3 

10 
16 

7 
16 


2 

5 
2 
6 

7 
8 


2 

3 

I 
I 

7 
I 
I 


I 

I 


I 


I 
I 






3 


14 
17 

6 

21 

43 
15 
25 






2 


I 


I 


17 
6 


3 

4 






3 

I 


I 












21 


6 












43 
25 


14 














19 































FAMILIES CONSISTING OF WHEATS ONLY 



5 

9 

12 


10 
10 
10 
10 
10 
10 
10 
10 
10 
10 
10 
10 






















II 

25 
13 
17 
21 

19 
38 
23 
29 
20 

33 
14 


II 
25 
13 
17 
21 

19 
38 
23 
29 
20 

33 
14 










































13 

18 










































20 






















22 






















27 

29 

31 

32 

33 



























































































































FAMILIES CONSISTING OP WHEATS AND SPELTS 



15- 
16. 

17- 
21 . 

23- 
24. 

25- 
26. 
28. 
30. 



Expectations 

Probable error 

DcAnation 

Ratio between deviation and probable error. 



3 
3 


20 


9 


2 












3 




I 


3 


4 


3 


2 


2 






2 


2 


.S 


2 


3 


2 




2 


2 


2 


I 


17 

5 


8 


I 


I 












5 


8 


7 


5 


6 


4 


2 


4 


3 


4 
7 
5 


H 


S 
2 


5 
I 


I 


3 




I 








I 










5 


4 


5 


2 


I 


I 


2 


2 


6 


7 


9 


3 


6 




4 




2 


2 


3 


2 


5 


I 






2 






I 




9 

7 


I 


4 
2 


4 

2 










24 
II 




2 


2 


2 


I 


3 


4 
6 


10 


I ^ 


S 
2 

rcrnlic 


I 




2 


2 






I 


6 

prnrx 


2 
T? r 










21 


>1ftTlt« 









34 

15 
20 
27 
44 
23 

4 
28 
29 

9 
33 
25 
33 
32 



356 

384 

±6.61 

18 

2. 72 



4 
4 
8 

IS 
10 

9 

24 
20 

17 
II 
12 
17 
3 



156 
128 



Nov. 12, 1921 Genetic Behavior of Spelt Form in Crosses 



353 



Table X. — Frequencies of spelt and wheat classes in F^ progeny of Fj individuals of 
series lj26oa, spelt X Turkey 

FAMILIES CONSISTING OP SPELTS ONLY 



Pedi^ee 

number 

F,. 


Class 
of F2 
parent 
plant. 


Classes of spelt inheritance (F3). 


Number 
of spelts 

and 
speltoids. 


Num- 
ber of 
wheats. 


Total. 


I 


2 


3 




5 


6 


7 


8 


9 


10 


8 


2 
2 

4 

I 

2 
2 
3 

3 


5 

10 

12 

4 

4 

2 

10 

9 


3 

I 

5 
8 
6 

I 
8 
6 


3 
















12 
II 
20 

13 
12 

3 
20 

16 






20 
















21 


2 

I 


















22 














13 

12 


23 

24 

27 

16 






I 


















3 
20 


I 
I 




























16 





















FAMILIES CONSISTING OF WHEATS ONLY 



I 


10 
10 
10 
10 
10 
10 
10 
10 
10 




















17 
14 
15 
15 
30 
20 
30 
12 
24 




17 
14 
15 
15 
30 
20 

30 
12 
24 


17 


a 




















4 

5 

II 




















15 
IS 
30 
20 
30 






































12 




















17 

18 






































26 




















24 























FAMILIES CONSISTING OP SPELTS AND WHEATS 



3 

6 


9 

I 
I 

3 
I 
2 
7 
5 
2 
2 

5 
I 







I 








I 


5 


6 


10 

4 
8 

7 
5 
7 
4 
4 

^ 

5 
2 
8 


13 
16 
20 
10 
18 
17 
13 
16 
26 

13 
12 
16 
22 


10 

4 
8 

7 
5 
7 
4 
4 
3 
4 
S 
2 
8 


23 


II 

13 

5 

10 
10 


5 
6 

3 
5 
3 








7 


I 
I 
3 














28 


9 


I 












17 
23 
24 
17 
20 

29 
17 
17 
18 


10 












13 




I 








3 
10 

2 


14 










3 

I 


IS 

19 


2 

9 
6 

I 
6 


II 
6 
I 

5 


2 
6 

I 
2 
3 


3 


4 


I 


I 


25 














28 

29 


I 
I 


2 


3 


I 


I 
I 
8 


I 
10 


30 




I 


2 






73 


4S 


20 


30 


ofhetet 
ous F2 ] 
Expectal 
Probable 
Deviatio] 
Ratio of 


ozyg- 
jlants. 
ions. . 


6 


7 


S 


5 


19 


32 


71 


212 

212. 25 

±4- 91 

■25 

•OS 


71 
70-75 


283 


error 




1 






deviation to crobable erro 


r 





















The F3 populations of the series segregating in the 3 to i ratio (13255a 
and 13260a) have a general tendency to produce the maximum frequen- 
cies at class I , sometimes at class 2. In only 4 families out of 15 are there 
individuals in classes above the fourth, 10 individuals in 248 receiving the 



354 Journal of Agricultural Research voi.xxii.no. 



higher classification. The few mdividuals found in these speltoid classes 
at present may be regarded as exceptions. Their significance will be 
considered later. 

The distributions of the families in the series segregating in the ratio 
of 15 to I (13 126a) (Table V) are entirely different. They do not take the 
general course described above. Some of them have very low frequencies 
at classes i and 2. Families 5, 7, and 12 have no individual in class i, the 
population of family 5 being composed of class 9 individuals only. Family 
25 produced its spelts in classes 6 to 9, inclusive. Of the 12 families 
being considered 6 produced class 9 individuals, while among the families 
segregating in the 3 to i ratio there is but one instance (13255a, family i) 
where class 9 individuals have been produced. 

The explanation of the increased variability of the constant spelt- 
producing families of series 13126a, as compared with series 13255a and 
13260a, will be found in the factorial explanations given for these two 
groups of crosses. 

Families 13255a and 13260a segregated in the simple monohybrid 
3 to I ratio. By hypothesis, all the spelts producing only spelts are sup- 
posed to have the genotypic composition SS. 

As to the cross 13126a which segregated in the 15 to i ratio, it was 
shown that there were five constant spelt forms, namely : 

SjvSiSaSg SiSiSjSs S1S1S2S2 SiSjSzSg S1S1S2S2 

Although these forms would keep on producing only spelts, they are 
not genotypically identical. 

The fact that in the families segregating in the 3 to i ratio there was 
only one genotypic spelt form and in the family segregating in the 15 to i 
ratio five such forms were present may account for the increased varia- 
bility among the pure-breeding spelts of the latter cross. 

The F3 progenies can not be separated into the five theoretical geno- 
typic groups just referred to because, among other things, there is positive 
evidence that modifiers also are concerned which have the tendency to 
shift the classes one way or the other. This phase of the subject will 
next be discussed. 

MODIFICATION OF THE DEGREE OF INHERITANCE OF SPELT CHAR- 
ACTERS, DUE TO THE PRESENCE OF MODIFYING FACTORS 

For the consideration of the subject of modifiers the analyses will be 
confined primarily to the spelt classes (i to 9, inclusive) of the progeny 
of the heterozygous F.^ individuals of series 13260a, shown in Table X. 
This family has been chosen because it represents a simple mode of 
segregation. Whatever is said about modifiers for this family will be 
found to apply just as well to the other families. 

It has been shown that only one spelt factor was concerned in the cross 
under consideration. All the Fj heterozygous plants had the formula 



Nov. 12, 1921 Genetic Behavior of Spelt Form in Crosses 355 

Ss and produced spelts and wheats in the 3 to i ratio. If these F3 
heterozygous plants have all the same genotypic constitution with 
regard to the spelt factor we would expect to see them produce similar 
F3 spelt distributions. These distributions obtained experimentally are 
far from being uniform. For instance, plant i326oa-3 (Table X) pro- 
duced individuals mainly of classes 8 to 9, and plant i326oa-6 produced 
its spelts in classes i and 2 only. The progeny of i326oa-i5, on the 
other hand, showed no definite grouping, the curve spreading from one 
extreme to the other. 

It may be argued that (i) these variations are insignificant; (2) they 
may be due to variations in soil and other external conditions; or (3) they 
are merely nonheritable fluctuations. 

These arguments may be answered easily: 

1 . In the first place, let us take the first two frequencies (classes i to 
9, inclusive), namely, those of i326oa-3 and i326oa-6 (Table X). The 
means are 8.00 ±0.29 and 1.3 1 ±0.08. The difference between the two 
means is 6.69 ±0.30. This difference, it is seen, is very significant. 
Similar striking differences will be found when the means of the other 
distributions are compared. 

2. With regard to variations due to soil conditions and external factors, 
it is only necessary to mention that these plants were grown on the same 
plot of the experimental field. The pedigree numbers following the 
family number represent the numbers of the rows in which the progeny 
of each of the Fj plants was grown. For instance, the progenies of 
plants i326oa-28, 29, 30 (at bottom of Table X) were grown in three 
adjacent rows, yet i326oa-28 was composed of individuals contributing 
to nearly all classes of spelting, i326oa-29 produced practically all typical 
spelts, and the progeny of i326oa-30 were nearly all speltoid forms 
approaching the wheat type. So this second objection may also be 
dismissed. 

3. As to the nonheritability of these variations, the objection may be 
settled by comparing the Fj and F3 generations in terms of the coefficient 
of heredity. If these variations are nonheritable fluctuations, there 
should be no correlation between the Fj and F3. Putting the statement 
in the affirmative, if there is a significant correlation in the degree of 
spelting of parent and offspring of the F2 and F3, then it is a direct and 
indisputable proof that these variations are transmitted to the follow- 
ing generations — that is, they are heritable. The accompanying corre- 
lation table (Table XI) has been prepared with a view of determining the 
validity of this last objection. The x-axis represents the mean classes 
of the F3 and the y-axis represents the classes of the F2 individuals which 
produced these Fg forms. In this table are included all the progeny of 
the heterozygous forms of series 13255a and 13260a in order to have a 
sufficient number of individuals. The coefficient of heredity as calculated 
is 0.880 ±0.029. As this coefficient is over 0.5 and over 10 times its 



356 



Journal of Agricultural Research voi. xxu.no. 7 



probable error it may be regarded as being very significant. As there 
is a significant correlation between these two generations, the variations 
under consideration are not fluctuations due to external conditions but 
are hereditary variations. 

Table XI. — Correlation table showing the classes of spelt inheritance o/Fj heterozygous 
plants, with the average degree of spelling of the F^ progeny of each F-^ plant (series Ij2j^a 
and 12260a) 

Classes of spelt inheritance (Fs) 







I 


2 


3 


4 


5 


6 


7 


8 


9 




I 






4 


I 














3 






I 


I 


2 




I 






a 


3 






I 




I 


I 








K 


































































^ 






















a 






















V 


4 










2 










rt 






















a 










































h 






















"S 


5 












2 


2 






M 






















a 






















































































Q 


6 














I 


















































K 
























7 












I 




2 


I 




8 






















9 
















I 


2 



CoeflBcient of heredity=o.88o±o.o29 

If there were no interference due to modifiers, the curve of the spelt 
F3 progeny of the heterozygous Fj individuals would follow the spelt 
curve of the Fj generation, as both the spelts of the Fj curve and those 
of the F3 curves of heterozygous Fg individuals consist of SS and Ss 
spelt plants in the proportion of i to 2, respectively. 

An examination of the F3 spelts of heterozygous individuals in Table X 
shows that the curves of the 13 families vary considerably from the curve 
of the F2 generation (Table I, series 13260a), although the curve for the 
totals is much the same. 

Again, if there were no genetic interference, all the F3 progeny curves 
produced by heterozygous F2 plants would be expected to follow more or 
less the same course. The experimental results exhibit wide differences, 
as the comparison of the classes of individuals i326oa-3, -6, -7, etc., 
will readily show. 



Nov. li, 1921 Genetic Behavior of Spelt Form in Crosses 357 

As the presence of multiple factors is entirely out of question, it being 
proved in this case that the parents of this cross differ in only one factor 
for spelting, the following explanation may be given to account for these 
variations. One or more sets of modifiers furnished by the spelt, by 
the wheat, or by both parents seem to be present where each set of 
modifiers was in a homozygous dominant condition in one parent and in 
the alternative condition or absent in the other parent. These modifiers 
in the presence of the S factor tend to intensify the spelt character. 

An example may be given to illustrate the effect which a modifier may 
produce in a spelt X wheat cross. The modifier which may cause dilu- 
tion of spelting may be represented by the factor D and may be assumed 
to be carried by the wheat parent. (If this factor were contributed by 
the spelt parent, the latter would have been a dilute spelt, which was not 
the case in these crosses.) The wheat parent will then be represented 
by ssDD and the spelt parent by SSdd. The genotypic forms of the 
successive generations will be as follows : 



Pi 




SSdd 
spelt. 


X 
SsDd 




ssDD 

wheat 




Fl 






semidilute 1 


spelt. 






1^2 


I SSdd. 




2 SSDd. 


I SSDD. 




I ssdd. 




2 Ssdd. 




4 SsDd. 


2 SsDD. 




2 ssDd. 
I ssDD. 




3 typical 




6 semidilute 


3 dilute 




4 nonspelts. 




spelts. 




spelts. 


spelts. 







12 spelts. 4 wheats. 

This represents a ratio of 3 spelts of different grades to i wheat. 

If we assume that the nature of the modifier were to produce intensifi- 
cation of spelt inheritance in the presence of factor S, which in this case 
may have been carried either by the spelt or by the wheat, we will have, 
in the F, generation, 3 intense spelts, 6 semi-intense spelts, 3 normal 
spelts, and 4 wheats. 

Some of these spelts (intense, normal, or dilute) will breed true to 
those conditions ; others will produce some or all of these forms in different 
proportions as expected on the factorial hypothesis. If more than one 
set of modifiers are present the types and their proportions naturally 
become rather complex. 

If, in the crosses 13255a and 13260a, a diluting modifier has been intro- 
duced, we would occasionally expect among the spelts (homozygous or 
heterozygous) some which are grouped in the dilute speltoid classes. 
The F3 population of i3i26a-5 and -25 (Table V); i3255a-26 (Table 
IX) ; i326oa-3, and -14 (Table X) ; and a number of others represent 
such cases. The progeny of i326oa-2o, -6, -10 and others may represent 
spelts carrying some intensifying factor. 



358 Journal of Agricultural Research Voi.xxn.No. 7 

In conclusion, it may be said with certainty tJbat besides tlie S factor 
in series 13255a and 13260a and the Si and So factors in 13126a, modifiers 
are present which tend to dilute or intensify the spelt character. 

In statistical studies of density in wheat, the junior author has found 
two characters whose mode of inheritance is almost identical with that 
of the spelt character. When a dense wheat (T. compactiim) is crossed 
with a lax wheat (T. vulgare) a 3 to i segregation is found in the Fj gen- 
eration. The F2 density curve consists of two distinct curves. One of 
these is a skew curve in the dense classes which contains 75 per cent of 
the individuals. After a gap, the other curve, which is composed of the 
lax segregates containing the remaining 25 per cent of the F2 population, 
begins. Although the F3 progeny of these heterozygous dense plants of 
the F2 generation invariably produce bimodal curves similar to that of 
the F2 just described, their modes or the means of the dense and lax 
curves shift at times considerably toward the lax classes and sometimes 
toward the denser classes, much in the same manner as does spelting. A 
similar phenomenon has been observed by Nilsson-Ehle (8) among his 
dense X lax wheat crosses. 

The other parallel case is density of the type just mentioned, but in 
this case the modifier is known to be the spelt factor itself. The curves of 
the progeny of the heterozygous dense individuals of dense wheat X spelt 
crosses have the general bimodal form, but the populations which, in 
addition to density, carry the S factor always have their density curves 
shifted toward the lax classes. 

With some spelts, the S factor shifts the density curves so much toward 
the lax classes that this S has to be regarded also as an inhibiting factor 
for density. 

Hull-lessness in oats, according to Love and McRostie (7), is inherited 
in a similar manner. While this character segregates in the simple 
mendelian ratio of i hulled to 2 intermediates to i hull-less, the inter- 
mediate forms vary appreciably as regards the percentage of hulled 
kernels they produce. By correlating the percentage of hull-lessness of 
parent and offspring, they have shown that these variations within the 
I to 2 to I ratio are hereditary. 

The mode of inheritance of the spelt character as shown in Tables IX 
and X closely resembles also Castle's (2) case of hooded rats, which had 
for a time aroused considerable controversy for and against the question 
of inconstancy of unit characters. In numerous crosses between rats 
having the hooded pattern and the wild (totally pigmented) or the Irish 
(white belly) types the hooded pattern behaved as a mendelian recessive. 
The ratio of nonhooded to hooded F2 offspring was 3 to i , showing that 
the hooded condition is dependent upon a single factor difference. 
Among the hooded individuals a considerable degree of variation was 
observed with respect to the degree of the extension of this pattern. By 
making selections for many generations in plus and minus directions 



Nov. 12. 1921 Genetic Behavior of Spelt Form in Crosses 359 

Castle was able to increase and decrease the pigmented area. His belief 
then was that the variations observed in the race of hooded rats were 
not mere fluctuations but were hereditary variations in the sense that 
the factor for the hooded condition had undergone alterations. 

The assumption of unit-factor inconstancy, which Castle applied to 
account for variation of pigmentation of his hooded rats of course can not 
be applied for the analogous variations in the groupings of the spelt 
individuals, for, if this were the case, variations in this same extent 
should have been present among the self -fertilized population from which 
the parental form was selected. The classes of spelt inheritance in the 
parental strain ranged from i to 4, the mode being between classes i 
and 2 . No departures nearly as great as those found in the homozygous 
extracted spelts of the F3 generation were observed among this parental 
population. The study of the F2 generation shows clearly that either 
some modifier or modifiers were introduced by the nonspelt parent or 
were carried by the spelt parent, but these modifiers were reduced to a 
recessive state as a result of crossing. 

PRODUCTION OF SYNTHETIC WHEATS BY CROSSING TWO SPELTS, AND 
SYNTHETIC SPELTS BY CROSSING TWO WHEATS 

The writers frequently have obtained synthetic spelts in interspecific 
crosses in wheat. No indication has been observed as to the possibilities 
of producing true wheats in crosses between two different nonwheat 
species. It is theoretically possible, however, that such forms eventually 
will be produced in crosses between certain kinds of spelts. This suppo- 
sition may be explained by taking as an example the results of one of the 
experiments discussed at length in this paper. 

In the case of cross 13126a it Avas shown in detail that two spelt factors 
Si and S2 were involved; that the F2 segregates which bred true to the 
spelt character were not all genetically identical; and that they were 
composed of five genotypic forms, namely: 

v3iOiv32^2 wiS202^2 W1D1V52S2 ^1^1^2'^2 Sj^Sj0202» 

As long as these forms are allowed to be selfed, as they are in nature, 
no wheats ever segregate; but, by hypothesis, in a number of crosses 
between these five forms, a certain proportion of wheats are expected 
to appear in the following manner : 

1. Crosses producing no wheats: 

SiSiSaSaXany other genotype; SiSiSoSaXSiSiSgSa; S1S1S2S2XS1S1S2S2. 

2. Cross where one out of every four F^ plants will produce 6^ per 
cent wheats : 

*-'l^l^2^2 X v5iSiv32W2 • 

3. Crosses where half of the F^ plants will produce 6^ per cent wheats: 

^l*-'l^2^2Xv3iSiv3202; Oiv3i02S2XSiSit52S2- 

4. Cross where all F^ plants will produce 6^' per cent wheats: 

^1^1^2^2XSiSiv32»32' 



360 Journal of Agricultural Research voi. xxii. no. 7 

If wheats segregate trom these crosses it will prove further the correct- 
ness of the two-factor hypothesis. It will also lead to the expectation 
that genotypic forms similar to the above, and other combinations as 
well, exist among the so-called pure commercial spelt forms and when the 
proper cross is made among these commercial spelts, a certain number 
of synthetic wheats may be produced in the Fj generation. 

It is easy to understand how the wheat character, being distinctly 
hypostatic, may be carried from generation to generation by the spelt 
type. But how can the spelt type segregate from a wheat X wheat cross? 
How can one conceive the spelt factor, which is so pronouncedly epistatic 
to the wheat character, as being carried by a wheat without being mani- 
fested phenotypically ? The explanation is simple. It was shown that 
modifiers are involved in these crosses. Common wheats occasionally 
carry modifiers which tend to dilute the spelt character. Some of these 
modifiers were shown to be so effective that they grouped all of the spelts 
in class 9. Most of the class 9 individuals, as recorded in the foregoing 
tables, resemble wheat so closely that no one would be likely to call 
them true spelts. 

If a certain diluting modifier can shift the spelts to class 9, a group of 
these may readily shift the spelt to classes between 9 and 10. If these 
diluting factors are reduced to a homozygous dominant condition, the 
dilute spelt which will be classified as 10 will breed true to type and be 
considered as wheat, although from a genetic standpoint such a form is 
a spelt. 

As long as such sorts are allowed to self-fertilize they will produce a 
so-called pure line consisting of a constant wheat type. Their spelt 
characteristics are exhibited only when crossed with a common wheat 
which carries the factor for dilution in a recessive state. In the second 
generation of this cross the segregates which carry the S factor with the 
factor for dilution in a recessive state; that is, SS dd, will be spelts. 

Fortunately, experimental evidence can be cited to support this 
statement. One of the writers has observed at the Kansas Agricultural 
Experiment Station over 20 spelts among Fj hybrid plants derived 
from a number of wheat X wheat crosses where one of the parents was 
a rust-resistant variety of winter wheat and the other was Preston, 
Marquis, or Haynes Bluestem, well-known varieties of spring wheats. 
These parental types and some of the spelt segregates are shown in 
Plate 33, B.^ 

None of the F^ plants in these crosses were spelts, or at least passed for 
spelts, although they might have shown some spelt characteristics in a 
weak form. In the Fj generation, however, depending upon the cross, 
the proportions of wheat to spelts varied roughly from 3 to i to over 

' The authors are indebted to Professors John H. Parker and L. E. Melchers for allowing them to photo- 
graph these forms and use them in connection with this paper. 



Nov. ra, igar Gefietic BehavloY of spelt Form in Crosses 361 

100 to I, with some sets of crosses producing no spelt at all. Besides 
the true spelts, a number of speltoid forms also segregated. 

The absence of the spelt type in the F^ generation shows that the 
absence of the spelt character in one of the parents was due to the pres- 
ence of an inhibitor in the parent plant which carried the S factor. If 
the appearance of the spelt form in the Fj generation was due to com- 
plementary factors furnished by both parents, the spelt should have 
appeared in the F^ generation. Such was not the case. 

The cultures consisted of over a thousand Fg plants. It is not now 
absolutely necessary to know how, or exactly in what proportion, these 
appeared. The purpose of citing these examples is to substantiate the 
views expressed above regarding the possibilities of producing synthetic 
spelts from wheat X wheat crosses, which might have been regarded 
as a mere speculation in the absence of this experimental evidence. 
The fact that there were no spelts grown near the P^ or F^ plants and 
that spelts appeared in more than one cross excludes the possibility of 
accidental or natural cross-fertilization. 

These observations show that common wheats may carry the spelt 
factor, but the latter can not express itself because one or more diluting 
or inhibiting factors are carried with it. Some of these diluting factors 
may be regarded as inhibiting factors which are not totally dominant 
but produce intermediacy in a heterozygous state. 

From the small percentage of spelts which appeared in the Fj genera- 
tion in some cases, it may be inferred that there is one and in some 
cases more diluting factors.^ 

The production of synthetic spelts in wheat X wheat crosses just con- 
sidered is similar to the synthetic production of T. dicoccum dicoccoides, 
the so-called "Wild Wheat" of Palestine, in the vulgar e X durum cross 
(6). The wild character, consisting of a number of interdependent 
specific characters, is strongly dominant over both the sativum and the 
durum types, as is the spelt type toward sativum. Yet in both instances 
the character showing strong dominance toward either of the parental 
forms was carried by one of the parents together with a factor inhibiting 
in one case the wild and in the other case the spelt characters. 

MODIFICATIONS OF MENDELIAN RATIOS 

The question of modifiers whose presence in the production of the 
spelt character was demonstrated in various ways brings us to the con- 
sideration of the modifications of mendelian ratios. As all spelts or all 
wheats are not alike with respect to their ability to intensify, repress, 
or inhibit the production of the spelt characters, it is natural to expect 

' Nilsson-Ehle (.8, q) and Kajanus (j, 4) have also observed the occurrence of spelts in certain sativum X 
sativum crosses. They support the correctness of the foregoing observations and tend to preclude the 
assumption of the possibility of the occurrence of some accidental or natural crossing of one of the sativum 
patents with a spelt. 



362 Journal of Agricultural Research voi. xxii. no. 7 

certain modifications of the 3 to i ratio. For instance, in certain spelt X 
wheat crosses, depending upon the potency of the diluting factors, the 
proportion of wheats to spelts may increase in certain amounts. In 
some instances, the increase of wheats will be slight, so that the obtained 
deviation from the 3 to i ratio, which will be on the side of excess for 
the wheat class, will be considered within or near the limits of the probable 
error and the inheritance will be regarded as simple mendelian. Perhaps 
series 13255a, where an increased proportion of both constant breeding 
wheats and F3 wheat segregates of heterozygous Fj plants are obtained, 
represents such a case. 

It also is possible that the ratio may fall between 3 to i and i to i in 
case the diluting factor is very strong. Here, then, will be an excess 
in the homozygous wheat class at the expense of the homozygous dom- 
inant spelt class. If factors of both dilution and of intensification 
are introduced in the same cross, the experimental ratios will defy any 
attempt at simple factorial explanations. The writers have obtained a 
cross where the Fg generation suggested a possible 3 to i segregation, 
but on examining the F3 generation, which was composed of a fairly 
large population, neither the individual segregations nor the totals of 
.these approached in any way 3 to i or i to 2 to i expectations. 

It naturally follows from the foregoing discussion that in spelt-wheat 
crosses wide departures from simple ratios occasionally may be expected. 

SUMMARY 

T. spelta and T. sativum are differentiated by a number of linked 
specific characters, which are present in one species and absent in the 
other. These characters, so far as observed, are not inherited independ- 
ently but are transmitted as a group. 

In crosses between a spelt and common wheats the F^ hybrid shows 
dominance of the spelt, but this character appears in a somewhat diluted 
form. In the second generation all classes of spelt inheritance are ob- 
tained. In order to classify these forms, 10 arbitrary classes were 
erected, class i representing the true spelt and class 10 the total absence 
of this character. The intermediate classes represent diff'erent grades of 
spelt inheritance. 

In most of the material studied there was but one factor difference for 
spelt, but in two cases two spelt factors were present. Both 3 to i and 
15 to I ratios were obtained. These ratios were verified after determin- 
ing the genotypic constitution of the Fj plants, which gave pure breeding 
spelts, inconstant spelts, and pure breeding wheats in the ratios of 
I to 2 to I and 7 to 8 to i , respectively. Of the constant spelts produced 
in crosses with this latter ratio, approximately half yielded (in the F3 
generation) spelts and wheats in the ratio of 15 to i and the other half 
in the ratio of 3 to i, as expected on the two factor hypothesis. 



Nov. 12, 1921 Genetic Behavior of Spelt Form in Crosses 363 

The speltoid form "Gatineau" when crossed with wheats gave also a 
3 to I segregation of spelts and wheats, but the curves showing the 
classes of spelts produced by this cross were entirely different from the 
curves produced by the spelt-wheat crosses which segregated in the same 
ratio. 

Aside from the factor or factors for spelting, there is positive evidence 
showing the presence of intensifying and diluting modifiers which tend 
to affect the degree of spelt characters without affecting to any extent 
the ratios of spelts to wheat. Some of the diluting modifiers tend to act 
as inhibitors. 

The progeny of all heterozygous spelts of the Ss type do not produce a 
similar spelt curve. There are wide discrepancies in the spelt inheritance 
of the progeny of these forms. These variations within the spelt classes 
have been found to be hereditary and to be caused by modifiers. 

The theoretical possibility of producing synthetic wheats in crosses 
between certain pure-breeding spelts is shown. 

Experimental evidence also is presented showing that, in spite of the 
fact that the spelt character is dominant over the wheat form, the former 
may be synthetically produced by crossing certain wheats, provided one 
of the wheats carries a spelt factor together with an inhibitor and that 
the other wheat carries neither. 

It is shown that, if intensifying, inhibiting, and diluting modifiers are 
introduced in a cross, wide departures may be expected from the 3 to i 
and 15 to I ratios. 

LITERATURE CITED 
(i) BosHNAKiAN, Sarkis. 

1918. THE MECHAmCAL FACTORS DETERMINING THE SHAPE OF THE WHEAT 
KERNEL. In Jour. Amer. Soc. Agron., v. lo, no. 5, p. 205-209, fig. 27. 

(2) Castle, W. E., and Phillips, John C. 

1914. PIEBALD rats and SELECTION ... 56 p., 3 pi. Washington, D. C. 
(Carnegie Inst. Wash. Pub. 195). Bibliography, p. 31. 

(3) Kajanus, Birger. 

1912. 0BER EINEN SPONTAN entstandenen weizenbastard. In Ztschf. 
Pflanzenziicht., Bd. i, Heft i, p. 13-24. 

(4) 

1918. KREuzuNGSSTUDiEN AN winTERWEIzen. In Bot. Notisef, 1918, Haftet 
5, p. 235-244. 
(5) Love, H. H., and Craig, W. T. 

1918. the relation between color and other characters in certain 

AVENA crosses. In Amer. Nat., v. 52, no. 620/621, p. 369-383. 

(6) 

1919. THE SYNTHETIC PRODUCTION oP WILD WHEAT FORMS. In Jonr. Heredity, 

V. 10, no. 2, p. 51-64, 10 fig. 
(7) and McRosTiE, G. P. 

1919. THE INHERITANCE OF HULL-LESSNESS IN OAT HYBRIDS. In Amer. Nat., 

V. S3, no. 624, p. 5-32, 7 fig. 



364 Journal of Agricultural Research voi. xxii, no. 7 

(8) NitSSON-EHLE, H. 

1911. KREuzuNGSUNTERSucHUNGEN AN hafer und weizen. II. In Lunds 
Univ. Arsskr., n. f., afd. 2, bd. 7, no. 6, 84 p. Literaturzerzeichnis zu 
den Einleitung, p. 20. 

(9) 

I917. UNTERSUCHUNGEN UBER SPELTOIDMUTATIONEN BEIM WEIZEN. In Bot. 

Notiser, 1917, Haftet 6, p. 305-329, i fig. 

(10) Saunders, Charles E. 

1907. report of the cerEaust. In Canada Exp. Farms Rpts., 1906, p. 235- 
256, I pi. 

(11) Surface, Frank M. 

I916. studies on oat breeding. III. ON THE INHERITANCE OF CERTAIN 
GLUME CHARACTERS IN THE CROSS AVENA FATUA X A. SATIVA VAR. 

KHERSON. In Genetics, v. i, ixO. 3, p. 252-286, pi. 2-3. Literattire 
cited, p. 285-286. 

(12) TscHERMAK, Erich von. 

1910. WEIZEN (TRITICUM). KORRELATIONEN. BASTARDIERUNG. /n Fruwirth, 

C. Die Ziichtung der landwirtschaftlichen Kulturpflanzen, Aufl. 2, Bd. 
4, p. 119-139, 164-187, fig. 11-12. Berlin. Bibliographical footnotes. 



PLATE 33 

A. — Wheat spikes showing different degrees of spelting. i and 2 are intense spelts; 
3 to 9 are intermediate forms, showing varying degrees of dilution of the spelt character; 
10 is a pure-breeding wheat (sativum) form, showing no trace of spelting. The numbers 
represent approximately the types falling in the 10 classes of spelting. 

B. — Snythetic spelts produced in Fj generation in wheat X wheat crosses; a, b, c, and 
d are the sativum parent plants; b-i spelt form of the progeny of a X b; c-i to c-5 
spelt forms of the progeny of a X c; d-i and d-2 spelt forms of the progeny a X d. 



Genetic Behavior nf the Spelt Form in Crosses 



Plate 33 



) 



\2 i<9 /^ 





Journal of Atjricultural Research 



Vol. XXII, No. 7 



PLUM BLOTCH, A DISEASE OF THE JAPANESE PLUM, 
CAUSED BY PHYLLOSTICTA CONGESTA HEALD AND 
WOLF^ 

By John W. Roberts, Pathologist, Fruit Disease Investigations, Bureau of Plant 
Industry, United States Departinent of Agriculture 

INTRODUCTION 

In June, 1905, W. M. Scott of the Bureau of Plant Industry, United 
States Department of Agriculture, collected near Fort Valley, Ga., fruits 
of the Japanese plum (Prunus triflora Roxbg.) affected with a disease 
very closely resembling the apple blotch, due to Phyllosticta solitaria 
E. and E. In the diseased areas were spore-bearing pycnidia which were 
found also on the leaves in gray papery spots resembling those on apple 
leaves caused by Phyllosticta solitaria. On May 27, 1908, the disease was 
again observed by Scott on both fruit and foliage of the Burbank plum at 
Montezuma, Ga. It was found to be rather common in several orchards 
about Montezuma, in some cases causing enough damage to injure 
seriously the market value of the fruit. In one orchard a large part of the 
fruit was affected, and many specimens bore from 15 to 20 spots each. 

On May 29, 191 7, the writer collected near the same locality Japanese 
plum fruits and leaves affected with the same disease. In the single 
orchard in which the disease was found, most of the fruit was heavily 
nfected and rendered nearly worthless. Considerable difficulty was 
encountered in finding the disease again, as the Japanese plum industry 
in Georgia had about passed out. Lack of demand for the fruit coupled 
with the susceptibility of all parts of the tree to various diseases and 
insect pests had caused growers either to eradicate their trees or to let 
them die. At present there are almost no Japanese plum orchards 
remaining in Georgia, and all of the trees in which plum blotch was found 
have been eradicated. So far as the writer knows, then, the disease no 
longer exists, though it is to be looked for throughout the South as far 
west as Texas. Should the growing of Japanese varieties of the plum 
be revived in the South, blotch may prove to be one of its most serious 
diseases, as it is very destructive, and probably would be exceedingly 
difficult to control. 

The varieties found to be affected were Abundance, Burbank, and what 
was apparently an unnamed seedling. 

1 A brief description of this disease was published as an abstract of a paper presented at the Ninth Annual 
Meetmg of the American Phytopathological Society. (Roberts, John W. plum blotch. (Abstract.) 
In Phytopathology, v. 8, no. 2, p. 74. 1918.) 



Journal of Agricultural Research, Vol XXII No 

Washington, D. C. Nov. i.. 1921 

^^ Key No. G-253 

70494°— 21 3 (365) 



366 Journal of Agricultural Research voi. xxu. No. 7 

DESCRIPTION OF THE DISEASE 

The infected parts on the unripe fruit appear as dark-colored raised 
areas with fringed margins and are somewhat roughened by the presence 
of small blisters and depressions (PI. 34, B). As in the case of apple 
blotch, the skin often becomes ruptured as the fruit increases in size. 

On the ripe fruit the blotched parts appear as irregular browned areas 3 
to 6 mm. in diameter and consist of an aggregation of from 4 to 20 sunken 
spots, each separate spot being i mm. or less in diameter. At this stage 
the spots have a peculiar light blue cast owing to the "bloom" of the 
ripe plum covering the browned epidermis. The diseased area is rather 
superficial, extending only slightly below the epidermis. The affected 
tissues become hardened and somewhat leathery and show no tendency 
to decay. 

Small, glistening pycnidia are produced in considerable numbers even 
in the younger spots. Quite commonly there are 25 to 30 scattered 
promiscuously about in each blotched area. Infection evidently takes 
place when the fruits are very young, since the spots found May 29 were 
well formed and bore pycnidia with mature spores. Judging from the 
writer's inoculation experiments, infection probably took place five to 
six weeks earlier, or about the middle of April. 

On account of its characteristic appearance on the fruit, the disease has 
been given the common name of "plum blotch." 

On the upper surface of the leaf blades (PI. 34, A), the spots are angu- 
lar, rather small (about 0.5 mm. across), brown when young, but later 
becoming gray or silvery in color. They may be numerous, as many 
as 200 sometimes appearing on a single leaf. Usually only a single 
pycnidium is present in each spot, except where two or more spots have 
coalesced to form a single large spot. Affected areas are also found on 
the petioles and on the veins of the lower surface, especially on the 
midrib. On these the diseased areas are much larger than on the upper 
surface of the blade and are black and sunken. Pycnidia, bearing 
spores, are present in great abundance. 

,, Pycnidia, apparently identical with those found on the fruit and 
leaves, were found also in small light-colored, often slightly sunken areas 
on the twigs; but, as spores were lacking, positive identification could 
not be made. It is possible that these pycnidia had discharged their 
spores early in the spring and had brought about the early infections on 
the fruit. 

CAUSE OF PLUM BLOTCH 

By comparison with type specimens, the organism involved in the pro- 
duction of plum blotch has been found to be identical with the fungus 
described by Heald and Wolf ^ as Phyllosticta congesta. Heald and 

1 Heaid, F. D., and Wolf, F. A. new spEaEs op texas fungi. /wMycologiav. 3, no. i, p. 8. 191 1. 

i 



Nov. 12, 1921 



Plum Blotch 



367 



Wolf found the fungus on the leaves only of Prunus sp. in Texas Their 
description is as follows : 

Maculis minutis, .5-.8 mm diam., brunneis numerosis, venis limitatis; pycnidiis 
solitariis in quaque area, 50-125 ix diam.; sporulis globulosis vel leniter elongatis, 
hyalinis 6-9 m- 

On Prunus sp. Boeme (Texas) 1554 (Type). 

On the upper surface of the leaf are very numerous brown areolae bounded by the 
veins of the leaf. The lower surface may not be discolored. These minute spots fuse, 
and each contains at its center a single black pycnidium. The pycnidia contain 
globular or slightly oval, clear spores. 




Fig. I. — Section through a pycnidium of Phyllosticta congesta, showing spores. Natural 
infection on plum fruit, Georgia 1917. X 340. 

Heald and Wolf do not mention the fact that the older spots become 
gray or silvery, though the type specimens as well as those collected by 
Scott and the writer show this to be the case. The spots on these 
leaves and those on Georgia specimens collected by the writer show a 
marked resemblance, and the fungi found upon them are morphologically 
the same. The spots on the leaves collected by the writer have a greater 
tendency to fall out. ^iof-:;, 

The pycnidia (fig. i) are glistening, lens-shaped, erumpent, on the 
leaves 65 to 120 /x in diameter, on the fruit 60 to 120 /x in diameter. 
On the average, pycnidia on the fruit are somewhat 
larger than those on the leaves. Spores on the leaves 
measured 7 to 9 ju in diameter, on the fruit 8 to 9 /i. 
Spores from younger spots were invested with gelati- 
nous envelops which were sometimes lengthened into 
appendages (fig. 2). Spores from older spots do not 
show these envelops, and they are not to be found in 
the dried herbarium specimens. The young spores 
of Phyllosticta solitaria have sucli an envelop. In fact, 
P. solitaria and P. congesta resemble one another so 
closely that on purely morphological grounds they 
might be considered as identical. Since the ascogenous stage of neither 
fungus is known, the writer prefers to retain the name P. congesta as a 
matter of convenience, unless it is shown by cross inoculations that the 
fungus on the apple and that on the plum are identical in every way. 




Fig. 2. — Spores of Phyl- 
losticta congesta, with 
the gelatinous envelops 
which are sometimes 
present. From pycni- 
dia on plum fruit, 
Georgia 1917. X 680. 



368 Journal of Agricultural Research voi. xxii. no. 7 

Of course the final test of identity would lie in whether or not the as- 
cogenous stages of the two fungi, assuming them to exist, would prove 
to be identical. 

Specimens of Phyllostica congesia on fruit and foliage of Prunus 
triflora have been deposited in the Pathological Herbarium, Bureau of 
Plant Industry, United States Department of Agriculture. 

It is not known how the fungus is carried over from one season to 
another. If it occurs on the twigs, as the writer is inclined to think, there 
would be good reason for believing that production of spores from twig 
lesions in the spring would constitute an important infection source. It 
is also possible that the fungus survives the winter on leaves and fruit. 

On all the ordinary culture media the fungus shows about the same 
type of growth. On corn meal agar, beef agar, prune agar, potato agar, 
and oatmeal agar growth is very slow, and on all these media it presents the 
same appearance. There is a dense black mass of closely woven hyphae 
forming a raised and irregular aggregation of shining bead-like bodies 
which may be considered as sterile pycnidia, since they are more or 
less hollow bodies containing oil drops. The margin of the growth is 
often fringed; in fact on the above-named media the growth is almost 
as blotch-like as it is on the fruit of the plum. On sterihzed stems of 
Melilotus the growth resembles that on the agars, but spores are often 
formed though very scantily. 

On Japanese plum twigs growth is also very slow. Pycnidia and 
spores are formed in about two months. Pycnidia are formed on the 
bark and may also be formed at the cut end of the twig, in which case 
they are densely aggregated. 

Sterilized apple twigs proved to be the best medium for the production 
of spores, though two to three months must elapse before spore produc- 
tion begins. On this medium the only sign of growth by the fungus is 
the formation of a dense mass of hyphae and pycnidia, closely aggregated 
at the upper end or at an abraided place on the side of the twig. On all 
the media used the type of growth exhibited by Phyllosticta congesia 
differs somewhat from that of P. solitaria. On sterile apple twigs, for 
instance, the latter produces pycnidia which are scattered over the bark, 
whereas the pycnidia of the former are formed only at the cut ends of 
the twigs. 

In 191 7 the fungus was isolated from both fruit and leaves by the 
poured plate method, using spores, and by planting bits of the diseased 
tissues in plates. 

In the spring of 19 18, no spores had been obtained in cultures, but 
inoculations were made by spraying the young fruits and leaves of 
Abundance and Burbank plums with bits of hyphae and sterile pycnidia 
suspended in sterile distilled water. The results were negative in every 
case. 



Nov. la, I92I Plum Blotch 369 

In 19 19, spores obtained from apple twig cultures and suspended in 
sterile distilled water were applied to fruit, foliage, and twig of Abundance 
plums on May 15. Where cultures originally obtained from plum fruits 
were used, two fruits were found with two typical blotches on each of 
them; three leaves were found with scattering spots, each spot typical 
of the disease and each bearing a single pycnidium with the character- 
istic spores of Phyllostida congesta. Like results were obtained by the 
use of cultures obtained from the leaves; one fruit showed three typical 
blotches with pycnidia and two others showed one; seven leaves were 
successfully infected. From all these artificially inoculated parts, the 
fungus was reisolated and proved to be P. congesta. 

No lesions were found on the twigs. 

Inoculations made upon Japanese plums with spores from pure cul- 
tures of Phyllostida solitaria gave negative results in 1918, 1919, and 1920, 
though the spores were applied to fruit, foliage, and twigs at frequent 
intervals throughout the spring. 

Though the inoculation experiments herein reported upon are suffi- 
cient to prove Phyllostida congesta the cause of plum blotch on leaves and 
fruit and show the fungus on the fruit to be identical with that on the 
leaves, they are not as complete as the writer should wish. All the inocu- 
lation work was done at Arlington, Va., under conditions probably un- 
favorable to the fungus, since it has been found naturally only in regions 
much farther south. 

It is planned to carry on further inoculation work with both the plum 
blotch and apple blotch Phyllostictas. The writer expects eventually to 
obtain successful inoculations on plum twigs using Phyllostida congesta as 
inoculum. 

CONTROL MEASURES 

No attempts to control plum blotch have been made. One would 
expect that control might be had by spraying with a strong fungicide at 
intervals beginning shortly after the petals have been shed as is the case 
with apple blotch. Dilute lime-sulphur solution and Bordeaux mixture 
injure Japanese varieties of the plum so severely as to preclude their use 
during the growing season. It is also doubtful whether or not dilute 
lime-sulphur solution would control severe cases of disease, since it will 
control only mild cases of apple blotch. Self-boiled lime-sulphur can be 
used with safety on the Japanese plum, but it is a fungicide which is 
even weaker than dilute lime-sulphur solution. It seems probable, 
therefore, that should this disease ever become an important one, its 
control will present a problem of considerable difficulty, though it is 
realized that the reasoning by analogy in which the writer has just 
indulged may easily lead to wrong conclusions. 



I 



370 Journal of Agricultural Research voi. xxii. no. 7 

SUMMARY 

Plum blotch, a hitherto unknown disease of the Japanese plum {Prunus 
iriflora), has been found in Georgia. In addition to the fruit, the leaves 
and possibly the twigs are affected. The lesions on fruit and leaves 
greatly resemble those of the apple caused by Phyllosiicta soliiaria 
E. and E. 

Varieties Abundance and Burbank were found to be susceptible. An 
unnamed seedling, probably also belonging to Prunus iriflora was found 
to be severely infected. 

From diseased fruits and leaves the fungus, Phyllosiicta congesta Heald 
and Wolf, was isolated and grown in pure culture. Spores obtained from 
cultures on sterile apple twigs when suspended in distilled water and 
sprayed on healthy fruits and leaves produced characteristic lesions of 
the disease. 

Phyllosiicta congesta Heald and Wolf is to be considered for the present 
as different from P. soliiaria E. and E-, though greatly resembling it. 
Inoculation experiments on plums using spores from pure cultures of P. 
soliiaria were negative. 

No attempts have been made to control plum blotch, but the possibili- 
ties of control are discussed. 



PLATE 34 

A. — Plum leaves affected with Phyllosticta congesta, Georgia, 1917. 
B. — Plum fruits affected with Phyllosticta congesta, showing the characteristic 
' blotches, " Georgia, 1917. 



I 



Plate 34 




Journal of Agricultural Research 



Vol. XXII ,No.7 



A COMPARISON OF THE PKCTINASE PRODUCED BY 
DIFFERENT SPECIES OF RHIZOPUS 

By ly. ly. Harter, Pathologist, and J. L. Weimer, Pathologist, Office of Cotton, Truck, 
atid Forage Crop Disease Investigations, Bureau of Plant Industry, United States 
Department of Agriculture 

INTRODUCTION 

Recent investigations by Harter and Weimer ^ showed that Rhizopus 
tritici Saito, an organism demonstrated to be parasitic on sweet potatoes, 
produces a powerful intracellular and extracellular pectinase which 
dissolves the middle lamella so that the cells readily separate without 
themselves undergoing any noticeable alteration. A suspension of 0.25 
gm. of the enzym powder in 25 cc. of water was found to completely 
macerate sweet potato disks i mm. in thickness in two to five hours. 
Furthermore, the solution on which the fungus grew was even richer 
in pectinase, maceration of the sweet potato disks being completed in 
one to three hours. 

Since the foregoing results have been published Harter, Weimer, and 
Lauritzen ^ have concluded experiments which showed that out of 1 1 
species of Rhizopus studied 9 were parasitic on the sweet potato. 
Furthermore, these investigators found that the species differed in 
degree of parasitism, both as regards the percentage of infection and 
the rapidity of decay. 

The present investigations, therefore, had for their object to determine 
(i) if pectinase is produced by all species of Rhizopus and, if so, to what 
extent and (2) if its production is any indication of the parasitism of 

the species. 

TECHNIC 

The methods employed in carrying out macerating experiments 
with the different species of Rhizopus were for the most part the same as 
those used in previous work to which reference ^ has already been made, 
although some slight modifications were necessary to meet certain phases 
of the problem. Three sets of experiments were carried out with each 
organism. All the species were included in a single experiment and the 
macerating action was determined for all at the same time, so that the 
results for each species are directly comparable for a single experiment. 
The culture medium was so prepared and in sufficient quantity as to make 

1 Harter, L. I,., and Weimer, J. L. studies in the physiology of parasitism with speciai, refer- 
ence TO THE SECRETION OF PECTINASE BY RHIZOPUS TRITICI SAITO. In Jour. Agr. Research, v. 21, no. 9, 
p. 609-625. 1921. Literature cited, p. 624-625. 

* Harter, L. I,-, Weimer, J. L-, and Laitritzen, J. I. the susceptibility of the different varie- 
ties OF SWEET potatoes TO DECAY BY RHIZOPUS NIGRICANS AND RHIZOPUS TRITICI. In Phytopathology, 
V. II. 1921. In press. 

Journal of Agricultural Research, Vol. XXII, No. 7 

Washington, D. C. Nov. 12, 1921 

aam Key No. G-354 

(371) 



1 

372 Journal of Agricultural Research voi. xxii. no. 7 

it uniform in all the flasks for all organisms. Four flasks (2 liters), 
containing 300 cc. of the culture medium, were inoculated with each 
one of the species in each experiment, and the cultures were incubated 
for three days. 

The macerating action was determined for the following species of 
Rhizopus: chmensis Saito, nodosus Namysl, tritici Saito, maydis Bru- 
derl, delemar (Boid) Wehmer and Hanzawa, arrhizus Fischer, oryzae 
Went and Pr. Geerligs, nigricans Ehmb., reflexus Bainier, artocarpi 
Racib., and microsporus v. Tieg. 

It has been shown ^ that the different species of Rhizopus do not all 
have the same optimum temperture for growth. Some species thrive 
at high temperatures, some at relatively low temperatures, and others 
at a temperature intermediate between the two extremes. Therefore, 
the 1 1 species studied have been separated into three groups with respect 
to their temperature relations. In all the experiments connected with 
the present investigations the same grouping of the different species 
has been observed, thus subjecting each organism to as nearly the 
optimum temperature for its growth as possible. 

The cultures of chinensis were incubated at 40° C, those of nodosus, 
tritici, maydis, delemar, arrhizus, and oryzae at 30°, and those of nigri- 
cans, reflexus, artocarpi, and microsporus at 20°. Although so far as 
temperature is concerned the results are not strictly comparable, pre- 
liminary experiments showed that more reliable data could be obtained 
by growing the different organisms at temperatures suited to their growth 
than by subjecting them all to a uniform temperature. Some of the 
species, as for example nigricans, which requires a relatively low tem- 
perature, make no growth or only a feeble growth at 30° and none at 35°. 
On the other hand, chinensis, a high temperature form, makes a reduced 
growth at 30° and a feeble growth at 20°. | 

At the close of the incubation period (three days) the mycelial growth 
was lifted from the culture flask and the substrate was filtered through a 
fine grade of muslin. The mycelium was treated subsequently by 
acetone and ether according to the method previously described.^ The 
solutions from the different flasks in which the same species had grown 
were made into a compound sample thoroughly shaken, and 25-cc. 
portions were used for maceration experiments. Likewise all the fun- 
gous felts of the same organism grown in the different flasks were brought 
together and treated as one sample, a weighed portion of the dried 
mycelium being used for maceration of the raw disks. Two types of 
controls were run with each set of experiments, as follows: (i) Sweet- 
potato decoction on which the fungus had grown for three days, which 
after the removal of the mycelium was steamed for 15 minutes to inac- 
tivate the enzym; (2) decoction which was identical with that used for 

1 Harter, X,. h; Weimer, J. L., and Lauritzen, J. I. op. cit. 

2 Harter, I^. 1,., and Weimer, J. 1,. op cix 



Nov. 12, 192 1 



Pectinase Produced by Different Species of Rhizopus 373 



inoculation purposes but which had not supported a fungous growth. 
Maceration by the enzym in the solution and in the mycelium was carried 
out at a temperature of 40° C. for all the species. Before the addition 
of the raw sweet-potato disks the solutions and suspensions of the my- 
celium were held for one hour at 40° in order to bring them to the tem- 
perature at which maceration was to take place. 



A/o^o,sa<s 
r/?/T/c/ 

Af/C/?OSPORUS 
A//(9/?/0/9A/iS 




O / 2 <^ ^ <5 e 7 

Fig. I. — Graph showing the time required by the different Rhizopus species to completely macerate 
sweet-potato disks by the enzym in the solution on which the fungi had grown for three days; also the 
comparative rate of maceration in the three experiments. 

Maceration with the mycelium was carried out by the use of 0.5 gm. 
ground in sand and suspended in 25 cc. of water. All the sweet potato 
disks (i mm. thick and 1.5 cm. in diameter) required for maceration in 
the solution on which the fungus grew and in a water suspension of the 
mycelium were cut from a single potato for an entire experiment of all 
the species. pJxism 



374 



Journal of Agricultural Research voi. xxn. Na» 



EXPERIMENTAL DATA 

The results obtained in the different experiments both as regards the 
maceration in the solutions and in a water suspension of the mycelium 
are represented graphically in figures i and 2. Each of the vertical lines 
of a single group represents the results obtained for a particular organism 
in a single experiment. The length of the vertical lines indicates the 



OE^£:M/P/r' 

C/y/ZV/ETA/iS/tS 
T/?/T/C/ 



m^mmm ^^mmm wammmm — — Mi— 1 — ■■ 1— oi— — aa» 

^mmm^ ■hbbkb watm^ 

wtm^ma — apif jawii wa^mmm i^^km mm^imm mw— hmmm 



O / 2 3 ^ s e 



7 e s /o 



Fig. 2.— Graph showing the time required by the differeut Rhizopus species to completely macerate 
sweet-potato disks by the enzym contained in Yi gr. of the mycelium; also the comparative rate of macera- 
tion iu the three experiments, 

length of time in hours required to complete maceration of the disks, 
I cm. being equivalent to one hour. From these figures a direct com- 
parison can be made of the results obtained from the different species as 
well as the variation in the results of the same species in different experi- 
ments. 



Nov. 12, 1921 Pectinase Produced by Different Species of Rhizopus 375 

DISCUSSION OF RESULTS 
VARIATION 

A comparison of the results as shown by figures i and 2 indicates that 
under the conditions of these experiments maceration was completed by 
the enzym exuded into the solution in a shorter length of time than by 
that contained in the mycelium. This difference, however, does not 
mean that the enzym is more powerful or more abundant in the solution 
than in the mycelium since no attempt was made to employ an amount of 
mycelium that would be equivalent in macerating power to the enzym 
of the solution. In these experiments maceration was regarded as 
complete when the disks pulled from opposite sides separated without 
any perceptible resistance. In completely macerated tissue coherence 
of the cells is entirely lost and the tissue can be readily pulped between 
the thumb and finger. 

The data show that a considerable amount of variation exists in the 
results obtained in the different experiments with a single species both in 
respect to the solution and the mycelium. There are probably several 
factors responsible for these variations. In the first place a dififerent 
supply of the sweet-potato decoction was prepared for each set of experi- 
ments. In spite of the fact that the dififerent solutions were prepared 
to be alike as nearly as it is possible to make them, it can not be said, in 
view of the fact that different potatoes were used each time, that the 
various preparations were identical. After the cultures were inoculated 
incubation was carried out at a temperature which varied very little 
but possibly enough to influence slightly the rapidity and volume of 
growth. At the close of the incubation period the substrate and mycel- 
ium were handled as nearly alike as possible in all the experiments, but 
in spite of such precautions some variations might result. It would seem 
that the greatest source of error might be attributed to a variation in the 
composition of the potatoes from which the raw disks were cut. In this 
connection preliminary experiments showed that different potatoes are 
macerated in a dififerent length of time the variations, however, being 
within relatively narrow limits. Furthermore, it is probable that the 
composition of the sweet potato is gradually changing with the increase 
in the length of time after digging. So far as their susceptibility to 
maceration is concerned it is interesting to note that a comparison 
between newly dug potatoes and those stored for several months showed 
that the latter are macerated in a shorter time than the former. Although 
the present experiments were carried out with a single variety the various 
experiments were conducted in sequence so that the later experiments 
were made on what might be termed older potatoes. 



376 Journal of Agricultural Research voi. xxii. no. 7 

COMPARISON OF SPECIES 
MACERATION OP DISKS IN THE SOLUTIONS 

It appears from figure i that two species, nigricans and artocarpi, 
macerate raw sweet potato disks more slowly than any of the others, 
followed by chinensis and reflcxus in the order mentioned. The other 
species complete maceration in a relatively short time, the most rapid 
being arrhizus, tritici, and maydis. 

MACERATION OF DISKS IN A SUSPENSION OF MYCELIUM 

With respect to maceration by the mycelial enzym, four species, 
nigricans, micros porus, chinensis, and artocarpi, stand out as being conspic- 
uously slow. So far as the mycelium is concerned delemar, a species inter- 
mediate between the slow and rapid forms, is less active than reflexus but 
more active than the latter when the solutions are used. Likewise the 
enzym contained in the mycelium of micros porus macerates slowly, while 
that in the solution, on the other hand, disintegrates the tissue rapidly. 
Chinensis shows a similar relationship existing between the enzym of the 
mycelium and that of the solution, although this species does not stand 
out as conspicuously as microsporus. 

From the few illustrations cited it is evident that there is no complete 
correlation between the activity of the mycelial enzym and the activity 
of that exuded into the substrate. An examination of figures i and 2 
shows that the different species do not secrete an equivalent amount of 
pectinase, since the completion of maceration by the enzym of both the 
mycelium and solution may vary greatly. It is conceivable and the 
results of these investigations seem to indicate that some species give up 
their enzym to the solution more readily than others. For example, the 
solution on which tnicrosporus grew is relatively rich in pectinase while 
the enzym contained in the mycelium acts slowly. Delemar is another 
outstanding example of the same phenomenon. 

PECTINASE PRODUCTION IN RELATION TO PARASITISM 

If a relationship between the production of pectinase by the different 
species of Rhizopus and their parasitism could be shown to exist, con- 
siderable light might be thrown on the physiology of parasitism, espe- 
cially among fungi which are characterized by their ability to dissolve 
the middle lamella in advance of their growth. In a previous publica- 
tion ^ it was pointed out that all the species of Rhizopus studied were 
parasitic on sweet potatoes with the exception of microsporus and chinen- 
sis. These two species were studied in connection with the others. 
They were given equal opportunity to cause decay, but in no case was 
there any evidence to indicate parasitism. However, both of these 

iHarter, L. L., Weimbr, J. L., and Lauritzen, J. I. op. cix. 



Nov. 12, 1921 Pectinase Produced by Different Species of Rhizopus 377 

species produced pectinase. The amount of pectinase in the myceHum at 
the end of the growth period was relatively small, but microsporus and 
to a lesser degree, chinensis, exuded enough into the culture solution to 
cause maceration in a much shorter time than either nigricans or arto- 
carpi, both of which are parasites. Maceration of sweet-potato disks 
by means of the mycelial enzym of the two parasitic and nonparasitic 
species just mentioned was completed in about the same length of time. 
Nigricans is the most commonly isolated species and is probably respon- 
sible for most of the loss to sweet potatoes caused by this group of fungi. 
At a suitable temperature it decays sweet potatoes and other vegetables 
rapidly. The middle lamellae of sweet potatoes decayed by this species 
are dissolved some distance in advance of the growth of the mycelium, 
so that coherence is lost. In the early stages, at least, the cells them- 
selves are not invaded by the fungus. The same may be said of artocarpi. 
However, in cultures nigricans and artocarpi, in contrast to the other 
species, exude a very small amount of enzym into the substrate and 
retain very little in the mycelium. Delemar, a species which readily 
decays sweet potatoes, seems to give up most of its pectinase to the 
substrate, so that maceration by means of the mycelium is relatively 
slow, at least within the limits of these experiments. All the other 
species are vigorous parasites, decaying the sweet potato within a few 
days under favorable conditions. They also produced large quantities 
of pectinase, relatively speaking, some of which is exuded into the solu- 
tion and some retained by the mycelium, as shown by the fact that 
maceration, by both the mycelium and solution, is comparatively rapid. 

SUMMARY 

(i) The secretion of pectinase by 11 species of Rhizopus has been 
studied. All the species were found to produce pectinase and to exude 
some of it into the culture solution. 

(2) The amount of pectinase produced varies with the species, grown 
under identical conditions. The mycelium of four species — nigricans, 
microsporus, chinensis, and artocarpi — and the solution on which two — 
nigricans, and artiocarpi — are grown is comparatively weak in pectinase. 
Chinensis and microsporus, whose mycelial enzym is weak, secrete it 
abundantly into the substrate. 

(3) Two species, nigricans and artocarpi, both of which are parasitic 
on the sweet potato, secrete a relatively small amount of pectinase. 
On the other hand, chinensis and microsporus, two nonparasitic species, 
while retaining a small amount of enzym in the mycelium, secrete a 
comparatively large quantity of enzym into the culture solution. 



ADDITIONAL COPIES 

OF THIS PUBLICATION MAY BE PROCURED FROM 

THE SUPERINTENDENT OF DOCUMENTS 

GOVERNMENT PRINTING OFFICE 

WASHINGTON, D. C. 

AT 

15 CENTS PER COPY 
Subscription Price, $3 Per Year 

V 



Vol.. XXII NOVEMBER 19, 1921 No. « 

JOUKNAI> OF 

AGRICULTURAL 

RESEARCH 



CONTENTS 

Pago 

Hemotoxins from Parasitic Worms - - - - - -379 

BENJAMIN SCHWARTZ 

( Contribution from Bureau of Animal Industry ) 

Ash Content of t^ie Awn, Rachis, Palea, and Kernel of Barley 
during Growth and Maturation - - - - - - 433 

HARRY V. HARLAN and MERRITT N. POPE 

<Contilbution rom Bureau of Plant Industry) 



PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE 

WITH THE COOPERATION OF THE ASSOCIATION OF 

LAND-GRANT COLLEGES 



WASHINGTON, D. C. 
GOVERNMENT PRINTING OFFICE 

1921 



EDITORUL COMMITTEE OF THE 

UNITED STATES DEPARTMENT OF AGRICULTURE AND 

THE ASSOCIATION OF LAND-GRANT COLLEGES 



FOR THE DEPARTMENT 

KARIy F. KELLERMAN, Chairman 

Physiologist and Associate Chief, Bureau 
of Plant Industry 

EDWIN W. ALI.EN 

Chief, Office of Experiment Stations 

CHARLES h. MARLATT 

Entomologist and Assistant Chief, Bureau 
of Entontology 



FOR THE ASSOCIATION 

J. G. LIPMAN 

Dean, State College of Agriculture; and 
Director , New Jersey Agricultural Experi- 
ment Station, Rutgers College 

W. A. RILEY 

Entomologist and Chief, Division of Ento- 
mology and Economic Zoology, Agricul- 
tural Experiment Station of the University 

of Minnesota 

R. L. WATTS 

Dean, School of Agriculture, and Director; 
Agricultural Experiment Station, The 
Pennsylvania Stale College 



All correspondence regarding articles from the Department of Agriculture should be 
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C. 

All correspondence regarding articles from State Experiment Stations should be 
addressed to J. G. Lipman, New Jersey Agricultural Experiment Station, New 
Brunswick, N. J. 



JOINAL OF AGRldlllAlRESEARCH 

Vol. XXII Washington, D. C, Nove;mber 19, 1921 ^\^^'y^^O- 8 
HEMOTOXINS FROM PARASITIC WORMS 

By Benjamin Schwartz ' 

Zoological Division, Bureau of Animal Industry, United States Department of 

Agriculture 

I. INTRODUCTION 

Aside from the purely mechanical injuries which parasitic worms may 
inflict upon their host as a consequence of their migrations, displacement 
of a certain amount of the host's tissue, bites and laceration of the mucosa, 
obstruction of ducts, and various other mechanical disturbances, it has 
been generally assumed that they may also produce harmful effects as a 
result of their toxic secretions. Despite the fact that the data on which 
the view that parasites secrete toxic substances is based, so far as they 
have been recorded heretofore in the literature, are somewhat contra- 
dictory, they have been accepted by a large number of investigators 
as affording a more plausible explanation of the cliemical pathology of 
helminthiasis than the data with reference to any other theory that has 
thus far been advanced. With reference to the subject of toxic products 
of parasitic worms. Wells {igiSy states: 

The subject has received much less consideration than its importance deserves, as 
we are quite in the dark as to how much of the effects produced by animal parasites 
are not merely mechanical, but are due to soluble poisons that they secrete or excrete. 
Some of the parasites probably cause harm mechanically and in no other way, but 
with most of them there is more or less evidence of the formation of poisonous 
substances. 

While it must be admitted that the evidence in favor of the view that 
parasites secrete products toxic to the host is as yet rather incomplete, the 
fact of the existence of such toxic products can not be denied. So 
far as they have been investigated, the serological reactions of hosts har- 
boring parasites afford proof that parasitic worms liberate products 
against which the host develops defense or "immunity" reactions. It 
has been known for a relatively long time that in cases of infestation 
with species of Trichinella, Schistosoma, Necator, and Anyclostoma a 

' Resigned December is, 1920. 

2 Dates in parenthesis refer to " Literature cited," p. 428-432. 

Journal of Agricultural Research, Vol. XXII , No. 8 

Washington, D. C. Nov. 19,1921 

aan Key No. A-62 

(379) 



380 Journal of Agricultural Research voi. xxii, no. s 

high eosinophilia is commonly present. An increase in the number of 
eosinophile leucocytes has also been observed, although not as regularly, 
in cases of infestation with species of Ascaris, Oxyuris, Strongyloides, 
and other nematodes. Similar conditions have also been encountered in 
cases of infestation with Taenia solium, T. saginata, Fasciola hepatica, 
Clonorchis sinensis, and other cestodes and trematodes. As a matter of 
fact, eosinophilia is so commonly associated with parasitic infestation 
tliat the finding of a high eosinophile content in the peripheral blood is 
generally considered as presumptive evidence of parasitic infection. In 
a recent extensive review of the literature on the subject of eosinophilia, 
Schwarz {191 4) states that an increase in the number of eosinophile 
leucocytes in the peripheral circulation in cases of parasitic infestation is, 
from an etiological viewpoint, the most clear-cut illustration of general 
eosinophilia.^ 

Aside from the cellular inmiunity reactions, as shown by the increase 
in the number of eosinophile leucocytes in the blood in cases of infesta- 
tions with parasitic worms, there appears to be evidence of a humoral 
immunity as well. In the case of hydatid (Echinococcus) disease of man 
and animals, it has been shown by a number of investigators that specific 
antibodies are present in the blood of the host, demonstrable by the 
technic of complement fixation and precipitate formation. That such 
immunity reactions are not limited to hydatid disease is the opinion of 
certain investigators, who support their views by experimental evidence 
which shows that specific antibodies are also present in cases of infesta- 
tions with species of Ascaris, Fasciola, Schistosoma, and other parasitic 
worms. ^ 

The facts cited in the preceding paragraphs appear to indicate that 
hosts harboring parasitic worms develop typical defense or "immunity" 
reactions to the absorption of foreign and presumably toxic substances 
of parasitic origin. A logical corollary to the study of the serological 
reaction of animals to secretions of parasitic worms is the study of the 
secretions themselves with reference to their chemical and physiological 
properties. This subject has recently received considerable attention 
in studies on the causes of pernicious or infectious anemia of horses, a 
disease of unknown etiology, which Seyderhelm and Seyderhelm (1914) 
attribute to a secretory product of an internal parasite (the larvae of 
Gastrophilus) . Although the assumption of the Seyderhelms has not been 
confirmed, numerous experiments by diff'erent investigators have shown 
that injection into animals of extracts of various parasitic Avorms may 
lead to serious consequences, frequently terminating in death. Despite 
the fact that these experiments are in a general way confirmatory of the 
work of earlier investigators on the physiological effects of extracts of 

• " Die Vermehrung der a-Zellen in peripheren Blut bei Anwesenheit von Parasiten aus dem Stamme 
der Wiirmer ist vielleicht die atiologisch am meisten klargestellte Form der allgemeinen Eosinophi'ie " 
' References to and a summary of this phase of the subject may be found in an article by G. Ghedini. 



Nov. 19,1921 Hemotoxins from Parasitic Worms 381 

parasitic worms, the experimental evidence on the subject is somewhat 
contradictory, due no doubt to the fact that different investigators 
experimented under different conditions. The study of the effects of 
extracts of parasites on hving animals presents numerous difficulties and 
complications and may lead to contradictory results unless suitable 
precautions are taken to control various extraneous factors. More accu- 
late studies on the effects of toxic products may be carried out in vitro, 
provided the toxic substance in question has affinity for tissues and cells 
available for such experiments. As is well known, red blood cells serve 
as excellent indicators of test-tube reactions in v/hich hemotoxic sub- 
stances are involved, and in the case of toxic products of parasitic origin, 
experiments with red blood cells are of great importance in view of the 
fact that in many parasitic infestations anemia is a characteristic symp- 
tom. The effects of extracts of parasitic worms on red blood cells, 
especially of extracts of those parasites that are known to cause anemia, 
are thus of interest with reference to the possibility that the parasites in 
question secrete specific toxins for the blood (hemotoxins). 

II. REVIEW OF LITERATURE ON HEMOTOXINS IN PARASITIC WORMS 

The same year in which Ehrlich (iSg8) announced the discovery of a blood toxin 
produced by Bacillus tetanus, Schaumann and Tallqvist (i8g8) reported the discovery 
of a blood toxin in the broad tapeworm of man {Diphyllohothrium latum). Ehrlich 's 
discovery in the field of bacteriology served as a stimulus to the study of bacterial 
hemolysins by various investigators and was followed by a series of investigations into 
the nature and action of these hitherto unknown products of bacterial growth. Al- 
though the discovery of Schaumann and Tallqvist did not arouse the same degree of 
activity in parasitology as Ehrlich 's discovery aroused in bacteriology, the results of 
their investigations were not without influence on subsequent researches in parasit- 
ology, the influence being especially marked in connection with studies on the causes 
of the anemia that occurs in cases of infestation with hookworms. 

The facts published by Schaumann and Tallqvist {1898) may be briefly summarized 
as follows: 

Macerated material of Diphyllobothrium latum contains a hemolytic substance active 
in vitro as well as in vivo. Peptic digestion liberates the hemolysin from the tissues 
of the parasites. The introduction of D. lattim material into dogs parenterally or peros 
leads to a marked reduction in the number of erytlirocytes. 

In a later paper Tallqvist {1907) gives a more detailed account of the nature of the 
hemotoxic secretions of Diphyllobothrium latum. The hemolytic principle is closely 
bound to the cells of the parasite and is but slightly soluble in water and physiological 
salt solution. By means of peptic digestion and alcohol or ether extraction it becomes 
dissociated from the tissues and goes into solution. The hemolysin is thermostabile 
and does not cause the development of antibodies when injected into animals. In 
these respects it resembles normal tissue hemolysins. Tallqvist fotmd, moreover, 
that D. latum contains not only a hemolysin but also a hemagglutinin. The latter is 
soluble in water in contrast to the water-insoluble lipoidal hemolysin. The hemagglu- 
tinin as well as the hemolysin is nonspecific. The potency of these agents varies, 
however, for different species of red blood corpuscles. 

Faust and Tallqvist (1907) studied the Diphyllobothrium hemolysin as to its chemi- 
cal nature. These investigators found that extraction of the entire worm material in 
ether removed all the hemolysin from the tissues of the parasite, since the removal of 



382 Journal of Agricultural Research voi. xxii.no. s 

the ether-soluble fraction left a fraction entirely devoid of hemolytic activity. The 
ether-soluble fraction was then freed from its lecithin and cholestrin content without 
injuring its hemolytic activity. In the remaining ether fraction (freed from lecithin 
an'd cholestcrin) I^aust and Tallqvist identified tlu-ee fatty acids, namely, paln:itic, 
stearic, and oleic acids. The first two substances did not exhibit any hemolytic 
properties, whereas oleic acid was found to be markedly hemolytic. These investi- 
gators therefore concluded that oleic acid is the active principle of Diphyllobothrium 
hemolysin. 

In a later paper Faust {1908) records the results of experiments on the effects of oleic 
acid on dogs when introduced per os. In brief, this investigator concluded that pro- 
Icftiged feeding of oleic acid to dogs produced anemia in the latter, as evidenced by a 
reduction in the number of red blood corpuscles. Beumer {igig), however, has 
found, on the contraiy, that animals may be fed daily with considerable quantities of 
oleic acid for long periods without permanent ill effects, and has failed to substantiate 
the harmfulness of oleic acid. 

In this connection it is of interest to recall the experiments of Dascotte (cited by 
Weinberg, 1912), who states that extracts of Taenia solium and T. saginata, cestodes 
parasitic in man, dissolve human red blood corpuscles. Dascotte found, moreover, 
that the hemolysin from these parasites is soluble in alcohol and resistant to heat at 
temperatures of 100° to 120° C. Calamida {igoi) found that extracts of two species of 
cestodes from carnivores {Dipylidiuvi canimim and Multiccps muUiceps) are hemolytic 
to the red blood corpuscles of rabbits and guinea pigs and that the hemolysin goes 
through the pores of a Berkefeld filter. According to Weinberg {igoy), physiological 
salt-solution extracts of two species of tapeworms parasitic in horses (Anoplocephala 
plicata and A. perfoliata) have no deleterious effects on the blood corpuscles of the 
horse. Tallqvist {igoy), on the other hand, denies the presence of hemolysins in ces- 
todes other than Diphyllobothrium latum. He states that he worked with a number of 
species, including T. saginata. He admits that he sometimes observed slight hemolytic 
effects of extracts of these parasites but expresses the opinion that they are to be 
ascribed to secondary degeneration products associated with acid formation. 

While Diphyllobothrium latum is capable of causing severe anemia in man, clinically 
indistinguishable from pernicious anemia and, according to many investigators, differ- 
ing from the former in one respect only, namely, by the disappearance of the symptoms 
and recovery of the patient after expelling the parasite, there are numerous cases on 
record in which the presence of D. latum in man was not accompanied by anemia. 
In fact, grave blood disturbances in cases of D. latum infection are, according to the 
observations on record , not as common as the incidence of infection with this tapeworm, 
a fact which has given rise to considerable speculation as to the causes of the occasional 
appearance of anemia in the course of infection witli this parasite. These speculations 
will be referred to elsewhere in this paper. 

In contrast to the occasional appearimce of anemia in cases of infection with Diphyl- 
lobothrium /atowinfections with hookworms (Necator and Ancylostoma) are generally 
accompanied by severe anemia. That the causes of anemia in hookworm disease are 
due to a toxin is a view which was adopted by a number of investigators on a purely 
a priori basis, because the direct abstraction of blood by these parasites, even when 
present in large numbers, fails to account for the severity of the clinical picture usually 
present in such cases. This fact was recognized early in the study of the disease, and 
led to the postulation of the "toxin theory." 

Luscana (iSgo) found that as a result of injecting rabbits with urine taken from 
patients suffering from hookworm disetise, the animals developed symptoms of anemia. 
It was not until 1905 , however, that the toxin theory received more direct experimental 
support from Calmeftte and Breton. These investigators found that salt-solution ex- 
tracts of the Old World hookworm of man (Ancylostoma duodenak) are hemolytic to the 
red blood cells of man. Alessandrini (zpo^) had already foxmd by direct microscopic 



Nov. 19, 1921 Hemotoxins from Parasitic Worms 383 

observation that human red blood corpuscles are destroyed when placed in contact 
with the cervical glands isolated from hookworms (species not given but presumably 
A. duodenale), but subsequent investigation showed that the hemolysin is not limited 
to the cervical glands. 

Loeb and Smith (1904) in the course of experiments wath salt-solution extracts of 
the dog hookworm (Ancylosioma caninum), found that these extracts showed no hemo- 
lytic properties and left the blood still intact and uncoagulated after being in contact 
with it for 17 hours. Reference to the work of these investigators on the anticoalgula- 
ting property of hookworm extract {A . caninum) will be made elsewhere in this paper. 

Liefmann (7905) found that in two out of three experiments salt solution extracts of 
Ancylosioma caninum produced slight hemolysis of dog blood. This writer observed 
intact erythrocytes in the intestines of the worms and therefore came to the conclusion 
that the parasites do not secrete a hemolysin. Liefmann fails to state whether or not 
he washed the blood corpuscles before testing them against hookworm extract. 

Preti {1908) foimd that the Old World hookworm of man {Ancylosioma duodenale) 
contains a hemolysin insoluble in salt solution but soluble in ether and alcohol. He 
states that tryptic digestion liberates the hemolysin and renders it soluble in water. 
He found the hemolysin to be resistant to boiling for three hours and nonspecific, since 
it was equally potent against the blood corpuscles of several other species of animals 
as well as man. 

In the course of his investigations concerning ancylostomiasis and beriberi, Noc 
{1908) found that physiological salt-solution extracts of the hookworm of man {Necator 
dmericanus) are hemolytic to the washed red blood corpuscles of man. He states that 
the hemolysin withstands a temperatiu-e of 80° C. for one hour without injury to its 
potency. Noc found that whereas the blood serum of patents suffering from severe 
ancylostomiasis and beriberi contained no antihemolysins, that of normal persons and 
of those recovering from these diseases was antihemolytic. 

De Blasi {1908) examined the blood serum of 12 human subjects infested with hook- 
worms {Ancylosioma duodenale) and found that after the serums were heated for 30 
minutes at 56° to 62° C. they acquired hemolytic properties. Before heating, the 
serums in question were not hemolytic. Heating the serum evidently destroyed some 
antibodies which neutralized the potency of the hemolysin. The heated serum of 
normal persons, according to this writer, did not contain any hemolysins. 

Whipple {1909) records tests of salt-solution eyLtractsoi A ncylostotna caninum, A. 
duodenale, and Necator americanus on unwashed citrated blood of man, dog, and rat. 
He states that he found a weak hemolysin in the three species of hookworms exhibiting 
similar properties, namely, nonspecificity, susceptibility to boiling which destroys it, 
and distribution in all parts of the body of the worms. According to Whipple, the 
hemolysin is only demonstrable in concentrated extracts, and probably bears no 
relation to the secondary anemia of ancylostomiasis. 

lyoeb and Fleisher {1910) state that a salt-solution extract of Ancylosioma caninum, 
containing as much as 5 mgm. of the powdered worm material in i cc. of salt solution 
did not produce any hemolytic effect on the washed erythrocytes of the dog. These 
writers also state that lecithin used in doses in which it alone produced no hemolytic 
effect failed to activate A. canimim extract. Loeb and Fleisher admit the possibility 
that the temperature at which the specimens were dried (42° to 50° C.) may have had 
an injurious effect on the hemolysin, but they do not consider this very probable. 

Recently Usami and Mano {1919) have studied the effects of hookworm extracts on 
red blood cells. These writers state that hookworm hemolysin is thermostabile, 
insoluble in water, and soluble in alcohol, ether, and acetone. 

It will be seen from the foregoing summary with reference to hookworm extracts 
that Loeb and Smith {1904) and Loeb and Fleisher {1910) are the only investigators 
who failed to observe hemolysis in the presence of these extracts. As will be shown 
elsewhere in this paper, the negative results of Loeb and Smith may have been due 



384 Journal of Agricultural Research voi. xxn. no. s 

to the antilytic action of normal blood serum. The negative results recorded by 
Loeb and Fleisher {1910) may have been due to insufficient or faulty extraction of 
the worm material, insufficient quantity of powder used in the experiments, or possibly 
to the destruction of the hemolysin by drying at temperatures between 42° and 50° C. 
The results recorded by Preti (iQoS) as regards the insolubility of the hemolysin in 
salt solution and its resistance to boiling are at variance with those of other investi- 
gators, and, as will be shown in the following pages, are not in harmony with the 
results obtained by the present writer. Moreover, Preti 's results can not be accepted 
as conclusive, owing to his failure properly to control his experiments. Alessandrini 's 
attempt {1904) to associate the secretisn of hemolysin with the cervical glands of the 
parasites is not sustained by Whipple {igog), who found the hemolysin in all parts 
of the worm. 

It is interesting to observe that the different species of hookworms referred to in 
the foregoing summary have the common biological property of secreting a substance 
destructive to red blood cells. Inasmuch as hookworm disease is characterized by 
severe anemia, the presence of a blood-destroying substance in the parasites is highly 
significant. 

In addition to the hemolytic substance which is present in hookworms, Loeb and 
his collaborators have shown that the hookworm parasitic in dogs {Ancylostoma 
caninum) also secretes a substance which inhibits coagulation of blood in vitro (IvOeb 
and Smith, 1904; Loeb and Smith, 1906; Loeb and Fleisher, 1910). The results of 
experiments by these investigators with reference to the anticoagulins of hookworms 
may be summarized as follows: In A. caninum a substance is present which retards 
coagulation of blood in vitro. This substance which is present in the anterior part of 
the worm and practically absent in the posterior part is not destroyed but is markedly 
weakened by boiling for 15 minutes. The substance does not resemble hirudin, a 
toxic constituent of the leech, but appears to resemble cobra venom so far as its physi- 
ological properties are concerned. It is of interest to note in this connection that 
Liefmann {1905), who rejects the view that the hookworm secretes a hemolysin, like- 
wise rejects the view that this parasite secretes an anticoagulin, since he obtained 
positive results in but one out of three experiments which he performed. Liefmann 
ascribes his positive results to substances from the intestine which may have adhered 
to the worms, namely, pancreatin and peptone. Loeb and Smith {1906) point out, 
however, that in view of the fact that they washed the wonns carefully and that 
neither peptone nor pancreatin is known to inhibit coagulation of dog's blood in 
vitro, and further, in view of the fact that the posterior parts of the hookworms showed 
but a slight anticoagulating effect on dog blood and that extracts of ascarids and tape- 
worms from dogs did not retard the coagulation of dog blood, Liefmann 's contention 
can not be sustained. 

The carefully controlled experiments of Loeb and his collaborators leave no room 
for doubt as to the presence of a hemotoxin in A ncylosioma canimtm which inhibits the 
coagulation of dog blood. Loeb and Smith ascribe etiological significance to this 
toxin and believe that it has the power of causing small hemorrhages in regions of the 
intestine that have been lacerated by the worms. 

The pathological role of the whipworm {Trichuris trichiura) parasitic in man is empha- 
sized by Askanazy (1896), who states that this parasite feeds on blood, basing his 
assertion on the presence of iron pigment in the intestine of the worm demonstrable 
by the Berlin blue reaction. Askanazy assumed, of course, tliat the iron found in the 
worm is obtained from the hemoglobin of the host's blood. Schultze (1905) rejects 
Askanazy 's interpretation and considers tliat the pigment in question is obtained from 
the host's intestine rather than from the blood. 

Guiart {1908) presents conclusive evidence as regards the bloodsucking habit of 
Trichuris trichiura, since he found blood-engorged specimens in a human patient. 
Guiart's observation has been confirmed by a number of investigators, including Garin, 



Nov. 19. 19" Hematoxins from Parasitic Worms 385 

Seidelin, and Leon (Guiart, IQ14). Guiart and Garin (iQog) found that the presence 
of Trichuris eggs in the feces of human subjects is correlated with the presence of blood 
in the feces as shown by a positive Weber test. 

As to the presence of hemotoxic secretions in whipworms, Whipple (1909), who ex- 
perimented with extracts of these parasites, found that they contained a hemolytic 
substance destructive to the red blood cells of the dog and of man. Whipple states 
that the hemolysin left some samples of human red blood cells intact but was destruc- 
tive to others. Garin (1913) performed similar experiments with Trichuris extracts 
and confirmed the presence of a hemolysin in these parasites. According to Garin, 
the whipworm hemolysin is tliermostabile, being destroyed by 30 minutes' heating at 
56° C. The inactivated hemylosin can not be reactivated by normal guinea-pig serum 
(complement) , according to this investigator. Garin states , furthermore , that whereas 
he obtained positive results with human red blood corpuscles the results of experiments 
with the erythrocytes of rabbits and guinea pigs were doubtful. 

A survey of tlie literature relating to tlie pathogenic role of Ascaris lumbricoides 
reveals the fact that this parasite may be responsible for anemia, which is sometimes 
mistaken for hookworm anemia or for pernicious anemia. The clinical reports of 
Demme (iSgi) have become a classic illustration of this fact. In brief, Demme found 
a child suffering from severe intestinal catarrh, with a high-grade pernicious anemia 
showing a red blood coiuit of 2 ,450,000 and a hemoglobin content of 40 per cent. Two 
weeks after numerous worms (A. lumbricoides) had been expelled from the child's 
intestine the red blood corpuscle count rose to 4,200,000 and the hemoglobin content 
reached 70 per cent. In a second case of apparent pernicious anemia, which resulted 
in death and in which the erythrocytes had diminished to 1,650,000 per cubic milli- 
meter, numeroiis ascarids were found on post-mortem examination which were appar- 
ently responsible for the death of the child . Kuttner {1865) found that in a girl aged 12 
blood destruction occurred and that this was cured by expelling a number of ascarids. 
According to Filatoff {189'/), Karaven cured a case of pernicious anemia in a child by 
expelling a number of ascarids from its intestine. Francois {1906) , in the course of his 
investigations on anemia of miners, found many cases of severe anemia in which hook- 
worms were not present but which showed numerous Ascaris eggs in the feces. A 
number of observations by different investigators on hogs and horses infested with 
ascarids and on man infested with A . lumbricoides bear out the fact that symptoms of 
anemia are frequently associated with such infestation. 

As to the manner in which species of Ascaris cause anemia two views have been 
advanced, which are not mutually exclusive. Guiart {1899), who accepts the view 
that worms of this genus secrete a hemolysin, inclines strongly to the view that they 
also lacerate the mucosa, thus causing hemorrhages. In support of this view Guiart 
describes and figures Ascaris conocephala attached to the stomach of a dolphin, the 
head of the parasite being deeply embedded in the mucosa. Guiart refers to the ob- 
servations of Leroiix, who found lesions in the intestine of a human being infested with 
ascarids resembling lesions produced by ascarids on the mucosa of the dolphin . Fried- 
berger and Frohner {1895) also support this view and state that dogs that harbor 
numerous ascarids show on post-mortem examination of the intestine numerous 
roimd, dark spots, surrounded by an inflamed zone, due, in their opinion, to bites of 
the worms. According to Garin {191 3), several observers, including Weinberg, have 
found ascarids attached to the mucosa. Garin admits, however, that despite the 
fact that he made numerous post-mortem examinations of human subjects infested with 
A. lumbricoides and of dogs and cats infested with ascarids, in the latter cases shortly 
after death, attached parasites were never observed by him. He confirms, however, 
the presence of reddish points sturounded by an ecchymotic area in the mucosa of the 
intestine of infested subjects, both human and animal. Thaler {1918) has recently 
reported a case of persistent intestinal hemorrhages in a human subject which did not 



k 



386 Journal of Agricultural Research voi. xxii. nos 

yield to symptomatic treatment and which was cured only after removing several 
ascarids. 

The view that Ascaris secretions contain hemotoxins was first advanced by Schim- 
melpfennig {1902), who found that in the presence of the coelomic fluid of Ascaris 
equorum red blood corpuscles of the horse became crenated and were ultimately 
destroyed. Schimmelpfennig furthermore discovered oxyhemoglobin in the coelomic 
fluid of the parasite , a fact which led him to regard this worm as a bloodsucker. Wein- 
berg {1907), Whipple (1Q09), and Alcssandrini (191 3) failed to observe any toxic 
efi'ect of salt-solution extracts of species of Ascaris on red blood cells. Flury (1912), 
on the other hand, records the presence of strong hemolysins in the coelomic fluid of 
species of Ascaris. Flury ascribes the hemolytic action of Ascaris secretions to free 
fatty acids, of which oleic acid is the most active principle. In the course of his 
studies on the pharmacology of salt-solution extracts of worms of the genus Ascaris, 
Brinda (1914) found that injection of the extracts into guinea pigs brings about a 
reduction in the number of erythrocytes and a diminution in the hemoglobin content 
of the blood. Recently Shimamura and Fujii {1917), in the course of their investiga- 
tions on "askaron," a toxic constituent of worms of the genus Ascaris, state that 
ether-soluble and alcohol-soluble fractions of Ascaris material contain a hemolytic 
agent. The present writer (Schwartz, 1919), in a preliminary paper on the hemolytic 
effects of Ascaris extracts, has briefly described the properties of the hemolysin. 

A number of investigators have found, moreover, that the coelomic fluid of worms 
belonging to the genus Ascaris contains a substance that inhibits the coagulation of 
blood. Weil and Boye {1910) found that as a result of injecting the fluid of Ascaris 
equorum into rabbits the blood of the latter when drawn remains uncoagulated for 20 
minutes longer than blood from a normal rabbit. Experiments with rabbit blood and 
Ascaris fluid in vitro yielded negative results, according to these investigators. Leroy 
{1910) likewise observed that the blood of dogs which had been injected with the body 
fluid of A. equorum coagulated more slowly than blood from normal dogs. Flury 
{1912) observed that Ascaris fluid delayed the coagulation of dog blood and of human 
blood in vitro. That Loeb and Smith {1904) failed to observe anticoagulins in extracts 
of dog ascarids that are active in vitro has already been mentioned. 

Worms belonging to the genus Strongylus (frequently referred to as Sclerostomum) 
are parasitic in the large intestine of horses. These nematodes attack the mucosa, to 
which they may be found adhering by means of their buccal capsule. In view of the 
fact that these parasites somewhat resemble hookworms in their attacks on the intes- 
tinal mucosa and in the effects which they produce on the host, Weinberg (1907) 
investigated their hemotoxic .secretions primarily with a view of throwing light on the 
causes of anemia due to hookworms. This investigator found that physiological salt- 
solution extracts of freshly collected vStrongylus material dissolves erythrocytes of 
horses, cattle, sheep, rabbits, and guinea pigs. The parasites secrete, therefore, a 
nonspecific hemolysin. Weinberg determined that the hemolysin is thermostabile, 
resisting heat at a temperature of 115° to 120° C. for 15 to 20 minutes. In addition to 
the hemolysin, Weinberg found that these parasites secrete a substance which inhibits 
the coagulation of horse blood in vitro. He also found that salt-solution extracts of 
worms of the genus Strongylus contain a substance which when brought in contact 
with the blood serum of the horse causes the formation of a precipitate. The precipi- 
tin, too, is nonspecific in its action, since it was found by Weinberg that it produces a 
precipitate when added to rabbit-blood serum. 

Bondouy {1908, 1910) studied the chemical composition of worms belonging to the 
genus Strongylus, with special reference to their hemolytic constituents, and con- 
firmed in the main the results obtained by Weinberg as regards the presence of a soluble 
hemolysin in these parasites. The new facts discovered by Bondouy may be briefly 
summarized as follows: The parasite contains soaps and free fatty acids which exert a 
destructive effect on red blood cells in vitro. Bondouy states, however, that the 



Nov. 19. 1921 Hemotoxins from Parasitic Worms 387 

presence of these substances in the parasite is due to its blood sucking habit, basing 
his assertion on the fact that blood serum contains neutral fats, fatty acids, and soaps. 
This writer found a lipolytic enzym in worms of the genus Strongylus which apparently 
converts the storage fat into fatty acid. It is of interest to note also that Bondouy 
found neither lecithin nor cholesterin in the parasite. Lecithin, as is known, has the 
property of activating certain hemolytic agents, namely, snake venoms, whereas 
cholesterin inhibits hemolysis of blood by active hemolysins. Contrary to Weinberg's 
experience (Weinberg, igoy), Bondouy found that Strongylus hemolysin is soluble in 
alcohol. From the alcohol-soluble fraction of the parasite this writer isolated an 
extremely active hemolysin which he identified as an alkaloid. He also found a 
ptomain in the parasites which exhibited hemolytic properties. 

Brumpt and Joyeux (quoted by Brumpt, ipio) found that a watery extract of the 
stomach worm of sheep (Haemonchus contortus) produced a slight hemolytic effect ^ 
after 2}^ hours and a total hemolysis after 12 hours. Cuill6, Marotel, and Panisset 
(iQii) state that extracts of sheep strongyles (species, of which apparently several 
were involved, not given) did not exert any effect on sheep red blood corpuscles from 
either healthy or sick animals. These writers also state that extracts of these parasites 
contained hemoglobin. 

According to Garin {1913) Graphidium sirigosum and Trichostrongylus retortaeformis, 
nematodes parasitic in the stomachs of hares and rabbits, secrete hemolysins. With 
reference to the hemolysin of G. strigosuvi, Garin found that it is secreted by the living 
worm in vitro. He found, furthermore, that the hemolysin is apparently a complex 
substance and acts on the blood not directly but in combination with complement. 
Heating at 55° C. for 30 minutes does not destroy but merely inactivates the hemolysin, 
which may be reactivated by normal serum, according to this investigator. In view 
of the limited number of experiments which Garin performed, his conclusions can be 
accepted anly with reservation. The work requires confirmation. As for the hemoly- 
sin from T. retortaeformis, Garin found it to be far less potent than that of G. sirigo- 
sum. He also states that the hemolysins from the two species have far greater affinity 
for the blood cells of rabbits than for those of other species of animals and are therefore 
relatively specific. 

Yagi {1910) found that salt-solution extracts of the blood fluke. Schistosoma japoni- 
cum, are hemolytic to erythrocytes of cattle, sheep, and rabbits. He found, fiuther- 
more, that this hemolysin is soluble in ether and concluded that it is probably a fatty 
acid. Yoshimura (191 3) experimented with salt-solution extracts of the same species 
and found them to be destructive to rabbit erythrocytes. Human blood cells, accord- 
ing to this writer, are refractory to these extracts. Yoshimura also experiment