XT ,0-=hS
JOURNAL OF
AGRICULTURAL
RESEARCH
Volume VI
APRIL 3, 1916— SEPTEMBER 25, 1916
BOTANIC LI
DEPARTMENT OE AGRICULTURE
WASHINGTON, D. C.
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICUIvTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Statiofis
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
Director, histitute of Animal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Patfiologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
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 Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine.
CONTENTS
Page
Relation of Carbon Bisulphid to Soil Organisms and Plant Growth.
E. B. Fred i
Climatic Conditions as Related to Cercospora beticola. Venus
W. Pooiv and M. B. McKay 21
Soluble Nonprotein Nitrogen of Soil. R, S. Potter and R. S.
Snyder 61
Oviposition of Megastigmus spermotrophus in the Seed of Douglas
Fir. J. M. Miller 65
Citrus Canker. Frederick A. Wolf 69
Determination of Stearic Acid in Butter Fat. E. B. Holland,
J. C. Reed, and J. P. Buckley, Jr loi
Life History and Habits of Two New Nematodes Parasitic on
Insects. J. H. Merrill and A. L. Ford 115
Insect Injury to Cotton Seedlings. B. R. Coad and R. W. HowE. 129
A Sex-Limited Color in Ayrshire Cattle. Edward N. Went-
woRTH 141
Watermelon Stem-End Rot. F. C. Meier 149
Effect of Pasteurization on Mold Spores. Charles Thom and
S. PIenry Ayers 153
Effect of Water in the Ration on the Composition of Milk. W. F.
Turner, W. H. Shaw, R. P. Norton, and P. A. Wright 167
Crowngall Studies, Showing Changes in Plant Structures Due to a
Changed Stimulus. Erwin F. Smith 179
Effect of Certain Species of Fusarium on the Composition of the
Potato Tuber. Lon A. Hawkins 183
Hyperaspis binotata, a Predatory Enemy of the Terrapin Scale.
F. L. Sim ANTON 1 97
Test of Three Large-Sized Reinforced-Concrete Slabs under Con-
centrated Loading. A. T. Goldbeck and E. B. Smith 205
Occurrence of Sterile Spikelets in Wheat. A. E. Grantham and
Frazier Groff 235
Effect of Cold-Storage Temperatures upon the Pupae of the Medi-
terranean Fruit Fly. E. A. Back and C. E. Pemberton 251
Effect of Climatic Factors on the Hydrocyanic-Acid Content of
Sorghum. J. J. Willaman and R. M. West 261
Egg and Manner of Oviposition of Lyctus planicollis. Thomas E.
Snyder 273
Hypoderma deformans, an Undescribed Needle Fungus of Western
Yellow Pine. James R. Weir 277
III
IV Journal of Agricultural Research voLvi
Page
Ornix gerniinatella, the Unspotted Tentiform Leaf Miner of
Apple. L. Haseman 289
A Western Fieldrot of the Irish Potato Caused by Fusarium
radicicola. O. A. PratT 297
Comparative Study of the Root Systems and Leaf Areas of Corn
and the Sorghums. K. C. Miller 311
Production of Clear and Sterilized Anti-Hog-Cholera Serum.
M. Dorset and R. R. Henley 333
Silver-Scurf of the Irish Potato Caused by Spondylocladium
atrovirens. Eugene S. Schultz 339
Woolly Pear Aphis. A. C. Baker and W. M. Davidson 351
Pathological Histology of Strawberries Affected by Species of
Botrytis and Rhizopus. Neil E. Stevens 361
Life Histories and Methods of Rearing Hessian-Fly Parasites.
C. M. Packard 367
Effect of Rontgen Rays on the Tobacco, or Cigarette, Beetle and
the Results of Experiments with a New Form of Rontgen Tube.
G. A. Runner 383
Stimulating Influence of Arsenic upon the Nitrogen-Fixing Or-
ganisms of the Soil. J. E. Greaves 389
Transmission and Control of Bacterial Wilt of Cucurbits. Fred-
erick V. Rand and Ella M. A. Enlows 417
Correlated Characters in Maize Breeding. G. N. Collins 435
Comparative Study of the Amount of Food Eaten b)'^ Parasitized
and Nonparasitized Larvae of Cirphis unipuncta. Daniel G.
Tower 7" 455
Aleyrodidae, or White FHes Attacking the Orange, with Descrip-
tions of Three New vSpecies of Economic Importance. A. L.
Quaintance and A. C. Baker 459
Relative Water Requirement of Corn and the Sorghums. Edwin
C. Miller 473
Availability of Mineral Phosphates for Plant Nutrition. W. L.
BURLISON 485
California Green Lacewing Fly. V. L. Wildermuth 515
Rape as Material for Silage. A. R. Lamb and John M. Evvard. . 527
Effect of Autolysis upon Muscle Creatin. Ralph Hoagland and
C. N. McBryde 535
Storage Rots of Economic Aroids. L. L. Harter 549
Experiments with Clean Seed Potatoes on New Land in Southern
Idaho. O. A. Pr.-^tt s73
Digestibility of Very Young Veal. C. F. L.'VNGWORThy and A. D.
Holmes 577
Influence of Calcium and Magnesium Compounds on Plant Growth.
F. A. Wyatt 589
Apr. 3-sept. 25, 1916 Contents
Page
Larval Characters and Distribution of Two Species of Diatraea.
T. E. Hollow AY 62 1
The Disease of Potatoes Known as "Leak." Lon A. Hawkins. . 627
Digestibility of Hard Palates of Cattle. C. F. Langworthy and
A. D. Holmes 641
Some Properties of the Virus of the Mosaic Disease of Tobacco.
H. A. Allard 649
Life Cycles of the Bacteria. F. Lohnis and N. R. Smith 675
A Respiration Calorimeter, Partly Automatic, for the Study of
Metabolic Activity of Small Magnitude. C. F. Langworthy
and R. D. Milner 703
Mottle-Leaf of Citrus Trees in Relation to Soil Conditions. Ly-
man J. Briggs, C. a. Jensen, and J. W. McLane 721
Vegetative Succession Under Irrigation. J. Francis Macbride. . 741
Agricultural Value of Impermeable Seeds. George T. Har-
rington 761
Mendelism of Short Ears in Sheep. E. G. Ritzman 797
Life-History Studies of Cirphis Unipuncta, the True Army Worm.
John J. Davis and A. F. Satterthwait 799
Infection of Timothy by Puccinia graminis. E. C. Stakman and
F. J. PiemeisEL 813
Control of the Powdery Dryrot of Western Potatoes Caused by
Fusarium trichothecioides. O. A. Pratt 817
Use of the Moisture Equivalent for the Indirect Determination of
the Hygroscopic Coefficient. Frederick J. Alway and
JOUETTE C. RUSSEL 833
Thersilochus conotracheli, a Parasite of the Plum Curculio. R.
A; CUSHMAN 847
Effect on Plant Growth of Sodium Salts in the Soil. Frank B.
Headley, E. W. Curtis, and C. S. Scofield 857
Observations on the Life History of the Army Cutworm, Chori-
zagrotis auxiliaris. R. A. Cooley 871
Aphidoletes meridionalis, an Important Dipterous Enemy of
Aphids. John J. Davis 883
Influence of Barnyard Manure and Water upon the Bacterial
Activities of the Soil. J. E. Greaves and E. G. Carter 889
Progressive Oxidation of Cold-Storage Butter. D. C. Dyer 927
Bacteriological Studies of a Soil Subjected to Different Systems
of Cropping for Twenty-five Years. P. L. Gainey and W. M.
GiBBS 953
Studies on the Physiology of Reproduction in the Domestic Fowl. —
XV. Dwarf Eggs. Raymond Pearl and Maynie R. Curtis. . 977
a-Crotonic Acid, a Soil Constituent. E. H. Walters and Louis
E. Wise 1043
Index 1047
ERRATA
Page 151, line 25, " Sphaeropsideae " should read "Sphaerioidaceae."
Page 156, "Peniciilium cameniberti, var. rogri" should read "Peiiicillium camemberti, var. rogeri."
Page 296, PI. XXXIII, figs. 3 to 15. The jnagnification of the illustrations should be half that stated in
the legend.
Page 303, line 17, " Plate XXXVI, figures i to 4" should read " Plate XXXVII, figures i to 4" and
"In figure 4. Plate XXXVI" should read "In figure 4, Plate XXXVII."
Page 3 18, Table IV, under head "General remarks," "rooting" should read "shooting."
Page 337, Table II, 4th column, "Phenolized defibrinated blood 3895 (unwashed)" should read "Phe-
nolized defibrinated blood 3895."
Page 377, last line, 2d paragraph, "winged" should read "wingless."
Page 384, Table I, 6th column, "Current (milliajnpere minutes) " should read "Current (milliamperes)."
Page 388, line 13 from bottom, omit "with humidity at 57."
Page 419, line 25, "The twelve-spotted (or squash) lady beetle" should read "The squash lady
beetle."
Page 419, line 28, " {Crepidodera cucu7neris)" should read "(Epilrix cucumeris)."
Page 459, lines 2 and 24, omit " Three."
Page 471, line 4, "Aleurodes inori Ckll." should read "Aleurodes mori, var. ari:yonensis Ckll."
Page 762, Table I, first column, "Medicago arbica" should read "Medicago arabica."
Page 791, Table XIV, ist column, "(p. 23)" should read "(p. 783)."
Page 865, legend under figure 5, "The solid black line, etc.," should read " The hatched line."
Page 866, legend under figure 6, omit sentences 2 and 3.
Page 881, line 6 from bottom, "April " should read " May."
VI
Page
ILLUSTRATIONS
PLATES
Relation ok Carbon Bisulphid to Soil Organisms and Plant
Growth
Plate I. Plate cultures of soil organisms growing on agar: Fig. i. — Colonies of
organisms from untreated soil. Fig. 2. — Colonies from soil treated with 2
per cent of carbon bisulphid. Fig. 3. — Colonies from soil treated with 2 per
cent of carbon bisulphid and evaporated. Fig. 4. — Colonies from soil
treated with 2 per cent of carbon bisulphid, evaporated, and reinoculated
with 5 per cent of soil from an untreated jar 20
Plate II. Fig. i. — Effect of varying amounts of carbon bisulphid on mustard;
A, B, soil untreated; C, D, soil treated with 0.5 per cent of carbon bisul-
phid; E, F, soil treated with i per cent of carbon bisulphid; G, H, soil
treated with 2 per cent of carbon bisulphid. Fig. 2. — Effect of carbon
bisulphid on clover in peat soil; .4. B, soil untreated; C, D, soil treated
with 2 per cent of carbon bisulphid. Fig. 3. — Effect of carbon bisulphid
on buckwheat in sand cultures; A, B, soil untreated; C, D, soil treated
with 2 per cent of carbon bisulphid 20
Climatic Conditions as Related to Cercospora beticoLa
Plate III. Cercospora beticola: Overwintering tests on the experimental field
at Rocky Ford, Colo., during 1912-13: Sugar-beet leaves infected with Cer-
cospora beticola (/) stored in soil in boxes, (2) buried in the groimd at dif-
ferent depths from i to 8 inches, and (j) left exposed above the ground in a
pile of hayed sugar-beet tops 60
Plate IV. Field stations for the collection of weather data at Rocky Ford,
Colo., in 1913: Fig. i. — Weather shelter, anemometer, and rain gauge at
edge of sugar-beet field . Fig. 2 . — Weather shelter among beet plants, show-
ing hygrothermograph and cog psychrometer. Fig. 3. — Weather shelter of
the local Weather Bureau station about 3 miles from sugar-beet field 60
OviPOsiTioN OF Megastigmus spermotrophus in the Seed of Douglas
Fir
Plate V. Oviposition of Megastigmus spermotrophus in the cones of Douglas fir:
Fig. I. — Type of cage in which the oviposition of Megastigmus spermotro-
phus was observed. Fig. 2, 3. — Female resting on cone with ovipositor
inserted 68
Plate VI. Oviposition of Megastigmus spermotrophus in the cones of Douglas
fir: Figs, i, 2. — Two positions of female on surface of cone with ovipositor
inserted. Fig. 3. — Female resting on cone with ovipositor inserted 68
Plate VII. Oviposition of Megastigmus spermotrophus in the cones of Douglas
fir: Fig. i. — Female in act of withdrawing ovipositor from cone. Fig. 2. —
Section through a Douglas fir cone on which a female has been killed while
in the act of ovipositing. Fig. 3. — A portion of same cone and dead female
with ovipositor inserted 68
vn
VIII Journal of Agricultural Research Voi.vi
Citrus Canker
Page
Plate VIII. Fig. i. — Grapefruit leaf showing young Citrus cankers. Fig. 2. —
Old Citrus canker on Satsuma leaves. Figs. 3, 4. — Seedling grapefruit
branches affected with Citrus canker. Fig. 5. — Severe canker infection of
branches of Citrus trifoliata 100
Plate IX. Fig. i. — View of lower side of leaves of seedling grapefruit artifi-
cially inoculated with Pseudomonas cilri. Fig. 2 . — Top view of plant shown
in figiu-e I. Fig. 3. — Spongy white cankers on leaf and twig of seedling
grapefruit produced by artificial inoculation. Fig. 4. — Citrus canker on
Satsuma leaves resulting from artificial inoculation with Pseudomonas citri.
Fig. 5. — Photomicrograph of section of young, open canker on grapefruit. . 100
Plate X. Fig. i. — Natural Citrus canker infection on leaves of Citrus trifoliata.
Fig, 2. — Matm-e cankers on fruit of Citrus decumana. Fig. 3. — Canker on
seedling grapefruit leaves, entrance having been effected through abrasions
made by thorns. Fig. 4. — Yotmg spongy cankers on fruit of Citrus decu-
mana. Fig. 5. — Phoma soda on cellulose agar showing dissolution of cellu-
lose. Fig. 6. — Mature cankerous areas on leaves of Dxmcan grapefruit. . . . 100
Plate XI. Fig. i. — Cankers on old grapefruit leaves which have enlarged
during the second growing season. Fig. 2. — Citrus canker resulting from
immersion of leaves in a bacterial suspension. Fig. 3. — Culture of Phoma
socia showing pycnidial formation in concentric rings. Fig. 4. — Dilution
poured plate of Pseudomonas citri on green-bean agar. Fig. 5. — Photo-
micrograph of pycnidium of Phoma socia taken in reflected sunlight.
Fig. 6. — Photomicrograph of pycnidia of Phoma socia taken in diffuse
light 100
Insect Injury to Cotton Seedlings
Plate XII. Fig. i. — Cutworm injury to cotton seedlings; produced in breed-
ing cages. Figs. 2, 3. — Cutworm injury to cotton seedling 140
Plate XIII. Fig. i. — Cutworm injury to cotton seedling. Fig. 2. — Tussock
larva feeding upon cotton leaf. Fig. 3. — Injury produced by a nearly
full-grown tussock larva when confined in a screen cage containing potted
cotton plants 140
Plate XIV. Cotton leaves showing grasshopper injury 140
Plate XV. Fig. i. — Underside of cotton leaf showing grasshopper injury.
Fig. 2. — Cotton leaf showing grasshopper injury 140
Plate XVI. Fig. i. — Injury to terminal bud of cotton by lepidopterous larva.
Fig. 2. — Two cotton plants from laboratory garden with leaves removed. . 140
Watermelon Stem-End Rot
Plate XVII. Watermelons, showing the effect of inoculation with species of
Diplodia: Fig. 1. — The upper four melons were held as checks; the
lower five are melons nine days after having been inoculated with a culttue
of Diplodia sp. Fig. 2. — A watermelon nine days after having been in-
oculated with a culture of Diplodia tuhericola E. and E 152
Crowngall Studies Showing Changes in Plant Structures due to a
Changed Stimulus
Plate XVIII. Teratoid cro\vngalls produced in Pelargonium spp 182
Plate XIX. Teratoid crowngalls produced in castor-oil plant. Fig. A. — A
red-stem variety. Leaves reflexed; axis distorted; and feeble shoots
developing out of the axillary tumors. Fig. B. — A green-stem glaucous
variety 182
Apr. 3-sept. 25, 1916 Illustratiotis IX
Plate XX. Teratoid crowiigalls produced in tobacco by inoculating Bacterium
tumefaciens 182
Plate XXI. The teratoid tumor strand of Plate XX, which gives rise during
its course to more than 30 small tumors. Top. — Cross section of outer part
of right side of stem of tobacco plant show-n on Plate XX. P, outer edge
of the phloem; E, epidermis; T St, tumor strand, which is bedded in the
normal cortex of the stem. Bottom. — Longitudinal section from upper
part of the above tumor strand. The pathological tissues are S T, sieve
tubes; C, cambium; Tr, trachei; Sp, spiral vessels 182
Plate XXII. Teratoid cro\\Tigalls produced in a tobacco plant by inoculating
Bacterium tumefaciens into the leaf axils 182
Plate XXIII. Teratoid cro\\Tigalls produced in tobacco leaves with the hop
strain of Bacterium tumefaciens by local inoculations — that is, inoculation
in places where shoot anlage are not known to exist. Fig. A. — Portion of
an upper leaf showing four shoot-bearing tumors growing from upper surface
of the inoculated midrib. Fig. B. — Same as A, but the leaf reversed and
the midrib stripped of its blade to show two other shoot-bearing tumors
which have developed from its under surface. Fig. C. — From middle
of another leaf on the same plant as A , but further magnified and photo
made on an orthochromatic plate to show the pale green character of
the shoot as contrasted with the dark green of the surrounding leaf 182
Hyperaspis binotata, a Predatory Enemy of the Terrapin Scale
Plate XXIV. Hyperaspis binotata: Fig. i. — Male, showing the characteristic
markings. Fig. 2. — Female, showing the dorsal view. Fig. 3. — Eggs and a
second-instar larva, a, Second-instar larva as disclosed by displacing the
host; b, larvae of the terrapin scale, Eulecanium nigrofasciatum; c, a dis-
placed scale ; d, eggs " in situ "; e, egg. Fig. 4. — First-instar larva. Fig. 5. —
Method of attacking the mature scales during the third and fourth instars . . 204
Plate XXV. Hyperaspis binotata: Fig. i. — Mature larva as it appears when
attacking the leaf -attached larvse of the terrapin scale , Eulecanium nigro-
fasciatum. Fig. 2. — Ventral yiewof mature larva. Fig. 3. — Dorsal view of
pupa, showing the last larval molt skin and the depressed segmented area.
Fig. 4. — Ventral view of pupa 204
Tests of Three Large-Sized Reinporced-Concrete Slabs under
Concentrated Loading
Plate XXVI. Fig. i. — Load-applying and load -measuring apparatus for test-
ing reinforced-concrete slabs, showing set-up for 4-point loading. Fig.
2. — Load -measuring apparatus and hydraulic jack for testing reinforced-
concrete slabs 234
Occurrence op Sterile Spikelets in Wheat
Plate XXVII. — Comparison of the number of sterile spikelets on bearded and
beardless varieties of wheat 250
Egg and Manner op Oviposition of Lvctus planicollis ,
Plate XXVIII. Lyctus planicollis: Fig. i. — Outline of the egg, showing strand
attachment. Fig. 2. — Greatly enlarged view of end of egg, showing granu-
lar appearance. Fig. 3. — Larva within egg, ready to hatch. Fig. 4.—
Sketch of egg in pore of wood on radial section of green-ash ladder-nmg
stock, showing longitudinal striae 276
X Journal of Agricultural Research voi. vi
Page
Plate XXIX. Lyctus planicollis: Larval burrows in an ash shovel handle 276
Plate XXX. Lydus planicollis: Pupal cells in "powder-posted" white-ash
shovel handle 276
Plate XXXI. Lydus planicollis: Exit holes of adults in ash shovel handles. . 276
Hypoderma deformans, an Undescribed Needle Fungus op the Western
Yellow Pine
Plate XXXII. Fig. i. — Needles of Pinus ponder osa infected with Hypoderma
deformans, showing the apothecia. Fig. 2. — Branches of Pinus ponderosa
deformed and broomed by Hypoderma deformans. Fig. 3. — A branch of
Pinus ponderosa, showing how it will hang vertically when supporting a
large broom caused by Hypoderma deformans 288
Ornix geminatella, the Unspotted Tentiform Leap Miner of Apple
Plate XXXIII. Omix geminatella Pack.: Fig. i. — Moth expanded. Fig. 2 —
Moth at rest on leaf. Fig. 3. — Egg on lower surface of leaf; also tunnel made
by miner on leaving the egg. Fig. 4. — Dorsal view of first larval stage ; be-
low, side view of head and thorax. Fig. 5. — Dorsal view of second larval
stage. Fig. 6. — Side view of second larval stage. Fig. 7. — Dorsal view of
third larval stage, showing edge of thoracic legs. Fig. 8. — Dorsal view of
fourth larval stage. Fig. 9. — Side view of fotirth larval stage. Fig. 10. —
Ventral view of pupa. Fig. 11. — Dorsal view of same. Fig. 12. — Lower
surface of leaf with numerous partly developed mines; also two cocoons,
one exposed. Fig. 13. — Portion of leaf showing a mine in process of devel-
opment. Fig. 14. — A small twig showing leaves badly curled and injvired
by numerous mines. Fig. 15. — Leaf much distorted with 10 mines almost
completed; also one cocoon appears at the tip of the leaf 296
A Western Fieldrot op the Irish Potato Tuber Caused by Fusarium
radicicola
Plate XXXIV. Fig. i, 2, 3, 4. — Types of stem-end blackrot, lenticel rot, and
eye rot in Idaho Rural potato tubers. Fig. 5, 6. — Longitudinal and cross
sections of an Idaho Rural tuber infected with blackrot 310
Plate XXXV, Fig. i. — Netted Gem potato tuber infected with jelly-end rot.
Fig. 2. — Stem-end view of a Netted Gem tuber infected with jelly-end rot.
Fig. 3. — Longitudinal section of a Netted Gem tuber infected with jelly-end
rot. Fig. 4. — Idaho Rviral tuber infected with stem-end and lenticel black-
rot, after having been kept 10 days in a moist chamber. Fig. 5. — Idaho
Rural tuber infected with lenticel blackrot after having been kept in a
moist chamber for 10 days 3 10
Plate XXXVI. Fig. i, 2. — Stem-end blackrot produced by stem-end punc-
tures with a needle carrying Fusarium radicicola. Netted Gem and Idaho
Rural potato tubers. Fig. 3. — Lenticel blackrot produced by spraying the
tuber with a spore suspension of F. radicicola. Netted Gem tuber. Fig. 4 —
Same tuber as shown in figure 3 ; after having been kept a few days longer
in the moist chamber. Fig. 5. — Stem-end blackrot produced by spraying
an Idaho Rural tuber with a spore suspension of F. radicicola. Fig. 6. —
Stem-end blackrot produced by the inoculation of the tuber stolon. Idaho
Rural tuber. Fig. 7. — Blackened vascular system produced by the inocu-
lation of the tuber stolon. Idaho Rural tuber 310
Apr. 3-sept. 35. 1916 Illustrations XI
Page
Plate XXXVII. Fig. i, 2, 3. — Stem-end and lenticel blackrot produced by
spraying the growing tubers with a spore suspension of Fusarium radicicola.
Idaho Rural potato tubers. Fig. 4. — Eye infection produced by spraying
the growing tuber with a spore suspension of F. radicicola. Netted Gem
tuber. Fig. 5, 6, 7. — Stem-end blackrot produced by the inoculation of
the stolons of growing Idaho Rural tubers. Fig. 8. — Stem-end rotof Netted
Gem tuber produced by inoculating the stolon of the growing tuber 310
Comparative Study of the Root Systems and Leaf Areas of Corn and
THE Sorghums
Plate XXXVIII. Fig. i. — Method used in isolating root systems in the field.
View of two soil prisms ready for washing. Fig. 2. — Method used in isolat-
ing root systems. Fig. 3. — Method of washing used in the isolation of the
root systems 332
Plate XXXIX. Fig. i. — Root sj^stem of a com plant that had reached a
height of 3 feet 6 inches. Fig. 2. — Root systems of two com plants with a
height of I foot 6 inches. Fig. 3. — Root system of a Dwarf milo plant at
the age of 4 weeks. Fig. 4. — Root systems of two BlackhuU kafir plants
I foot in height 332
Plate XL. Fig. i. — Root systems of two mature com plants. Fig. 2. — Root
system of a com plant at the time of " shooting " 332
Plate XLI. Fig. i.— Root systems of two BlackhuU kafir plants at the time
they had reached a height of 6 feet and were blooming. Fig. 2. — Root
system of two Dwarf milo plants at the time the seed was in the milk stage. 332
Plate XLII. Fig. i. — Portion of a primary root of Pride of Saline com, show-
ing the number and relative size of the secondary roots. Fig. 2. — Portions
of the primar)' roots of BlackhuU kafir, showing the number and relative
size of the secondary' roots '. 332
Plate XLIII. Fig. i. — Pride of Saline com. Dwarf milo, and BlackhuU kafir
plants, showing their relative leaf and sheath areas at 4 weeks of age.
Fig. 2. — Pride of Saline com, Dwarf milo, and BlackhuU kafir plants,
showing their relative leaf and sheath areas at 6 weeks of age 332
Plate XLIV. Fig. i. — Pride of Saline com, Dwarf milo, and BlackhuU kafir
plants, showing their relative leaf and sheath areas at 8 weeks of age.
Fig. 2. — Pride of Saline com, Dwarf milo, and BlackhuU kafir plants,
showing their relative leaf and sheath areas at 10 weeks of age 332
Silver-Scurf of the Irish Potato Caused by Spondylocladium atrovirens
Plate XLV. Fig. i. — Potato tubers showing shriveling and a silvery appear-
ance caused by Spondylocladium atrovirens. Fig. 2. — Tuber naturally
infected by 5. atrovirens, showing the segregated area type of infection.
Fig. 3. — Immatiu-e potato tuber artificially inoculated with conidia of 5.
atrovirens, July, 1913, at Houlton, Me 350
Plate XLVI. Fig. i. — Photomicrograph of Spondylocladium atrovirens on
corn-meal agar, showing method of development of conidiophores and
conidia in the early stages. Fig. 2. — Photomicrograph of 5. atrovirens in
hanging-drop culture, showing development of conidiophore and conidia
in mature stages. Fig. 3. — Negative heliotropism of S. atrovirens on corn-
meal agar exposed on one side to daylight from April 8 to April 24, 1915,
in laboratory at room temperature 350
Plate XLVII. Effect of temperature upon mycelial development of Spon-
dylocladium atrovirens in pure culture on corn-meal agar at end of four
weeks 350
XII Journal of Agricultural Research volvi
Page
Plate XLVIII. Contact infection. A part of the new tubers becoming in-
•fected with Spondylocladium atrovirens by means of contact with the in-
fected mother tuber 350
Pathological Histology of Strawbermes Affected by Species op Botrytis
AND Rhizopus
Plate XLIX. A-H, Strawberry cells attacked by Botrytis sp.: A, Hypha
growing partly between and partly within strawberry cells; B, hyphae
inside strawberry cells in which remnants of the protoplasm may still be
distinguished; C, hypha passing from one cell into another across a short
intercellular space; D, E, F, G, H, hyphae entering cells in various ways
(in G the hypha has pushed back a portion of the cell wall before breaking
through). I-M, Strawberry cells attacked by Rhizopus. sp.: /, Hypha
growing between the epidermis and the adjacent layer of storage cells; J,
hypha growing over the stu^ace of the strawberry; K, hyphae growing
underneath the epidermal layer and between the storage cells; L, Rhizopus
sp. growing between epidermal cells; M, germinating spore in cavity
formed by a seed ' 366*
Plate L. Strawberry cells attacked by Rhizopus sp. A, Normal storage cell
of strawberry; B, storage cell showing a slight contraction of the proto-
plasm; C, D, E, F, G, progressive contraction of protoplasm of host cells
near hyphae; H, I, J, strawberry cells near hyphae in which the cell wall
has crumpled with the contraction of the protoplasm; K, M, hyphae inside
cells; L, hyphae growing between cells of the strawberry; K, L, M are
drawn from berries which had been rotted in the desiccator 366
Life Histories and Methods of Rearing Hessian-Fly Parasites
Plate LI. Fig. i. — Egg of Eupelmus allynii. Fig. 2. — Egg of Eupelmus
allynii in situ. Figs. 3, 4. — Pupa of Eupelmus allynii. Fig. 5. — Egg of
Merisus destructor. Fig. 6. — Pupa of Merisus destructor. Fig. 7. — Egg of
Micromelus subapterus. Fig. 8. — Pupa of Micromelus subapterus 382
Plate LIL Fig. i. — Mandibles of full-grown larva of Eupelmus allynii. Fig.
2. — Larva of Eupelmtis allynii. Fig. 3. — Mandibles of full-grown larva of
Merisus destructor. Fig. 4. — Larva of Merisus destructor. Fig. 5. — Man-
dibles of full-grown larva of Micromelus subapterus. Fig. 6. — Larva of
Micromelus subapterus 382
Transmission and Control of Bacterial Wilt of Cucurbits
Plate LIII. — Two wilted cucumber plants which contracted bacterial wilt at
beetle gnawings of the leaves marked x. Three healthy, uninjured plants
in same hill are also shown 434
Plate LIV. — Plots in field i, East Marion, Long Island, N. Y., 1915. Fig. i. —
Plot sprayed with Bordeaux mixture and lead arsenate, begiiming June 25.
Fig. 2. — Plot sprayed with Bordeaux mixture and lead arsenate, beginning
July 19, after most of the stiiped-beetle injury had occurred . Fig. 3 . — Plot
sprayed with Bordeaux mixture and lead arsenate, beginning July 27.
Fig. 4. — General -view of field, showing cages and meteorological-instrument
shelter 434
Correlated Characters in Maize Breeding
Plate LV. — Typical plant of the Waxy Chinese variety of maize, showing
numerous tassel branches, erect leaf blades, one-sidedness, and curved
tassel 454
Apr. 3-Sept. 25, i9j6 IllustrationS XIII
Page
Plate LVI. Fig. i. — Uppermost leaf sheaths of Chinese maize plant, showing
the one-sided arrangement of leaf blades and absence of hairs. Fig. 2 . —
Leaf sheath of the Waxy Chinese variety of maize, showing the transverse
lines and absence of hairs 454
Plate LVII. — A plant of the Esperanza variety of maize, showing the drooping
leaves, few tassel branches, and elongated intemodes characteristic of the
variety 454
Plate LVIII.— Leaf sheaths of the Esperanza variety of maize, showing the
maximum development of tuberculate hairs 454
Plate LIX. — A leaf sheath of a second-generation hybrid maize plant 454
Plate LX. — A first-generation plant of Chinese X Esperanza maize hybrid. . . . 454
Plate LXI. — A second-generation plant of a Chinese X Esperanza maize hybrid 454
Plate LXIL — A second-generation plant of a Chinese X Esperanza maize hybrid 454
Plate LXIII. — A second-generation plant of a maize hybrid, showing the
"goose-neck" character that appeared for the first time in this hybrid 454
Aleyrodidae, or White Flies Attacking the Orange, with Descriptions
OF Three New Species of Economic Importance
Plate LXIV. — Akurocanthus woglumi: Eggs, larvae, and pupa cases on orange
leaves 472
Plate LXV. — Aleurocanthtis woglumi: Fig. i. — Colony on an orange leaf.
Fig. 2. — Eggs and pupa cases 472
Plate LXVI. — Fig. i. — Dialeurodes citri: Pupae. Fig. 2. — Male and female
adults of an aleyrodid. Fig. 3. — Aleurolobus marlatti 472
Plate LXVIL Aleurothrixus howardi: Larvae and pupa cases on an orange
leaf 472
Plate LXVIII. Aleurothrixus porteri: Larvae and pupa cases on A/jr/wj sp 472
Plate LXIX. Fig. i. — Tetraleurodes inori, var. arizonensis: Larvae and pupa
cases on an orange leaf. Fig. 2. — Tetraleurodes mori: Pupa cases on a mul-
berry leaf 472
Relative Water REQtnREMENT of Corn and the Sorghums
Plate LXX. Fig. i. — General view of the screened inclosure and the scale
house. Fig. 2. — Method of moving the cans. Fig. 3. — General view of
the plant shelter and the surrounding country at Garden City, Kans 484
Plate LXXI. Fig. i. — Dwarf milo, grown May 22 to September 3, 1915. Fig.
2. — Dwarf Blackhull kafir, grown May 22 to September 11, 1915. Fig. 3.—
Feterita, ^rown May 22 to September 6, 1915 484
Plate LXXII. Fig. i. — Sudan grass, grown May 22 to September 14, 1915.
Fig. 2. — Pride of Saline com, grown May 22 to August 25, 1915. Fig. 3. —
Blackhull kafir, grown May 22 to September 18, 1915. Fig. 4. — Method of
sealing the lids with tape and the wax seal around the plants 484
Availability of Mineral Phosphates for Plant Nutrition
Plate LXXIII. Effect of varying quantities of Tennessee brown rock phos-
phates on plant growth: Fig. i. — Spring wheat. Fig. 2. — Sixty-Day oats . . 514
Plate LXXIV. Effect of varying quantities of Tennessee brown rock phos-
phate on plant growth: Fig. i. — Barley. Fig. 2. — Timothy 514
Plate LXXV. Effect of varying quantities of Tennessee brown rock phos-
phate on plant growth: Fig. i. — Cowpeas. Fig. 2. — Soybeans. Fig. 3. —
Red clover. Fig. 4. — Alfalfa 514
Plate LXXVI. Effect of different kinds of mineral phosphate applied in
different quantities for red clover 514
XIV Journal of Agricultural Research voi. vi
Page
Plate LXXVII. Cowpeas, showing the comparative effect of Tennessee brown
rock phosphate alone and in combination with dextrose 514
Plate LXXVIII. Cowpeas, showing the comparison of their growth when
treated with Tennessee brown rock phosphate, phosphate and dextrose,
and phosphate, dextrose, and calcium carbonate 514
Plate LXXIX. Effect of different substances on the growth of cowpeas: Fig.
I. — Growth after the addition of varying quantities of raw rock. Fig. 2. —
Growth after the addition of dextrose and soluble phosphate 514
Plate LXXX. Effect of various substances and combinations on the growth of
cowpeas: Fig. i. — Effect of adding lime, phosphate rock, dextrose and
lime, and phosphate rock, dextrose, and lime to the soil. Fig. 2. — Effect of
adding nothing, lime, phosphate rock, and phosphate rock and lime to the
soil 514
Storage-Rots op Economic Aroids.
Plate LXXXI. Fig. i. — A dasheen conn {Colocasia esculenta) showing Java
blackrot produced by Diplodia tubericola. Fig. 2. — ^A conn of Alocasia sp.
showing Java blackrot produced by D. tubericola. Fig. 3 . — A dasheen tuber
partially decayed by Sclerotium rolfsii 572
Plate LXXXII, Fig. i. — A tuber of Colocasia esculenta showing a powdery
grayrot caused by Fiisarium solani. Fig. 2 . — A tuber of Xanthosoma sagitti-
folium showing partial decay by Fusarium solani. Fig. 3. — ^A tuber of
C. esculenta softened throughout by Sclerotium rolfsii 572
Plate LXXXIII. A corm of Colocasia esculenta from Brooksville, Fla., mostly
rotted away by Bacillus carotovortis 572
Influence op Calcium and Magnesium Compounds on Plant Growth
Plate LXXXIV. Fig. i. — Growth of wheat in sand containing varying quan-
tities of calcium and magnesium. Fig. 2. — Growlh of alfalfa in sand con-
taining varying amotmts of calcium and magnesium 620
Plate LXXXV. Fig. i. — Growth of soybeans following a crop which had al-
ready absorbed most of the readily available calcium and magnesium.
Fig. 2. — Growth of soybeans in soil treated with magnesium 620
Plate LXXXVI. Fig. i. — Comparative growth of soybeans in brown silt
loam and dolomite, showing that the loam would have been improved by
the addition of some limestone or dolomite. Fig. 2. — Soybeans in sand
treated with magnesium, showing that their growth increases inversely
with the quantity of magnesium applied as sulphate 620
Plate LXXXVII. Comparative root production of wheat grown in the chlo-
rids, sulphates, and carbonates of magnesium and calcium 620
Plate LXXXVIII. Fig. i. — Comparative growth of wheat in sand and in
dolomite. Fig. 2. — Comparative gro\\lh of wheat in magnesium chlorid
and magnesium sulphate 620
Larval Characters and Distribution op two Species op Diatraea
Plate LXXXIX. Fig. i. — Diatraea saccharalis cramhidoides: Larva, summer
form, dorsal view. Fig. 2. — D. zeacolella: Larva, summer form, dorsal
view. Fig. 3. — D. saccharalis cramhidoides: Larva, summer form, side view.
Fig. 4. — D. seacolella: Larva, summer form, side view. Fig. 5. — D. sac-
charalis cramhidoides: Larva, winter form, dorsal view. Fig. 6. — D. zeaco-
lella: Larva, winter form, dorsal view 626
Apr. 3-sept. 25. i9j6 Illustrations
XV
The Disease op Potatoes Known as "Leak"
Plate XC. Potatoes affected with potato leak: Fig. i, 2. — Natural infection
from fork woimd. Fig. 3. — Rot produced by inoculation with Pythium
debaryanum. Fig. 4. — Rot produced by inoculation with Rhizopus nigri-
cans
Page
640
Some Properties of the Virus of the Mos.^ic Disease of Tobacco
Plate XCI. Livingstone atmometer porus cup as used for filtration 674
Life Cycles op the Bacteri.a
Plate A. Fig. i. — Azotobacter 11. Mannite-nitrate solution, 5 days old.
Types A and La. Some cells in conjunction. Fig. 2. — Azotobacter 21.
Contact preparate from a colony on mannite agar, 4 days old. Types A, L.
Most cells in conjunction. Fig. 3. — Azotobacter 23. Contact preparate
from a colony on mannite-agar, 4 days old. Types A, B, I, Ka, and many
conjunct cells. Fig. 4. — Azotobacter 13. Jannite-nitrate solution, 17 days
old. Type KX. Fig. 5. — Azotobacter 14. Mannite-nitrate solution, 5
daysold. Type B forming I. Fig. 6. — AzotobacterS. Beef bouillon. Type
B forming types I and J '. 702
Plate B. Fig. 7. — Azotobacter 21. Mannite-agar colony, 4 days old. TypeC
forming types D and I. Fig. 8. — Azotobacter 22. Mannite-agar colony,
4daysold. Type C forming D, also A in conjtmction. Fig. 9. — Azotobacter
II. From a filter paper strip in mannite-peptone solution, 16 days old.
Types A and B forming D. Fig. 10. — Azotobacter 3. Mannite-peptone
solution, 24 days old. Types L and M forming D. Fig. 11. — Azotobacter
II. Mannite-peptone solution, 16 days old. Type D (stained) resulting
from Type C. Fig. 12. — Azotobacter 6. From condensation water of
mannite-agar slant, i day old. Type D (vmstained) containing regenera-
tive units 702
Plate C. Fig. 13. — Azotobacter 24. Mannite-nitrate solution kept 5 days after
having been heated i minute at 96° C. Types I and F developing from
D. Some I germinating in conjunct stage and inclining to form spores.
Fig. 14. — Azotobacter i. Mannite-nitrate solution, 10 days old. Types B,
K/3, E, and Fa developing from stained and unstained type D. Fig. 15. —
Azotobacter 15. From condensation water of a mannite-nitrate agar slant,
2 days old. Types Fa and F/3 developing from type D. Fig. 16. — Azoto-
bacter 17. Mannite-soil-extract agar, 2 months old. Type E, Fa, K^, and
G developing from type D. Fig. 17. — Azotobacter 17. Mannite-nitrate
agar, 10 days old. Preparate treated with hot aqueous fuchsin. Type G,
partially dissolved; also K^. Fig. 18. — Azotobacter 7. Mannite-soil-
extract solution, 14 days old. Type H forming D 702
Plate D. Fig. 19. — Azotobacter 2. Mannite-nitrate agar, 23 daysold. Spores
formingtypeD. Fig. 20. — Azotobacter 2 . Mannite nitrate-agar, 6 daysold.
Types L and F, endospores and exospores and dissolving of spores to type
D. Fig. 21. — Azotobacter 18. From a filter paper strip in mannite solu-
tion, 25, days old. Type L with gonidia, forming B (type JX). Fig. 22. —
Azotobacter 7. Mannite-soil-extract agar, 2 months old. Types E and F
forming B. Fig. 23. — Azotobacter 7. Mannite-soil-extract agar, 2 months
old. Type B, formed by types E and F, germinating to type G. Fig. 24. —
Azotobacter 7. Mannite-soil-extract agar, 2 months old. Type K7 702
55854°— 16 2
XVI Journal of Agricultural Research voi. vi
Page
Platk E. Fig. 2$.— Bacillus sublilis (No. 31). Beef agar, 2 days old. Types
I and D formed by spores. Fig. 26.— Bacillus sublilis (No. 31). Beef agar,
6 days old. Formation of type I. Fig. 27.— Bacillus sublilis (No. 31).
Beef agar, 8 days old. Type I forming H and stained D. Spores forming
unstained type D. Fig. 28.— Yellow bacillus (No. 41). Peptone-glycerin
solution, 2 days old. Type I germinating from D, stretching to type L.
Fig. 2g.— Bacterium bulgaricum (No. 49). Whey-yeast agar, 6 days old at
40° C. Types C, D, E, F, G, I, and K. Fig. 30. — Bacterium fluorescens
(No. 40). Ammonium-citrate-glycerin solution, 11 days old. Types D
and H 702
Plate F. Fig. ^i.—Sarcinaflava (No. 43)- Beef agar, i day old. Type I in
conjimction and forming D. Fig. ^2.— Streptococcus lactis (No. 48). Pep-
tone lactose solution, 5 days old. Type D, with regenerative tmits, forming
type I. Fig. ^^.—Streptococcus lactis (No. 48). Milk, 3 days old. Types
D and I in casein. Fig. 34. — Bacillus radicicola(No. ^g). Types D and I.
Preparate made from a root nodule in 1908. Fig. 35. — Spirillum sp. from
Great Salt Lake (No. 46). Beef broth plus 3 per cent of sodium chlorid, 14
days old. Budding and branching forms; stained and tmstained regenera-
tive bodies. Some cells in conjunction. Fig. 36. — Spirillum sp. from
Great Salt Lake (No. 46). Beef broth plus 3 per cent of sodium chlorid,
14 days old. Type I germinating 702
Plate G. Fig. 37. — Micrococcus candicans from soil (No. 45). Ammonium-
citrate-glycerin solution, 6 days old. Irregular, thick-walled type I.
Fig. 38. — Micrococcus candicans from milk (No. 44). Ammonium-citrate-
glycerin solution, 2 days old. Irregular, thick-walled type I. Fig. 39. —
Yellow bacillus (No. 41). Beef agar, i day old. Budding gonidia, forma-
tion and germination of type I. Fig. 40. — Bacteriiim fluorescens (No. 40).
Ammonium-citrate-glycerin solution, 2 days old. Budding gonidia, for-
mation and germination of type I. Fig. 41 .—Bacterium fluorescens (No. 40),
Beef agar, 4 days old. Filterable gonidia germinating. Fig. 42. — Bac-
terium fluorescens (No. 40). Beef agar, 4 days old. Types D and F formed
bj"^ filterable gonidia. Dark field 702
A Respiration Caloiumeter, Partly Automatic, for the Study op Meta-
bolic Activity of Small Magnitude
Plate XCII. — General view of the respiration calorimeter: A, Chamber
inclosed in heat-insulating cover. B, Tension equalizer to maintain atmos-
pheric presstire in the air of the chamber. C, Absorber table. D, Rotary
pump to maintain air circulation. E, Motor to drive pump. F, Bottles
containing sulphuric acid to remove water vapor from circulating air. G,
Large U-tube, containing soda-lime to remove carbon dioxid from the air.
H, Bottle containing sulphuric acid to catch the water vapor from the soda-
lime. /, Bottle containing cotton to catch sulphuric acid vapor. /, Small
absorbers for determining water vapor and carbon dioxid in residual air. K,
Meter to measure the sample of residual air. L, Reservoir to maintain a
constant pressure of water in the heat absorber in the chamber. M, Tank
to catch water flowing from the heat absorber. A^, Pump to raise water
from the tank to the reservoir. O, Devices for automatically controlling
and recording temperatures 720
Apr. 3-sept. 25. 1916 Illustrations xvil
Page
Plate XCIII. — Chamber with part of outer covering removed: -4 , Double metal
wall chamber. B, Heat-insulating outer cover. C, Window to chamber.
D, Outlet providing passage for pipes, wires, etc., through the walls of the
chamber. The exterior ends of the resistance thermometers for ingoing and
outgoing water are seen projecting from the outlet. E, Removable top of
chamber. F, Device for heating the air entering the respiration chamber.
G, Small pipe carrying water for cooling the outer metal wall of the chamber.
H, Electric-resistance wire carrying current for heating the outer wall 720
Plate XCIV. — Apparatus connected with the respiration calorimeter: A,
Tension equalizer. B, Mixing bottle for equalizing the temperature of
water entering the heat absorber. C, Device for heating air entering the
respiration chamber. D, Preheater, and E, final heater, for raising the
temperature of water entering the heat absorbers. There is an electric-
heating coil in the lower half and an electric-resistance thermometer in the
upper half of the final heater. F, Temperature indicator comprising part
of the apparatus for controlling the temperature of the water entering the
heat absorber. G, Multiple-point switch for connecting the resistance
thermometers for the metal walls and air of the chamber with the Wheat-
stone bridge for measuring their temperatures. H, Tube conducting air
from the respiration chamber to the rotary air pump. /, Tube conducting
air from the purifying devices to the respiration chamber 720
Plate XCV. — Devices for controlling and recording temperatures: A, Mechan-
ism for shifting the contact on the rheostats controlling the current for heat-
ing the outer walls of the calorimeter chamber and the ingoing air. B, Ratio
coils for the four bridges governing the action of the shifting mechanism A
are combined in this box, together with means for checking the constancy of
the resistance of the coils and for correcting slight inequalities in them and
also to compensate for small differences in the pair of resistance thermome-
ters forming the other arms of each bridge. C, Mechanism for shifting the
contact on the rheostat controlling the current in the heating coil in the
final heater, shown at E in Plate XCIV. D, Temperature-difference
recorder (self-balancing Wheatstone bridge) for continuously recording the
difference between the temperature of the water entering and that leaving
the heat absorber. E, " Check box" containing the ratio coils of the bridge
for temperature difference measurements and coils for extending the range
of differences measured, with means for checking the constancy of the resis-
tances of the coils and the accuracy of the recorder readings and also for
compensating for slight differences in the resistance of the thermometer
coils when they are at the same temperature 720
Mottle-Leaf of Citrus Trees in Relation to Soil Conditions
Plate H. — Various stages in mottle-leaf of the orange 740
Plate XCVI. — Orange leaves showing mottle-leaf 740
Plate XCVII. — A more advanced stage of mottle-leaf of orange, showing the
reduction in the size of the leaves 740
Vegetative Succession Under Irrigation
Plate XCVIII.— Rock Creek Valley, near Rock River Station 760
Plate XCIX. — A nearer view of the bench slope; the same tree shown in Plate
C, figure 1 760
Plate C. Fig. i. — Where upland and lowland meet. Fig. 2. — Characteristic
draw; the stream valley beyond. Lupin, wheat-grass, white sage, and
gaillardia in the foreground 760
XVIII Journal of Agricultural Research voi.vi
Page
Platu CI. Fig. I.— The bench. The cotirse of Rock Creek is indicated by the
distant trees. Fig. 2. — Part of a reservoir on the Rock Creek ranch 760
Plate CII. Fig. i. — Lupin recessive and cat 's-f&ot becoming dominant. Gay's
sedge subphase in background. Fig. 2 .—Wheat-grass phase 760
Plate CIII. Fig. i.— Squirrel-tail phase. A few grindelias in the foreground.
Fig. 2.— Rush-sedge phase (the darker areas) replacing wheat-grass phase. 760
Plate CIV. Fig i.— Hair-grass phase. Fig. 2.— Natural meadow 760
Plate CV. Fig. i .—Field of oats on bench. Cat's-foot and other upland plants
in foregroimd. Fig. 2.— Alfalfa field one year after sowing. Cat's-foot and
bench grasses in foreground 76°
Agricultural Value op Impermeable Seeds
Plate CVI. Fig. i. — A row of alsike clover from impermeable seeds between
two rows from permeable seeds. Fig. 2 . — A row of white clover from imper-
meable seeds between two rows from permeable seeds 796
Life-History Studies op Cirphis unipuncta, the True Army Worm
Plate CVII.— ^, Cages for rearing Cirphis unipuncta; B, leaves glued together
after the eggs have been deposited; C, characteristic leaves partly eaten by
first-instar larvae; D, full-grown larva; E and F, characteristic pupal cells. 812
Control op the Powdery Dryrot of Western Potatoes Caused by
Fusarium trichothecioides
Plate CVIII. Fig. i. — A potato tuber infected with powdery dryrot, showing
the wrinkled condition of skin due to the decay of underlying tissues.
Fig. 2. — A potato tuber infected with powdery drj'^rot. Advanced stage.
Fig. 3. — Section through a potato tuber infected with powdery dryrot,
showing the internal cavities filled with the mycelium and the spores of the
fungus 832
Thersilochus conotracheli, a Parasite op the Plum Curculio
Plate CIX. Thersilochus conotracheli: Fig. i. — Adult female. Fig. 2. — a,
Adult male ; b, side view of abdomen 856
ApHIDOLETES MERIDIONALIS, AN IMPORTANT DIPTEROUS EnEMY OP ApHIDS
Plate CX. Aphidoletesmeridionalis: Fig. i. — ^Adult female : a, Antenna of male,
showing structure; h, tip of male abdomen. Fig. 2. — Larva attacking a
pea aphis 888
Progressive Oxidation of Cold-Storage Butter
Plate CXI. — Gas apparatus used in the extraction and analysis of the air
confined in butter 952
Studies on the Physiology of Reproduction in the Domestic Fot\t.. —
XV. Dwarf Eggs
Plate CXII. — A collection of dwarf eggs with a normal egg in the center of
the group 1042
Plate CXIII. — Fig. i. — Ovarian follicles {b-f) and the dwarf egg a from case 27.
Fig. 2. — Shell of a compound egg Avhich was composed of two albumen
masses partly separated at the level of the seam in the shell by an incom-
plete egg membrane. Fig. 3. — Dwarf egg containing a mass of yolk not
in a yolk membrane, but separated from the albumen by a membrane-
like layer of chalazal threads. Fig. 4. — Dwarf egg formed around an
artificial yolk of agar which was inserted into the oviduct, o, Complete
egg; b, agar yolk 1042
Apr. 3-sept. 23, 1916 Illustrations xix
n'-CROTONic Acid, a Son, Constituent
Page
Plate CXIV. Fig. i. — o-Crotonic acid from soil. Fig. 2. — Synthetic a-cro-
tonic acid 1046
TEXT FIGURES
Climatic Conditioxs as Related to Cercospora beticola
Fig. I. Cercospora beticola: A, Section of ovenvintered sugar-beet leaf show-
ing embedded sclerotia-like body, o, with a mass of old conidio-
phores, b, from which a new conidium, c, was produced. D, Produc-
tion of rather typical conidiophores, b, and conidia, c, from a sclero-
tia-like mass, a, taken from overwintered host material and placed
in hanging-drop cultures 22
2. Curves of the maximum and minimum soil and air temperatures for
the period from December 5, 1912, to March 13, 1913, at Rocky Ford,
Colo., and air temperatures from December 5, 1913, to March 13,
1914, at Madison, Wis., together with the periods that snow covered
the ground 28
3. Curves of the maximum and minimum soil and air temperatures for
Rocky Ford, Colo., from March 13 to June 17, 1913, and for Madison,
Wis. , from March 13 to June 17, 1914 29
4. Curves of the maximum and minimum temperatures and relative
humidities and the number of hours that the humidity remained
above 60 from noon of the preceding to noon of the given day among
sugar-beet plants and in the air 5 feet above the field, together v/ith
the field rainfall and irrigation records. June 11 to August 2, 1913,
at Rocky Ford, Colo 32
5. Ciu-ves of the maximum and minimum temperatiwes among sugar-beet
plants and at the Weather Bureau station, and the seasonal rainfall
records at Madison, Wis., in 1914, and the number of hours that the
humidity remained above 60 among the sugar-beet plants in the
field at Madison, Wis., in 1914, and at Rocky Ford, Colo., in 1913 .... 34
6. Ciu-ves of the leaf spot history series, showing the production of conidia
on different dates from June 24 to September 19, 1913, at Rock}'
Ford, Colo 48
7. Curves of the maximum and minimum temperattues and humidities,
the number of hours that the humidity remained above 60 from
noon of the preceding to noon of the given day among the plants,
and rainfall and irrigation records, taken in a medium-early sugar-
beet field from June 10 to September 22, 1913, at Rocky Ford, Colo. . 51
8. Curves of the comparative production of conidia on the upper and
lower surfaces of the leaf spots, representing series E, K, N, and
G of Table V and figure 6. Rocky Ford, Colo., 1913 52
9. Curves of the 2 -day average increases in the number of leaf spots per
plant in a. medium-early and an early sugar-beet field, from June
18 to September 19, 1913, at Rocky Ford, Colo 54
10. Curves of the maximum and minimum temperatiu-es and relative
humidities and the number of hours that the humidity remained
above 60 from noon of the preceding to noon of the given day among
the sugar-beet plants of a medium-early and an early field. August
2 to September 21, 1913, at Rocky Ford, Colo 58
XX Journal of Agricultural Research voi. vi
Citrus Canker
Page
Fig. I. Diagrammatic representation of young open type of Citrus canker of
half the diameter of the one shown in figure 2. pp, Palisade paren-
chyma; ue, upper epidermis; le, lower epidermis; d, diseased tis-
sues; a, air space arising from tensions due to the enlargement of
cells and disintegration of tissues 72
2. Diagrammatic representation of canker on old Citrus leaf; pp, Palisade
parenchyma; ue, upper epidermis; le, lower epidermis; p, pycnidium
of Phonia socia; d, diseased tissues; a, air space arising from tensions
due to the enlargement of cells and disintegration of tissues 73
Pseudomonas citri: a. Stained with carbol fuchsin; b, stained with
Williams's flagellar stain (adapted from Hasse); c, stained with
anilin gentian violet 76
4. Early stage of Citrus canker in cross section on a young leaf of seedling
grapefruit 80
5. Pseudomonas citri: (a). In the mesophyll tissue and (6) in the palisade
parenchyma 81
6. Drawing of a stained section of a natural canker on grapefruit 82
7. Cross section in outline of a spongy canker on the rind of a fruit of
Citrus decumana, showing ruptured epidermis and hypertrophy of
the rind tissues, the cells of which are loosely attached 83
8. a. Cross section of a pycnidium of Phoma socia from a grapefruit leaf.
b, Germination of conidia of Phoma socia after 24 hours in water, c,
Mycelium of this fungus in old cultures 86
Determination of Stearic Acid in Butter Fat
Fig. I. Exterior of constant-temperature crj'^stallization tank 104
2. Interior of constant-temperature crystallization tank 105
LiPE History and Habits of Two New Nematodes Parasitic on Insects
Fig. I. Diplogaster labiata: A, Mating; B, mature female reared in water cul-
ture, a, lip region, 6, esophagus, c, median bulb, d, cardiac bulb,
c, intestine,/, ovaries, g, egg, h, genital pore, i, rectum, k, anus;
C mature male reared in water culture, a, lip region, b, esophagus,
c, median bulb, d, cardiac bulb, e, intestine, k, anus, m,, spicula;
D, at time of hatching; E, female during process of molting; F, dead
female with yoimg nematode which hatched within her body 116
2. A~H, Diplogaster labiata: Developmertt of the egg; /, Diplogaster
aerivora: mature male reared in moist soil; /, Diplogaster aerivora:
mature male reared in water culture, a, lip region, b, esophagus, c,
median bulb, d, cardiac bulb, e, intestine, k, anus, m, spicula; K,
Diplogaster aerivora: dead female with yotmg which hatched within
her body; L, Diplogaster aerivora: mating 117
3. Diplogaster aerivora: A, Form found in termite; B, at time of hatch-
ing; C, female reared in water culture, not quite mature, a, lip
region, b, esophagus, c, median bulb, d, cardiac bulb, e, intestine,
/, ovaries, h, genital pore, i, rectum, k, anus; D, mature female
reared in moist soil; E, mature female reared in water culture;
F-M, development of the egg 122
Effect of Pasteurization on Mold Spores
Fig. I. Curi'e of the number of species of molds siu^iving pasteurization of
milk for 30 minutes at a series of temperatures 157
2. Curve of the number of species of molds surviving flash pasteurization
at a series of temperatiu^es 160
Apr. 3-sept. 2s. 1916 Illustrations xxi
Page
Fig. 3. Curve of the number of species of molds sxu^iving dry heat for 30 min-
utes at a series of temperatures 164
Effect of Water in the Ration on the Composition of Milk
Fig. I. Grouping of cow and kind of ration fed cows 23, 24, 25, and 27 173
Hyperaspis binotata, a Predatory Enemy of the Terrapin Scale
Fig. I. Map showing the distribution in the United States of Hyperaspis
binotata igg
Tests op Three Large-Sized Reinforced-Concrete Slabs under Con-
centrated Loading
Fig. 1. Diagram illustrating the method of obtaining "effective width" in
reinforced-concrete slab tests 207
2. Diagram showing location of strain -gauge points on top of slab 835 ... . 211
3. Diagram showing location of strain-gauge points on top of slab 930. ... 212
4. Diagram showing location of strain-gauge points on top and bottom of
slab 934 213
5. Concrete deformation curves for concentrated center load on slab 835. . 214
6. Concrete deformation curves for slab 930 215
7. Deformation cxu^'es for slab 934 216
8. Deformation ctirves for slab 934, computed from first zero reading. ... 216
9. Concrete deformation curves for slab 835 with 2 -point loading 217
10. Concrete deformation ciu-ves for slab 934 with 2-point loading 217
11. Steel deformation curves for slab 934 with 2-point loading 218
12. Concrete deformation curves for slab 835 with 4-poLnt loading 219
13. Concrete deformation ctu-ves for slab 934 with 4-point loading 219
14. Deflection curves for slab 934 on first application of load 220
15. Deflection curves for slab 934 on second application of load 221
16. Deflection curves for slab 934 with 2-point loading 222
17. Diagram showing effect of breaking load on slab 835 223
18. Diagram showing effect of breaking load on slab 930 223
19. Diagram showing effect of breaking load on slab 934 224
20. Concrete deformation ctu^j^es for a 10,000-pound concentrated center
load on slab 934. Variation in deformations plotted parallel to
supports 225
21. Concrete deformation curves for a io,ooo-potmd concentrated center
load on slab 934. Variation in deformations plotted perpendicular
to supports 226
22. Concrete deformation curves for slab 934. Lateral deformations
plotted both parallel and perpendicular to supports 227
23. Iso-deformation lines for slab 934 imder concentrated center load .... 228
24. Concrete deformation ciu-ves for slab 934 under 40,000-pound 4-point
loading. Deformations measured perpendicular to supports. Vari-
ation of deformations plotted parallel to the supports 229
25. Concrete deformation etudes for slab 934 under 40,000-pound 4-point
loading. Deformations measured perpendicular to supports. Vari-
ation of deformations plotted perpendicular to supports 230
26. Iso-deformation lines for slab 934 xmder 40,000-pound 4-point loading.
Deformations measured perpendicular to supports 231
27. Ctu*ves showing effective width versus load (concentrated center load) . 231
28. Ciu-ve showing effective width versus thickness 232
XXII Journal of Agricultural Research voi. vi
Effect of Climatic Factors on the Hydrocyanic-Acid Content of
Sorghum
Page
Fig. I. Curves showing the hydrocyanic-acid content of sorghum on the various
plots 264
2. Curves showing the rate of growth of the sorghum on the various plots. . 265
3. Curves showing the precipitation, temperatm-e, and humidity' relations
at the various experiment stations during the growing season of 1915 . . 266
4. Curves showing the contemporary climatic conditions at the various
plots, together with crop data and hydrocyanic-acid content 267
Hypodbrma deformans, an Undescribed Needle Fungus of the
Western Yellow Pine
Fig. I. A side view of two apothecia of Hypoderma deformans on needles of
Pinus ponderosa, showing the longitudinal medial split 278
2. Asci, spores, and paraphyses of Hypoderma deformans 279
3. Cross section of an apothecium of Hypoderma deformans on a needle of
Pinus ponderosa, showing mature asci with spores, the point of first
rupture, and the tissues of the leaf most seriously affected by the
mycelium of the fungus 280
4. The upper portion of a young ascus of Hypoderma deforynans, showing
the formation of the pore at the tip through which the spores are
expelled 281
CojiparativE Study of the Root Systems and Leaf Areas op Corn
and the Sorghums
Fig. I. Evaporation from a free water surface (tank) at Garden Cit>% Kans.,
diuing the growing .seasons of 1914 and 1915 314
2. Comparison of the leaf areas of Pride of Saline com. Black-hull kafir,
and Dwarf milo at four stages of the growth of these plants diuing
the season of 1914. 328
3. A graphic illustration of the sheath areas of Pride of Saline com, Black-
hull kafir, and Dwarf milo at four stages of the growth of these plajits
during the season of 1914 339
Woolly Pear Aphis
Fig. I. Comparative structure of antennae and wax pores of Eriosoma spp.:
A , distal segments of antenna of winged viviparous female of E. pyri-
cola; B, distal segments of antenna of winged viviparous female of
E. ulmi; C, distal segments of antenna of wingless viviparous female
of E. americanum; D, distal segments of an tenna of wingless vivip-
arous female of E. lanigerum; E, distal segments of antenna of
wingless viviparous female of E. pyricola; F, distal segments of
antenna of winged viviparous female of E. atnericanum; G, distal
segments of antenna of winged viviparous female of E. lanigerum;
H, compound wax pore of E. lanigerum; I, compound wax pore of
E. pyricola; J, distal segments of antenna of first instar wingless
viviparous female of E. pyricola 359
Stimltlating Influence of Arsenic t;pon the Nitrogen-Fixing Organ-
isms op the Soil
Fig. I. Graph showing the action of five compounds of arsenic on nitrogen
fixation in dry soil 392
2. Graph showing the effect of aeration on the nitrogen-fixing activit}''
of soil containing compounds of arsenic 408
Apr. 3-sept. 25, 1916 Illustrations xxiii
Page
Fig. 3. Graph showing the effect of heat on the nitrogen-fixing power of soil
treated and not treated with arsenic 409
4. Effect of various arsenic compounds in the ratio of 400 parts of the
compound to r, 000,000 parts of soil on the activity of various soil
organisms 411
5. Graph showing parts per million of various arsenic compounds in the
soil at which the greatest stimulation occtu-red 412
Transmission and Controi, o? Bacterial Wilt op Cucurbits
Fig. I. Comparison of the amount of wilt with striped-beetle prevalence and
with meteorological phenomena in three fields, East Marion, Long
Island, N. Y., season of 1915 421
2. Comparison of relative wilt control of Bordeaux mixture plus lead
arsenate, Bordeaux mixture alone, and lead arsenate alone in field i.
East Marion, Long Island, season of 1915 430
3. Curves showing relative wilt control of Bordeaux mixture and lead
arsenate with date of first application as a variant in field i, East
Marion, Long Island, season of 1915 431
Aleyrodidae, or White Flies, Attacking the Orange, with Descrip-
tion OP Three New Species of Economic Importance
Fig. I. Aleurocanthus citriperdus: A, Pupa case; B, egg; C, polygonal mark-
ings of egg; D, vasiform orifice of pupa case; E, spine from dorsum
of pupa case; F, margin of pupa case; G, genitalia of adult male;
H, forewing of male; /, antenna of pupa case; /, leg of pupa case;
K, L, marginal teeth, much enlarged; M, central swollen spine from
dorsal area 460
2. Aleurocanthus woglumi: A, Egg; B, polygonal markings of egg; C, pupa
case; D, margin of pupa case; E, vasiform orifice of pupa case;
F, forewing of adult female; G, same, showing variation in markings;
H, costal margin at base of wing of female; /, forewing of male;
/, male genitalia 462
3. Aleurothrixus porteri, A. howardi, and A. floccosus: A, Akurothrixus
porteri: Larva, first instar. B, A. porteri: Caudal spine of pupa
case. C, A. porteri: Clasper of male. D, A. porteri: Egg. E, A.
howardi: Caudal spine. F, A. porteri: P*upa case. G, A. porteri:
Forewing of adult. H, A. floccosus: Vasiform orifice of pupa case.
7, A . porteri: Vasiform orifice of pupa case. /, A . houardi: Vasiform
orifice of pupa case. K, A. porteri: Margin of pupa case. L, A.
Porteri: Margin of early larva 467
Relative Water Requirement of Corn and the Sorghums
Fig. I. Curves of the evaporation at Garden City, Kans., for the growing period
of 1915 474
California Green Lacewing Fly
Fig. 1. The California green lacewing fly: Adult 515
2. Chrysopa californica: Eggs 518
3. Chrysopa californica: First instar 519
4. Chrysopa californica: Third instar 521
5. Chrysopa californica: Pupal case 522
6. Chrysopa californica: Pupa 522
7. Chrysopa californica: Pupa freshly emerged from its cocoon 523
XXIV Journal o} Agricultural Research voi. vi
Storage-Rots op Economic Aroids
Page
Fig. I. Spores of different storage-rot organisms: A, Diplodia tubericola from
dasheen; B, Diplodia tubericola from sweet potato; C, Diplodia
gossypina from cotton; D, Diplodia sp. from Mangifera indica:
E, Diplodia maclurae from Toxylon pomiferum; F, Fusarium solani. . 552
Larval Characters and Distribution op Two Species op Diatraea
Fig. I. a. Average angle formed by imaginary lines through bases of setae of
Diatraea saccharalis crambidoides; b, average angle formed by imagi-
nary lines through bases of setae of D. zeacolella 622
The Disease op Potatoes Known as "Leak"
Fig. I. Microscopical appearance of Pythium debaryanum isolated from potatoes
affected with potato leak: a, Cell of a potato tuber showing fungus
filaments therein; b, oogonia and antheridia; c, mycelium; d, ger-
minating conidia 630
Life Cycles op the Bacteria
Fig. I. IMg. cycle oi Bacillus azotobacter. The broken straight lines divide the
different types of growth indicated by the letters A to M. The
Greek letters a to X refer to subdivisions. The single and double
pointed arrows indicate the development of one form from another.
The four circles confine, in every case, all those forms which repre-
sent together a rather constant mode of life and which have been
usually considered as bases for establishing separate species 678
Mottle-leap op Citrus Trees in Relation to Soil Conditions
Fig. I. Graphical presentation of the relationship between humus content of
soil and percentage of mottled orange leaves 731
2. Graphical presentation of the relationship between the ratio of organic
carbon to humus in the soil and the percentage of mottled orange
leaves 734
3. Graphical presentation of the relationship between the ratio of humus
to lime in the soil and the percentage of mottled leaves 735
4. Graphical presentation of the relationship between the mineral car-
bonates in the soil and the percentage of mottled lemon leaves 737
Agricultural Value of Impermeable Seeds
Fig. I. Curves showing the rate of softening of impermeable red-clover seeds
of different degrees of maturity 767
2 . Curves showing the rate of softening and of germination of impermeable
red-clover seeds of different lots 768
3. Curves showing the changes in the permeability of seeds in wet blotters
and in dr>- storage for various periods 773
4. Curves showing the rate of softening of impermeable red-clover seeds
under different temperatiu"e conditions 780
5. Curves of the seedling production in the field in 16 to 18 days and of
the germination in chamber in 8 days 787
6. Ciurves of the seedling production in the field and of the germination
in chamber in i2>^ months 790
Apr. 3-Sept. 25, 1916
Illustrations xxv
Lepe-History Studies op Cirphis vnjpuncta, the True Army Worm
Page
Fig. I. Diagram of relative amounts of foliage eaten in each larval instar by
Cirphis unipuncta 801
2. Posterior extremity of male and female pupa: c, Male ; h, female 812
Thersilochus conotracheli, a Parasite of the Plum Curculio
Fig. I. Tliersilochus conotraclieli: Cocoon 852
2. Tliersilochtis conotracheli: Egg 852
3. Tliersilochus conotracheli: Larva of first instar 852
4. Thersilochus conotracheli: Ventral siu^ace of head of first larval instar:
a, Labrum; b, maxilla; c, labium; d, mandible 853
5. Thersilochzis conotracheli: Full-grown larva of first instar 853
6. Thersilochus conotracheli: Larva of second instar 853
7. Tliersilochus conotracheli: Full-grown larva of fifth instar 854
8. Thersilochus conotracheli: Face of full-grown larva: a, Labrum; b, man-
dible; c, maxilla; d, maxillary palpus; e, labium; /, labial palpus. . . 854
g. Thersilochus conotracheli: Pupa of female, and apex of abdomen of male
pupa 854
Effect on Plant Growth of Sodium Salts in the Sou,
Fig. I. Diagram of the percentage of sodium carbonate added to the soil in
experiment i (1913), with the percentage of carbonate and bicar-
bonate recovered and the total green weight of wheat obtained 860
2. Diagram of the percentage of sodium carbonate added to the soil in
experiment 2 (i9i4\ with the percentage of carbonate and bicar-
bonate recovered and the green weight of wheat obtained 862
3. Diagram of the percentage of sodium carbonate added to the loam soil
in experiment 3 (1915), with the percentage of carbonate and bicar-
bonate recovered and green weight of wheat 864
4. Diagram of the percentage of sodium carbonate added to Monterey sand
in experiment 3 (19 15), with the percentage of carbonate and bicar-
bonate recovered and the total green weight of wheat obtained 864
5. Diagram of the percentage of sodium bicarbonate added to the soil in
experiment 4 (1914), with carbonate and bicarbonate recovered,
together with the total green weight of wheat obtained 865
6. Diagram of the decrease in growth of wheat seedlings in experiments 2
and 4 as affected by the total carbonate salts recoverable from the soil . 866
7. Diagram of the quantity of sodium chlorid added to the soil, with the
quantity of chlorid recovered, and the total green weight of wheat
obtained 867
8. Diagram of the quantity of sodium sulphate added to the soil in experi-
ment 6, the quantity recovered, and the total green weight of wheat
obtained 868
Aphidoletes meridionalis, an Important Dipterous Enemy of Aphids
Fig. I. Aphidoletes meridionalis: Eggs in situ on leaf of rape; a, egg 884
2. Aphidoletes meridionalis: Larva, dorsal view 886
3. Aphidoletes meridionalis: a, Cocoon formed on surface of ground; b,
cocoon formed on a leaf blade 886
4. Aphidoletes meridionalis: Pupa, lateral view 887
XXVI Journal of Agricultural Research voi.vi
Influence of Barnyard Manure and Water upon the Bacterial Activ-
ities OF THE Soil
Page
Fig. I. Curves of the ammonifying powers of soil in pots with varying quanti-
ties of manure and water 899
2. Curves of the nitrifying powers of soil in pots with varying quanti-
ties of manure and water 900
3. Ctu^cs of the number of colonies of bacteria developing from fallow soil
with varying quantities of manure and water 904
4. Curves of the ammonifying powers of fallow soil with varying quanti-
ties of manure and water 907
5. Ctu^es of the nitrifying powers of fallow soil with varying quantities of
manure and water 910
6. Curves of the number of colonies of bacteria developing from cropped
plats with varying quantities of manure 912
7. Curves of the ammonifying powers of soil of cropped plats with varying
quantities of manure and water 915
8. Curves of the nitrifying powers of soil of cropped plats with varying
quantities of manure and water 918
9. Diagram of the influence of manure on the yield and bacterial activities
of a soil 920
ID. Diagram of the influence of irrigation water on the yield and bacterial
activities of a soil 921
Progressive Oxidation of Cold-Storage Butter
Fig. 1. Diagram of gas apparatus used in the extraction and analysis of the air
confined in butter 931
Bacteriological Studies of a Soil Subjected to Different Systems of
Cropping for 25 Years
Fig . I . Curves of the nitrate formation in soil of fertility plats at Columbia, Mo. ,
in 1913-14, after 28 days' incubation with the addition of 60 mgm. of
nitrogen as cottonseed meal, but without the addition of calcium car-
bonate 962
Studies on the Physiology of Reproduction in the Domestic Fowl. —
XV. Dwarf Eggs
Fig. I. Diagram showing the percentage of the yearly total egg production
(8-year average, 1899-1907) and the percentage of the total dwarf-egg
production (8-year average, 1908-1916) which occurred during each
month 1004
2. Diagram showing for the years 1911-12 and 1914-15 combined the per-
centage of the yearly total egg production and dwarf-egg production
which occurred during each month and 100 times the percentage of
the eggs produced each month which were dwarf 1005
3. Diagram showing the number of dwarf eggs which occurred in each
tenth of a litter loii
4. Diagram showing the number of dwarf eggs which occurred in each fifth
of a litter 1014
Vol. VI APRIL 3, 1916 No. 1
JOURNAL OF
AGRICULTURAlv
RESEARCH
CONTENTS
Page
Relation of Carbon Bisulphid to Soil Organisms and Plant
Growth . . . , . . , ... 1
E. B. FRED
Climatic Conditions as Related to Cercospora beticola . .21
VENUS W. POOL and M. B. McKAY
DEPARTMENT OF AGRICULTUEE
WAS HINGTON , D.G
WA8HIN0TON : 0OVEHNMENT PRINTINO OFFICE ; 1916
PUBI.ISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KEIvI/ERMAN, Chairman RAYMOND PEARL
Physiologist and Assisian/ Chief, Hnreaii
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Assistant Ctiief, Bureau of Entomology
Siologist, Maine Agriciiltt'.ra! Experinient
Station
H. P. ARMSBY
Directdrr, Instiiuie of Animal Nutrition, The
Pennsylvania State Cot'i'i,'
M. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
All correspondence regarding articles from the Department of Agriculture
should be addressed to Karl F. Kellerinan, Journal of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Ra\Tnond Pearl, Journal of Agricultural Research, Orono, Maine.
JOim OF AGRiaJLTlAL MARCH
DEPARTMENT OF AGRICULTURE
Voiv. VI Washington, D. C, April 3, 1916 No. i
RELATION OF CARBON BISULPHID TO SOIL ORGAN-
ISMS AND PLANT GROWTH^ .
By E. B. Fred,
Agricultural Bacteriologist, Agricultural Experiment Station
of the University of Wisconsin
INTRODUCTION
In a previous publication concerning the action of carbon bisulphid
(CS2) on bacteria and plants data were presented to show the beneficial
effect of this substance on the soil flora (i)." The increased plant growth
following the addition of carbon bisulphid in many cases is enormous.
For example, a small application often causes an increase in yield from
100 to 200 per cent. It is impossible to account for this remarkable gain
on the assumption that the only action of the carbon bisulphid is that of
added plant food. It was found, as has been noted by many investi-
gators (5, 6, II, 12), that this volatile antiseptic exerts a very decided
effect on the micro-organisms of the soil. As measured by plate counts,
there is at first usually a great decrease in numbers, followed by a period
of excessive increase, the total numbers far exceeding those that ordinarily
exist. In certain cases carbon bisulphid has not only failed to cause an
increase in plant growth, but has, on the contrary, caused a decrease.
Search has been made by many investigators for a satisfactory ex-
planation of this peculiar action of carbon bisulphid. Many theories
have been advanced. Concerning these theories so much has been
written that a detailed discussion of the literature seems unnecessary.
Indeed, it would be impossible within the limited scope of this paper to
present a summary of the various explanations. One point is very
prominent in nearly all of the publications : The action of carbon
bisulphid is varied. Because of the interest attached to this problem,
it was arranged to study some of the factors that might influence
the action of carbon bisulphid. The experiments described in this paper
are discussed under three main heads : First, the effect of varying
amounts of carbon bisulphid; second, the effect of carbon bisulphid on
various plants; and third, the effect of carbon bisulphid in various soils.
In all of this work fresh field soil and commercial carbon bisulphid
were used. vSome of the experiments represent a combined study of the
effect on both the lower and higher forms of plant life.
1 Published with permission of the Director of the Wisconsin Agricultural Experiment Station.
* Reference is made by number to "I,iterature cited," p. i8-ig.
Journal of Agricultural Research, Vol. VI, No. i
Dept. of Agriculture, Washington, D. C. Apr. 3, 1916
cu (l) * Wis.— 5
Journal of Agricultural Research
Vol. VI. No. 1
EXPERIMENTAL METHODS
Commercial carbon bisulphid was poured into small holes in the soil,
and these were covered immediately. The soil Vv'as sieved and potted in
2-gallon jars and the moisture maintained at half saturation. Changes in
the soil flora were determined at regular intervals by plate counts of the
number of bacteria and dilution counts of the number of active protozoa.
The formation.of ammonia and nitrates was measured at regular intervals.
The following plants were used: Buckwheat {Fagopyrum fagopyrum),
clover (Trifolium pratense), corn (Zeamays), mustard (Sinapis alba), oats
(Avena sativa), and rape {Brassica napus). In many of the experiments
a first and a second crop were grown.
EFFECT OF CARBON BISULPHID ON THE NUMBER AND ACTIVITY OF
SOIL ORGANISMS
Eight jars were filled with Miami silt-loam soil from the Experiment
Station farm. These were arranged in duplicate and treated as follows:
(i) Control, untreated; (2) 2 per cent of carbon bisulphid; (3) 2 per cent
of carbon bisulphid, evaporated; (4) 2 per cent of carbon bisulphid,
evaporated, and reinoculated with 5 per cent of the original soil.
Twenty-four hours after treatment the soil in the evaporated series
was spread out on sterile paper and the volatile antiseptic allowed to
escape. At the end of the second 24-hour period the soil was put back
into the jars. In order to prevent any contamination, the jars were
covered with a double layer of cheesecloth and nonabsorbent cotton.
This cover should allow free access of air without much danger of con-
tamination. At regular intervals the covers were removed and samples
drawn for analysis. The results of these determinations are presented
in Tables I and II.
NUMBER OF ORGANISMS
Bacteria. — In Table I are shown the number of bacteria in i gm. of
soil at different times and under the different conditions.
TabIvE I. — Effect of carbon bisulphid on number of bacteria
Time.
Bacteria per gram of dry soil.
Control.
2 per cent of
carbon bisul-
phid.
2 per cent of
carbon bisul-
phid evapo-
rated.
2 per cent of
carbon bisul-
phid evapo-
rated + 5 per
cent of soil
from control.
Days
I
3
5
9
13
21
25
29
60
II, 496, 000
22, 010, 000
20, 635, 000
14, 739. 000
16, 115, 000
19, 508, 000
18, 272, 000
15.346,000
12, 372, 000
I, 965, 000
23.975.000
25.253,000
36, 651, 000
90. 473. 000
60, 149, 000
68, 276, 000
90, 645, 000
58, loi, 000
2, 260, 000
8, 254, 000
27, 416, 000
61, 904, 000
98, 850, 000
71, 257, 000
86, 483, 000
84, 272, 000
60, 000, 000
2,358,000
12, 480, 000
95. 499. 000
80, 420, 000
52, 495, 000
64, 570, 000
38, 495, 000
30, 000, 000
Apr. 3, i9i6 Relation of Carbon Bisulphid to Plant Growth 3
At first the antiseptic causes a great reduction in the number of organ-
isms capable of developing on Heyden agar. The period of depression
lasts for only a short time — in this experiment about five days. From
that time until the end of the test the number of organisms in the treated
series far exceeded that of the control. The highest number in the
carbon bisulphid evaporated and unevaporated soil occurred about the
thirteenth day; while the carbon bisulphid evaporated soil plus control
soil gave the highest count on the fifth day. At the time of the last
count, 60 days after carbon bisulphid was added, the organisms in the
treated series far exceeded those in the original soil. Apparently the
effect of carbon bisulphid on the number of bacteria is noticeable for a
long period of time.
If the results of the counts with carbon bisulphid unevaporated are
compared with those of carbon bisulphid evaporated, it appears that no
very marked difference exists. The greatest reduction in numbers
occurred in soils with the carbon bisulphid evaporated. It is significant
that soil with carbon bisulphid evaporated should prove more injurious
to micro-organisms than the unevaporated. This agrees with Gainey (2,
p. 592), who reports that the combined effect of the two processes seemed
more injurious to nitrification than treatment with carbon bisulphid
unevaporated.
After the thirteenth day the treated and reinoculated soil did not
show as many organisms as the treated series. This difference is shown
very distinctly in Plate I, which is reproduced from a photograph of
a number of colonies developing on agar. Four parallel plates were
made from the same dilution of each soil.
On this date samples were also drawn for ammonification tests. The
purpose of this was to measure the rate of the decomposition of casein
in the various series, and i per cent of casein was added to the soil and
the ammonia determined after 12 and 24 hours. The beneficial effect
of carbon bisulphid on ammonification is very evident. If after 12 hours
the untreated is 100, then carbon bisulphid unevaporated is 154, carbon
bisulphid evaporated is 212, and carbon bisulphid reinoculated is 190.
After 24 hours the untreated is equal to 100, carbon bisulphid un-
evaporated is 149, carbon bisulphid evaporated is 171, and carbon
bisulphid reinoculated is 153. The data show very clearly that casein
is decomposed more rapidly in treated than in untreated soils. This
difference is most prominent in the 12 -hour tests.
Protozoa. — Counts at the beginning showed the presence of protozoa
in dilutions representing i to 1,000 gm. of soil (13, p. 626). Two weeks
after treatment the soils were recounted. At this time numerous small
flagellates were found in dilutions of i to 1,000. It is evident that the
different treatments with carbon bisulphid had not seriously injured this
group of organisms.
Journal of Agricultural Research
Vol. VI. No. I
AzoTOBACTER. — One month after treatment with carbon bisulphid,
qualitative tests were made. The Azotobacter organisms were found in
all soils. The brown film of Azotobacter from the treated soils was not
so profuse as that from the original soil.
ALiGM. — In order to estimate the number of algae, dilution tests were
made. These cultures were incubated for 30 days. The smaller forms
were found in great numbers in all of the soils.
The important facts in these data are (i) that the volatile antiseptic
fails to remove these larger soil organisms and (2) that the smaller forms
of bacteria are only temporarily reduced. The decrease in numbers is
soon followed by a period of excessive growth.
ACTIVITY OF ORGANISMS
A rapid multiplication of bacteria should naturally be followed by a
parallel increase in decomposition products. Accordingly samples for
analysis were drawn from the jars used in the previous experiment.
The results of these periodic analyses are presented in Table II.
Table II. — Effect of carbon bisulphid on ammonia and nitrate content of soil
Time.
Nitrogen per loo jjm. of dry soil.
Ammonia.
Control.
2 per cent
of carbon
bisulphid.
2 per cent
of carbon
bisulphid
evapo-
rated.
2 per cent
of carbon
bisulphid
evapo-
rated+5
per cent of
soil from
control.
Control.
2 per cent
of carbon
bisulphid.
2 per cent
of carbon
bisulphid
evapo-
rated.
2 per cent
of carbon
bisulphid
evapo-
rated+s
per cent of
soil from
control.
Days.
At beginning
30
45
60
75
90
Mgm.
60
68
38
59
85
94
Mgm.
I. 60
5-27
8.40
5-43
5.60
4. 06
Mgm.
I. 60
4. 06
Mgm.
1. 60
4.71
4.90
2.31
2. 10
2. 24
Mgm,.
2.66
4. 00
Mgm.
2.66
2. 50
2. 70
2.81
2. 40
5.00
Mgm.
2.66
2. 00
2.50
2.50
2. 60
3-32
Mgm.
2.66
55
66
5°
5. 00
6.66
In the soils treated with carbon bisulphid there is a very decided accumu-
lation of ammonia nitrogen. If the figures of Table I are compared
with those of Table II, ammonia production, it will be seen that an
increase in the number of bacteria within a certain range results in a
gain in ammonia. After 30 days the amount of ammonia nitrogen in
the treated soils averaged more than three times that in the original soil.
After 60 days the ammonia content in the carbon bisulphid and car-
bon bisulphid evaporated soil was about double that of the control,
while in the carbon bisulphid evaporated plus 5 per cent fresh soil it was
Apr. 3, 1916
Relation of Carbon Bisulphid to Plant Growth
less. From the data it appears that rcinoculation prevents large accu-
mulations of ammonia. This is no doubt due to the oxidation of ammo-
nia by the nitrifying bacteria. The figures of the last column (nitrate
accumulation) support this statement. A stimulation of ammonification
is still noticeable at the end of 3 months.
The nitrate-forming bacteria apparently do not recover so rapidly
from carbon bisulphid treatment as the ammonia-producing organisms;
consequently, there is no increase in nitrates until the end of 3
months. An exception to this is noted in the reinoculated soil. Here
the activity of the nitrifying bacteria is evident 30 days after inoculation.
In order to ascertain, as nearly as possible, the effect of carbon bisul-
phid on the soluble nitrogen of the soil, the figures of Table II, ammo-
nia and nitrate nitrogen, -were combined in Table III.
Table III. — Effect of carbon bisulphid on soluble nitrogen
Time.
Ammonia and nitrate nitrogen per loo gm. of dry soil.
Control.
Days,
At beginning
30
45
60
75
90
Mgm,
4. 26
5-03
5-13
6-59
5-05
6.94
2 per cent
carbon
bistilphid.
Mgm.
4. 26
8.47
II. 10
8.24
8.00
9. 06
2 per cent
carbon
bisulphid
evai>orated.
Mgm.
4. 26
7-41
10. 20
7.82
7-38
3 per cent
carbon
bisulphid
evaporated
+ S per cent
of soil from
control.
\Igm.
4. 26
10. 26
10. 56
6.87
7. 10
8.90
From the data in this table it is very evident that carbon bisulphid
causes a large increase in ammonia and nitrate nitrogen. There seems
to be very little difference between the effect of the various treatments of
carbon bisulphid on the formation of ammonia and nitrate nitrogen.
When compared with the control soil, it will be seen that 45 days after
treatment the carbon-bisulphid soils contain more than twice as much
soluble nitrogen. The higher ammonia and nitrate content is very
marked 90 days after treatment. A repetition of this experiment gave
similar results.
A review of the data in Tables II and III shows very clearly that carbon
bisulphid in Miami soil increases the total soluble nitrogen — namely,
ammonia and nitrates. One interesting fact that appears from a com-
parison of the ammonia and nitrate content is that these two substances
are to a certain degree inversely proportional.
Journal of Agricultural Research
Vol. VI, No. I
EFFECT OF CARBON BISULFHID ON THE HIGHER AND LOWER FORMS
OF PLANT LIFE
From the results of the preceding experiments it seems that carbon
bisulphid should exert a beneficial effect on the growth of higher plants.
At first this should be most marked with ammonia-feeding plants, and
later with nitrate-feeding plants. Unfortunately it is not possible to
secure plants that feed entirely on nitrates or ammonia. For this reason
it was thought best to study the relation of carbon bisulphid to the
growth of several different plants. Accordingly a combination study of
the effect of carbon bisulphid on higher plants and on bacteria was made.
A wide range of soil types, as well as different higher plants, was used.
Before entering upon a study of the relation of carbon bisulphid to
soil type and various plants, it was desired to obtain some idea of the
influence of various amounts of carbon bisulphid on plant growth. The
procedure was as follows : Ten kgm. of field soil (Miami silt loam) were
placed in each of sixteen 2 -gallon jars. The carbon bisulphid was added
in varying amounts, from 0.5 per cent to 2 per cent. It was poured into
holes in the soil. These holes were closed immediately and the water
increased to half saturation. In order to overcome the injurious effect
of carbon bisulphid, the jars were then allowed to stand for two weeks
before planting.
CORN AND MUSTARD IN MIAMI SII^T LOAM
The results of the test with corn and mustard are given in Table IV.
It is evident from the data of the table that these plants do not re-
spond alike to carbon bisulphid.
Table IV. — Effect of varying amounts of carbon bisulphid on the growth of corn and
mustard
Soil.
Carbon
bisulphid
added.
Weight of corn.
Weight of mustard.
No.
Green.
Dry.
Average.
Green.
Dry.
Average.
I
Miami.
. ..do. ..
Per cent.
Control.
Control.
0-5
•5
I
2
2
Gm.
75
80
82
83
132
22
85
125
Gm.
20
25
Gm.
} 22.5
Gm.
I 49
{ ?,
{ n
/ 105
I 112
Gm.
9-5
9
13
12
13
12
16
17
Gm.
} 9-25
2
■}
. . .do. ..
f. } -5
} 12. 50
A
...do...
C
...do...
} 12. 50
6
...do...
7
...do...
} 16.50
8
...do...
In all concentrations except 2 per cent, carbon bisulphid injured the
growth of corn. Mustard, on the other hand, was greatly benefited by
the carbon-bisulphid treatment. An increased growth was observed
from all concentrations. The maximum gain was noted with 2 per cent
of carbon bisulphid. This beneficial effect on mustard is very evident
from Plate II, figure i. If this increase in growth is due to the larger
Apr. 3. 1916 Relation of Carbon Bisulphid to Plant Growth 7
amount of soluble nitrogen as ammonia or nitrate, then corn and mustard
should behave much alike. The nitrogen-feeding power of these plants
has been studied by Kriiger (8), Gerlach and Vogel (3), and others. It
is supposed that both corn and mustard are heavy nitrogen-feeding
crops, able to take nitrogen either in the form of ammonia or nitrate.
BUCKWHEAT, CORN, AND OATS IN MIAMI SII.T LOAM
In order to decrease the factor of individual variation, four parallel
jars of Miami silt loam were used in each series in the following experi-
ment. For the second crop these were subdivided into sets of two
each. After the first crop was harvested, the soil and roots were thor-
oughly mixed and the jars replanted. The rotation was as follows:
First crop, buckwheat; second crops, corn and mustard; first crop, com;
second crop, buckwheat; first crop, oats; second crops, com and mus-
tard. In Tables V, VI, and VII are presented the results of these
experiments.
TabIyE V. — Effect of carbon bisulphid on the growth of buckwheat and corn
No.
SoU.
1
Miami
2
. . .do. .
5
. . .do. .
4
. . .do. .
e
. . .do. .
6
...do..
7
...do..
8
...do..
Carbon
bisulphid
added.
Per cent.
Control.
Control.
Control.
Control.
W eight of first crop, buckwheat. Weight of second crop , com
Gm.
90
97
121
126
124
145
127
126
Dry.
Average. Green
Gm.
19
24- 5
Gift,
152
160
169
136
Dry.
Gm.
28. 5
33-5
34
31-5
Average.
Gm.
32.7
The yields of buckwheat and corn are given in Table V. The weights
of the mustard were lost. Buckwheat gave an increase in the treated
soil, while corn (the second crop) did not show any improvement. Deter-
minations of ammonia present at the time the buckwheat was cut (three
months after treatment) resulted as follows: Ammonia — if control is 100,
then carbon bisulphid treated is 192. Nitrate — if control is 100, then
carbon bisulphid treated is 28. The antiseptic increases ammonia, but
decreases the nitrate content of soil. The results of investigation show
that buckwheat feeds largely on nitrate nitrogen (9), while com is sup-
posed to be able to take its nitrogen in the form of ammonia. A difference
in nitrogen-feeding power can not be used to explain the unequal behavior
of these plants toward carbon bisulphid. Although the weights of the
mustard crop were not kept, the action of the carbon bisulphid was
evident. There was a decided gain in the growth of plants in the treated
series.
From the data of Table VI it is obvious that carbon bisulphid has very
little effect on corn (first crop) or buckwheat (second crop).
Journal of Agricultural Research
Vol. VI, No. I
Table VI. — Effect of carbon bisulphid on the growth of corn and buckwheat
No.
I
Miami
2
. . .do. .
■}
...do..
A ...
. . .do. .
C
. . .do. .
6
...do..
7
. . .do. .
8
...do..
.Soil.
Carbon
bisulphid
added.
Per cent.
Control .
Control.
Control .
Control.
Weight of first crop, com.
Green.
Gnt.
480
440
500
410
380
460
410
460
Dry.
Gm.
82
90
82
77
85
83
86
Average.
Gm.
\-l
83
Weight of second crop, buck-
wheat.
Green.
Dry.
Gm.
26
27
22
19
27
17
20
18
Average.
Gm.
23-5
20.5
Table VII gives the effect of this volatile antiseptic on oats (first crop)
and corn (second crop). The former showed an increase in growth in
the treated soil; the latter was not affected.
Table VII. — Effect of carbon bisulphid on the growth of oats and corn
I
Miami .
2
...do...
...do...
A
...do...
C
. . .do. . .
6
...do...
7
...do...
8
...do...
Soil.
Carbon
bisulphid
added.
Per cent.
Control.
Control.
Control.
Control.
Weight of first crop, oats.
Green.
Gm.
172
184
171
182
200
205
197
192
Dry. Average.
Gm.
46.5
51
46.7
49
59
59
57-7
57-5
Gm.
48.3
57-8
Weight of second crop, com.
Dry.
w.
166
Gm.
40
132
31
118
180
29
45
155
161
37
38
152
135
37
36
Average.
Gm.
36
37
A general consideration of the data shows that corn in this soil type is
apparently indifferent toward carbon bisulphid. Buckwheat, oats, and
mustard were all benefited by the antiseptic.
BUCKWHEAT, MUSTARD, OATS, AND CORN IN DIFFERENT SOILS
The experiment with buckwheat, mustard, corn, and oats was a com-
bination study of the effect of carbon bisulphid on bacterial activity
and plant growth in three different soils. The first series contained
Miami silt loam, the second series Miami soil diluted one-half by volume
with sand, and the third series sand alone. According to chemical anal-
ysis, Miami silt loam is fairly rich in organic matter, nitrogen, potas-
sium, and phosphorus. Of the three fertilizing elements, phosphorus
perhaps is present in the smallest amount. The quantity of soil and its
treatment was similar to that of the preceding experiment except that
the treated jars were kept tightly covered with parchment paper. One
month after the carbon bisulphid was added, these were removed. By
Apr. 3, 1916
Relation of Carbon Bisulphid to Plant Growth
this means it was hoped to prevent a rapid volatilization of the anti-
septic. The jars were not planted until three months after treatment.
At the beginning and at intervals of one, two, and three months
bacterial activity was measured. Naturally, under the conditions of this
experiment, carbon bisulphid proved very drastic. A great reduction in
the number of bacteria, without any increase until the second month,
was noted. The relation of carbon bisulphid to the number of bacteria
was about the same in all three series. In the more compact type,
Miami silt-loam soil, the carbon bisulphid proved most injurious to num-
bers, and consequently the period of increase was much later. Of the
three soils, the treated sand showed the greatest proportional gain in
number of bacteria.
Because of the severe nature of the carbon-bisulphid treatment, it was
thought that probably the protozoa would be destroyed or the number
greatly diminished. This was not the case, however, as protozoa were
found in great numbers in both the treated and untreated soil.
Three months after treatment the jars were divided into two series and
planted. The weights of the first and second crops are given in Tables
VIII and IX.
Table Ylll.— Effect of carbon bisulphid on the growth of buckwheat and mustard in
different types of soil
No.
Soil.
Carbon bi-
sulphid
added.
Weight of first crop, buckwheat.
Weight of second crop,
mustard.
Green.
Dry.
Average.
Green.
Dry.
Aver-
age.
I
2
Miami silt loam
do
Per cent.
Control.
Control.
2
2
Control.
Control.
2
2
Control.
Control.
2
2
Gm.
123
107
119
114
74
72
76
78
20
21
21-5
21-5
Gm.
22. 5
20.5
25.0
23.0
15-5
15.0
18.0
r6. 0
2-5
3-0
3-0
3-0
Gm.
I 21.5
\ 24. 0
1 15-25
} 17-0
} 2.75
I 3- 00
Gm.
12. 0
10. 0
24-5
41-5
21.0
17.0
21.5
17.0
4.0
4-5
17-5
5-5
Gm.
3-4
5-3
5-2
9-5
3-75
3.6b
4-5
4.0
0.4
0-5
I. 2
.6
Gtn.
} 3-3
3- 67
3
4
5
6
do
do
Half Miami silt loam,
half sand.
do
7
8
do
} 4.25
}■-
} 90
do
9
10
Sand
do
II
do
12
do
The figures of the buckwheat crop show the same general increase as
noted in a previous experiment. Although not great, the gain in the
treated series is consistent in all three soils.
The residual crop of mustard responded to a very marked degree to
the carbon bisulphid treatment. In Miami silt loam the yield from the
treated soil exceeded that of the control by more than loo per cent. The
gain in weight of oats in the treated soils was not so great, while the
second-crop corn showed a loss (Table IX).
lO
Journal of Agricultural Research
\(A. VI, No. I
Table IX. — Effect of carbon bisulphid on the growth of oats and corn in different types
of soil
Soil.
Carbon bisul-
phid added.
Weight of first crop, oats.
Weight of second crop, com.
Green.
Dry.
Average.
Green.
Dry.
Average.
I
2
Miami silt loam
do
Per cent.
Control.
Control.
2
2
Control.
Control.
2
2
Control.
Control.
2
2
Gm.
162
180
157
190
82
85
85
82
Gm.
47
48.5
45-5
52-5
26.5
28
27-5
28.5
Gm.
} 47-7
} 49
} 27-2
Gm.
(114
1 103
/ 77
I 87
/ 76
I 56
/ 52
I 64
Gm.
25
23
18. 5
20
16
14
13
15
Gm.
I 24
3
4
S
do
f 19- 2
do
Half Miami silt loam,
half sand
} ^^
6
do
7
8
do
do
} ^^
9
lO
II
12
Sand
do
do
do
18
18
15
6
5-8
5-2
6
} 5-5
12
(14
4
5
4
4
} 4-5
The results of the nitrate determinations agree with those obtained in
previous experiments. At the time of planting the carbon-bisulphid soils
were lower in nitrate but higher in ammonia than the original soil.
The data from Tables VIII and IX show that carbon bisulphid has a
much more beneficial efi"ect on mustard than on any other crop. Buck-
wheat and oats are benefited, but not so markedly as mustard. Corn
fails to show any improvement from treatment with carbon bisulphid.
EFFECT OF CARBON BISULPHID ON BUCKWHEAT AND RAPE IN VARIOUS
SOILS
The five soil types selected for the study of the effect of carbon bisul-
phid on buckwheat and rape in various soils ranged all the way from a
very compact red clay to an open, sandy soil. After treating with 2
per cent of carbon bisulphid the soils were allowed to stand for three
months before planting. Bacteria counts and nitrate determinations
were made at the beginning and after two and three months. The effect
of the carbon bisulphid on the total number of bacteria is very evident.
In every case the carbon-bisulphid soil contained the most bacteria. The
maximum gain occurred in the clay-loam soil, the minimum in the Norfolk
sand. The increase due to the treatment was greatest after two months.
Here, again, the treated soils gave a much lower nitrate content than
the controls. It seems safe to say that a rapid increase in numbers of
bacteria in a carbon-bisulphid soil is followed by a decrease in the
amount of nitrates.
Three months after treatment the soils were planted to buckwheat.
Growth was slow at first, especially in the carbon-bisulphid series. The
crop was harvested when 60 days old. The results of this experiment
are shown in Table X.
Apr. 3, 1916 Relation of Carbon Bisulphid to Plant Growth
II
Table X. — Effect of carbon bisulphid on the growth of buckwheat in different types of soil
No.
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
Soil.
Cecil clay
do
do
do
Porters clay
do
do
do
Clay loam
do
do
do
Hagerstown loam
do
do
do
Norfolk sand
do
do
do
Carbon
bisulphid
added.
Per cent.
Control.
Control.
2
2
Control .
Control.
Control.
Control.
Control.
Control.
2
2
Control.
Control.
Weight of first crop.
Green.
Gm.
5
15
10.5
14-5
14-5
5-
12.
28
30-
27
25
17-
40.
32
49.
12
17-
Dry.
Average.
Gm.
I. 2
1-5
3
4
2
3-5
3-7
3.7
I. 2
2
6.5
8
6.2
S
4-5
7-5
8.7
10. 2
3
3-5
Gm.
1-35
3-5
2-75
1.6
7-25
5-6
6
9-45
3-25
With one exception, Norfolk sandy soil, the carbon-bisulphid series
gave a larger yield. This was most marked in the case of clay-loam soil.
The data on p]^t growth agreed with the plate counts.
The buckwheat was followed by a crop of Dwarf Essex rape. Unfortu-
nately the young rape plants suffered seriously from insects. Although
the tissue was too badly infested to save, a decided difference in growth
could be seen. The beneficial effect of carbon bisulphid on rape was
noted in every soil type.
EFFECT OF CARBON BISULPHID ON VARIOUS CROPS IN ACID SOILS
In order to study the effect of carbon bisulphid on the growth of higher
plants in acid soils, a series of experiments was made. Four types of
soil were selected for this work: Miami silt loam, Sparta sand, Colby
silt loam, and Marshfield peat. The neutral Miami silt loam was used
as a check for the acid soils. According to the Truog acidity test, Sparta
sand requires 0.5227 gm. of calcium carbonate per 100 gm. of soil, Colby
silt loam 1.021 gm., and Marshfield peat 4.43 gm. Four weeks after
treatment with carbon bisulphid, the soils were planted.
RED CLOVER
The effect of carbon bisulphid on medium red clover in acid soils is
clearly seen from the figures of Table XL The clover grew luxuriantly
in all soils except the untreated acid peat. Two crops were cut. Carbon
bisulphid in peat soil caused an enormous gain in the growth of clover.
This was very striking in both the first and second crop.
12
Journal of Agricultural Research
Vol. VI, Xo. I
Table XI. — Effect of carbon bisulphid on the growth cf red clover in acid soils
No.
I
3
3
4
5
6
7
8
9
lo
II
12
13
14
IS
16
Soil.
Miami silt loam .
do
do
do
Sparta sand . . . .
do
do...
do
Colby silt
....do
do
do
Peat
do
do
do
Carboo
bisulphid
added.
Per cent.
Control.
Control.
Control.
Control.
Control.
Control.
Control.
Control.
Weight of first crop, clover.
Green.
Gni.
138
140
158
124
36
33
19
31
95
87
153
^33
4
2
83
79
Dry.
Gm.
(a)
(«)
(a)
(«)
Average
green.
Gm.
139
141
34
25
91
143
3
81
Weight of second crop,
clover.
Green.
Gm.
129
145
168
131
58
, 48
18
: 43
no
85
108
82
6
53
46
Dry.
Gm.
19
21
26
20
13
ID
4
8
20
14
15
12
2.8
2
9
8-5
Aver-
age.
Gtn.
■ 20
• 23
• II
■ 6
• 17
■ 13
• 2.4
• 8.7
a Lost.
Plate II, figure 2, shows the relative growth of clover in the treated
and untreated soils.
Each figure for Miami silt loam in Table XI represent^ the average of
triplicate jars. Because of the individual variation, it was decided to use
12 jars for this experiment. Six of these were used as controls and six
treated with 2 per cent of carbon bisulphid. It is evident from the data
that medium red clover in Miami soil is benefited both in the first and
second crop by the antiseptic. In the Sparta sand a decrease was noted
with each crop. The Colby silt loam gave a decided increase with the
first crop, but not with the second.
Previous tests with these soils showed that the clover bacteria were
present in sufficient numbers to produce good inoculation. In view of the
large amount of carbon bisulphid applied, it was thought that this sub-
stance would probably injure nodule formation. However, examination
of the root systems showed this was not the case. The plant roots were
thoroughly inoculated, both in the treated and tmtreated soils. Appar-
ently the plants in carbon bisulphid soils contained the greater number
of nodules.
Because of the remarkable action of carbon bisulphid in peat soil, this
part of the previous test was repeated. In addition to carbon bisulphid,
the effect of flowers of sulphur was studied. If the data in the previous
experiment are correct, the carbon bisulphid should greatly increase the
growth of clover. A glance at the results in Table XII confirms this
statement.
Apr. 3, 1916 Relation of Carbon Bisulphid to Plant Growth
13
Table XII. — Effect of carbon bisulphid and sulphur on the growth of red clover in peat
soil
No.
Soil.
Treatment.
Weight.
Green.
Dry.
Average.
I ...
Peat
Control
Gm.
34
30
95
no
loS
90
8
4
Gm.
9
8.2
21-5
22
23
19-5
3- 5
I
Gm.
\ 8.6
1
2
do. .
. . do. .
■2*
do
I percent of carbon
bisulphid.
do
4.
do
\ 21.7
c
do
2 per cent of carbon
bisulphid.
do
1
6
do
> 21. 2
7 .
do
0.3 percent of sul-
phvu".
do
> 2. 2
8
do
Carbon bisulphid causes a remarkable increase in the growth of clover
on peat soil. There is apparently no decided difference in the action of i
or 2 per cent of carbon bisulphid. Just why the volatile antiseptic should
stimulate so markedly the growth of clover in the peat soil is not known.
A more detailed study of the action of carbon bisulphid in peat is now
underway. Flowers of sulphur at the' rate of 0.3 per cent proved very
injurious. In view of the high sulphur content of carbon bisulphid, it
was thought that possibly free sulphur in peat might have somewhat the
same effect.
CORN AND MUSTARD
The action of carbon bisulphid on corn and mustard in acid soils was
studied in an experiment the results of which are given in Table XIII.
Table XIII. — Effect of carbon bisulphid on the growth of corn and mustard in acid soils
SoU.
Carbon bisul-
phid added.
Weight of com.
Weight of mustard.
No.
Green.
Dry.
Average.
Green.
Dry.
Aver-
age.
I
2
3
4
5
6
Miami silt loam
do
do
do
Sparta sand
do
Per cent.
Control.
Control.
2
2
Control.
Control.
2
2
Control.
Control.
2
2
Control .
Control.
2
2
Gin.
190
360
315
320
65
70
100
120
393
390
385
160
130
165
Gm.
50
79
70
60
17
18
21
26
83
86
85
77
28
20
25
20
Gm.
} 64. 5{
} -{
} ^^"4
} 84. sj
}-{
}-{
} ^^'H
Gm.
83
145
159
13
19
II
12.5
18
21
27
24
2-5
3-5
2
3
Gm.
} 19-5
} 25-5
} ^
} 2. 5
} 9-6
7
8
do
do
9
10
Colby silt
do
67
62
10
9.6
II
do
12
do
13
Id
Peat
do
do
0
0
0
0
IS
t6
do
14 Journal of Agricultural Research voi. vi, no. i
It is clear from the data that carbon bisulphid does not materially
benefit corn. An exception to this was seen in the case of Sparta sand;
in this instance the treated series showed a slight improvement.
A comparison of the growth of mustard in acid and in neutral soil
shows that this crop grows best in a neutral soil. In Sparta sand and
Colby silt loam the yield of mustard in the treated soil was below that
of the control, while in the peat soil it failed entirely. It seems very
probable that the acid reaction of the soil inhibits the growth of mus-
tard. For instance, Kossovich (7) reports that mustard is sensitive to
acidity. The addition of 2 per cent of carbon bisulphid to Miami soil
stimulated the growth of mustard. This agrees with the results of pre-
vious tests. An increase in the growth of mustard has been noted in all
four experiments with carbon bisulphid in Miami soil.
One series of jars, corn on Miami silt loam, was replanted to buck-
wheat. As previously reported, buckwheat showed a distinct improve-
ment in the carbon-bisulphid soil. If the control weights are taken as
100, the treated series is equal to 115.
A review of all the data on the effect of carbon bisulphid on higher
plants shows very clearly that carbon bisulphid does not produce the
same effect on all plants. In almost every case (except acid soils) the
carbon bisulphid favors in a decisive way the growth of mustard. Next
in order of their response to carbon bisulphid come rape, red clover,
buckwheat, oats, and com. In acid soils, especially those rich in organic
matter, the growth of clover is greatly favored by the carbon-bisulphid
treatment.
The majority of the evidence indicates that carbon bisulphid is most
beneficial to the growth of higher plants in peat or in open, sandy soils.
EFl^ECT OF CARBON BISULPHID ON THE GROWTH OF PLANTS IN SILICA
SAND
If carbon bisulphid is a plant stimulant, then the addition of the proper
amount to a nutrient solution for plants should exert a beneficial effect
on the growth of higher plants. To test this a series of experiments was
performed on different plants.
BUCKWHEAT AND OATS
Eight jars were filled with pure silica sand (99 per cent pure quartz),
and the following ingredients added to each jar :
Water (HgO) 500 c.c.
Potassium nitrate (KNO3) 5 g™-
Ferrous phosphate (Fe3(P04)2) 1.25 gm.
Calcium phosphate (Ca3(P04)2) i. 25 gm.
Calcium sulphate (CaS04) i. 25 gm.
Magnesium sulphate (MgS04) i. 25 gm.
Apr. 3, 1916 Relation of Carbon Bisulphid to Plant Growth
15
In addition to the soluble plant food, half of the jars received 2 per
cent of carbon bisulphid. After treatment the jars were held for two
months before planting to buckwheat and oats. The results of the test
are given in Table XIV.
Table XIV. — Effect of carbon bisulphid on the growth of buckwheat and oats in silica
sand
No.
Carbon
bisulphid
added.
Weight of buckwheat.
Weight of oats.
Green.
Dry.
Average.
Green.
Dry.
Average.
I
Per cent.
Control.
Control.
2
2
Gm.
15.8
8-5
37-5
21
Gm.
3-2
1.4
7
4.2
Gm.
} -
Gm.
3.4
Gm.
1-5
I. 2
Gm.
) -
} <-
2
4
{ 2X
6.5
It is apparent from the data that carbon bisulphid in silica sand exerts
a beneficial effect on the growth of both buckwheat and oats. This
agrees with the results of Koch (6) — that carbon bisulphid stimulates
the higher plant growth. Although the duplicate jars do not agree very
closely, the highest yield of the control was lower than any of the treated
groups. For some unexplainable reason, the oats in jar 3 failed to grow.
The young seedling died soon after germination. Plate II, figure 3, is a
reproduction of a photograph of the buckwheat series.
CLOVER, BUCKWHEAT, AND MUSTARD
The foregoing experiment was repeated, using 3-kgm. jars and Tollen's
medium. Only i per cent of carbon bisulphid was added. The jars
were planted 30 days after treatment. The yields of the different crops
are presented in Table XV. From the beginning clover and mustard
began to show the favorable effect of carbon bisulphid.
Table XV. — Effect of carbon bisulphid on the growth of buckwheat, clover, and mustard
in silica sand
Carbon
bisulphid
added.
Weight of buckwheat.
Weight of clover.
Weight of mustard.
No.
Green.
Dry.
Aver-
age.
Green.
Dry.
Average.
Green.
Dry.
Aver-
age.
I
2
Per cent.
Control.
Control.
I
I
Gm.
49
41
49
45
Gm.
7-5
6.6
7.8
7-S
Gm.
Gm.
{ :5
Gm.
I
1.8
2. 2
2-3
Gm.
} -
> 2. 2
Gm.
{ ''
Gm.
9
Gm.
3
4
[ 52
I 98
5-8
"•5
As compared with the results shown in Table XIV, the increase in the
growth of buckwheat with carbon bisulphid was much smaller. The
clover crop was about doubled in the presence of carbon bisulphid. Mus-
i6
Journal of Agricultural Research
Vol. VI. No. I
tard did not do well in sand cultures ; growth was very irregular. Because
of the size of the jars and the irregular growth of the crops it will be
necessary to repeat the experiment.
EFFECT OF CARBON BISULPHID IN REINOCULATED SOIL
In the first part of this paper it has been shown that if soil treated with
carbon bisulphid is reinoculated with fresh soil the bacterial processes
are altered. The increase in number of bacteria attains a maximum
much sooner and begins to decline earlier than in soil treated with carbon
bisulphid but not reinoculated. This is also noted in the formation of
soluble nitrogen. In order to record the effect on plant growth, the fol-
lowing experiment was planned. Six jars with 9 kgm. each of Miami
silt-loam soil were used. Two months after treatment with carbon
bisulphid, 2 per cent of untreated soil were added to jars 5 and 6. An
equal amount was removed before the original soil was added. All of
the jars were kept for another month before planting.
Plate"counts three months from the date of treatment showed a decided
increase in number of bacteria in the carbon-bisulphid soils. No appre-
ciable difiference existed between the carbon bisulphid and the carbon-
bisulphid reinoculated soil.
The effect of treatment on nitrate content is evident from the following
figures: If the nitrate nitrogen at the beginning is 100, then the control
after three months is 370, carbon bisulphid is 50, and carbon bisulphid
plus 2 per cent of the original soil is 44. Here, again, the inverse rela-
tion of number of bacteria and nitrate content is noted.
Protozoa were found in all of the soils and apparently in about the
same number two months after treatment as in the original soil.
The effect of this treatment on the growth of oats and com may be
seen from the figures in Table XVI.
Table XVI.
-Effect of carbon bisulphid on the growth of oats and corn in reinoculated
soil
No.
Soil.
Miami
...do..
...do..
...do..
...do..
...do.
Treatment.
Control
....do
2 per cent of carbon bisul-
phid.
do ..,
2 per cent carbon bisulphid
plus 2 per cent of the orig-
inal soil.
do
Weight of first crop, oats.
Green.
Gm.
168
178
178
185
178
215
Dry.
Gm.
SI
50-75
50
54
51-5
61.5
Aver-
age.
Gm.
^50. 9
56.5
Weight of second crop,
com.
Green.
Dry.
Gm.
28
30
2,2,
26
26
Aver-
age.
Gm..
■ 29
29-5
■ 24
Apr. 3. 1916 Relation of Carbon Bisulphid to Plant Growth
17
The average dry weight of oats in soil treated with carbon bisulphid
was slightly greater than that of the control. This difference was most
noticeable in the case of reinoculated soil. It appears that the reinocula-
tion benefits the action of carbon bisulphid on the growth of oats. The
second crop of corn gave the opposite results. The com in untreated
soil gave the highest yield.
EFFECT OF CARBON BISULPHID ON THE ACCUMULATION OF SULPHATES IN
SOIL
Very soon after the jars were planted it was observed that the surface of
carbon-bisulphid soil was partly covered with needle-like crystals. Quali-
tative tests showed that these were made up largely of sulphates, possibly
magnesium sulphate. The occurrence of salts was noted in several of
the soils treated with carbon bisulphid. Possibly a part of the carbon
bisulphid was oxidized to sulphates. It has been reported that a small
portion of the carbon bisulphid may be converted into sulphates (4, p.
247-251; 10, p. 151-152).
Samples of the treated and untreated soils were analyzed for sulphates.^
The results are shown in Table XVII.
Table XVII. — Effect of carbon bisulphid on the accumulation of sulphates in the soil
Time.
Treatment.
Sulphur as
sulphates.
Months.
3-
4-
5-
6.
Untreated
2 per cent of carbon bisulphid
Untreated
2 per cent of carbon bisulphid
Untreated
2 per cent of carbon bisulphid
Per cent.
0.023
.038
.018
•039
. 019
. 060
It is apparent from the data in this table that the addition of carbon
bisulphid tends to increase the sulphate content of the soil.
CONCLUSIONS
The addition of carbon bisulphid to soil exerts a decided effect on the
fauna and flora of the soil. This is characterized by a temporary reduc-
tion in the number of micro-organisms. Later, an enormous multiplica-
tion of bacteria takes place and an almost parallel increase in production
of by-products or soluble nitrogen is noted. The ammonia content seems
to follow the curve of bacterial growth and later gives way to larger
amounts of nitrate. From the evidence it seems that carbon bisulphid
in soil produces an increase in soluble compounds of nitrogen and sulphur.
• The author is indebted to Prof. W. E. Tottingham, of the Department of Agricultural Chemistry, for
the analyses.
27469°— 16 2
i8 Journal of Agricultural Research voi. vi. No. i
In Miami soil carbon bisulphid benefited the growth of buckwheat,
oats, and mustard. No relation seems to exist between plant stimula-
tion with carbon bisulphid and the form of the soluble nitrogen. In non-
acid soils carbon bisulphid is most beneficial to sulphur crops. Mustard
offers a good example. In all of the experiments, except acid soils,
mustard showed an increased growth from the use of carbon bisulphid.
Carbon bisulphid in peat soil greatly benefits the growth of red clover.
In sand cultures plus soluble plant food carbon bisulphid favors the
growth of certain plants.
The data show clearly that carbon bisulphid does not act alike in aH
soils or toward all crops.
LITERATURE CITED
(i) Fred, E. B.
191 1. Uber die Beschleunigimg der Lebenstatigkeit hoherer und niederer
Pflanzen durch kleine Giftmengen. In Centbl. Bakt. [etc.], Abt. 2,
Bd. 31, No. 5/10, p. 185-245, 4 fig. Literatur, p. 242-245.
(2) Gainey, p. L.
1914. Effect of CS2 and toluol upon nitrification. In Centbl. Bakt. [etc.],
Abt. 2, Bd. 39, No. 23/25, p. 584-595, 2 fig. Literature, p. 595.
(3) Gerlach and VoGEL.
1905. Ammoniakstickstoff als Pflanzennahrstoff. In Centbl. Bakt. [etc.],
Abt. 2, Bd. 14, No. 3/4, p. 124-128, 2 fig.
(4) Heinze, B.
1907. Einige weitere Mitteilungen iiber den SchwefelkohlenstoflF und die CS2-
Behandlung des Bodens. In Centbl. Bakt. [etc.], Abt. 2, Bd. 18,
No. 1/3, p. 56-74, I fig.; No. 7/9, p. 246-264, 2 fig.; No. 13/15, p. 462-
470; No. 19/21, p. 624-634; No. 24/25, p. 790-798.
(5) HaTNER, Lorenz, and Stormer, Kurt.
1903. Studien iiber die Bakterienflora des Ackerbodens, mit besonderer
Beriicksichtigung ihres Verhaltens nach einer Behandlung mit
SchwefelkohlenstofE und nach Brache. In Arb. Biol. Abt. Land- und
Forstw. K. Gsndhtsamt., Bd. 3, Heft 5, p. 445-545, 4 fig., pi. 9-10.
(6) Koch, Alfred.
1899. Untersuchimgen iiber die Ursachen der Rebenmiidigkeit mit besonderer
Beriicksichtigimg der Schwefelkohlenstoff behandlung. Arb. Deut.
Landw. Gesell., Heft 40, 44 p., 5 pi.
(7) KossoviCH, P. S., and Ai^Thausen, L.
1909. Influence of CaCOs and MgCOs on the soil and plants. (Abstract). In
Exp. Sta. Rec, v. 23, no. 3, p. 226. 1910. Original article appeared
in Trudui Mendelyevsk. Syezda Obshch. i Prikl. Khim., god i,
1907, p. 490-493. 1909. Not seen.
(8) Kruger, Wilhelm.
1905. tJber die Bedeutung der Nitrifikation fur die Kulturpflanzen. In
Landw. Jahrb., Bd. 34, Heft 5, p. 761-782, pi. 13-15.
(9) Lehmann, Jul.
1875. Ueber die zur Emahrung der Pflanzen geeignetste Form des Stickstoffes.
In Centbl. Agr. Ckem., Bd. 7, p. 403-409.
(10) Moritz, J., and SchERPE, R.
1904. Uber die Bodenbehandlung mit Schwefelkohlenstoff und ihre Einwirkung
auf das Pflanzen wachstum. In Arb. Biol. Abt. Land- u. Forstw.
K. Gsndhtsamt., Bd. 4, Heft 2, p. 123-156.
Apr. 3. 1916 Relation of Carbon Bisulphid to Plant Growth 1 9
(11) Russell, E. J., and Hutchinson, H. B.
1909. The effect of partial sterilisation of soil on the production of plant food.
In Jour. Agr. Sci., v. 3, pt. 2, p. 111-144, 4 fig., pi. 8-9.
(12)
19 13. The effect of partial sterilisation of soil on the production of plant food.
II. The limitation of bacterial numbers in normal soils and its con-
sequences. In Jour. Agr. Sci., v. 5, no. 2, p. 152-221, 7 fig.
(13) Sherman, J. M.
1914. The number and growth of protozoa in soil. In Centbl. Bakt. [etc.],
Abt. 2, Bd. 41, No. 18/23, P- 625-630.
PLATE I
Plate cultures of soil organisms growing on agar:
Fig. I. — Colonies of organisms from untreated soil.
Fig. 2. — Colonies from soil treated with 2 per cent of carbon bisulphid.
Fig. 3. — Colonies from soil treated with 2 per cent of carbon bisulphid and evapo-
rated.
Fig. 4. — Colonies from soil treated with 2 per cent of carbon bisulphid, evaporated,
and reinoctilated with 5 per cent of soil from an untreated jar.
(20)
Relation of Carbon Bisulphid to Plant Growth
Plate
Journal of Agricultural Research
Vol. VI, No. 1
Relation of Carbon Bisulphid to Plant Growth
Plate
Journal of Agricultural Research
Vol. VI, No. 1
PLATE II
Fig. I. — Effect of varying amounts of carbon bisulphid on mustard; A, B, soil
untreated; C, D, soil treated with 0.5 per cent of carbon bisulphid; E, F, soil treated
with I per cent of carbon bisulphid; G, H, soil treated with 2 per cent of carbon
bisulphid.
Fig. 2. — Effect of carbon bisulphid on clover in peat soil; .4, B, soil untreated;
C, D, soil treated with 2 per cent of carbon bisulphid.
Fig. 3. — Effect of carbon bisulphid on buckwheat in sand cultures; A, B, soil
untreated; C, D, soil treated with 2 per cent of carbon bisulphid.
CLIMATIC CONDITIONS AS RELATED TO CERCOSPORA
BETICOLA ^
By Venus W. Pool, Assistant Pathologist, and M. B. McKay, Scientific Assistant,
Cotton and Trtick Disease Investigations, Bureau of Plant Industry
INTRODUCTION ^
Climatic conditions of both mnter and summer bear an important
relation to the vitality and development of Cercospora heticola. During
cold weather certain conditions enable the fungus to overwinter, while
certain other conditions are inimical to its growth, a fact which has an
important bearing on the control of the disease, as the earliest infections
on growing sugar beets {Beta vulgaris) originate from the overwintered
fungus. In the early summer, after infection occurs, temperature, rela-
tive humidity, rainfall, and wind directly affect the development of the
fungus, the rapidity of conidial production, and subsequent infection.
OVERWINTERING
From the investigations here described it seems evident that under
ordinary field conditions of winter the conidia of C. heticola usually live
but a short time, although under ordinary herbarium conditions desic-
cation takes place only after exposure for several months. The sclerotia-
like bodies (fig. i , A, a), or masses of mycelium, the most resistant part
of the fungus, which are embedded in the infected areas of the leaf blades
and petioles, however, live over the winter under favorable conditions and
in the spring produce conidia from the remnants of the old conidiophores
(fig. I, A, h), or both conidiophores and conidia (fig. i, A, c) may be
formed anew. For the purpose of making direct microscopical observa-
tion of such development sections of infected tissue which had been
stored throughout the winter under favorable conditions were placed in
hanging-drop cultures of bean agar. New conidiophores (fig. i, B, h)
grew from the masses of embedded mycelium, and although somewhat
abnormal they produced rather typical conidia (fig. i, B, c), thus show-
ing that such material may be a source of early infection of growing
plants.
> The investigations were carried on entirely in the field. Preliminary work was conducted during 191 1
and 1912 at Rocky Ford, Colo. The detailed data were collected during 1912 and 1913 at Rocky Ford,
which is in the Arkansas Valley of Colorado, a semiarid region under irrigation, and during 1914 near
Madison, Wis., where the rainfall and average humidity were greater.
2 The writers are indebted to Mrs. Nellie E. Fealy, of the Bureau of Plant Industry, for aid in editing
and revising the manuscript.
Journal of Agricultural Research, Vol. VI, No. i
Dept. of Agriculture, Washington, D. C. Apr. 3, 1916
cr G— 7S
(21)
22
Journal of Agricultural Research
Vol. VI, No. I
_~^ i/O/A.
Fig. I. — Cercospora beticola: A , Section of overwintered sugar-beet leaf showing embedded sclerotia-like
body, a, with a mass of old conidiophores, b, from which a new conidium, c, was produced. B. Produc-
tion of rather typical conidiophores, b, and conidia, c, from a sclerotia-like mass, a, taken from over-
wintered h"st material and placed in hanging-drop cultures.
Apr. 3, 1916
Climatic Conditions and Cercospora beticola
23
CONIDIA
Thiimen (1886, p. 50-54) ^ believed that the spores of Cercospora beti-
cola are able to live for a certain length of time in the soil and retain their
viability and produce new infection, and Pammel (1891, p. 238-243) and
Massee (1906, p. 52-53) accord with this view. In the investigations
here considered it was found that when kept dry, as in the case of her-
barium material, the conidia remained viable for 8 months (Table I,
tests 10 to 13), but soon after that no growth occurred. Only rarely
were conidia found on the infected areas of the leaves which were exposed
to outdoor weather conditions, and such conidia seemed to lose their
vitality soon after harvest. No germination was found to take place
under optimum conditions in the case of conidia which had been thus
exposed from i to 4 months (tests 14 and 15). However, conidia occa-
sionally found on spots that had been well protected, for instance in the
interior of a pile of hayed beet tops, retained their viability for from 5 to
less than 12 months (tests 16 and 17). Since the coUidia are rarely found
after a short time even on infected material that has been well protected
and since they rarely germinate after being exposed outdoors for even i
month after harvest, it would seem that under ordinary field conditions
they play no important part in the overwintering of the fungus.
Table I. — Viability of the conidia of Cercospora beticola as affected by desiccation
Test
No.
Environment. .
Period of exposure.
ViabiUty.
Stored, dry
14 years
None.
do
Do.
3
.do
Do.
....do
S years
Do.
s
6
....do •.
Do.
do
Do.
. .do
Do.
8
do
Do.
do
Do.
.do
8 months
Slightly viable.
do
7 months
Extremely viable.
do
Do.
do
Do.
None.
Do.
16
Stored inside pile of hayed sugar-beet leaves
... .do
Extremely viable.
None.
i
SCLEROTIA AND MYCEUUM
Various investigators have attempted to determine whether different
fungi live in the soil over winter and the manner in which they over-
winter. Treboux (191 4) found that the mycelia of several different rusts
overwinter on host material freely exposed to climatic conditions. Stew-
art (191 3) placed in boxes of soil potato leaves and tubers infected with
Phytophthora injestans, exposed them to outdoor winter conditions, and
found that plants grown on such soil developed no blight. However,
' Bibliographic citations in parentheses refer to "Literature cited," p. 60.
24 Journal of Agricultural Research voi. vi, no. i
temperature and moisture conditions in boxes of soil exposed above-
ground to winter conditions are much more varied than in soil at different
depths in the field where normal overwintering usually occurs.
In the overwintering experiments here described the host material was
kept in an environment comparable to ordinary field conditions. The
experiments at Rocky Ford, Colo., were started about the middle of Octo-
ber, 1 91 2, and continued for 1 1 months. In these experiments some of the
infected material was mixed with soil, placed in boxes, and exposed above-
ground during the winter (PL III, i); a second portion was buried from
I to 8 inches in the ground (PI. Ill, 2), wire netting being used above and
below the infected material to insure ready location when examinations
were made for cultural tests (Pool and McKay, 191 5); a third portion of
the infected tops was placed in a pile on top of the ground (PI. Ill, 3).
During the experiment records were kept of soil and air temperatures, the
former being taken at a depth of 5 inches and the latter being obtained
from the Weather Bureau station at Rocky Ford.^
The experiments carried on near Madison, Wis., were started the last of
November, 191 3, and continued through the winter. Infected sugar-beet
tops were buried in the soil at depths of 5 and 8 inches, while seed-beet
stalks were left under ordinary conditions in the field. In this experi-
ment also records were kept of soil and air temperatures, the former being
taken from March until June at a depth of 5 inches and the latter obtained
from the Weather Bureau station at Madison.
The effect of desiccation on material kept under herbarium conditions
was to kill probably all life of the fungus within 12 months, as already
shown, but material kept under an environment having more or less
moisture accompanied by the disintegrating action of various organisms
was affected in an entirely different manner, as will be shown. All cul-
tures from the infected material used in the two experiments above out-
lined were made from definite leaf-spots. Although the diseased tissue
was the last to be completely disorganized and consequently could be
found as long as any portion of the leaf remained, it became more and
more difficult to obtain such tissue as time went on.
The fungus was unable to survive six months' outdoor exposure in boxes
of soil (Table II, experiment 2), and this was also true of the fungus on
leaves which had been freely exposed to outdoor conditions — for instance,
on the outside of a hayed pile of sugar-beet tops (experiment 3), and
on leaves buried 6, 7, and 8 inches in the ground (experiments 19 to 23).
In cultures from infected mother-beet stalks and leaves that had been
left in the field for a time and then plowed under or stored there was no
growth, or only an indefinite growth, of the fungus after 7 months (ex-
periments 8 to 10), while in infected material that had been protected
in the interior of a pile of hayed beet tops (experiment 4) and in material
' All the records included in this paper from the Weather Bureau station at Rocky Ford, Colo., were
kindly furnished by Mr. P. K. Blinn, the local observer.
Apr. 3, 1916 Climatic Conditions and Cercospora beticola 25
that had been slightly covered or buried from i to 5 inches in the ground
the life of the fungus was entirely extinct after 12 months (experiments
II to 18). The death of the fungus in material plowed under is due in
all probability to the rapid disorganization which results under favor-
able temperature and moisture conditions, such, for instance, as those
which prevailed at Rocky Ford through the winter of 191 2-13. During
that period there was insufficient moisture to permit severe freezing,
but there was a daily extreme variation of soil temperature, indicating
that the air temperature produced the changes through the more or less
dry soil. In the experiments at Madison there was only a partial dis-
integration of the buried beet tops six months after harvest, but other
factors impaired the vitality of the fungus and its life appeared to be
entirely extinct; consequently, notwithstanding the great differences in
soil factors, comparable results as to the life of the fungus were obtained
from the experiments at both places.
Table II. — Effect of desiccation and overwintering on the viability of Cercospora beticola
in infected sugar-beet tops under field conditions at Rocky Ford, Colo., and Madison,
Wis.
Number
Ex-
peri-
ment
No.
of spots
Environment of sugar-beet-top
material.
Period of
exposure.
from
which
cultures
Number
of viable
spots.
Condition of leaves.
were
made.
Ol
Dried, stored:
Illinois, Iowa
14 years. ..
II years. . .
10 years. . .
S years
4 years. . . .
3 years
10
10
10
10
10
10
0
0
0
0
0
0
Good.
Connecticut
Do.
New York
Do.
Wisconsin
Do.
Iowa
Do.
Maryland
Do.
Colorado
2 years ....
10 months.
10
10
0
3
Do.
New Jersey
Do.
Colorado
9 to II
months.
20
10
Do.
2 months..
IS
15
Do.
3 months..
7
7
Do.
(Stored in soil in boxes and left free im-
\ der outdoor conditions, Colorado.
4 months. .
13
4
Do.
5 months. .
12
2
Do.
S% months
12
0
Do.
7 months..
6
0
Do.
3
jFrom the outside of "hayed" pile of
\ sugar-beet tops, Colorado.
(^ months..
\ 10 months.
6
0
Do.
13
0
Do.
2 months..
10
10
Do.
3 months. .
66
64
Do.
/From the interior of "hayed" pile of
\ sugar-beet tops, Colorado.
4 months..
5 months..
7 months..
29
27
Do.
Do.
4
40
i8
40
12
Do!
10 months.
IS
10
Do.
12 months.
25
0
Do.
2 months. .
10
10
Do.
5
In field, Colorado
5 months..
8 months..
11
10
8
2
Do.
Do.
6
Leaves from "mother beet" stalks
free in field, Wisconsin.
5 months..
21
15
Do.
fFirst-year sugar-beet leaves free in
\ field, Wisconsin.
f 5 months. .
\8 months. .
32
3
Do.
7
40
0
Partially disintegrated.
8
f Spots on " mother beet" stalks free in
\ field, Wisconsin.
(4 months. .
\7 months..
7
67
Good.
35
4?
Do.
9
Spots on " mother beet" stalks free in
field 6 months, then plowed under
I month, Wisconsin.
7 months. .
10
0
Somewhat softened.
a Herbarium specimens for this test were furnished by Barrett, Illinois; Clinton, Connecticut; WheUel,
New York; Pammel, Iowa; Norton, Maryland; and Cook, New Jersey.
* 39S colonies.
26
Journal of Agricultural Research
Vol. VI- No. I
Table II. — Effect of desiccation and overwintering on the viability of Cercospora beticola
in infected sugar-beet tops under field conditions at Rocky Ford, Colo., and Madison,
Wis. — Continued
Number
Ex-
peri-
ment
No.
of spots
Environment of beet-top material.
Period of
exposure.
from
which
cultures
were
Number
of viable
spots.
Condition of leaves.
made.
lO
Spots on ' ' mother beet ' ' stalks free in
field 4 months, then stored dry 3
months, Wisconsin.
7 months. .
10
5?
Good.
[6 months..
21
0
Partially disintegrated.
II
Buried i inch in ground, Colorado. . . .
1 7 months. .
1 10 months.
10
8
2
3
Do.
Greatly disintegrated.
1 12 months.
24
0
Entirely disintegrated.
( 5 months..
14
2
Partially disintegrated.
12
Buried 2 inches in ground, Colorado. .
l6 months..
1 10 months.
19
II
10
0
Do.
Greatly disintegrated.
I12 months.
28
0
Entirely disintegrated.
(6 months..
21
10
Partially disintegrated.
13
Buried 3 inches in ground, Colorado . t
■I 10 months.
II
I
Greatly disintegrated.
I12 months.
18
0
Entirely disintegrated.
14
/Broken and buried 3 inches in ground,
\ Colorado.
(6 months..
•^7 months..
1 10 months.
14
12
16
0
5
0
Greatly disintegrated.
Do.
Entirely disintegrated.
15
Buried 4 inches in ground, Colorado. .
it months..
\io months.
19
19
5
I
Partially disintegrated.
Greatly disintegrated.
6 months..
20
4
Do.
16
Buried 5 inches in ground, Colorado. .
< 10 months.
15
0
Entirely disintegrated.
12 months.
20
0
Do.
4 months. .
30
0
Good.
17
Buried 5 inches in ground, Wisconsin.
Is months..
1 6 months..
30
81
3?
0
Do.
Do.
1 7 months..
80
0
Partially disintegrated.
fPlowed under in field, aboitt 5 inches,
\ Wisconsin.
fs months..
\s months. .
30
3?
Do.
18
20
0
Do.
[s months..
8
0
Greatly disintegrated.
19
Buried 6 inches in ground, Colorado. .
l6 months..
1 10 months.
20
12
4
0
Do. •
Do.
12 months.
IS
0
Do.
20
/Broken and buried 6 inches in ground,
\ Colorado.
6 months. .
< 7 months.,
lio months.
18
10
16
0
0
0
Do.
Do.
Entirely disintegrated.
[6 months..
15
0
Do.
21
Buried 7 inches in grotmd, Colorado. .
I7 months..
1 10 months.
10
12
0
0
Do.
Do
12 months.
22
0
Do
6 months. .
20
0
Do
22
Buried 8 inches in ground, Colorado. .
■ 7 months..
10
0
Do.
1 10 months.
12
0
Do.
(4 months..
30
0
Good; ground frozen.
S months. .
35
2?
Good.
23
Buried 8 inches in ground, Wisconsin ,
<6 months..
S7
0
Good ; leaves have sour
odor.
18J4 months
100
0
Partially disintegrated.
These experiments and observations made in the field during several
spring and summer months showed that on leaves slightly protected on
or near the surface of the ground during the winter C. beticola can live a
sufiEicient length of time to be a source of infection for the succeeding
sugar-beet crop and that the fungus is entirely killed by planting time
when the infected material is plowed under to a depth of 6 to 8 inches in
the fall.
AIR AND SOIL TEMPERATURES AT ROCKY FORD, COLO.
WIS.
AND AT MADISON,
Comparison of the air and soil temperatures which prevailed during
the experiment at Rocky Ford and Madison showed a wide difference.
Apr. 3. 1916 Climatic Conditions and Cercospora beticola 27
One of the most striking characteristics of these temperatures at Rocky
F'ord was the wide range between the maximum and the minimum, and
this range may be observed throughout the entire records (fig. 2, 3). In
the case of the soil temperatures especially, the wide range appeared to
be due to a lack of moisture, the extreme variations being greater than if
more moisture had been present. A comparison of the records shows that
the variation in air temperature was much less and the mean daily
temperature constantly lower at Madison than at Rocky Ford, notwith-
standing the fact that the daily minimum temperature was usually lower
at Rocky Ford. A comparison of the soil temperatures at the two points,
however, shows that at Madison it probably remained more constant and
was never as low as at Rocky Ford. This was due apparently to the
greater amount of moisture in the soil at Madison and consequently its
continued frozen condition. After March 23, the date on which the
record was begun at Madison, the soil temperature at that place was
never below 29° F., notwithstanding the fact that the air temperature
was as low as 15° on April 8, while the minimum soil temperature at
Rocky Ford was 22° on December 21 and 25° on February 8. However,
as the air temperature on these dates was lower here than at Madison,
comparisons can not be drawn too closely.
In view of the presence of snow on the ground, which, as is well known,
protects the soil from the extreme variations of air temperature, and the
prevailing low air temperatures, as shown by the records, it may be
assumed that the soil temperatures at Madison during January and
February and the early part of March varied but little from freezing.
This assumption is supported by Frodin's experiments (191 3), which
showed in general that when the air temperature was much lower than
that of the soil the soil temperature in ground covered with snow was
higher than in bare ground. He found that temperatures taken at a
depth of ID cm. in the former were the same as those taken at a depth
of 27.4 cm. in the latter. After the early part of April the minimum soil
temperatures at Rocky Ford and Madison agreed closely, although the
minimum air temperature at the former place remained generally the
lowest of the temperatures recorded.
Temperatures obtained from the interior of a pile of hayed sugar-beet
leaves by means of a soil thermograph buried in the pile varied less than
temperatures taken outside the pile, as shown by the following records
made on May 8,1913, and as was probably the case during the entire winter
season : Temperature inside pile, maximum, 67° F. ; minimum, 58°; differ-
ence, 9°. Temperature outside pile, maximum, 84° F.; minimum, 45°;
difference, 39°.
In view of the fact that the fungus lived twice as long inside the pile as
it did on the outside it would seem that a more uniform temperature
might be regarded as one of the controlling factors in the life of the fungus.
28
Journal of Agricultural Research
Vol. VI, No. 1
\
■^
^^
"*^
"^
^^
•-,
■^•^
"^
~>
1
^.
Iv'
C3
^.
S
■>
y
v
^
>
''
^
^
...^
>
■*\
''^
j^
^>~
——
'~
:.Tr.-.
•TVX».
.«
r
—
" /
/^-<
.'— ■
•ii-^
p--*^
~-^
':;;^
«<.'._
-.^
"
H
'-^
s^
~«..
V^:-'
};
i^'r
_^
:»
?/
-*««
^
<t--
TT."?
^"*'
— •
■ ■■<
'v
^___
■
r"
"^
K
••.-.:
>';'
^■'
k'
\
. ^
\
"•^
>
t"""
""v
— >
^
L.
^
^Nfc,
\
i
/■^
*"*'
■^
---.
-TB**
y
>
1?»-
-^
"**•*.*
'"^.
•<;
:^
JB*.
■^s
•-V
•-...
,.^-
_>
~i
'1^
>^
-f jr-«
.■^"i
«:»•
^.-
.s
>>
$t
-
"=.'.'
— .
•"""•—,
;«•
1
■• —
-^
►•
'K..
^
-.,.
....
/
,. — ■
^
^•^'
— --
...-
-••■-
_^_^
T!f^
?<r
....'
<*'■
<r
— —
\"
---
X
~~
\
— --
•'•V.
V.
•^ •
— s
n
i!
— ..
-f-
.-'•
<;
^=^
i
^
-^
s ,
~^
-■••.;
x*'
;;:::=
-
^
-;■
^
»'•»"
-==:
_^_^
^
^
.\
J>
•'
^^
— ■
—
'<
^!r
r
s?::
* — '
k
^
\
.—•■•'
•■■'
>
\
■"m^
^i!*
^'f.-.-
■ ■
—
~?^
""
— ^
»-\-
— .".
'^'■n.
"•«
-•^A
>
---
-Tir
-V!
— — ■
-_.
^.
-«f:
.*.-">■
V^
i~:
■ -^
i^ —
.-«»'
\^
=^
r— ■
v-_
1
••
-••"
«=
1
^
V
-—
1
^
^>
K
V
»
>
•5r
--:;
1
nil
1 1
11
1
.
=c:
""^
<
;^"^
•>»,
\
-ir*^-
.;,.>
>
1 b
^
^^
"
)
1
N
\
"v.
■-•:
^
^ h
-
- —
\
>
>
'
^
*
r
1
*0
>^
s,
*s
p
__^
X^'
, *i»
f''
:;.■:::
^
;
7^
t^
-
V
,<-•
y...
::-•••-
^
r
i:::^
■:^<
S
=-...
.
! 1
I !
1 1
1^
<■
-^
-f
-*/
A
■"
— »«.
— -
r\'
;^
\
-•-
- — '
J>
1
>
^'
•-^'»
••-.
-^
;>
1
"^
;:^
^C
•"'
-Jw
= 'x
>■
K <o Vi tk CIO) fV* N
> <Vi aA//70i/S
e/ — I
//
s- t
e/
%
oe
\9/\
7#/
o/
e
ee
se
c/
//
"I
•2 J=
12
.a u
a o
Apt. 3, 1916
Climatic Conditions and Cercospora beticola
29
c
— s
^
.-•••
""■>
^/ —
S/
e/
//
■\
c
/
09
9^
^^
■^^
^^
9/
s^\
■^/\
e/
0/
9
9
e
se
9^
^^
<?<?
oe
9/
<?/
9
9
rf'
c'
/:^
eA
s/
c/
^
5
i^
'"•
/'•
>
S^
••*
^^'•:
'^'
<
^
■■■
/
\
"n^
p.
.--
...
■
^.
^
r^
1 —
'•^
"^
lllillll-
llliii-
1^
<'
r"'
—
T
^
.^
?-
i.^
\
^**''
"i.
....••
-C
v>
S \
,.-'
<;
^1
•
.-•»=^
''^.
/
r -^
1 "
C".
<i'
,
L.
(
""
> ^
^^
'^
'•>«.
r'^
-.^
i->
{
^
.^
—
—
>^
k. ••r-
^'
N
'\^
ij-
*
k ,j
<♦;
"
=^
-7
v.
'^•
*J?-
(
»^
t>
->«»
^ '
V
11:
11;
1 1 :
i 1 '
i!!-
i: 1 1.
*-*
v^
^
V.
'•■**?
■stv-'
S7
r
\
"^•
.^ /
...^
*
dC.
■-.^
^x
,^
"».;.«.
.■?!»'•
*>-•
;^ss
^-T
<.
'*•:<
.^
<
<
1
\
>
>
>
>-
^^
A'
/ -
■•••.,
^
>--
* ^_
3*'
Z"^*
"*ii
-"v-
/^
>
r
1'-
'?
•
K
/
>^
^^••■^
.'-^;
\
^
^
1
•"^
■^
■•••■
-..
^
:::;:^
^
<
I..I
•^
-■•..
;;.
^
^
b
.»•
}?.T:
^•
■■■
'^^
iTT
^^
"T^-
-.,
.,_
^/<
._
__
ptH
^
^'■Jst
J.'-* —
._
_•'
•<
-i"^
„.^
"^^
"«*v
'<\
<
■<
V^
\
V,
V
^
-^
"•^^
^C
v^
'•^^
•.-.
ts
^^
— -
^T"''
F-^
-.,
* } ■■•
■
^
f'i
J .^'
■--
L.^
1
•
5=S!
<'"
—
<-
.^-
--•'
L
U
r
^
""«
-i-;
X'}
r^;^"
' .T.v
^
^
^
• ** __
'J^
iiX
•' -K-
/
'V
k
s"
^
%
i-
'
c
r^
/
■ >
-'!!'
l
— ^
^
\
'"*■•■
4
■I.
\
?*•
t-^
1-
•■-
..^
:^
■^x
T
- '.~-.
^*V
-=:
::::tl
is^
■■-T)!
-ts.-
V
-^iS
-N...
T.-.-
■'-i
c
— "
L
'-^
«> ""^ ■■
__/
1
■
""■
■Abh
r:*
s^
'•'''
> ^'
_.-
■~-
>B9
2-
rrs
Wrt
f
9
30 Journal of Agricultural Research voi. vi, No. x
Low temperatures are not entirely inhibitive, as was shown by thermal
tests of artificial cultures. After such cultures had been exposed to
temperatures averaging 0.9° C. for 48 days and then kept at 28° C,
numerous colonies developed. Also, heavily infected leaves kept at
0.9° C. for 97 days yielded good growth when cultures were made and held
at favorable temperatures. Had the cultures been exposed to freezing
temperatures or to extreme variations in temperature, the effect would
doubtless have been more pronounced.
Although the temperature variations and the amount of soil moisture
at Rocky Ford and Madison differed greatly, the effect on the life of the
fungus was apparently the same at both places. It may be concluded
that conditions of the soil which favor the process of disintegration are
the most important factors in the control of the disease, and these ex-
periments indicate that these processes are most active at a depth of
6 to 8 inches.
SUMMER CLIMATIC CONDITIONS
The summer climatic conditions here considered were recorded during
1 91 3 in fields of first-year sugar beets grown at Rocky Ford, these fields
being an example of the usual progress of the disease where neither ro-
tation nor sanitation at the preceding harvest time had been practiced.
A study of the temperature and humidity records taken at different
places in a beet field at Rocky Ford and at the Weather Bureau station
3 miles from the field was made to determine their comparative values
in making important correlations. The records made in the sugar-beet
field were taken by means of hydrothermographs kept in meteorological
instrument shelters 5 feet above the ground (PI. IV, fig. i) and among
the plants (PI. IV, fig. 2) . These were checked at frequent intervals with a
sling and cog psychrometer (Shaw, 1914), respectively, and under Col-
orado conditions were found to be accurate. The records of the Weather
Bureau station were taken by means of maximum and minimum ther-
mometers kept in an instrument case about 5 feet above the ground in
an open space (fig. 3).
The daily maximum and minimum temperatures and humidities,
together with the total number of hours the humidity was above 60
from noon of the preceding day to noon of the given day, are used in
the present interpretations. It has been found that when a high relative
humidity prevails, the stomata of the sugar-beet leaves are usually open;
and as the fungus enters the leaves only through the open stomata, the
length of time they remain open is a fundamental factor in determining
the possible occurrence of infection (Pool and McKay, 191 6).
AIR TEMPERATURE AND RELATIVE HUMIDITY
The temperature and relative humidity taken with hygrothermographs
placed near the ground among the plants varied widely from those taken
with hygrothermographs in the air above the field and also from those
Apr. 3.I9I6 Climatic Conditions and Cercospora beticola 31
taken at the Weather Bureau station ; hence, the place where the records
were taken for use in the present correlations with the development of
the disease is an important consideration.
Air temperature. — At Rocky Ford the maximum temperatures taken
among the plants near the surface of the ground from June 13 to 30
ranged from 2 to 19 degrees higher and the minimum temperatures
generally from i to 14 degrees lower than those taken at 5 feet above
the ground (fig. 4). This was due to the fact that, the plants were small
during this period and covered only a portion of the ground; conse-
quently during the daytime the temperature of the soil became higher
than that of the air, and in turn the temperature of the air near the
ground became higher than that of the air a few feet above. During the
night the reverse occurred, the surface soil losing its heat by radiation
and conduction faster and finally reaching a lower temperature than
that of the air in contact with it, after which the heat of the latter
gradually passed into the soil and as a result the temperature of the air
immediately above the ground eventually became lower than that a few
feet higher up. It is possible that convection currents also tended to
lower the temperature of the air immediately above the ground; for, as is
well known, when it is not disturbed by other factors, the coolest air
settles to the lowest levels.
The maximum temperatures of the air near the ground, as shown by
the records, were higher for a longer period during June than at any
time during the season, varying from 100° to 106° F. on nine different
days between the 14th and 26th of that month and rising above 100°
only once thereafter, on August 16. The maximum temperature of the
air 5 feet above the ground, on the other hand, was lower during June
than during the middle of the season, ranging from 90° to 93° on six
different days during the month, while it was above 90° and some-
times as high as 100° on 12 different days during July.
As shown by the records, the temperature of the air near the ground
among the plants was lower during the middle than during the early
part of the season. This was probably due to the difference in the size
of the plants, the larger plants practically covering the ground in mid-
season and preventing the heating of the surface soil, while early in the
season the smaller plants covered the ground but sparsely and con-
sequently afforded less protection against heating. Comparison of the
records also shows that during the middle of the season the temperature
of the air among the plants near the ground was practically the same as
that of the air 5 feet above the field and that throughout the entire period
the latter was quite comparable \vith the temperatures taken at the
Weather Bureau station (fig. 4).
A similar marked variation was shown at Madison, the maximum tem-
perature there being almost constantly higher and the minimum tempera-
27469°— 16 3
32
Journal of Agricultural Research
Vol. VI. No. 1
k so
I
\
so
\f
*■
>
\
f
^4. J
A
■
W
v^
\
j^
\
•'/S X.*!
\
rf
»
h
i
1^
d
\
1
V
/
A
._
\
i
i^tf
;
<-4
^0^
-
V
V
■ '
V
/v
^1
Vi
\
h
\
\
(
i
V.
^Oh
V
\
\
/'
\
^
1
"4
f
F
y
}
2
1
:••
>
\;,
L
/
^
\
\
\
A
y?
<
/
\
iK
\/
J
\
.^
^.
>>/
>
M
V'
/\
/
C!s
^
/v
\aN
)S.
A
\/
M
r
^.
<^
S3
'J
^^
V
V
**
u
"'
o-
1
1 '
_
L
J
_j
a to ^
<u .2 9
> t; Pi
.Q .Sf -M
C3 Ifl "
a =3 «-
>..s 3
G cj O
3 <a +->
J 1-1
3 ^ -a
O CI 3
V JS l-i
^ -■ -
Apr. 3. 1916 Climatic Conditions and Cercospora beticola 33
ture usually lower among the beet plants than the temperature shown by
the Weather Bureau records, which were taken on top of a four-story
building about a mile from the sugar-beet field. These wide variations
between the air temperature taken near the ground among the plants
and that taken 5 feet above the field and between the former and the
temperature taken at the Weather Bureau stations show that for correla-
tion with fungous activities only the records taken among the plants
should be used.
Relative; humidity. — There was also a wide variation in the humidity
near the ground among the plants and 5 feet above the field. For instance,
the daily minimum humidity at Rocky Ford from June 13 to 29, with two
exceptions, was higher and remained above 60 generally for a longer
period in the air above the plants than among the leaves near the ground
(fig. 4), owing to the higher temperature at the surface of the ground as
a result of the small amount of covering afforded by the young plants.
During this period the daily variation of humidity among the leaves was
extreme, ranging from 99 to 10 on June 13, from 99 to 16 on June 25, and
from 100 to 8 on July 2. After June 29, on the other hand, the minimum
humidity was generally higher, the humidity remained above 60 for a
longer time among the leaves than in the air above, and the daily varia-
tion among the leaves was less extreme than earlier in the season. These
conditions were due mainly to the greater amount of covering afforded by
the larger plants and consequent longer retention of moisture among the
leaves. The humidity both among the plants and in the air 5 feet above
the field remained, on an average, above 60 for a longer time each day
during midsummer than during June, owing in part to the increased use
of irrigation water as the season advanced and the increased amount of
moisture in the surrounding air resulting from the increased transpiration
of the larger plants.
Comparison of the Madison and the Rocky Ford records (fig. 5) of the
number of hours that the relative humidity remained above 60 each
day among the sugar-beet plants shows that throughout the season it was
higher, on an average, at Madison. Here it remained above 60 for a
longer time each day during the latter half of June, when the records
were started, and for a shorter time each day during August than during
any other summer month. This was due to difference in the amount of
rainfall, there being frequent rains during the former period and com-
paratively dry weather during the latter. At Rocky Ford the facts were
reversed, the humidity remaining above 60 for a longer time each day
during midseason than during the latter half of June or the first part of
September. This was probably due to more frequent irrigation and the
increased covering afforded by the larger plants of midseason.
34
Journal of Agricultural Research
Vol. VI, No. I
Table III showvS the average number of hours a day that the relative
humidity remained above 60 at Madison and at Rocky Ford.
Table III. — Average number of hours a day that the relative humidity was above 60 at
Madison, Wis., and Rocky Ford, Colo., during ihesummer of igi4and igij, respectively
June 16 to 30
July
August
September i to 6
Seasonal average
Madison,
Wis. (1914).
Hours.
19.4
17-3
16.6
18.0
17.4
Rocky Ford,
Colo. (1913).
Hours.
10.8
14.2
14. I
II- 3
13-4
— /oo
I.
s eo
so
A^
"o,
sSy^
1 M 1
f
~
/
/^
r-
r
A
^^■"^
-""i-^^
^]
\
.
\
/
v
A
\
-.
\
\
<.
u
'i
f
k^
i
\\
ll
'\\
n
'\
/,
, >
\
f
"^
\i'
t
A
K
\
,f
1
If
If
\
1
1
v\
r
\
/
i^
Ir
3'
\.
A
k
^
4
/
V
-
\
1
La
/
'
V
•3
\
1
1
I
1
»
1
1
V
^/
1
,'
\
/^'
W
>•
'■<^'^^.
."
1
^,
A
'
»
\
1
\
r
A
\
\
k
^^
l\
'n
-A
^
fl
V
^
A
A
/
ly
>^>>
V
V
-r
f
A
V
1
'\
^
\/i
/
>*1
1
W
\
/
\\
1
I
f
\
■>
s
1
1
i'
""
v:
\
«
V
'
^
^
V
j
i
^
'
/
I
1
c
W
->
n
\
V
\
1
V,
"C
'\
^
^',
7
\
f
1
\
/
\
/
7
1
II-
c
J
\
'
1
/
\
n
1
1
>
i
]
i
A
A
, A
h
\
f\
/
./
fy
\
1
I
\
1
1
1
V
f
i\
A
/
I
A
1 \
/'
^!
^
1
\/
V
I
^
»'
\l
/S
,1
I
\
yl
/
/
7
1
\
^
\
1
j
1\
/)
-/
K
^
\
'h
1
1 1
/
1
V
V
\
\
>
V
■^
1
1
1
\
)
\/
>
/
1
1
1
'•,
/T
1
1
1
/
1
V
V
\\,'
V
*.__
,<
1
t
1
1
1
1
1 1
'
\
1
v'
A
^A
0/
sc
^
V
1
,.
1
-
1
r
1.
1
7"
,r
H
T
T
r
r
I
7-
r
- ■
1
Fig. s. — Curves of the maximum and minimum temperatures among sugar-beet plants and at the Weather
Bureau station, and the seasonal rainfall records at IMadison, Wis., in 1914, and the number of hours
that the humidity remained above 60 aipong the sugar-beet plants in the field at Madison, Wis., in
1914, and at Rocky Ford, Colo., in 1913.
The greater average number of hours of high humidity at Madison
accounts for the periods of extreme infection which occurred there when
the fungus was present. Here leaves badly infected with Cer cos para
heticola and entirely covered with conidia were found at times, but this
condition was rarely seen at Rocky Ford. There were numerous cases
of cotyledon infections also at Madison, the high humidity early in the
season favoring their occurrence; but no such infections were found at
Rocky Ford.
Apr. 3 1916 Climatic Conditions and Cercospora heticola 35
RAINFALIv AND IRRIGATION
The rainfall records made during the summer season of 191 3 in the
beet field at Rocky Ford in which infection was studied in detail (fig.
4, 7) where obtained by means of a rain gauge placed at the edge of the
sugar-beet field (PI. IV, fig. i). Most of the rain was in the form of local
showers, the amount varying greatly within a radius of less than 2 miles;
but occasionally general rains fell. The efifect of the increased relative
humidity resulting from rainfall usually lasted longer among the leaves
than in the air 5 feet above (fig. 7).
The effect of irrigation on humidity was found to be similar to the
effect of rain. On July 2, before the field was irrigated, its humidity
was as low as 8 and on July 3 and 4 remained above 60 for 7 and 6 hours,
respectively. On July 4 and 5 the field was irrigated and the humidity
remained above 72 on the 4th and above 60 during 23 hours of the 5th.
On July 27 the field was again irrigated and the humidity remained
above 60 for 15 hours that day and 21 hours the following day. On
August 19 and 20 the field was irrigated the third time and the humidity
remained above 60 for 12 and 13 hours, respectively, and the next day 21
hours. The general humid conditions necessary for leaf spot infection,
however, are developed much better by rain than by irrigation, because
of the latter being comparatively local and unaccompanied by the
atmospheric conditions attending rainfall.
WIND
Records of wind velocity at Rocky Ford were taken by means of an
anemometer placed 6 feet in the air at the edge of the beet field (PI. IV,
fig. i), the readings being made at irregular intervals and the velocities
computed being the hourly averages from one reading to the next. As
the records were not made daily, accurate hourly velocities for different
intervals during the day can not be obtained from the records. They
show, however, that the average seasonal velocity from June 12 to Sep-
tember 22 was 5.3 miles per hour. Occasionally two daily readings
were made, one in the morning and the second late in the afternoon.
These show that the average velocity of the wind was always higher
during the day than at night, the greatest velocity usually prevailing
in the afternoon during the period of lowest humidity. While no gen-
eral dissemination of conidia was correlated with high wind velocity,
the afternoon combination of highest wind with lowest humidity ap-
parently favored the dissemination of conidia. In fact, in the case of
air cultures made at different times during several days, it was found
that the fungus grew usually only on those exposed during the afternoon.
36 Journal of Agricultural Research voi. vi. no. i
SUMMER INFECTION CYCLES
The thermal relations of the fungus are closely linked with the effect
of various climatic factors on the production and dissemination of co-
nidia and on infection cycles. With a view to determining these relations
the fungus was grown in Petri-dish cultures in thermostats at different
and varied temperatures. At first the moisture was probably more or
less constant, but as time went on it became relatively low. The effect
of different temperatures, however, was comparable with that observed
under existing field conditions.
THERMAL RELATIONS OF THE FUNGUS IN CULTURES
Tests of the fungus on string-bean agar were made at Washington
during November and December, 191 3, and January, 191 4. The cul-
tures were obtained from isolations made at the time of the tests from
infected sugar-beet leaves collected at Rocky Ford during the preceding
September. One colony of the first isolations was macerated in 10 c. c.
of sterile water, and one platinum loop of this suspension was used for
each tube of medium. Three poured plates were used for each single
test. The cultures were exposed to different constant temperatures and
to varied constant temperatures (high and low changed to low and high,
respectively) . Exposures were also made for 8 hours at the higher tem-
peratures and then for 16 hours at lower temperatures, and, after a short
interval of exposure in a certain number of these tests, both temperatures
were lowered, it being possible in this way to approximate night and
day temperatures in the field under normal conditions.
Series A (different constant temperatures). — When the cultures
were held at different constant temperatures, the abundance and size of
the individual colonies gradually increased, while the time necessary for
development decreased with the temperatures 12.5°, 17.3°, 19.2°, 20°,
and 30.8° C. The best growth was made at a temperature of 30.8°, but
this in all probability was slightly above the optimum constant tempera-
ture, as no growth took place in cultures held for 9 days at 34.7°, 35.8°,
and 40.6°, respectively (Table III, series A).
Series B (varied constant temperatures). — ^Although no growth
of the fungus took place in cultures held at constant temperatures of 34.7°
and 35.5° C, a small percentage of normal colonies developed in cultures
exposed for three days to these temperatures and then for several days
to a temperature of 30.8°, while in cultures exposed for three days to
40.5° no growth occurred when subsequently held at 30.8°. On the
other hand, in cultures exposed for three days to a temperature of 30.8°
there was almost a normal development of the colonies for three days after
they were exposed to 34.7°, but at the end of five days the inhibitive
effect of the latter temperature became manifest. In the case of cultures
37
Apr. 3. 1916 Climatic Conditions and Cercospora beticola
changed from 30.8° to 40.5° only a very slight increase in growth was
apparent during the first three days at the higher temperature, and after
that it ceased entirely (Table III, series B).
Table III. — Comparative diameter of colony growth (in millimeters) of Cercospora
beticola at different constant temperatures, o^ at decreasing and increasing constant tem-
peratures, and at daily varied temperatures
SERIES A, CONSTANT TEMPERATURES
Maximum
Minimum
Average
Period of growth
3 days
6 days
9 days
II days
14 days
18 days
23 days
Diameter of colony growth at temperature (°C.) of —
a. I
03
7-9
4.0
8.9
14.4
II. S
19. I
16.3
20. 2
18.6
21.0
19-5
33- o
29. 7
30.8
36. 2
33-8
37-4
34- S
41. O
40.0
40.6
o. 65
1-5
2.8
4.2
o. 67
2-3
4.6
6.4
3-6
6.4
7
" The temperatures of each thermostat for all tests were averaged from two daily readings continued
throughout the time of the experiment.
^ No growth occurred in these plates when held at 28° C. for 10 days.
SERIES B, DECREASING AND INCREASING CONSTANT TEMPERATURES.
Diameter of colony growth at tempera-
ture CO.) of—
°34.7 "35-5 "40-S *30.8 "30.8
Period of growth
3 days
6 days
8 days
'^8.3
3-6
'9-3
3-6
6
6.4
" Temperature changed to 30.8* C. after three days.
f> Temperature changed to 34.7° C. after three days.
« Temperature changed to 40.5° C. after three days.
<* Only 19.2 per cent of the normal number of colonies developed.
* Only 12.8 per cent of the normal number of colonies developed.
SERIES C, DAILY VARIED TEMPERATURES
16 hours at.
8 hours at .
Diameter of colony growth at temperature (°C.) of —
14.5
19. 2
14- S
21. 6
30.8
35-8
40. 6
Period of growth:
3 days
sdays
7 days ,
9 days
0.4
• 7
2.7
0.5
1.9
3-6
0.7
3-4
5-3
1-7
6
8.8
0.6
3.6
4.8
o. 6
3a
S-4
Series C (daily varied temperatures). — In these tests the tempera-
tures were made to correspond closely with summer outdoor temperatures
of night and day by holding the cultures for 1 6 hours at the lower and for
8 hours at the higher. After seven days' exposure the growth of colonies
38 Journal of Agricultural Research voi. vi. No. i
on cultures exposed to temperatures of 14.5° and 19.2° C. averaged 2.7
mm. in diameter; after exposure for the same length of time to 14.5° and
21.6°, 14.5° and 28°, and 14.5° and 30.8° the growth gradually increased
until it reached a maximum diameter of 8.8 mm; but when exposed to
higher temperatures (20° and 34.7° or 20° and 35.8°) the growth grad-
ually diminished until finally it equaled approximately that attained
under 14.5° and 28°. There was no growth on cultures exposed for nine
days or longer to 20° and 40.6° (Table III, series C).
SERIES D (high varied changed to low varied temperatures) . — A
plate culture exposed for three days to temperatures of 20° and 40.5° C,
being held 16 hours at the lower and 8 hours at the higher, and then for
six days at 20° and 30.6°, developed 23 colonies, averaging 10.3 mm. by
the end of the latter period, while a check plate exposed constantly to a
temperature of 30.6° developed 100 colonies by the end of the latter
period. A plate exposed to the higher temperatures — 20° and 40.5° —
for five days and then held at 20° and 30.6° for six days developed six
colonies at the end of the latter period, while a plate exposed to 20° and
40.5° and then held at 20° and 30.6° for seven days developed no growth
of the fungus.
Later on in this paper the fact that high minimum and maximum tem-
peratures inhibit the growth of the fungus, as brought out by these tests,
is correlated with the effect of existing high field temperatures, with
their consequent accompanying factors, on the leaf spot. Although
the optimum temperature variations — 20° and 30.8° C. — were found to
be very favorable to the development of leafspot in the field, little or
no increase in the disease was observed to follow high night and day
field temperatures — 20° and 40.5°, respectively.
It was also observed that different temperatures affect conidial septa-
tion. The normal average septation varies from 6 to 11, but during
warm, humid periods the conidia were usually found to be many septate,
sometimes as high as 20-septate, while after a cooler period, such as
usually occurs in September, they were only from 2- to 4-septate.
relation of conidial production and dissemination to climatic
conditions
For the purpose of studying the relation of temperature and relative
humidity to the production and dissemination of conidia, detailed life
histories of a large number of individual spots on 10 plants in the medium-
early field at Rocky Ford were kept during the season of 191 3. The tem-
perature and humidity records used in these correlations were those
taken among the beet leaves near the ground and, together with rainfall
and dates of irrigation, are shown in figure 7.
Beginning with the outermost or oldest, the leaves were tagged and
numbered consecutively, and the location of the spots on each was indi-
Apr. 3, 1916
Climatic Conditions and Cercospora beticola
39
cated on diagrams. As new leaves developed, they were included in the
observations, and this was true also of new spots, until they became too
numerous, after which only a few representative ones on each leaf were
studied in detail. During the period from the 24th of June to the 19th
of September 330 spots were studied, both surfaces being examined at
frequent intervals with a hand lens. For the purpose of getting a basis
for comparison of rates of development at different stages in the life
history of the disease, percentage values were assigned to each stage as
follows, the spots being grouped and averaged later (Table IV) :
Percentage value. Stage of development of fungus.
5.0 Spot first noticed. Neither conidia nor conidiophores present.
12.5 Conidiophores present.
19.7 Very few conidia.
25 Few conidia.
31.2 Conidia fairly numerous.
37.5 Conidia numerous.
43.7 Conidia fairly abundant.
50 Conidia abundant.
The value of the spot is the sum of the values of the two sides— ^that
is the value of a spot on which there were but few conidia (25) on one side,
and abundant conidia (50) on the other, is 75. Again, the value of a
spot on which conidiophores only (12.5) were present on one side, and very
few conidia (19.7) on the other, is 32. '^
Table IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of 191 3 b
Plant
No.
Leaf
No.
Spot
No.
Date.
Graph
values.
Data on life histories.
**2
3
5
1913-
July 7
10
No conidiophores on either surface. July 8, no change.
10
37
Conidiophores above, conidia few below.
12
5°
Conidia few on both surfaces. July 14, no change.
16
SO
Conidia very few above, fairly numerous below.
21
100
Conidia abundant on both surfaces. July 23, no change.
25
62
Conidia fairly numerous on both surfaces.
28
37
Conidia none above and few below.
2
4
I
3
10
No conidiophores on either surface. July 7, no change.
8
24
Conidiophores on both surfaces.
10
69
Conidia very few above and abimdant below.
12
75
Conidia few above and abundant below.
14
100
Conidia abundant on both surfaces. July 16, 21, 23 (leaf yel-
low), July 25, no change.
28
63
Conidia numerous above and few below.
a
4
3
7
10
No conidiophores on either surface. July 8, no change.
10
25
Conidiophores oa both surfaces.
12
62
Conidia fairly numerous on both surfaces.
14
82
Conidia numerous above and fairly abundant below.
21
100
Conidia abundant on both surfaces.
2
4
4
7
10
No conidiophores on cither surface.
8
25
Conidiophores on both surfaces.
10
75
Conidia few above and very abundant below. July 14, no
change.
16
57
Conidia very few above, numerous below.
o For convenience the decimal fractions, which make only a negligible difference in the averages, are
omitted.
<> In Table IV asterisks (*), daggers (t), and section marks (§)are used to designate definite leaf spots 10
which reference is made in the text.
40
Journal of Agricultural Research
Vol. VI, No. I
TablB IV. — Data on life histories of a representative number of leaf spots studied on
lo plants in a tnedium-early sugar-beet field near Rocky Ford, Colo., durinq the season
of jpij — Continued
Plant
No.
No.
Spot
No.
Graph
values.
Data on life histories.
1913.
July 7
June 24
July
10
62
12
100
16
69
21
25
7
10
10
25
12
100
23
94
25
5°
7
10
10
25
12
100
16
94
21
100
25
5°
7
10
10
25
12
100
16
94
21
100
25
75
23
20
25
44
28
50
June 24
July 8
12
62
14
50
June
10
10
July
7
30
10
25
la
100
10
30
12
50
14
50
t6
SO
Conidiophores forming above, nothing below.
Conidiophores above, conidia very few below.
Heavy production of conidia on both surfaces. July 12, 14, 16,
21, 23, no change.
Conidia few above and numerous below.
No conidiophores on either surface. June 25, 26, 27, 28, 30,
July 1,2,7 (leaf yellow), 8, no change.
No conidiophores on either surface. June 25, 26, 27, 28, 30,
July I, 2, 7, 8, no change.
No conidiophores on either surface.
Conidiophores forming only on lower surface.
Conidiophores abundant above and conidia abundant below.
Conidia abundant on both surfaces. July 14, no change.
Conidia very few above and abundant below.
No conidia on either surface.
No conidiophores on either surface. July 8, no change.
Conidiophores abundant on both surfaces.
Conidia abundant on both surfaces. July 14, 16, 21 (center of
spot gone), no change.
Conidia fairly abundant above and abundant below.
Conidia few on both surfaces, leaving and leaf yellowing.
No conidiophores on either surface. July 8, no change.
Conidiophores abundant on both surfaces.
Conidia abundant on both surfaces. July 14, no change.
Conidia fairly abundant above and abundant below.
Conidia abundant on both surfaces. July 23, no change.
Conidia few on both surfaces.
No conidiophores on either surface. July 8, no change.
Conidiophores abundant on both surfaces.
Conidia abundant on both surfaces. July 14, no change.
Conidia fairly abundant above and abundant below.
Conidia abundant on both surfaces. July 23, no change.
Conidia few above and abimdant below.
Conidiophores numerous above and few below.
Conidia few above and very few below.
Conidia few and matted together on both surfaces.
No conidiophores on either surface. June 25 (leaf dying), 26
(leaf dead), 27, 28, 30, July i, no change.
No conidiophores on either surface.
Conidia fairly numerous on both surfaces.
Conidia leaving, few on both surfaces. July 16, 21 (leaf dead),
no change.
No conidiophores on either surface. July i, 2, no change.
Conidia forming on both surfaces. July 8, no change.
Only conidiophores on both surfaces.
Conidia abundant on both surfaces. July 14, no change.
Conidia fairly abimdant on both surfaces. July 21, 23, no
change.
Conidia few on both surfaces.
Conidia few on both surfaces. July 23. no change.
Conidia numerous on both surfaces.
No conidiophores on either surface. July 2, 7, 8, no change.
No conidiophores on either surface. July 7. no change.
Conidiophores forming only on upper surface.
Conidiophores above and conidia forming below.
Conidia few on both surfaces.
Conidia very few above, and fairly nimierous below.
Leaf dead, no change in spot.
Apr. 3, 1916
Climatic Conditions and Cercospora heticola
41
Table IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Plant
No.
Leaf
No.
Spot
No.
Data on life histories.
1913-
July 7
June 24
27
July 7
16
21
88
75
100
63
75
100
No conidiophores above, very few conidia forming below.
Conidiophores on upper surface, few conidia on lower.
Conidia numerous above and very abundant below.
Conidia few above and abundant below. July 14, no change.
Conidia abundant on both surfaces. July 23 (leaf yellow).
July 25, no change.
Conidia few above and numerous below.
Conidiophores on both surfaces.
Conidia few above and conidiophores below. July 12. no
change.
Conidia few on both surfaces.
Conidia abundant on both surfaces. July 23. no change.
Conidia few on both surfaces.
No conidia on either surface.
Conidiophores none above and forming below.
Conidiophores on both surfaces.
Conidia few above and very abundant below. July 12, 14, 16,
no change.
Conidia abundant on both surfaces. July 23, no change.
Conidia very few above and numerous below.
Conidia none above and few below.
10 No conidiophores on either surface.
17 Conidiophores none above and forming below.
100 Conidia abundant on both surfaces. July 23, no change.
Conidia abundant above and few below.
s6
62
100
IS
75
5°
100
Conodiophores on both surfaces.
Conidia numerous on both surfaces.
Conidia very few on either surface but more on lower. July 8,
no change.
Conidia few above and abundant below.
Conidia abundant on both surfaces. July 14. 16, 21. no change.
Conidiophores forming above and conidia few below. July 8,
no change. .j. .. j ^
Conidiophores well developed above and conidia abundant
below.
Conidia very few above and abundant below.
Conidia few above and abundant below. July 16, no change.
Conidia abundant on both surfaces. July 23, no change.
No conidiophores on either surface.
Conidiophores above and none below.
Conidia abundant on both surfaces. July 23, no change.
Conidia few above and numerous below.
Conidiophores on both surfaces.
Conidia very few above and abundant below.
Conidia few above and numerous and matted below.
No conidiophores on either surface.
Conidiophores very few above and abundant below. June 26.
no change.
Conidiophores above and conidia very few below. June 28.
30, July I. 2 (leaf dying), no change.
Conidiophores, but no conidia present.
No conidiophores on either surface.
Conidia few above and fairly numerous below.
Conidia none above and abundant below.
Conidia abundant on both surfaces. July 14, 16, 21, 23. ^5.
no change.
Conidiophores forming on upper surface only.
Conidia abundant above and few below.
Conidia few on either surface. July 16, no change.
Conidia abundant on both surfaces. July 33, 25, no change.
No conidiophores on either surface.
Conidia abundant on both surfaces.
July 23, 25, no change.
42
Journal of Agricultural Research
Vol. VI, No. I
Table IV. — Data on life histories of a representative number of leaf spots studied on
lo plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Plant
No.
Leaf
No.
Spot
No.
Date.
Graph
values.
Data on life histories.
*7
6
8
1913-
July 21
56
Conidia few above and fairly numerous below. July 23, no
change.
25
100
Conidia abundant on both surfaces.
*7
■ 8
5
21
37
Conidia very few on both surfaces. July 23, no change.
25
75
Conidia numerous on both surfaces.
28
50
Conidia few on both smfaces. July 30, no change.
7
8
6
21
50
Conidia few on both surfaces.
23
56
Conidia few above and fairly numerous below.
25
63
Conidia numerous above and few below.
28
SO
Conidia few on both surfaces. July 30, Aug. i, no change.
*7
8
7
July 23
IS
Conidiophores none above and few below.
25
75
Conidia few above and abundant below. July 28, 30, no
change.
Aug. I
SO
Conidia none above and numerous below.
7 8
12
July 25
25
Conidiophores on both surfaces.
28
45
Conidia few forming on both surfaces.
30
SO
Conidia few on both surfaces. August i , no change.
7
IS
3
28
25
Conidiophores on both surfaces.
30
63
Conidia few above and numerous below.
Aug. I
88
Conidia numerous above and abundant below.
ttts
8
5
July 9
10
No conidiophores on either surface.
10
100
Conidia aburrdant on both surfaces.
*8
12
I
23
SO
Conidia few on both surfaces.
25
75
Conidia numerous on both surfaces. July 28, no change.
30
88
Conidia numerous above and abundant below. August i, no
change.
Aug. s
75
Conidia numerous on both surfaces. Aug. 7,9. n, no change.
8
23
13
5
15
Conidiophores few above and none below. Aug. 7, no change.
9
25
Conidiophores on both surfaces.
II
57
Conidia numerou s above and very few below.
13
100
Conidia abtmdant on both surfaces.
t8
23
18
7
10
No conidiophores on either surface. Aug. 9, no change.
II
60
Conidia abundant above and conidiophores few below.
I?
63
Conidia nmnerous above and few below. Aug. 15, 18, no
'
change.
22
100
Conidia abundant on both surfaces.
t8
24
I
July 28
75
Conidia numerous on both surfaces.
30
62
Conidia fairly numerous on both surfaces. Aug. i, no change.
Aug. 5
44
Conidia few above and very few below.
7
75
Conidia numerous and matted above and numerous below.
8
25
3
I
10
No conidiophores on either surface.
s
44
Conidia very few above and few below.
»
7
S6
Conidia few above and fairly numerous below.
**-+8
29
I
I
10
No conidiophores on either surface.
s
62
Conidia fairly numerous on both surfaces.
7
75
Conidia numerous on both surfaces.
9
50
Conidia few on both surfaces.
II
S6
Conidia few above and numerous below.
13
100
Conidia abundant on both surfaces. Aug. 15, no change.
18
75
Conidia few above and abundant below.
22
88
Conidia numerous above and abundant below. Aug. 25,28,
no change.
30
63
Conidia few above and numerous below. Sept. i, no change
except conidia matted above.
Sept. 3
57
Conidia very few above and numerous below.
8
29
3
Aug. 5
25
Conidiophores on both surfaces.
7
9
30
5°
Conidiophores above and conidia forming below.
Conidia very few above and fairly numerous below.
11
100
Conidia abundant on both surfaces. Aug. 13, 15, 18. 22, 23, no
change.
Apr. 3, 1916
Climatic Conditions and Cercospora beticola 43
Table IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Pknt
No.
*** +3
***8
***f 8
Leaf
No.
Spot
No.
Graph
values.
Data on life histories.
1913-
Aug. s
Sept.
Aug.
tt8
Sept.
75
100
75
69
100
87
75
69
62
13
88
94
18
88
22
100
2S
81
5
35
7
SO
9
88
II
37
13
75
9
44
13
75
18
88
22
63
25
88
28
75
30
88
I
82
3
69
6
5°
5
81
9
69
II
63
13
IS
100
81
22
63
28
30
SO
56
13
lO
IS
18
IS
37
22
25
28
44
62
69
3
62
6
8
50
37
Conidiophores very few on both surfaces. Aug. 7, no change.
Conidia numerous on both surfaces. Aug. 11, no change.
Conidia abundant on both surfaces. Aug. 15, no change.
Conidia few above and abundant below.
Conidia fairly numerous above and numerous below.
Conidia abundant on both surfaces.
Conidia fairly abtmdant on both surfaces.
Conidia numerous on both surfaces.
Conidia fairly numerous above and numerous below.
Conidia fairly numerous and matted above and fairly numerous
below.
No conidiophores on either surface. Aug. 7, 9, no change.
Conidia very few above and nothing below.
Conidia abundant above and numerous below.
Conidia fairly abvmdant above and abundant below.
Conidia numerous above and abundant below.
Conidia abundant on both surfaces.
Conidia abimdant above and fairly numerous and matted
below.
Condiophores few above and conidia few_ below.
Conidiophores niunerous above and conidia numerous below.
Coaidia numerous above and abundant below.
Conidia few above and none below.
Conidia numerous and matted above and numerous below.
Aug. 15, no change.
Conidia numerous above and abundant below.
Conidiophores few on both surfaces.
Conidia very few above and few below.
Conidia numerous on both surfaces. Aug. 15, no change. .
Conidia numerous above and abundant below.
Conidia very few and matted above and fairly abundant below.
Conidia numerous and matted above and abundant below.
Conidia few above and abundant below.
Conidia numerous and matted above and abundant below.
Conidia numerous and matted above and fairly abvmdant
below.
Conidia few and matted above and fairly abundant below.
Conidia very few and matted above and fairly numerous and
matted below.
Conidia none above and fairly numerous and matted below.
No conidiophores on either surface.
Conidia fairly numerous and matted above and abundant
below. Aug. 7, no change.
Conidia fairly numerous and matted above and numerous
below.
Conidia numerous above and few below.
Conidia abundant on both surfaces.
Conidia fairly numerous and matted above and abundant
below. Aug. 18, no change.
Conidia few and matted above and numerous below. Aug. 23,
no change.
Conidia none above and numerous below.
Conidia none above and fairly abimdant below.
No conidiophores on either surface.
Conidiophores few above and none below.
Conidia very few on both surfaces.
Conidia very few above and few below.
Conidia fairly numerous on both surfaces.
Conidia numerous above and fairly numerous below. Aug. 30.
Sept. I, no change..
Conidia fairly numerous and matted above and fairly numerous
below.
Conidia few and matted on both surfaces.
Conidia very few on both surfaces. Sept. 10, no change except
matted on both surfaces.
44
Journal of Agricultural Research
Vol. VI, No. I
TabIvB IV. — Data on life histories of a representative number of leaf spots studied on
JO plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of jpij — Continued
Plant
No.
Leaf Spot
No. No.
Date.
Graph
values.
Data on life histories.
tt,§8
hs
10
1913-
Aug. 13
10
No conidiophores on either surface.
IS
15
Conidiophores very few above and none below. August 18, no
change.
22
37
Conidia very few on both surfaces.
25
56
Conidia fairly numerous above and few below.
28
KO
Conidia fairly numerous above and very few below.
30
88
Conidia numerous and matted above and abundant below.
September i, 3, 6, no change.
Sept. 8
50
Conidia few and matted on both surfaces.
10
37
Conidia very few and matted on both surfaces.
n,§8
§37
I
Aug. 11
24
Conidia very few above and nothing below.
13
100
Conidia abundant on both surfaces. August 15, no change ex-
cept conidia matted above.
18
94
Conidia fairly abundant and matted above and abimdant be-
low.
Conidia numerous above and abundant below. August 25, no
22
88
change.
28
75
Conidia numerous on both surfaces.
30
69
Conidia fairly numerous above and numerous below. Septem-
ber I, 3, no change.
Sept. 6
37
Conidia very few on both surfaces.
8
25
Conidia none on either surface. September 10, 13, is.nochange.
tts
37
2
Aug. II
10
No conidiophores on either surface.
13
100
Conidia abundant above and abundant and matted below.
IS
94
Conidia abundant above and fairly abundant below.
18
87
Conidia fairly abimdant on both surfaces.
22
100
Conidia abundant on both surfaces. August 25, no change ex-
cept slightly matted on both surfaces.
28
94
Conidia abundant and matted above and fairly abundant
below.
30
100
Conidia abundant and matted on both surfaces.
Sept. I
87
Conidia fairly abundant and matted on both surfaces. Sep-
tember 3, 6, no change.
8
S7
Conidia very few above and fairly numerous below. Septem-
ber 10, no change.
13
37
Conidia none above and few below. September 15, no change,
conidia still matted on both surfaces.
tt8
37
3
Aug. II
10
No conidiophores on either surface.
13
18
Conidiophores very few on both surfaces.
15
44
Conidia very few above and few below.
18
50
Conidia very few above and fairly numerous below.
22
62
Conidia fairly numerous and matted above and fairly numerous
below.
23
62
Conidia fairly numerous on both surfaces.
28
75
Conidia numerous on both surfaces. August 30, no change,
conidia matted on both surfaces.
Sept. I
62
Conidia fairly numerous and matted on both surfaces.
§8
37
6
Aug. 18
10
No conidiophores on either surface.
22
5°
Conidia few on both surfaces.
25
62
Conidia fairly numerous on both surfaces.
28
75
Conidia fairly abundant in center on both surfaces.
30
70
Conidia numerous in center above and fairly abundant in center
blow. September I, no change.
Sept. 3
69
Conidia fairly numerous above and fairly abundant in center
below.
6
87
Conidia fairly abundant on both surfaces.
8
75
Conidia fairly numerous and matted above and fairly abundant
below. September 10, no change.
13
94
Conidia fairly abimdant above and abundant below. Septem-
ber 15, no change.
17
50
Conidia few on either surface.
8
38
I
Aug. II
10
No conidiophores on either surface.
13
69
Conidia numerous above and fairly numerous below.
15
69
Conidia fairly numerous above and numerous and matted be-
low.
Conidia fairly abundant above and numerous and matted be
low.
Conidia numerous and matted above and fairly numerous
18
82
22
69
below.
8
39
6
Aug. iS
10
No conidiophores on either surface.
22
63
Conidia numerous above and few below.
25
82
Conidia fairly abundant above and numerous below.
Apr. 3, 1916 Climatic Conditions and Cercospora beticola
45
TablB IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Plant
No.
n,{
Leaf
No.
tts
tt,§8
'Spot
No.
1913-
Aug. 3
Sept.
Aug.
2'i
10
28
24
30
44
Sept.
8
50
10
56
J3
44
15
69
Graph
values.
Aug.
25
28
30
Sept
8
10
17
19
Aug.
25
28
30
Sept
6
10
IS
17
19
Aug.
22
25
28
30
Sept.
3
10
13
IS
17
19
Aug.
Sept.
25
28
30
I
6
8
10
10
69
100
56
Data on life histories.
Conidiophores above and conidia very few below.
Conidia fairly numerous on both surfaces.
Conidia numerous above and fairly abundant below.
Conidia numerous above and abvmdant below. September i.
no change.
Conidia fairly numerous above and fairly abundant below.
Conidia fairly numerous and matted above and numerous and
matted below. September 8, 10, no change.
Conidia very few and matted above and numerous below.
September 15, 17, no change.
Conidia none above and numerous below.
No conidiophores on either surface.
Conidia very few above and nothing below.
Conidia few above and very few below. September i, 3, 6, no
change.
Conidia few on both surfaces.
Conidia fairly numerous above and few below.
Conidia very few above and few below.
Conidia fairly numerous and matted above and numerous
below.
Conidia very few and matted above and few below.
Conidia none above and very few below.
Conidiophores forming above and nothing below.
Conidia fairly abundant in center of both surfaces.
Conidia abundant in center of both surfaces. September 1,3,
6, no change.
Conidia fairly numerous above and numerous below.
Conidia abundant on both surfaces. September 13, 15, no
change.
Conidia few and matted above and fairly numerous and matted
below.
Conidia very few and matted above and fairly numerous and
matted below.
No conidiophores on either surface.
Conidia fairly abvmdant in center on both surfaces.
Conidia fairly abimdant in center above and abundant in
center below. September i, 3, no change.
Conidia fairly abundant and matted above and fairly abimdant
below. September 8, no change.
Conidia fairly abundant above and abimdant below. Septem-
ber 13, no change.
Conidia numerous and matted above and abundant below.
Conidia fairly numerous and matted above and fairly abund-
ant below.
Conidia few and matted above and numerous below.
Conidia very few on both surfaces.
Conidia few above and fairly numerous and matted below.
Conidia numerous and matted above and fairly abundant
below.
Conidia numerous above and abundant below. September i,
no change.
Conidia fairly numerous and matted above and abundant be-
low. September 6, 8, no change except conidia matted
below.
Conidia fairly numerous and matted above and fairly abundant
and matted below.
Conidia few and matted above and fairly abundant and matted
below.
Conidia none above and numerous and matted below.
Conidia none above and very few and matted below.
Conidia none on either surface.
No conidiophores on either surface.
Conidia few above and conidiophores forming below.
Couidia very few above and nothing below.
Conidia very few above and few conidiophores below. Septem-
ber 3, no change.
Conidia few above and very few below.
Conidia few and matted above and few below.
Conidia few and matted above and very few below. Septem-
ber 13, 15, 17, no change.
Conidia none on either surface.
46
Journal of Agricultural Research
Vol. VI. No. I
Table IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Plant
No.
Leaf
No.
Spot
No.
Date.
Graph
values.
Data on life histories.
8
45
s
1913-
Aug. 25
10
No conidiophores on either surface.
28
56
Conidia few above and fairly niunerous below. August 30,
September i, 3, no change.
Sept. 6
65
Conidia few and matted above and fairly abundant in center
below.
8
58
Conidia few and matted above and numerous in center below.
10
69
Conidia few and matted above and fairly abundant below.
13
63
Conidia few and matted above and numerous in center below.
September is, no change.
8
49
I
Aug. 28
37
Conidia very few on both surfaces.
3°
5°
Conidia few on both surfaces. September i, 3 , no change.
Sept. 6
69
Conidia numerous above and fairly numerous below.
8
62
Conidia fairly numerous and matted abwve and fairly numer-
ous below.
10
75
Conidia numerous and slightly matted on both surfaces.
13
56
Conidia fairly numerous and matted above and few below.
8
49
3
Aug. 28
10
No conidiophores on either surface.
30
25
No conidiophores above and conidia very few below, Septem-
ber 1,3, no change.
Sept. 6
56
Conidia few above and fairly numerous below.
8
SO
Conidia very few above and fairly numerous below. Septem-
ber 10, 13. no change.
8
49
3
Aug. 28
10
No conidiophores on either surface.
30
IS
Conidiophores few above and very few below.
Sept. I
30
Conidiophores few above and conidia very few below. Sep-
tember 3, no change.
6
62
Conidia fairly nimierous on both surfaces.
8
65
Conidia fairly numerous above and numerous in center below.
10
75
Conidia numerous above and numerous and matted in center
below .
13
62
Conidia fairly numerous on both surfaces.
8
49
6
I
10
No conidiophores upon either surface. September 3, no
change.
6
44
Conidia very few above and few below. September 8, no
change.
Conidia few on both surfaces. September 13, no change.
10
5°
8
49
7
I
10
No conidiophores on either surface. September 3, no change.
6
24
Nothing above and conidia very few below.
8
44
Conidia very few above and few below. September 10, 13,
no change.
8
49
8
I
10
No conidiophores on either surface. September 3, 6, no
change.
8
20
Nothing above and conidia forming below.
10
44
Conidia very few above and few below. September 13, no
change .
8
49
9
3
10
No conidiophores on either surface. September 6, no change.
8
20
Nothing above and very few conidia forming below.
10
44
Conidia very few above and few below. September 13, no
change.
8
49
10
3
10
No conidiophores on either surface. September 6, 8, no
change.
10
3 7
Conidia very few on both surfaces. September 13, no change.
8
51
I
3
10
No conidiophores on either surface. September 6, no change.
8
24
Conidia very few above and nothing below. September 10,
no change.
13
37
Conidia very few upon both surfaces. September 15, 17, no
change.
8
51
a
6
10
No conidiophores on either surface.
8
37
Conidia very few on both surfaces.
10
6S
Conidia numerous in the center upon both surfaces. Sep-
tember 13, 15. no change.
17
44
Conidia very few above and few below.
8
SI
3
6
10
No conidiophores on either surface. September 8, no change.
10
37
Conidia very few on both surfaces.
13
44
Conidia few above and very few below. September 15, 17,
no change.
Apr. 3, 1916 Climatic Conditions and Cercospora beticola
47
Table IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of igij — Continued
Plant
No.
Leaf
No.
Spot
No.
1913-
Sept. 6
17
June 24
July 9
June 30
July
Graph
values.
Data on life histories.
87
100
37
No conidiophores on either surface. September 8, 10, no
change.
Conidiophq^'es forming above and conidia very few below.
September 15, 17, no change.
No conidiophores on either surface. September 8, 10, 13, 15, 17,
no change.
No conidiophores on either surface. September 8, 10, 13, no
change.
Conidiophores none above and forming below.
Nothing above and very few conidia below.
No conidiophores on either surface. September 19, no change.
No conidiophores on either surface. September 19, no change.
No conidiophores on either surface. Jime 25, 26, 27, 28, 30;
July I, 2, 7, no change.
No conidiophores on either surface.
Conidiophores abundant on both surfaces.
Conidia fairly abundant on both surfaces. July 14, no change.
Conidia abundant on both surfaces.
Leaf dead, conidia very few on both surfaces. July 25, 28, no
change.
Conidia few on both surfaces. July 23, 25, no change.
Conidia none above and few below. July 30, no change.
Conidiophores above and none below.
Conidia numerous on both surfaces. July 28, 30, no change.
No conidiophores on either surface.
Conidia few above and conidiophores abundant below.
Conidia abundant above and fairly abundant below. July 14,
16, no change.
No conidiophores on either surface,
change.
No conidiophores on either surface.
Conidiophores few above and abundant below.
Conidiophores above and conidia few below.
Conidia numeroxis on both surfaces. July 12, 14, no change.
Conidia abundant on both surfaces. July 21, 23, no change.
Conidia few above and abundant below.
Conidia none above and numerous below. Leaf dead. July
30, no change.
July I, 2, 7, 9, 10, 12, no
21
25
28
5°
56
25
21
37
23
44
25
28
37
25
3
10
7
.9
31
62
16
56
12
10
14
16
IS
37
31
100
Conidia few on both surfaces. July 23, no change.
Conidia fairly numerous above and few below.
Conidia none on either surface. July 30, no change.
Conidiophores above and conidia few below.
Conidia very few above and few below.
Conidia very few on both surfaces.
Conidia none on either surface. July 30, no change.
No conidiophores on either surface.
Conidia very few above and conidiophores below.
Conidia fairly numerous on both surfaces. July 10, 12, 14, no
change.
Conidia few above and fairly numerous below.
No conidiophores on either surface.
Conidiophores none above and very few below.
Conidiophores above and conidia few below.
Conidia abundant on both surfaces. July 23, 25. 28, 30, no
change.
27469'*— 16-
48
Journal of Agricultural Research
Vol. VI. No. I
TablB IV. — Data on life histories of a representative number of leaf spots studied on
10 plants in a medium-early sugar-beet field near Rocky Ford, Colo., during the season
of ipij — Continued
Plant
No.
Leaf
No.
Spot
No.
Date.
Graph
values.
Data on life histories.
»*jo
9
I
1913.
July 7
50
Conidia few on both surfaces. July 9, no change.
10
75
Conidia few above and abundant below.
12
94
Conidia fairly abimdant above and abundant below. July 14,
no change.
16
69
Conidia very few above and abundant below.
21
100
Conidia abundant on both surfaces. July 23, no change.
25
75
Conidia numerous on both surfaces.
tio
9
2
7
37
Conidiophores above and conidia few below. July 9, no change.
10
75
Conidia few above and abundant below.
12
69
Conidia few above and fairly abundant below. July 14, no
change.
Conidia very few above and abundant below.
16
69
21
lOO
Conidia abundant on both surfaces. July 23, 25, no change.
lO
9
3
23
25
Conidiophores on both surfaces.
25
75
Conidia numerous on both surfaces. July 28, no change.
*IO
10
I
23
10
No conidiophores on either surface.
25
75
Conidia numerous on both surfaces.
38
75
Conidia abimdant above and few and matted below. July 30,
no change.
lO
II
I
2S
75
Conidia numerous on both surfaces.
28
75
Conidia abundant above and few and matted below. July 30.
no change.
lO
ir
3
25
50
Conidiophores above and conidia numerous below.
28
75
Conidia numerous on both surfaces. July 30, no change.
_ co/v/o/^ /OO.O
jrfl/t/ O C c/
C0A^/£/0P»O^£^
/oo
"T~r T7F 11
"±^ ^13^ It i^--^
^
^ *!«->'
vP«S'\
1 3 lB:;/ ^, \
s
KT^ I \_
ic."^ TBpL ^ —i
/■
^ ^ S ZN
/
/ \^^ li T^
i//\ / / »i \ / \ In 1
^J
- 3 j>^ L Z^
N
^z--— g::^ ^J iitt^^xixi
/ ^ r n
N^
2 J\ u
i \
"^ "1 :3?,: 'TO TmX
y'^ \\-
\K.
^^IL 411 r \tLl.U^
i
T AlLj^
=> I 2
^i: iJT. J ., 11
/I
\ A '
, 55^_
J i- -P // 41-
J J^
^^i-
t. t --- /- U--
/ 1-
A -/" A-1
^-^
^U J '/ ^H -. I -H
^ I .
-4 L ^ h-
-^ t
p—-t-j' I 'i 4-i 7-t-t
-l4 l %
t -//' it "^i-i t-t t---t
-i
-P -!-!
r A I ^Lt 'ii^ - -- '2
1 .
I ^lA
te c 0 e r Iv /r «? /v 0 f'
L ± ±
X
s V N <\| Oj
>o N 0, s <\, » V, H) 5
<\, e(, n, 0) \
i^j RJ -Xl <0
Fig. 6. — Curves of the leal spot history series, showing the production of conidia on different dates from
June 24 to September 19, 1913, at Rocliy Ford, Colo.
After the values were all assigned, the spots which appeared on or about
the same day were brought together in 24 groups and averaged (Table V
and fig. 6). The temperature and humidity records used in the correla-
lations with the leaf spot histories were taken among the sugar-beet leaves
near the surface of the ground.
Apr. 3. 1916 Climatic Conditions and Cercospora beticola
49
S N
lO
S'ti,
\^
t
1
>
•0
a
a
1
X
0
0
^
0 yn<o iH « 0 •
11 w « e<) t<5 to •
«
>
0 0 Ovmd»t^«^m
,.
ID
0 0 -0-0 -c 0 •
t^
,.
f-i
0*6\'^'*<nmOO* •
00
'T
m
Mmmmm-o^or-OOm-*
Ov
Pi
\ '.'.'.'.'.'.'.'.'.'.'.'. 0 ^ \ri r^OO Ovd^OvOONt^Oro
H -^o 000000 t^r^i-ONO m'^tro
■*
a
1 '. ! ! ! 1 '. ! I ! ! M in tosd m m ro moo « oc A
M«(*i'«t'Oor*r*ONO^to
=s
(U
* '. looooot^t^H'^t^mmmoo*^*^
M mo NO r-oooooo t- t^ *^ m m CO CO
r^
0
0
'. '. '. '. '. '. * !oOwNO»"^wOwoomMt^mmmm
MM\oo»ooooo*oocor>.t>-t>.r^Nw«N
2
; '. '. !N'00**CNTfoO-^MOOd*oOr^'t
MWrO'C000N0000O»0C.00r^'O'^P0
:?
^
'. '. '. '. io^^mmdinOOt^t^OiMO*^^
H M -"tmso r^oo 0*0000 t^i^mmm
t^
hJ
i-'^Nm6*t~^t^-'-;;;
o.
W
^ • !oo6ro6md\00 • • • •
- . .Mrn'Ot^r^omm- ■ • •
-
i-»
. . iH m-o <> ^ ;::;::: ;
«
"
i M S ri. I^^oo t^ t^ t^ t^ : : : :
«
W
OM«osi^>-'6
M t^ t^NO ^ »^ *^ ::::::: :
«
0
MMMMMM;.
Mf^QO*'--
■*
fe
:::::::: :22^SS
m m •;;•■;;■ ■
sO'O'^'^
O'^fO'O;;;;;;;;;;;
r-
w
:::::;: :nS :^SK
00 On OnO ''i ;;;;;;;;; ;
"S
0
i-irO*'^«^t^*^
OmOmOOO-
00O\OO\mm---;;;;;;
a
u
* *. '. '. -oooo^^jogo
r^At^O
oooooom '.'.'.'.
0
P3
1/1 in \n \ri ^ ^ '•'••■ •
: :^ ::;:::::::: :
^ 1
<!
mm'0^««'Om'^'*' ■ • ;
.«■
0
TT m-o r^oo d M « t^oo Oi 0 « r
0 w romoo 0 Mm*^0**^ pomoo « '
""""""1
h
1
0
k
\
0-C
oc
c
'
,.
^i
^ di
P
<
^o Journal of Agricultural Research voi. vi.no. i
CoNiDiAL PRODUCTION. — Under favorable conditions conidia are
produced apparently much more readily by young than by old leaf-
spots. For instance, as will be seen in Table IV, leaf spots (*) ^ 2 to 4
days old showed a marked increase in conidial production from July 23
to 25, while during the same period spots (**) 14 to 24 days old in most
cases showed a decrease. However, spots (***) 2 to 3 weeks old on
green leaves showed an increased production from August 18 to 25, the
conditions being favorable, and some (f) even produced a second and
third crop, although usually but one crop (ft) is produced and this while
the spots are comparatively young. It was also found that under favor-
able conditions a spot (fft) may produce abundant conidia on both
surfaces in one day. Usually the maximum production is reached within
10 days after the spots appear (fig. 6), and sometimes under very favor-
able conditions the production may increase after this period (fig. 6,
curves D and E, July 17 to 23), but the older spots do not always respond
to favorable conditions in this way (fig. 6, curves C and F). In no case
was a new growth of conidia observed on spots on yellow or dying leaves
on green plants in the field. The fungus seemed to lose its vigor much
sooner on such leaves than on green leaves which remained attached
to the crown at harvest time. From the standpoint of control of the
disease this is a very important point, from the fact that at harvest time
the green leaves, on which the fungus is vigorous, are removed with
the crowns and stored in the silo, while the yellow and dying leaves,
on which the fungus may be too weak to overwinter, break off and
remain on the ground.
During the greater part of August and September, when the precipita-
tion was light (fig. 7), many of the conidia had a shrunken appearance
and were massed together on the leafspot areas (Table IV,§). When
placed in water, these conidia did not germinate; consequently this
desiccation of the conidia may also be an important factor in connection
with the vitality of the fungus on the host.
The position of the leaf on which the spot studied was located was also
found to be an important factor in conidial production, an abundance of
conidia being frequently observed on leaves protected from the sun,
while at the same time few were observed on those exposed to the sun
the greater part of the day. This difference in production is thought to
be due mainly to the difference in humidity of the protected and the
exposed locations.
A study of the comparative production of conidia on the upper and
the lower surfaces of the spots was also made, the conidia on the spots
included in series E, K, N, and S (fig. 6) being tabulated for this purpose.
Generally a more abundant conidial production was found on the
lower than on the upper surface (fig. 8), and this was due apparently to
I The asterisks (*). daggers (t), and section marks (§) refer to particular leaf spots in Table IV.
Apr. 3, 1916
Climatic Conditions and Cercospora beticola
51
the probably higher humidity of the former. Only during a very favor-
able period (fig. 7, July 19 to 21) or where the leaves were turned up or
protected by other leaves was the conidial production on the upper
surface equal to that on the lower surface (fig. 8, series E, July 21). At
times, conidia were formed more abundantly on the upper surface than on
the lower (fig. 6, series N, August 11, and series S, August 28). Because
of the spongy parenchyma and the greater number of stomata on the
lower surface, it might be supposed that conidiophores could be produced
more readily on this than on the upper surface; but, as above indicated,
humidity would seem to be the controlling factor in this connection.
S-'^
. T _ _ _
/vi"/ .. -
V'/-.'-'rn\ !\j\/\f^^'\i \ ,-'^'Na/>^/ ^Yf^■V•^^' ■"->-\/^-'i
1 V VM^^-i^Al^ ^ 1^=, ^^ ^ i CiW-U sSl^
i^ ^ t A-iV"^^^ TT.^rC^^^-"^ ^ ^L ^ vx ^^^ 4t
L li J ^ iA^'^^y ±5 ^ ' "L.2L:
aI- i-^^ S -r ^-ut
^1 W -^ I I ffi^
^ Xlk^ I . UmI- A-.A -L
Y^y S j^-y^-$^^ ^ u ^a
S dS:^S:, 3CL2T ^^u^Cs,^^ClAt a .^^4^1
jioz J^i L^j L 4^ ^ it jQ J ^' xH
Vr J\ ! ' '■' / !■' ' I'V, " /\
V ^ •• i 'r ■ ' 'l' ' ''1 J ■ , ',L '\
\ 1 \ \ 1 / '• ; \ V .' ' \ ./ 1 .< h V rA,
\ I \ J 1 ) • ; '' A / K \\ r V , / '. ;'-'
i y« ' \ ,' I 1 ' ' ^' ■ < ; /^/ "■''','''
L/j. "J II "^ '. L ''jjt^
\rr'~ "7 ir L J 1 n^
1/ f ■
------------Hr-'---Ifl7~^"]^"X
A l\ r U /'l /; ^n -)
Aii w A/i r / ^T ^^ ^ iM A 7^/^
-il-4t4fP^ft ---- J---------~---i
_XF_ Jff ----- -tf— -- - - t — -L____J __
y 1
•\%
\ % '\i ^ \ ^
Fig. 7. — Curves of the maximum and minimum temperatures and humidities, the number of hours that the
humidity remained above 60 from noon of the preceding to noon of the given day among the plants,
and rainfall and irrigation records, taken in a medium-early sugar-beet field from June 10 to September
22, 1913, at Rocky Ford, Colo.
A comparison of the conidial production as shown in Table IV and
figure 6 and the climatic data shown in figure 7 indicates many definite
relations. When the spots were first found, on June 20 and 24, conidia
were fairly numerous (fig. 6, curve A) on all except six spots, which had
evidently just developed on the latter date, as no conidia were present
at this time and conidiophores only were produced the next two weeks.
The following week there was but little increase, and during the next few
days many of the conidia were disseminated. The small production of
conidia was evidently due directly to the high temperature and the low
humidity which prevailed during this period (fig. 7), as conidia were pro-
duced in great abundance from July 9 to 12 (fig. 6, curves C, D, E, F),
when the temperature was lower and the humidity higher (fig. 7). Dur-
52
Journal of Agricultural Research
Vol. VI, No. I
ing this time and the few days just preceding, the humidity remained
above 60 for a longer time on an average and the minimum humidity
did not become so excessively low nor the temperature so excessively
high as during the time previous to July 4.
The next period of pronounced increase in conidial production was
from July 19 to 23 (fig. 6, curves D, E, G), when the conditions were more
favorable than during any period of similar length through the summer,
the humidity ranging above 60 on an average of 19.4 hours each day and
not falling below 52 (fig. 7), and the temperature ranging from 60° to
90° F.
Conidial production was again above the average (curves M, N, and O)
from August 9 to 13, during which period the humidity remained above
60 from 13 to 20 hours each day and there was a small amount of rain
which seemed to aid in maintaining the necessary humid conditions.
Production was checked on August 16, on which date the temperature
was 102° and the average humidity low, and was again inhibited after
coff/o/o^MOMrs SO
.
rs
\
1
-'
^
..
_
..
_,
/•
/
\
\
^
\
/
(
\
/
\
y
\ .
^
_
/
\
■<
s
N
i
v
/
■•^
\
»>
^■
.
t
i
/*
\
-
-J
^('
s
"7
a
^
^j
''
^
\
^
t
i
y
^^
V
i
?
r
9
^
V
\
S
>
j«
V
1
k
\
,;
\
''J
^
M
/^
%
N
I
ih
7
«<
J,
k
^
—i
If
A
»V
^
1
^^
s
i"
ff
1
J
«
Fig. 8.— Curves of the comparative production of conidia on the upper and lower surfaces of the leaf spots,
representing series E. K, N, and G of Table V and figure 6. Rocky Ford, Colo.. 1913.
September ii, subsequent to which date the minimum temperatures
ranged from about 30° to 45° and the maximum from about 65° to 83°,
while the humidity remained above 60 for 12.4 hours per day, on an
average.
The general conclusion from these tests is that conidial production is
greatly influenced by temperature and relative humidity, or speaking
specifically —
(i) A temperature of 100° F. or over is detrimental to conidial pro-
duction, directly perhaps because it is inimical to the growth of the
fungus and indirectly because humidity is ordinarily excessively low at
such an extreme temperature.
(2) Conidial production is greatly checked at daily temperatures
ranging below 50° as a minimum and 80° as a maximum.
(3) The most favorable temperature for conidial production is 80° to
90° in the daytime and not below 60° at night.
(4) The temperature being favorable, the largest conidial production
occurred at the higher humidities. A good production occurred when
Apr. 3, 1916 Climatic Conditions and Cercospora heticola 53
the humidity remained above 60 for not less than 15 to 18 hours, but
very few were produced when the humidity remained above 60 for less
than 10 to 12 hours daily.
With a view to determining the approximate number of conidia pro-
duced on a sugar-beet plant under a favorable temperature and humid-
ity, one representing a heavy infection in August was selected. After
the infected leaves were measured a representative portion of conidia
were carefully washed off into sterilized water and counted. The count,
which was made by means of a dilution method, showed 250,000,000
conidia on the plant at that time.
CoNiDiAL DISSEMINATION. — That a period of low humidity, with its
accompanying factors, is favorable to the dissemination of conidia was
frequently observed (fig. 6, curve R). For instance, it was found that
the amount of conidia diminished on September i, 6, 14, and 15, when
the humidity remained above 60 for 5, 6, 10, and 4 hours, respectively;
while, on the other hand, there was no diminution in the amount present
on September 3 to 5 and 8 to 10, during which periods the humidity
remained above 60 for 12 to 16 hours.
Rainfall is also an important factor in the dissemination of conidia, as
was noted in several instances. On July 19 (fig. 6, curve F) rain fell,
and as a result many conidia were washed off, and the same was true in
the case of rains on July 23 (curves C, D, E, G, H), August 9 (curve K),
September 4 (curves N, O, R), and September 16 (curves Q, S, T, U, V,
W). After rains on July 19, August 9, and September 4, however, there
were more conidia present than before, but this was probably due to the
fact that more we re produced under the favorable humid conditions attend-
ing these rains than were washed oft'. It was also found that the conidia
were disseminated more rapidly from the upper than from the lower sur-
face of the spots (fig. 8). This was due probably to the greater exposure
of the former to wind and rainfall.
RELATION OF INFECTION CYCLES TO CLIMATIC CONDITIONS
For the purpose of determining the relation of infection cycles to
climatic conditions, a study was made of the increase and spread of dis-
ease in a field of sugar beets planted about May i ^ at Rocky Ford and
one planted about two weeks later. Both fields had been in beets for two
or three years, and as very few, if any, of the tops were removed after
the harvest of 191 2, infection appeared early in 191 3 and was generally
distributed.
Three plants in the early field (Table VII) and ten in the medium-
early (Table VI) were selected, the leaves tagged and numbered con-
secutively, beginning with the outermost or oldest and continuing with
the new ones as they appeared. The spots on each leaf were counted at
' ConidJal production and dissemination were also studied in this field.
54
Journal of Agricultural Research
Vol. VI, No. 1
frequent intervals,^ and the average actual increase of spots per plant
computed (Table VIII and fig. 9). It was found that frorn 400 to i ,000
spots on a leaf, depending on its size, killed it within a few days.
— sooo
fOOO
I
%30do
\
'^£000
}
1
1;
|\
f\
11
1 1
f
^
/\
1 1
1 1
1 1
s
1
\/
1
1
1
\
1
I
I
',»■'''*
f
\
f
V
\
1
/
/
V
V,
/N
,-
I
,1
AT^
1
■<^
y
Fig. 9. — Curves of the 2-day average increases in the number of leaf spots per plant in a medium-early and
an early sugar-beet field, from June i8 to September 19, 1913, at Rocky Ford, Colo.
Table VI. — Average infection cycle of Cercospora beiicola in a medium-early sugar-beet
field with poorly developed foliage and with a consequent low humidity early in the season
at Rocky Ford, Colo., in igij
July 2.
8.
10.
12.
lA.
16.
21.
23.
25
2.~v
Aug. I
5
9
II
13
IS
18
22
2S
28
30
Sept. 3
6
8
10
13
IS
17
19
Total
number
of
leaves
marked.
14. 2
16.8
17.4
18. s
19.7
21-3
22.6
26.3
27-5
29
30. S
33
34
35-5
36.5
37- S
38. S
40
41- S
43
4S
46
48
48. S
SO
SO-S
Si-S
S2.5
53
S3-S
Total
num-
ber of
leaves
in-
fected.
5-8
6
6.6
10.8
17.6
27
35- S
38
41
41
41-5
42- S
42. S
44- S
44-5
44- S
Total
num-
ber of
leaves
dead.
4-3
4.6
S-I
s-s
6.1
8.9
13
13-5
IS- 5
IS-S
IS- 5
IS- 5
IS- 5
15- S
17
23- S
25-5
26. 5
28.5
30
32. S
34- S
3S
Num-
ber of
leaves
killed
by Cer-
cospora
beticola.
Num-
ber of
infected
green
and
dying
leaves.
4.4
4-7
4-S
4-3
4-3
4.1
4.4
4-7
4-4
6.3
10. 7
18. s
19- s
20.5
20
20. 5
22. 5
24- 5
27
29- S
28.5
25- S
23- S
21- S
22. S
20.5
19
17- s
IS-S
Number of func-
tional leaves —
In-
fected.
2.9
4.1
4.4
4-3
4
4
3-8
3-5
18. s
17- S
19. S
19
20
20.5
22. s
24- S
27
28
27- S
22.5
21.5
20.5
20.5
19
16.5
15- S
IS
Unin-
fected.
8.9
8.7
9.6
10. 6
12.7
IS- 7
15-4
16.6
IS- 4
II. 7
4-S
5-5
4-S
Total
num-
Total
ber of
number
func-
of leaf
tional
spots per
leaves.
plant.
10.9
5
13
10.4
13- 1
II. 4
13-9
II. 2
14.6
10.8
15-8
10.8
16.5
10. 4
19-2
13-3
19- 5
17- 2
20. I
17-5
21. 2
20. s
22
42
22
93-5
23
292
23
355
24
402
23
455-5
24- 5
593
26
875
27- 5
I
ISS-S
29-5
2
045-5
30- 5
3
420
31
9
oSi. s
30-5
10
856
26.5
12
371-5
25
II
493
25
10
92S. 5
24
10
7S2
23
10
003.5
21
9
585-5
19-5
8,
504-
20
7
597-5
Aver-
age
num-
ber of
leaf
spots
per leaf.
1.6
2-3
2.4
2-4
2-S
2-5
2-5
3
3-6
3-9
3-2
3-9
4-7
15-3
19. 1
20. 6
22. 2
29. 6
42.6
51-3
8s-9
126. 6
307. 8
380.9
485-1
489
S08.3
479-2
487.9
504- S
485- 9
490.1
' For convenience and uniformity, 2-day averages were used in making the comparisons, the coimts
being made usually at 2-day intervals.
Apr. 3. 1916 Climatic Conditions and Cercospora beticola
55
Table VII. — Average infection cycle of Cercospora beticola in an early sugar-beet field
in which there was a heavy production of foliage and a conseqtient high humidity
early in the season at Rocky Ford, Colo., in IQ13
Date.
Total
num-
ber of
leaves
marked
Total
num-
ber of
leaves
in-
fected.
Total
num-
ber of
leaves
dead.
Nimi-
ber of
leaves
kiUed
by Cer-
cospora
beticola.
Num-
ber of
infected
green
and
dying
leaves.
Number of func-
tional leaves —
In-
fected.
Unin-
fected.
Total
number
func-
tional
leaves.
Total
number
of leaf
spots per
plant.
Aver-
age
num-
ber of
leaf-
spots
per
leaf.
July 7-
9-
12.
14.
16.
21.
23-
2S-
29.
Aug. I
5
7
9
II
13
15
13-5
14
14-5
II- S
13
18.5
23
22
22. 5
28
26. s
26. s
26.5
24- S
23-5
23
21. s
22
24- 5
25-5
24- 5
24- S
23- S
20. s
18
10. s
10. 5
5
6
S-5
7
8-5
19- S
20.5
22
22.5
26
26. s
27- 5
30
29
3°- 5
30-5
30
30
529
596. S
61S
554-5
574- S
l>ii3-5
2>2l6. 5
3> 776- s
7.045.5
1 1 , 966. 5
12,638
13.905
14, 228. 5
16,386
16, 693
14.993
44
49- 7
51-2
48-2
44-2
60. 3
96-3
171. 6
313- 1
427-3
476.9
524- 7
536-9
668.8
710.3
651. S
Table VIII. — Actual and 2-day average increase in the number of leaf spots per plant
in a medium,-early and an early sugar-beet field from June 18 to September ig, at
Rocky Ford, Colo., in igij
Date.
Increase in medium-
early field.
Increase in early
field.
Date.
Increase in medium-
early field.
Increase in early
field.
Actual.
2-day
average.
Actual.
2-day
average.
Actual.
2-day
average.
Actual.
2-day
average.
03
•5
0
•3
- I
•4
2-3
4-9
1-4
0-3
-5
0
-3
. I
•4
2-3
1.9
2.8
Aug. 5
7
9
II
13
15
18
22
25
28
30
Sept. I
3
6
8
10
13
15
17
19
203
6S
127
S3
190
282
280
895
1,379
5,664
2,572
2,582
1,634
799
771
578
732
857
63s
543
102
65
127
53
190
282
186
447
919
3.776
2,572
2,582
1,634
532
771
578
488
857
635
543
3,996
2. 167
2,373
4,208
1,831
l,4SO
2,450
2,286
2,582
2,944
1,998
2, 167
2,373
26
28
1,831
July 2
1,633
1,143
7
8
70
70
-9
. 2
-7
•7
3-3
3-3
1-9
4-9
-9
. 2
-7
•7
1-3
3-3
1-9
3-2
18.5
24-5
20.5
672
1,272
l>723
12. 2
24.5
20. 5
269
I, 272
1,723
14
16
23
25
28
29
30
Aug. I
3,282
1,641
5-1
44
5- I
44
4.978
3,318
The period of incubation of the fungus being from 11 to 13 days, as
shown by artificial infection experiments, a corresponding increase in the
number of spots on the leaves would not necessarily follow immedi-
ately after a period during which conditions favorable for infection pre-
vailed.
Notwithstanding the early appearance of the spots in the medium-
early field — on June 20 — and a consequent expectation of an epidemic
of the disease, the increase in infection was very light (Table VIII)
56 Journal of Agricultural Research voi. vi, no. i
during the latter part of June and early part of July. This was doubtless
due to the fact that during this period the stomata were closed against
the fungus the greater part of the day on account of the excessively high
temperature, which was generally above 100° F., and the excessively
low humidity, which at one time fell to 10 and which was only above
60 from 6 to 15 hours a day, and also to the fact that a temperature as
high as 95° inhibits the growth of the fungus and kills it after a few days.
After July 5 the temperature was lower and the humidity higher than
during the period above mentioned. As a result, numerous conidia were
produced from July 9 to 12, and at the end of the period of incubation —
July 21 to 25 — there was a slight increase in the number of spots (Table
VIII). Rains between July 19 and 35 and the resulting high humidity
caused a rather marked increase in the number of spots in late July and
early August, these spots appearing on many leaves hitherto uninfected
(Table VI). Prior to this period the number of leaves showing spots
were comparatively few, but after July 30 the majority showed spots,
and the proportion of infected to uninfected leaves gradually increased
until August 28, after which it decreased. During the period from July
30 to August 28 the humidity was comparatively high, remaining above
60 on an average of 14.6 hours on all except three days, on which it
remained above 60 for 10 hours; and the maximum and minimum
temperatures generally were not above 90 nor below 55, respectively.
The increased proportion of infected to uninfected leaves during this
period, however, was not necessarily due to increasingly favorable
climatic conditions but to the cumulative effect of the organism, the
amount of viable conidia and consequent new infections increasing as
the number of spots increased, as shown by the enormous increase of
3,776 spots per plant on August 27 and 28. After September 3 the
increase in infection was considerably less (fig. 9) . This was due appar
ently to the fall in temperature, the maximum being rarely above 76° F
and the minimum seldom above 50° after September 8, while the humidit}
was comparatively favorable.
The increase in infection through the season was considerably higher in
the early than in the medium-early sugar-beet field, as shown by the
total amount of the disease (Tables VI and VII) and the actual increase
(Table VIII and fig. 9). This was due to the fact that the foliage was
heavier in the early than in the medium-early field (Tables VI and VII,
functional leaves) , and consequently the humidity was higher and the
infection greater in the former than in the latter (Table VIII and fig. 9).
The maximum increase in spots was reached on August 1 1 in the early
field and on August 28 in the medium-early sugar-beet field. The period
of greatest increase in the disease is not its period of greatest destruc-
tiveness, however, as the plant is not immediately affected by the dis-
ease, some time being required for the leaves to be killed.
Apr. 3, 1916 Climatic Conditions and Cercospora beticola 57
Prior to August i , only isolated records of humidity were made in the
early field,^ but after this date continuous records of both humidity and
temperatures in both fields were available for comparison (fig. 10).^
The temperatures prevailing in the two sugar-beet fields were quite com-
parable, but the humidity was generally different. For instance, from
August 2 to 23 the humidity remained above 60 for a longer time, and
the maximum humidity was, as a rule, higher in the early than in the
medium-early field; from August 23 to September i the maximum
humidity was lower in the early than in the medium-early field; after
the latter date strikingly lower, the difference ranging from 5 to 1 5 units ;
after September 5 the humidity remained above 60 for a shorter time
in the former than in the latter field; but from September 6 to 21 the
range of humidity in the two fields was much closer than during the
periods previously mentioned.
The difference in the humidity of the two fields seemed to be due to
the difference in the amount of foliage present. Early in the experiment
the foliage was heavier in the early than in the medium-early field, but
owing to an extremely severe infection, which developed between July
29 and August 13 (fig. 9), the relative proportion of foliage in the two
fields was reversed after that period. As a result of this reversal, less
moisture was retained and the humidity was lower in the early than in
the medium-early field during September, and consequently at that time
the relative increase in infection was less in the former than in the latter.
Speaking more specifically, early in the experiment there was an aver-
age of 29 functional leaves per plant in the early field and 22 in the
medium-early; on August 15 there was an average of 26 leaves per plant
in both fields, while later on there were fewer per plant in the early
than in the medium-early field. On the other hand, on August 13 there
was an average of 23.5 infected leaves per plant, with an average of 710
spots per leaf in the early field, and on September 8 there was an average
of 21.5 infected leaves per plant, with an average of 508.3 spots per leaf,
in the medium-early field.
A comparison of the death rate of the leaves in the two fields before
and after the disease appeared shows its destructiveness. For instance,
in the early and medium-early fields, from July 7 to 29 and from July 2
to August 25, when no leaves were killed by the fungus, the death rate
from normal causes was approximately one leaf per plant in three and
four days, respectively; while from July 29 to August 15 and August 25
to September 19, when the disease was most severe in the two fields, the
death rate averaged one leaf per plant in nine-tenths of a day and one
and three-tenths days, respectively.
1 These and the later continuous records indicate that prior to August i the humidity was generally
higher in the early than in the medium-early field.
* The temperature records taken at the Weather Bureau station were included in the comparisons and
were found to agree closely with those obtained in the two fields.
58
Journal of Agricultural Research
Vol. VI, No. I
30
I
so
K JO
_/o
I
I o
'^
V
V
^
^rK\
f
*>
^
^
^
J
iv
[r
\
-V
1 ''
V
A
'<c_
^
1
V
V
\
k
V
i
Af,
IX,
f^C
'M
I
,/
s^
i<
t^
1
\!
1
t-
•s
1
>
''a
X3
il
7
u
3
y
1
/
1
\
1
\
/
/
\
\
»
\
1
1
II
^
,
i
4
1
y
j
\/
-/
^-'
v/
/
V
1
"^
rf--^
■:>
fi
¥
',
Y\
J
w
/
V
i
\
A
I
'
^
\
J
/
1
1
\
A
1
1
'V
V
\\
/
/i
\j
1
1
A7
(M
n-n
//^
V
v
/
i
V
\
1
/
1/
'V
\
^^'i
f
'/
s'
\
V
\
\
\
h
<'
\
\
^
J
\
1
t^
J
y
,
\
-^^^/(/y.
r
V
!
,\
i\
^1
l)
1
y
1
1
'I
.
/\
i
1
1
L
/\
/
I
^
/
\
J
J
\
/
r
-o)
/\
7
\/
)
\
'Is
ij
V
V
T
A
^
/
V
>
\
V
-
^
<
->
y
' i
g a
1 1
as
.^ a
■r) 01
^ §
V tn
H a
.3 a
5 o u
s fe ^■
.a "a
S g«
p a o
ca cs "
u S i>
> a OT
'■^ d O
"So"
<V » >o
yfi/Gusr I
'*i'oNO)^'^'o\0)>
Ill
3 i* •
■^ -? "S
S.a 4)
2 M «
a c a
■S a '5
0 3 5
§^ s
a s a
a M i,
>< n ja
« il -i
t3 O 0)
^ ° .S
^ c t:
m o a
y *j H
b M 2
s .a ■:;;
V 7 -3
6 t °
Apr. 3, 1916 Climatic Cotiditions and Cercospora beticola 59
SUMMARY
(i) The life of the fungus Cercospora beticola overwintering in sugar-
beet-top material varies with different environment. When exposed to
outdoor conditions, the conidia die in from one to four months; but when
kept dry live as long as eight months. The sclerotia-like bodies, which are
more or less embedded in the tissues of the host, are more resistant than
the conidia, living through the winter when slightly protected, as, for
instance, in the interior of a pile of hayed sugar-beet tops or buried in
the ground from i to 5 inches, and become a source of infection for the
succeeding crop. Notwithstanding the difference in temperature and
soil-moisture conditions, similar results from the overwintering experi-
ments were obtained at Rocky Ford, Colo., and Madison, Wis.
(2) Climatic conditions and the development of the leafspot can be
correlated only when all records are taken at the same relative positions,
as shown by comparisons of the Weather Bureau records and the records
taken among the plants and 5 feet above the field.
(3) The maximum temperature is much higher near the ground than
5 feet above early in the season, but the difference diminishes as the
season advances.
(4) Throughout the season the maximum relative humidity was higher
among the leaves than 5 feet above the field. Early in the season, while
the plants were small, the humidity remained above 60 longer each day
5 feet above the field than among the plants near the ground ; but after
the plants attained a good size this condition was reversed. Because of
this difference, only records collected among the leaves should be con-
sidered in correlating climatic conditions and conidial production and
infection.
(5) The effect of rainfall and irrigation on the increase of relative
humidity and its duration is apparently much the same.
(6) Thermal tests with artificial cultures showed (a) that exposure to
constant temperatures of 35° and 36° C. is fatal to the growth of the
fungus ; (b) that growth occurred when cultures after exposure for 3 days
to either of these temperatures were changed to 30.8°, and also when
they were held at either for 8 hours and then at 20° for 16 hours; and (c)
that a temperature of 40.5° was fatal in all combinations tested.
(7) Temperature and relative humidity influence the production of
conidia and infection in much the same way. A temperature of 80° or 90°
F., with a night minimum preferably not below 60°, is most favorable to
conidial production, while it is checked by a temperature of 100° or
higher and greatly checked by a range from below 50° to 80°. A maxi-
mum humidity ranging above 60 for not less than 15 to 18 hours each
day induces a good growth of the fungus.
(8) Because of the higher humidity on the lower than on the upper
surface of the leaf, the conidia are generally more abundant on the lower
surface of the spots, but because of the action of rain and \\ind they
disappear more rapidly from the upper surface.
6o Journal of Agricultural Research voi. vr, no. i
LITERATURE CITED
Frodin, John.
1913. Beobachtungen iiber den Einfluss der Pflanzendecke auf die Bodentem-
peratur. In Lunds Univ. Arsskr., n. f. afd. 2, bd. 8, no. 9, 16 p., i fig.,
4 pi. Literatur, p. 15.
Masses, George.
1906. Plant diseases. — IV. Diseases of beet and mangold. In Roy. Gard. Kew,
Bui. Misc. Inform., 1906, no. 3, p. 49-60, 5 fig.
PammEL, L. H.
1891. Fungus diseases of sugar beet. In Iowa Agr. Exp. Sta. Bui. 15, p. 234-254,
7 pi.
Pool, Venus W., and McKay, M. B.
1915. Phoma betaeon the leaves of the sugar beet. In Jour. Agr. Research, v. 4,
no. 2, p. 169-178, pi. 27.
1916. Relation of stomatal movement to infection by Cercospora beticola. In
Jour. Agr. Research, v. 5, no. 22, p. 1011-1038, 6 fig., pi. 80-81.
Shaw, H. B.
1914. An improved cog psychrometer. In Plant World, v. 7, no. 6, p. 183-185,
2 fig.
Stewart, F. C.
1913. The persistence of the potato late-blight fimgus in the soil. N. Y. Agr.
Exp. Sta. Bui. 367, p. 357-361.
ThumEn, Felix von.
1886. Die Bekampfimg der Pilzkrankheiten unserer Culturgewachse. 157 p.
Wien.
Treboux, O.
1914. tjberwinterung vermittels Mycels bei einigen parasitischen Pilzen. In
Mycol. Centbl., Bd. 5, Heft 3, p. 120-126.
PLATE III
Cercospora beticola: Overwintering tests on the experimental field at Rocky Ford,
Colo., during 1912-13:
Sugar-beet leaves infected with Cercospora beticola (z) stored in soil in boxes, (2)
buried in Hie ground at different depths from i to 8 inches, and (j) left exposed above
the ground in a pile of hayed sugar-beet tops.
Climatic Conditions and Cercospora beticola
Plate III
Journal of Agricultural Research
Vol. VI, No. 1
Climatic Conditions and Cercospora beticola
Plate IV
^ .,'«-•:.-., <^^>.^;:^*
^^^'
Journal of Agricultural Research
Vol. VI, No. 1
PLATE IV
Field stations for the collection of weather data at Rocky Ford, Colo., in 1913:
Fig. I. — Weather shelter, anemometer, and rain gauge at edge of sugar-beet field.
Fig. 2. — Weather shelter among beet plants, showing hygrothermograph and cog
psychrometer.
Fig. 3. — Weather shelter of the local Weather Bureau station about 3 miles from
sugar-beet field.
27469°— 16 5
ADDITIONAL COPIES
OF THIS PUBLICATION MAT BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
•WASHINGTON, D. C.
AT
20 CENTS PER COPY
Subscription Price, 3.00 Per Yeap.
V
Vol. VI AIPRIL lO, lOlG No. 2
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS
Soluble Nonprotein Nitrogen of Soil 61
R. S. POTTER and R. S. SNYDER
Oviposition of Megastigmus spermotrophus in the Seed of
Douglas Fir 65
J. M. MILLER
Citrus Canker 69
FREDERICK A. WOLF
DEPARTMENT OF AGRICULTURE
WASHINGTON, D.C.
'. U'WUBHamBOB
WASHINOTON : OOVERNMCNT PRINTINa OFFICI : tBI»
PUBUSHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR TBnS ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL
' Physiolozist and Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of Anitnal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
tfie University of Minnesota
All correspondence regarding articles from the , Department of Agriculture
should be addressed to Karl F. KeUetman, Journal of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine.
JOWiNAL OF AGKICHTIAL RESEARCH
DEPARTMENT OF AGRICULTURE
Vol.. VI Washington, D. C, Aprii. io, 191 6 No. 2
SOLUBLE NONPROTEIN NITROGEN OF SOIL
By R. S. Potter, Assistant Chief in Soil Chemistry, and R. S. Snyder, Assistant in
Soil Chemistry, Iowa State College Experiment Station
INTRODUCTION
Dilute alkali dissolves a larger proportion of the organic material of soil
than any of the other relatively mild reagents. A still larger percentage
is extracted from soil previously treated with i per cent of hydrochloric
acid (HCl), and this latter reagent dissolves but little of the organic mate-
rial. The term "humates" is fast disappearing from current scientific
literature, yet one often reads that the reason the preliminary washing
with acid renders the organic matter more soluble in the alkali is that the
calcium of the calcium humates is dissolved out, making the free humic
acids soluble in the alkali. To say that the proteins are rendered more
soluble by the removal of the calcium and the heavy metals would
explain the solubility just as well and would be more correct scientifically.
As pointed out by Lipman (4, p. 251), much of the organic nitrogen of
the soil must be protein in nature. The chief sources of the nitrogen are
crop residues, manures, and bacterial cells, and in these much of the nitro-
gen is in the form of protein. Investigations carried out in this laboratory
(5) have shown that soils contain a large quantity of the so-called humin
compounds. These have a great tendency to be adsorbed by such com-
pounds as magnesium oxid and calcium hydroxid, and therefore removal
of calcium from the soil by acid would tend to make these more soluble.
Upon the acidification of the alkali extract a precipitate is obtained which
has been called humic acid. This term also is no longer taken seriously.
It would seem that the rational explanation of this precipitate would be
simply that it was made up of proteins thrown down, as salts of the
precipitant, as salts of organic acids, such as nucleic acid (7), or resinous
acids (6), both of the latter substances having been isolated from the acid
precipitate. It would also contain, no doubt, some free organic acids.
In analyses of the solution obtained by the prolonged boiling of soils
with strong acids and of the hydrolyzed humic acids by the Van Slyke
method (8), it was found that the results for the humic acids did not
differ markedly from the results for the organic matter of the soils as a
whole from which they were derived. Since that time it has occurred to
Journal of Agrictiltural Research, Vol. VI, No. a
Dept. of Agriculture, Washington, D. C. Apr. lo, 1916
cy Iowa — »
(61)
62 Journal oj Agricultural Research voi. vi, no. 2
the Avriters that this would hold for the material precipitated by acid
from the alkali extract, but perhaps this would not be true of the organic
nitrogen compounds remaining in solution. It has been pointed out by
Shorey (7) that many organic compounds have been isolated from the
alkali extract of soil, which, though relatively quite soluble in water, can
not be detected in or isolated from the water extract of soils. Therefore
it has seemed that information might be obtained relative to the degree
of decomposition of the organic matter in the soil by determining the
proportion of nitrogenous compounds left in solution after the precipita-
tion of the proteins by a suitable reagent v/as completed. It was with
these problems in mind that the preliminary investigation was carried out.
EXPERIMENTAL METHOD
The general procedure followed was to determine the nitrogen in the
alkali extract of soil with and without added material and the determina-
tion of nitrogen in the filtrate from the precipitate of the proteins in the
alkali extract of soil with and without added material. The recent crit-
ical examination of a few protein precipitants by Greenwald (3) led us to
use trichloracetic acid as our precipitant.
The detailed procedure was as follows: Soil, ground to pass a sieve of
100 meshes to the inch, was extracted with i per cent of hydrochloric acid
until no calcium was found in the wash water. After air drying, 100
gm. were placed in an 800 c. c. bottle and 500 c. c. of a 1.5 per cent solu-
tion of sodium hydroxid added. After shaking the mixture for 2 hours
it was centrifuged for 5 minutes in a bowl centrifuge having a speed of
18,000 revolutions per minute. Two 25 c. c. portions of the clear but
deeply colored extract were analyzed for total nitrogen by the Gunning-
Arnold method. Two 25 c. c. portions were neutralized with sulphuric
acid, sufficient trichloracetic acid in solution added to give a 2.5 per
cent solution of the acid and a total volume of exactly 30 c. c. After
centrifuging, 10 c. c. portions were taken from each tube and analyzed
for nitrogen by the Bock and Benedict (i) modification of the Folin and
Denis method (2). This was called the soluble nonprotein nitrogen.
The same procedure was used where material was added to the soil.
In the case of guanin, hypoxanthin, and glucosamin,* weighed portions
of the compounds were added to the soil, which was then shaken with
the alkali. The hydrolyzed casein was prepared as directed by Green-
wald (3), which consisted, in brief, of boiling the casein for 40 hours with
hydrochloric acid, the removal of the acid under diminished pressure,
neutralization with sodium hydroxid (NaOH), and filtration. After
mixing the filtrate with animal charcoal it was again filtered and final
filtration carried out after crystallization of the tyrosin. For all the
remaining materials solutions were prepared, sometimes with the aid of
a little Nlio acid or Njio alkali. vSuitable amounts of the solutions were
* We wish to express our thanks to Prof. P. A. Kober for the samples of the hypoxanthin and guanin,
«nd to Dr. A. W. Dox for the sample of glucosamin.
Apr. lo, 1916
Soluble Nonprotein Nitrogen of Soil
63
added to the soils and then sufficient alkali added to make 500 c. c. of
a 1.5 per cent solution. In all cases, with the above-noted exceptions,
the purest commercially available compounds were used, but analyses
for nitrogen were run on the solid material when it was used, and when
solutions were employed aliquots were analyzed. These determinations
were also made by the micro method. It should be mentioned here
that 6 minutes was found to be quite an inadequate digestion period for
some of the compounds. It is believed that in some cases when appar-
ently more than 100 per cent of the added substance was extracted from
soil, faulty analysis of the substance was the cause. Insufficiency of
material precluded repeating tests with many of the materials.
The soil used for all these tests was a silt loam containing 0.30 per
cent of nitrogen. Samples A and B, as shown in Table I, differ only in
that they were not taken from the field at the same time.
Table I. — Analyses of 5-gni. portions of soil for alkali-soluble and soluble nonprotein
nitrogen
Soil
sample.
Substance added.
Nitro-
gen
added.
Nitro-
gen in
the
alkali
ex-
tract.
Nitrogen of the
added sub-
stance recov-
ered in the
alkali extract.
Soluble
nonpro-
tein
nitrogen.
Soluble non-
protein nitro-
gen recovered.
Nothing
Hydrolyzed casein .
Amino benzoic acid.
Glutamic acid
Hippuric acid
Glutamic acid imid .
Succinimid
Guanidin sulphate . .
Urea
Uric acid
Caffein
Theobromin
Guanin
Hypoxanthin
Skatol
Nucleic acid
Cadaverin
Amygdalin
Peptone (Witte)
Casein
Edestin
Egg albumin
Glucosamin
Nothing
Asparagin
Acetanilid
Benzamid
Creatinin
Mgm.
Mgm.
Mgm.
2. 09
2. 14
2. 27
2. 42
2.36
2-53
1.97
2.48
2. 42
2.32
I. 96
2.49
2.52
1. 26
2. 00
2. 00
2. 42
2-43
2. 01
i.q8
2. II
2. 14
2. 17
2. 07
93
99
08
26
40
36
45
725
41
34
93
90
45
34
19
92
92
41
76
42
58
56
08
51
62
66
69
06
2. 06
2.15
2-33
2.47
2-43
2. 52
1-975
2.48
2. 41
1. 00
1.97
2. 52
2. 42
I. 26
1.99
1.99
2.48
2.83
1.49
•65
1.63
2-15
2. II
2.15
2. 18
1-55
Per ci.
9»-4
100. 4
102. 6
102. o
103.0
99-5
40.4
100. o
99.6
43- I
100.5
loi. 2
96. o
100. o
99-5
99-5
100. o
94.9
61.6
26.8
81.3
108.5
Mgm.
I. 29
3-33
3-41
3-48
3-73
3-59
3.86
Mgm.
100. O
100. 5
100. 5
74.8
3-27
2-325
3-42
1.80
1. 67
2.45
3-42
2. 12
I. 29
I. 29
1. 29
3-3^
1-25
3-24
3-24
3-27
2. 71
2. 04
2. 12
2. 19
2.44
2.30
2-57
0-5S
2-37
2.49
.92
1.98
1-035
2. 13
•51
•38
1. 16
2. 13
-83
o
o
o
2. 07
1.99
1.99
2. 02
I. 46
Per ct.
97-5
99.0
96-5
100. 8
97-5
loi. 5
28.0
95-6
103.0
39-7
101. o
41. 6
84.6
41. 2
19. o
58.0
85-7
33- S
o
o
o
104.5
94.2
93- o
93-2
70-5
It is not thought that all the compounds used are actually present in
soil. The substances were chosen rather to represent classes of com-
pounds which conceivably might be in soils. Guanin, hypoxanthin,
nucleic acid, peptone, and creatinin have been isolated from soil. It is
64 Journal of Agricultural Research voi. vi, N0.2
realized that the Ust is very incomplete. As opportunity to make or to
purchase more compounds presents itself, the investigation will be con-
siderably extended. From the data presented, it is observed that quite
varying proportions of the pure proteins, which in reality are soluble in
dilute alkali, are extracted. This seems to be a confirmation of the con-
tention that the alkali extract as a whole does not represent a definite
class of nitrogenous compounds. Of the simpler compounds, it is seen
that the more acid and more closely neutral compounds are completely
extracted and remain in the soluble nonprotein portion. An exception
to this is found in the case of nucleic acid. This is to be expected from
its tendency to combine with protein compounds to give insoluble
nucleins. The action of the purin compounds is interesting. In general
the more basic the compound the less the quantity recovered.
CONCLUSIONS
(i) If the results with the pure proteins be considered, it is probable
that the alkali extract as a whole contains no definite group of compounds.
(2) From the results obtained by the precipitation of the alkali extract
with trichloracetic acid it would seem that the soluble nonprotein frac-
tion may contain most of the simpler nitrogenous compounds, and
therefore its determination would give an index of the degree of decom-
position of the organic matter in the soil.
LITERATURE CITED
(i) Bock, J. C, and Benedict, S. R.
191 5. An examination of the Folin- Farmer method for the colorimetric estima-
tion of nitrogen. In Jour. Biol. Chem., v. 20, no. i, p. 47-59, i fig.
(2) Folin, Otto, and Denis, W.
1912. New methods for the determination of total non-protein nitrogen, urea
and ammonia in blood. In Jour. Biol. Chem., v. 11, no. 5, p. 527-536.
(3) Greenwald, Isidor.
1915. The estimation of non-protein nitrogen in blood. In Jour. Biol. Chem.,
V. 21, no. I, p. 61-68.
(4) LiPMAN, J. G.
191 1. Microbiology of soil. In Marshall, C. E. Microbiology for Agricul-
tural and Domestic Science Students, p. 226-291, fig. 66-72. Phila-
delphia,
(s) Potter, R. S., and Snyder, R. S.
191 5. Amino-acid nitrogen of soil and the chemical groups of amino acids in
the hydrolyzed soil and their humic acids. In Jour. Amer. Chem.
Soc, V. 37, no. 9, p. 2219-2227.
(6) ScHREiNER, Oswald, and Shorey, E. C.
1910. Chemical nature of soil organic matter. U. S. Dept. Agr. Bur. Soils Bui.
74, 48 p., I pi.
(7) Shorey, E. C.
1913. Some organic soil constituents. U. S. Dept. Agr. Bur. Soils Bui. 88,
41 p., I pi.
(8) Van Slyke, D. D.
19 11. The analysis of proteins by determination of the chemical groups charac-
teristic of the different amino-acids. In Jour. Biol. Chem., v. 10,
no. I, p. 15-55, 2 fig.
OVIPOSITION OF MEGASTIGMUS SPERMOTROPHUS
IN THE SEED OF DOUGLAS FIR
By J. M. Miller,
Assistant in Forest Entomology, Bureau of Entomology
The larva of a seed chalcidid, Megastigmus spermotrophus Wachtl, has
been very commonly recorded from the seeds of Douglas fir (Pseudotsuga
taxijolia) , but most of these records apply only to mature seed. The method
by which the larvae of this insect get into the seeds has not been previ-
ously described. The oviposition of the female, the period of the growth
at which the seeds are infested, and the subsequent development of the
larvse are matters on which we have no published data.
The following is an account of the oviposition of this species observed
at the Forest Insect Seed Station of the Bureau of Entomology at Ash-
land, Greg., during the season of 191 5.
During the season of 1914 a heavy emergence of adults of Megastigmus
spermotrophus from Douglas fir seed of the 191 3 crop occurred in the vicin-
ity of Ashland. From stored seed kept in a rearing box the male adults
began to emerge on April 12, and the females on April 16; 2,897 adults
emerged from 6^ ounces of seed, the period of maximum emergence occur-
ring between April 23 and May 11. A number of these adults were lib-
erated in a small cage kept in the laboratory. It was found that the
adults would not live any length of time unless fed. Pieces of blotting
paper saturated with sugar solution were hung in the cage and on this
the adults were frequently seen feeding. Young Douglas fir cones were
kept in the cage with the adults for a period of about three weeks; and
although copulation was frequently observed, no attempts were noted
on the part of the females to oviposit in the cones.
During the season of 191 5 another effort to secure a record of oviposi-
tion in a rearing cage met with far better success. Various lots of in-
fested Douglas fir seed were kept at the station, and the emergence of
the adults from this seed was quite similar to that observed in 191 4.
The maximum period of emergence in the laboratory occurred between
April 20 and May 2. From cones which were kept caged over winter
under outdoor conditions at the same elevation, the maximum emergence
occurred between May i and 16. At elevations of 3,000 to 4,000 feet
the emergence occurred during the latter part of May, and above 4,000
feet much of the emergence occurred in June.
Many adult chalcidids were liberated at different dates between April
18 and May 20 in a cage considerably larger than that used in 191 4 (PI. V,
Journal of Agricultural Research, Vol. VI. No. ■
Dept. of Agriculture. Washington. D. C. Apr. lo. 1916
da K-aS
(65)
66 Journal of Agricultural Research voi. vi. no. 2
fig. i). This was kept outdoors in a partially shaded position. The
adults were fed as before with sugar solution.
A Douglas fir branch bearing cones about 3 weeks old was placed in
this cage on April 18. The young cones were then about i}4 inches long,
the scales were still soft, and the seeds had the milky interior and un-
hardened coat. The base of the branch was kept in a jar full of moist
earth. Fresh branches were placed in the cage at intervals until May
15, when the cones were estimated to be about half mature.
Mating was observed in this cage and in the emergence vials of the
rearing boxes during the entire period. The first oviposition of a female
on a cone was observed on April 20 at about 3.30 p. m. A female was
observed crawling about over the bracts and feeling the scales with her
antennae. This lasted for several minutes; then the female paused on
one of the exposed scales with her head pointed toward the base of the
cone. After resting quietly for a moment the abdomen was lifted and
at the same time the posterior end was doubled under so that the sheath
of the ovipositor was brought forward between the legs until the tip
rested on the surface of the cone scale at a point directly under the
insect's head. The point chosen for the insertion of the ovipositor was
close to the outer edge of the scale on which the female rested. The
sheath of the ovipositor was then withdrawn and assumed its normal
position back of the abdomen, while the ovipositor was slowly forced
down into the cone. The abdomen was gradually lowered as the ovi-
positor was thrust into the cone until finally the entire body rested close
to and in a line parallel with the surface of the cone scale (PI. VI, fig. 3).
In this position the female rested for about a minute and then withdrew
the ovipositor. This was accomplished by raising the body and doubling
the abdomen until it assumed a position similar to that in which the
oviposition was started (PI. VII, fig. i). This allowed the ovipositor to
be withdrawn and returned to its sheath.
The oviposition of two females was recorded on April 22 and that of the
same number on April 23. Between this date and April 26 no oviposition
and very little activity on the part of the seed chalcidids were observed.
On the morning of April 26 a female was observed ovipositing, and this
operation was recorded four times during the day. On April 27 about
the same activity occurred. April 28 was a warm, sunny day and great
activity on the part of the females occurred. The cage at this date con-
tained 10 cones and about 50 females. At almost any hour during the
day from one to three females could be seen either ovipositing on the
cones or preparing to do so. From April 29 to May 2 cool rainy weather
prevailed, and almost no activity on the part of the chalcidids occurred
in the cage. May 3, 4, and 5 were warm, sunny days, and the oviposi-
tion could be w^itnessed at any time during the day. Oviposition in the
10 cones in the cage on these dates was in progress continuously during
the day, at which time the best observations of the act were obtained.
Apr. lo, 1916 Oviposition of Megastigmus spermotrophus 67
A spell of rainy weather persisted from May 8 to 25, and no further
records were secured. The subsidence of emergence after the latter date
made it impossible to obtain adults for liberating in the rearing cages,
and efforts to secure further records were not attempted.
Difficulty was encountered in securing photographs, as females will not
oviposit if even slightly disturbed. If a cone was jarred in any way
while a female was in the act, the ovipositor would be withdrawn as rap-
idly as possible. Even though the ovipositor was inserted deep in the
cone the female would struggle to disengage it and fly away. However,
by raising the glass on the front of the cage it was possible to focus a
camera directly on the cones, and several pictures were obtained in this
way. For the purpose of further study and dissection, a number of females
were captured and killed with the ovipositor thrust into the cones. This
was best accomplished by quickly immersing the cone on which the female
rested in a graduate filled with chloroform. This killed the female so
quickly that her efforts to withdraw the ovipositor were seldom suc-
cessful. Several of the females which were killed in this position were
photographed (PI. VI, fig. i, 2).
The time required for oviposition varies from two to five minutes.
The same female was observed to oviposit five times on the same cone,
and it is probable that the operation is repeated many times before the
egg-laying capacity is exhausted. The point selected for the insertion of
the ovipositor was always on the surface of a scale, never on a bract,
and may be either on the margin or near the center of the scale. The
female always assumed a position with head pointed toward the base of
the cone. As Douglas fir cones were pendent at the time of oviposition,
this allowed the female to stand with her head pointed upward (PI. V,
%• 2, 3)-
In cones which were dissected with the ovipositor of the female inserted
it was found that the ovipositor reached the seed in a few cases only.
Apparently where successful the ovipositor passes through the scale
nearest the surface and underlying bracts until it reaches the second or
third scale from the surface. It then follows down through the center
of the last scale nearly to its base and then turns forward into the seed
just ahead of it (PI. VII, fig. 2, 3). The fact that the ovipositor was
seldom found in the seed in the cones dissected is doubtless due to the
fact that the female partly withdrew her ovipositor in the death struggle.
It would seem that successful oviposition occurs only when the egg is
deposited in the seed, as the larvae have never been found to work their
way through the tissues of the cone, and their development is confined
entirely to the interior of one seed.
Numerous cases were found in which the ovipositor did not penetrate
even as far as the base of the scale. This occurred most frequently where
the cones were of such an age that the scales had hardened. In these
oases the tough tissues of the scales seem to bend the ovipositor out of
68 Journal of Agricultural Research voi. vi, No. 3
its course, and in a number of the dissections the ovipositor was bent and
twisted around to a course directly opposite to that intended. This con-
dition was not encountered where the cones were still young and soft.
In fact, after the cones become hardened it is difficult to realize how they
can become infested at all by the chalcidids.
Actual oviposition in the field by the seed chalcidids was observed only
once — on May 28 by Entomological Ranger J. B. Patterson. While
collecting cones he noted a female on one cone with her ovipositor
inserted. The insect withdrew the ovipositor and left the cone very soon
after it was noticed.
PLATE V
Oviposition of Megastigmus spermotrophus in the cones of Douglas fir:
Fig. I. — Type of cage in which the oviposition of Megastigmus spermotrophus was
observed. This cage was kept under outdoor conditions. A branch bearing young
cones of Douglas fir was set in a jar of moist soil and kept in the cage with the females.
Fig. 2, 3. — Female resting on cone with ovipositor inserted. Photographed from
life. On left, original; on right, enlargement of same to show exact attitude of the
female.
Oviposition of Mee:astigmus spermotrophus
Plate V
Journal of Agricultural Research
Vol. VI. No. 2
Oviposition of Megastigmus spermotrophus
Plate VI
Journal of Agricultural Research
Vol. VI, No. 2
PLATE VI
Oviposition of Megasiigmus spermotrophus in the cones of Douglas jfir:
Fig. I, 2. — Two positions of female on surface of cone with ovipositor inserted.
Photographed from dead females which had been killed in this position. Enlarged.
Fig. 3. — Female resting on cone with ovipositor inserted. Photographed from life.
Enlarged.
PLATE VII
Oviposition of Megastigrmis spermotropkus in the cones of Douglas fir:
Fig. I. — Female in act of withdrawing ovipositor from cone. Photographed from
life. Enlarged.
Fig. 2. — Section through a Douglas fir cone on which a female has been killed
while in the act of ovipositing.
Fig. 3. — A portion of same cone and dead female with ovipositor inserted. Slightly
retouched to show course followed by ovipositor in reaching the seed.
Oviposition of Megastigmus spermotrophus
Plate VII
Journal of Agricultural Research
Vol. VI, No. 2
CITRUS CANKERS
By Frederick A. Wolf,^
Plant Pathologist, Alabama Agricultural Experiment Station
INTRODUCTION
The ravages of certain insect pests and plant maladies have, in a con-
siderable number of instances, been so severe as to cause intense alarm.
It has been feared in the case of several crops that their culture was no
longer possible in certain sections because effective means of preventing
the losses resulting from such ravages were not then known. Within the
last two years it has been realized that a new disease known as Citrus
canker has been introduced into the Citrus-growing sections of the Gulf
Coast States. This disease, beyond all doubt, is the most destructive
malady affecting species of Citrus, and when it was realized that its con-
trol and eradication were so difficult, alarm concerning the future pro-
duction of Citrus fruits became almost an hysteria. Those who have
never seen Citrus canker under field conditions regard the reports of the
highly infectious nature of this disease, of its destructiveness, and of the
difficulties experienced in its eradication as the results of an overwrought
imagination. The severity of Citrus canker has not been exaggerated,
however, and growers should lose no time in preventing its further dis-
semination and in effecting its eradication.
HOSTS OF THE ORGANISM
Citrus canker has been found to affect many of the varieties and species
of Citrus, and in all probability none of the species of this genus are
entirely immune. It is perhaps productive of more serious injury to the
varieties of grapefruit, or pomelo {Citrus decumana), than to any other
of the Citrus fruits. Seedling grapefruits appear to be more susceptible
to canker than the budded varieties. Some regard the injury to the
hardy or trifoliate orange {Citrus trijoliata) , which is extensively used as
the stock upon which to bud other species of Citrus, as equally severe.
Certain of the varieties of round oranges {Citrus aurantium) are known
to be very susceptible to Citrus canker and under favorable conditions
suffer as severe injury as grapefruits. The disease occurs also on varie-
ties of the sweet orange. Oranges of the mandarin group {Citrus nobilis) ,
• Published with the permission of the Director of the Alabama Experiment Station.
3 The -writer is greatly indebted to his colleague, Dr. J. S. Caldwell, for suggestions and material aid
diuing the progress of this investigation and for assistance in the preparation of the manuscript. Much
of the chemical portion of the investigation would have been impossible but for the skillful and arduous
assistance of Messrs. A. C. Foster and C. W. Culpepper, formerly laboratory aids in the Department of
Botany of the Alabama Poh-tcchnic Institute. To each of these gentlemen grateful appreciation for the
several services is hereby acknowledged.
Journal of Agricultural Research, Vol. VI . No. :
Dept. of Agriculture. Washington. D. C. Apr. lo. 1916
cz Ala. -3
(69)
70 Journal of Agricultural Research voi. vi. No. a
including mandarins, tangerines, and Satsumas, have also been found to
be diseased. The disease has been observed, too, on several varieties
of lemons {Citrus medico) and limes {Citrus limetia). Thus far Citrus
canker in Alabama has not been found to attack kuraquats, the four
species of which Swingle (17) ^ regards as belonging to the genus Fortu-
nella. It has been observed, however, on the leaves and twigs of the
kumquat in Louisiana. Swingle (18) reports its occurrence on this host
in Japan.
HISTORY OF THE DISEASE
Citrus canker is not of American origin, but beyond doubt was intro-
duced into the Gulf States from Japan. This statement is supported by
the fact that it is known to occur in Japan and the Philippine Islands (18),
and, so far as can be learned, it appeared in the United States several
years ago simultaneously with the importation of Satsuma and trifoliate
stock into Texas in order to supply the large demand for trees for Citrus
plantings. Whether it is indigenous to Japan is not known, but it is
probably native of parts of eastern Asia. Since its introduction into
Texas it has been disseminated by the shipment of diseased trees to other
States and has further been introduced by shipments to these States
direct from the Orient, so that it now occurs in parts of Florida, Alabama,
Mississippi, Louisiana, and Texas.
Citrus canker had probably been present in the United States for
five or six years before it was recognized as a new Citrus disease. Speci-
mens were first collected in September, 1912, but it was not until July of
the following year (i) that the Office of Nursery Inspection of Florida
realized that these specimens did not represent an unusual manifestation
of scab caused by Cladosporium citri. This mistake in diagnosis had
also been made by inspectors in other Gulf States and by officers of State
Experiment Stations and of the Federal Department of Agriculture,
Japanese authorities had also mistaken this disease, since specimens
received at the Florida Agricultural Experiment Station (2) from Japan
had been identified as scab. The disease was brought to the writer's
attention in February, 1914, and has been interruptedly studied by him
since that time.^
A number of publications upon Citrus canker, all preliminary in nature,
have appeared. These papers call attention to the presence of the
disease in the several States, briefly describe its appearance, and recom-
mend concerted cooperation in its eradication. The disease was first
regarded as of fungoid origin, and the first claim that bacteria are the
primary cause of the disease was made by Hasse (6). The present publi-
1 Reference is made by ntunber to " Literature dted," pp. 98-99.
2 The writer severed his connection with the Alabama Agricultural Experiment Station on January i,
1916. This study therefore is incomplete, time not having been afforded for verification of all portions of
the study, and certain probleras which have appeared in connection with the work have not been inves-
tigated. However, it was deemed advisable to record the results of the studies thus far conducted.
Apr. lo, 1916 Citrus Canker 71
cation has for its purpose the recording of studies which are in part
confirmatory of previous studies (2, 5, 6, 16, 19) and which further
contribute to our knowledge of this disease.
ECONOMIC IMPORTANCE
The serious nature and unusual virulence of Citrus canker and the
jeopardy in which it has placed the Citrus industry can best be realized
when it is recalled that the Federal Horticultural Board, on January i,
191 5, placed a quarantine on the importation from all foreign countries
of Citrus nursery stock, including buds, scions, and seeds, in order to
prevent further introduction of the disease into the United States. It is
difficult to obtain figures as to the number of nursery and orchard trees
which have been destroyed in an effort to eradicate the disease from the
Gulf coast. It is equally difficult to obtain accurate figures on the amount
of money which has already been expended by the Federal Government,
together with the State horticultural boards, liberally aided by various
organizations and by private subscriptions, in an effort to stamp out
Citrus canker. Suffice it to say that the actual cost in money for eradica-
tion and for trees destroyed has been enormous.
SYMPTOMS OF THE DISEASE
Citrus canker affects the leaves, twigs, larger branches, and fruits in a
characteristic manner. Upon any of these parts the diseased areas are
light brown in color and project more or less above the surrounding
tissues. The cankerous areas consist of a corky mass of cells covered by
a lacerated grayish membrane. It can be determined with certainty
without a microscopic examination in case one has typical diseased
material, and in case one has seen the disease in the various stages of
development under field conditions. It is sometimes impossible to be
certain whether meager specimens such as are sometimes sent in for
identification are affected with canker or with some other leaf trouble.
This is especially true in the case of canker on the Satsuma orange. If,
however, one is permitted to make a field examination, and can thus
learn of the origin of the trees, and can also observe adjacent trees, typical
material may be found if Citrus canker is present.
OCCURRENCE ON THE LEAVES
The first evidence of canker on the leaves is the appearance of very
small oily or watery dots on the lower leaf surface. They may appear
on either surface, but are more commonly found on the lower leaf
surface. They are of a darker green color than the surrounding leaf
tissue and may at this stage be mistaken for oil glands (fig. 1,2). The
diseased areas are slightly convex, however, and within a few days will
have extended through the leaf, appearing on the upper surface as
72
Journal of Agricultural Research
Vol. VI, No. 2
greenish yellow spots. By continued development the convex surface
of the spots comes to be more and more elevated until the epidermis is
broken by the increased tension and the subjacent tissues are thus ex-
posed to desiccation. The exposed tissues then become corky, darken-
ing with age. The ruptured epidermis is turned back irregularly and
persists as a lacerated membrane. The margin of the diseased area
maintains an oily appearance even after the spots have ceased to increase
in size. Mature spots (Pi. XI, fig. i) vary in size from very minute to
a quarter of an inch in diameter and are typically circular in outline.
They may occur singly; or when they are very numerous, fuse, thus
forming large, irregular areas. Cankered areas are typically elevated on
both leaf surfaces. In the case of canker on Satsumas (Pi. VIII, fig, 2,
and PI. IX, fig. 4), however, there is little or no elevation of the upper
leaf surface. Neither is the oily margin so evident on this host, espe-
cially in case of old cankers,
in which diseased tissues have
become dark brown, simulat-
ing the appearance of mela-
nose. The appearance of the
disease on leaves of Citrus
trifoliata as shown in Plate
X, figure I, is very similar to
that on grapefruit. Stevens
Fig. I.— Diagrammatic representation of young open type of (2) rCpOrtS that he haS neVCr
Citrus canker of half the diameter of the one shown in fo^^d CitrUS Cankcr On trifo-
figure 2. pp. Palisade parenchyma; ue, upper epidermis; _
le, lower epidermis; d, diseased tissues: a, air space aris- Hate OraUgC IcaVCS. The Un-
ing from tensions due to the enlargement of cells and dis- j^vadcd tisSUCS SUrrOUndiug
mtegration of tissues. °
the cankers are paler green
than the normal tissue and gradually form a chlorotic or yellowish zone
(PI. VIII, fig. I, and PI. X, fig. 6), which may invade all the tissues
not actually occupied by the cankers. At this stage considerable
defoHation, especially in the case of grapefruit and trifoliate oranges,
may occur. Cankers on the leaf petioles cause defoliation even though
the leaves are otherwise uninvaded.
OCCURRENCE ON THE TWIGS AND BRANCHES
Limb canker appears more commonly on very young twigs because
of the absence of any considerable suberization, but larger branches are
subject to infection. Growing cankers have been observed on limbs
>^ to K inch in diameter (PI. VIII, fig. 3, 4), The disease has been found
on branches of grapefruit, trifoliate oranges, lemons, Satsumas, and
certain varieties of round oranges. Cankers on twigs are first apparent
as small, circular, watery spots. They rapidly enlarge, become blister-
like and the epidermis ruptures, exposing the cankerous tissue below.
At this stage they project more or less prominently and are very similar
Apr. lo, 19 16
Citrus Canker
73
in appearance to the spots on the foHage. Isolated cankers remain
circular in outline. When the spots originate close together, however,
large irregular, variously cracked or fissured cankers are developed,
which may involve an area several inches in length. The epidermis
persists as a grayish broken membrane at the margin of these cankers
(PI. VIII, fig. 5). Twigs and larger branches may be completely girdled,
resulting in the death of the distal parts. Affected trees exhibit a
stunted growth, and numerous branches may be developed below the
d5dng tips.
The disease is very severe upon stems of grapefruit and trifoliate
oranges. On the latter host the thorns are abundantly cankered and the
base of the thorns appears commonly to be the initial seat of im'ection.
Fig. 2. — Diagrammatic representation of canker on old Citrus leaf: pp. Palisade parenchyma; uc. upper
epidermis; le, lower epidermis; p, pycnidium of Pkoma socia; d, diseased tissues; a, air space arising
from tensions due to the enlargement of cells and disintegration of tissues.
lyimb cankers on trifoliate oranges oftentimes are zonate with different
shades of brown, especially if the outer membranes have not yet been
ruptured.
OCCURRENCE ON THE FRUIT
The cankerous areas on the fruits are quite similar in appearance to the
leaf cankers, differing mainly in the larger size of the former. They are
scurfy elevations, for the most part circular in outline and surrounded by
a zone of chlorotic rind tissues. The corky diseased tissues are quite
superficial; and if the spots unite, large scaly areas are formed (PI. X,
fig. 2). In this case the fruits may crack open because of their increase
in size owing to the growth of the fruits and may become prematurely
yellow and drop. Fruits which are badly cankered and have burst open
are, of course, subject to invasion by various organisms of decay. Even
if they remain on the tree, they are rendered very unsightly and are
unsalable.
OCCURRENCE ON THE BUDS
Nurserymen experience considerable losses from failure of Citrus buds
to unite with the stock. In some cases when Citrus irifoliala seedUngs
affected with canker are used as stock, losses of over 50 per cent have
74 Journal of Agricultural Research voi. vi, no. a
been sustained. The operation of budding either directly conveys the
organisms into the wounded tissues or they are subsequently washed into
them from cankers above the insertion of the bud before union has been
effected.
ETIOIvOGY OF THE DISEASE
The primary cause of Citrus canker is a bacterial parasite, Pseudomonas
ciiri Hasse (6). Hasse isolated this organism from cankers on grapefruit
and proved it to be pathogenic to grapefruit seedlings. This claim was
established at a time when the disease was regarded as of fungus origin.
Hasse further pointed out the fact that a number of fungi were isolated
from old Citrus cankers. The writer had found a fungus, as had also
Prof. H. E. Stevens, of the Florida Agricultural Experiment Station,
belonging to the form genus Phoma, commonly associated with cankerous
tissues. The writer's initial inoculations were made not with pure
cultures of Phoma, as has subsequently been learned, but with cultures
which had overrun the bacterial parasite. Successful infections reported
in the previous publication (19) are thus accounted for. Consideration will
be given in another part of the present report to the part which Phoma
spp. and certain other fungi play in the production of Citrus canker.
PATHOGENICITY
Pseudomonas ciiri has repeatedly been isolated during the past season
from cankers on grapefruit, trifoliate orange, lemon, and Satsuma oranges.
The strains from these different hosts present the same cultural charac-
ters. Because of this, together with the added fact that no difficulty has
been experienced in making cross inoculations, the strains are regarded
as identical.
The plants used in making the inoculation experiments were grown in
the greenhouse at Auburn, Ala. Typical cankers have been produced on
McCarty and seedling grapefruits (PI. IX, figs. 1,2), pineapple oranges,
Satsuma oranges, and seedling trifoliate oranges. Infections on all these
species were as readily secured, whether the organism had been isolated
from Citrus trifoliata, Satsuma, grapefruit, or lemon. Neither was there
any evident difference in virulence of any of the strains. A suspension
of the organism taken from pure cultures grown either on potato cylinders
or in bouillon was used in making the inoculations. This suspension
when applied with an atomizer resulted in a high percentage of successful
inoculations. A greater number of successful inoculations were secured,
as would be expected, when the plants were covered with bell jars to pre-
vent the too rapid evaporation of the moisture. When the inoculum was
introduced into the tissues of leaves, stems, or fruits through needle punc-
tures, cankers developed in all cases. In some cases the suspension was
applied to leaves with the fingers. They were dipped into the suspension
Apr. lo, 1916 Citrus Canker
75
and the material was then applied by gently rubbing the leaves between
the thumb and fingers. In a few cases it was arranged so that twigs
bearing young leaves could be immersed for an hour or two in a bacte-
rial suspension. Leaves inoculated in this manner are shown in Plate
VIII, figure 2. It is to be noted that the infections are so numerous as
to involve the greater part of the lower leaf surface.
The period of incubation appears to vary, depending on temperature,
moisture, and age of the plant tissues. Very definite signs of the disease
have been noted within 72 hours after inoculation. In other cases 10
days were required before the infections were evident to the eye. The
longest periods were secured on Satsumas.
An organism of the same color as Pseudomonas citri and similar in
appearance on certain media, but which does not exhibit the character-
istic growth of P. citri on potato cylinders, has commonly been isolated
from old cankers. This organism has not been found to be pathogenic
on species of Citrus, however. There 'can be little doubt of the pathoge-
nicity of the organism concerning which Hasse made her preliminary report
(6). It is to be noted that her Plate X, figures A, B, represent natural
infections and Plate X, figure C, artificially produced cankers. These,
however, are regarded as identical in appearance. Artificially inoculated
seedlings are represented also in Plate IX. As can readily be seen, the
artificial cankers are much more prominently projecting than natural ones,
are evidently greenish white in color, and there has been no discoloration
of the leaf tissue surrounding the spot. The writer has never, under field
conditions, seen specimens which resembled these artificial inoculations
represented in Hasse's Plates IX and X, and he, furthermore, has exam-
ined fresh specimens in various stages of development sent from Florida,
Alabama, Mississippi, Louisiana, and Texas. However, cankers similar
in appearance to Hasse's artificial cankers have been produced in the
greenhouse. Following her suggestion that the open, spongy type of
canker is due to favorable conditions of moisture and temperature, seed-
ling grapefruit which had been atomized with a suspension of P. citri
were kept continuously covered with a bell jar. They were watered
sufficiently often so that the air under the bell jar was maintained at a
high relative humidity. Within 10 days the cankers shown in Plate
IX, figure 3, had developed. These are regarded as similar in appear-
ance to those previously produced by Hasse and represented in her Plates
IX and X.
DESCRIPTION OF PSEUDOMONAS CITRI
The primary cause of Citrus canker is a yellow, i -flagellate organism.
Its motility can be observed when taken directly from young cankers
and examined in a drop of water. In this case it will be found to occur
singly or in pairs. On solid media it may form into chains of six or
more elements. It is quite variable in shape and size. When taken
27470°— 16 2
76 Journal of Agricultural Research voi. vi, no. 2
from young cankerous tissues it is usually a short rod with rounded ends
which measures from 1.5 to 2.5 by 0.5 to 0.75/i. In old cultures the ele-
ments may be ellipsoidal. No endospores have been demonstrated; nor
have involution forms been observed.
The organism stains readily with solutions of carbol fuchsin, analine
gentian violet, and methylene blue. Only negative results have been
secured with Gram's stain. When the organism has been grown on
potato cylinders and is stained with anilin gentian violet, it has an
apparent capsular portion (fig. 3). This capsular portion gives rise, no
doubt, to the viscidity which characterizes its growth on steamed potatoes.
The slime on old potato cultures can be drawn out an inch or two and
does not dissolve readily in liquid cultures.
Young cultures of this organism on steamed potato cylinders have a
very characteristic appearance. The growth is bright yellow, smooth,
moist, glistening, and raised, with a narrow white zone along the margin
of the bacterial growth. This white margin does not persist, since by
its rapid growth the organism covers the en-
tire surface of the medium. It acts very
strongly on potato starch, as indicated by the
entire absence of an iodin reaction on steamed
potato cylinders 6 to 8 weeks old. The mid-
dle lamellae in such old cultures have been
dissolved, and the empty cells can readily be
separated from one another.
The organism has been grown on nutrient
agar made by adding a water extract of corn
Pig. 3— Pseudomo7ias citri: a. stsiined nieal, bean m^eal, green beans, cowpeas, pota-
with carbol fuchsia; 6, stained with , . . . „ i ^ i
wSiams's flagellar stam (adapted toes, nce, oraugc jmcc, or orangc leaves and
from Hasse); c, stained with aniUn stemS, but the grOWth On nOUC of thcSC media
gentian violet. .^ characteristic No attempt was made to
titrate any of these media to determine their acidity or alkalinity.
Colonies appear on the second day in poured plates of green-bean agar
kept at room temperature. Within four or five days the surface colonies
in poured plates will have become 2 or 3 mm. in diameter. The margin
of the colonies is entire, and they are opaque yellow in color. They are
appreciably raised and have a smooth, wet-shining surface. The char-
acter of the margin and of the surface is shown in Plate XI, figure 4.
It will be noted that the reflection of the two windows in the room in
which the exposure was made is shown in each of the colonies.
A filiform growth, following the Hne of the stroke and widening at the
base of the slant, is formed in stroke cultures on green-bean agar. The
growth does not penetrate the agar and does not give rise to the produc-
tion of any stain or odor. In stab cultures on this medium a filiform
but otherwise nontypical growth is produced, which when viewed from
4
yjoo
;^^
Apr. lo, 1916 Citrus Canker
77
above appears like the surface colonies in poured plates. Growth in
stab cultures on various media is always or nearly always best at the
surface of the media.
On nutrient-gelatin plates the colonies are circular in outline, slightly
raised, entire margined, and yellowish. In gelatin stabs a filiform growth
appears along the line of puncture, with the greatest growth at the
surface of the medium, and with a rather slow liquefaction.
The organism is regarded as a facultative anaerobe. No gas is formed
in fermentation tubes containing a 2 per cent solution of Witte's peptone.
With this as a basal solution, five solutions were made by adding i per
cent of one of the following carbon compounds: Saccharose, dextrose,
lactose, maltose, and glycerin. All inoculated tubes developed a slight
cloudiness, which extended into the closed end of the tube by the second
day. More vigorous growth occurs in the open end of the tube, however,
and after four or five days the cloudiness is very marked. Yellowish
flocculent particles appear later in the open end, and a yellowish ring
is formed at the surface. No gas formed in any of the solid media in
which the above-mentioned carbon compounds were added to the nutrient
agar.
In stab cultures on litmus-dextrose, litmus-lactose, litmus-saccharose,
and litmus-glycerin agar no gas formation was apparent in lo-day old
cultures. It is not known whether acidification will occur in old cultures
on these media.
In sterile tubes of litmus milk there is a rather slow reduction of the
litmus. After five days there is a slight increase in the blue color. The
reddish whey is gradually formed on the surface, and the casein is
precipitated.
There is no reduction of nitrates in Witte's peptone solution containing
a trace of potassium nitrate. Phenoldisulphonic acid was used as a
reagent 10 days after the date of inoculation, at which time both the
check and the solution in which the organism was growing were colori-
metrically alike.
Only negative tests for indol were secured in peptonized beef-bouillon
cultures. A very conspicuous clouding occurs in this medium within 24
hours after inoculation. As these cultures get older they become some-
what flocculent, and a yellowish ring is formed at the surface of the
media.
The thermal death point, as found in preliminary tests, was between
58° and 70° C. In order to determine more nearly the point, tests were
made by exposing the organism taken from potato cylinder cultures, and
transferred to tubes of bouillon. The tubes were then placed in a water
bath for 10 minutes at some given temperature between these Hmits.
The temperature of the bath was kept constant during the period of
exposure. The tubes were subjected to room temperature for several
78 Journal of Agricultural Research voi. vi, no. 2
days to observe the development of cloudiness. In order to be certain,
however, of the viability of the organism, loops of bouillon from these
tubes were transferred to planted plates of nutrient agar, and the sub-
sequent development noted. No growth occurred in the tubes exposed
at temperatures above 65° C.
No attempts have been made to determine the exact degree of tolerance
of this organism to acids. When transfers were made to dextrose-
peptone agar + 10, + 20, and + 40 Fuller's scale, it was found at the end
of three days to have grown in the first two, but growth was completely
inhibited in + 40 acid. Hydrochloric and citric acids were employed in
acidification.
The organism seems to exhibit a very considerable resistance to drying.
In the desiccation experiments bacteria from vigorous pure cultures on
potato plugs were smeared by means of a sterile platinum needle on clean
miscroscopic slides in moist chambers. The moist chambers containing
the microscopic slides were sterilized prior to transferring the bacterial
smear to the slides. These preparations were made on June i , and placed
in a wall closet in the laboratory. On July i , August i , and September i
several of the microscopic slides were removed from the moist chambers
and placed in sterilized Petri dishes, using proper aseptic precautions
in making the transfers. Tubes of melted nutrient agar which had been
cooled almost to the point of solidification were poured upon these
smeared slides. No growth occurred in the case of those tested on
September i, but those tested on July i and August i were still alive.
From this it is believed that the organism can retain its viability for
about two months.
The group number according to the descriptive chart of the Society of
American Bacteriologists is 221.3332513.
LIFE raSTORY OF THE ORGANISM
Pseudomonas citri, so far as is known, passes its entire life cycle under
natural conditions within the tissues of the host. New infections appear
in spr,ing shortly after the new growth has begun. In southern Ala-
bama the first appearance of Citrus canker in the field was noted on May
II, in 1914, and on May 27, in 1915. Old diseased areas on the foliage
together with the cankers on the twigs and larger limbs are undoubtedly
the source of infection in the spring. New leaves formed near old twig
cankers are especially liable to become diseased first. Infections are not
confined to the new growth, however. Old diseased areas on leaves and
branches may enlarge by the renewed growth of the organism which has
remained dormant on the margin of the old cankers. New cankers may
also develop on old foliage and twigs, especially near the old, actively
growing cankers. Under favorable conditions new infections may appear
at any time throughout the growing season of the host. In one instance
Apr. 10, 1916 Citrus Canker 79
new infections are known to have appeared abundantly under field con-
ditions during November, 1914. Old leaves on the ground may possibly
harbor the organism and there it may remain viable for a long time.
Unsuccessful attempts, however, have been made to recover the organ-
ism from leaves kept in the laboratory from September, 1914, to May,
1 91 5; nor has recovery been possible in the case of twig cankers kept
under laboratory conditions from March to October, 191 5.
It is believed, moreover, that the organism survives the winter in
fallen leaves and that these fallen leaves constitute a very important
source of infection in the following spring, especially in the case of nursery
trees which have been planted between diseased grove trees.
There is every reason to believe also that the organism can remain
alive in soil. This is evidenced by numerous instances in which new
sprouts have come up from the roots of diseased trees which had been
burned. A large percentage of these sprouts are early found to be diseased.
Furthermore, the leaves on the lowermost branches or those in actual con-
tact with the soil are commonly the first to become diseased.
The fact that the stomata, or breathing pores, on species of Citrus
occur only on the lower leaf surfaces and that infections developed only
on the lower surface of the leaves in all of the inoculation experiments in
which the plants had been sprayed with bacterial suspensions led to the
inference that the canker organism must gain entrance to the leaves
through the stomata. That such is the case was established by leaf
sections which were fixed 72 hours after inoculation and which were sub-
sequently properly infiltrated, cut, and stained (fig. 4). Lenticels very
probably serve as portals of entrance for the organism into the stems.
A film of moisture on the surface of the leaf, twig, or fruit enables the
organism to move about and thus to gain entrance into the substomatal
cavity. Under ordinary conditions inoculation will be successful only
in the presence of moisture. Wounds or abrasions from any cause may
afford an entrance to the bacteria. Inoculations not infrequently occur
through wounds made by thorns. Inoculations on leaves made by thorn
scratches are shown on Plate X figure 3. Thorns which come in con-
tact with limbs near by may inflict wounds which have subsequently
been observed to be the point of origin of limb cankers. Cankers have
also been found at the point of contact of limbs which rub together
through movement by the wind.
When once the bacteria have passed through the stomata into the
substomatal cavities, they multiply rapidly and effect a passage between
the host cells to the intercellular spaces which become filled with solid
masses of bacteria. As the bacteria continue to multiply, the cells
farther away from the substomatal chambers become involved seriatim.
In this way an area circular in outline and extending entirely through the
leaf comes to be invaded. Various stages of invasion of the leaf tissues have
8o
Journal of Agricultural Research
Vol. VI, No. 2
been observed in serial paraffin sections. Within three to five days after
inoculation the disease is evident in the form of oily or watery spots.
Within another week with favorable weather conditions and on young
leaves the epidermis will have ruptured on one or both surfaces and open
cankers will have formed. At this stage, before the exposed cells have
become desiccated, the greatest danger of spreading the infection exists.
Young tender tissues seem to be more susceptible to infection at this time
than mature tissues. The disease progresses more
rapidly, too, in young tissues than in older parts.
RELATION BETWEEN PARASITE AND HOST
No Study has been made other than the prelimi-
nary account of Hasse (6) of the effects of Psevdo-
monas citri on Citrus tissues. She states (p. 98)
that—
There is a rapid development of cells, and the tension re-
sulting from the abnormal growth quickly ruptures the
epidermis. The cells are found to be filled with short rod
bacteria. All the cells exhibit more or less enlargement.
In later stages in the development of the canker some of the
cells disintegrate, and lesions are formed. The organism
appears to act more vigorously on the cell contents than on
the cell walls, and in due time the cell contents are exhausted.
The cell walls which remain become suberized.
This problem was first attacked by making a
histological study of the diseased tissues. For this
purpose cankers in various stages of development
on fruits and leaves were cut out so as to include
some of the surrounding healthy tissue. Cankers
which had developed under conditions of very high
relative humidity (PI. IX, fig. 3, and PI. X, fig. 4)
and which were consequently of the spongy type and
white in color yielded especially interesting results.
This white color is due to the presence of air between the cells and can be
made to disappear if the cankers are immersed in water. These excised
cankers were then killed in strong alcohol, embedded in paraffin, sec-
tioned, and stained with carbol fuchsin. This stain renders the bacteria
bright red, making it easily possible to determine their position within
the tissues.
Contrary to Hasse's observation, the bacteria teem around and
between the host cells, being present in especially large numbers in
the intercellular spaces (fig. 5). When the organism occurs within the
Fig. 4. — Early stage of Cit-
rus canker in cross section
on a young leaf of seedling
grapefruit. The leaf was
inoculated by immersion
in a suspension of Pseu-
dorrwnas citri from pure
culture. The material was
collected 72 hours after in-
oculation. It was then
killed in strong alcohol,
embedded in paraffin, sec-
tioned, and stained with
carbol fuchsin. The or-
ganism entered the leaf
through the stoma, multi-
plied in the substomatal
chamber, and spread to
adjacent intercellular
spaces. Drawing made
with a camera lucida.
X600.
Apr. lo, 1916
Citrus Canker
81
cells, one is led to conclude, since they appear to be confined to such cells,
that entrance was effected after some mechanical rupture of the host
cells.
A microscopic examination of sections of young spongy cankers in
which there has been no desiccation from contact with the air shows that
the host cells are not killed at first. Instead, they are considerably
hypertrophied and become lightly attached to each other, as shown in
figures 6 and 7. In fact, if fresh cankers are cut off with a sharp razor
and mounted on a sHde in a drop of water, some of the host cells separate
intact and of their own accord from the mass of cankerous tissue. Little
if any hyperplasia is believed to occur. It is highly improbable that cell
division would occur in cells in which such profound changes Vv^ere taking
place. It is evident from figures 6 and 7 that the enlargement of cells
already present would account for the production of the cankerous tissues.
The same is believed to be true in Plate X, figure i , illustrating Hasse's
observations. It is
not clear, however,
from her explanatory
statement that
"there is a rapid de-
velopment of cells"
whether hypertrophy
or hyperplasia is
meant. Death of
cells in the later
stages of develop-
ment of canker is
probably caused by
drying (PI. IX, fig.
5). The dried canker-
ous tissues gradually
become suberized.
To explain the enlargement of the cells and their separation from each
other, two hypotheses are advanced : First, the middle lamellae are dis-
solved by an enzym, pectinase, secreted by the bacteria; second, osmotic
pressure of the colloidal cell contents is modified so that the cells have a
greater affinity for water. Evidence in support of both hypotheses has
been secured which in part, at least, explains these interesting phenomena.
An attempt was made to demonstrate the secretion of pectinase by
Pseudomonas citri by the following method: Six flasks of bouillon were
inoculated with pure cultures of the organism. It was realized that the
production of enzyms is largely dependent on the nature of the culture
medium and that pectinase might be formed only within the host tissues.
For this reason grapefruit leaves were placed in three of these flasks of
bouillon prior to their sterilization and inoculation. After the organism
Fig. s. — Pseudomonas citri: (a). In the mesophyll tissue and (6) in the
palisade parenchyma. This material was fixed in strong alcohol, infil-
trated with paraflBn, sectioned, and stained with carbol fuchsin. Out-
lined with a camera lucida.
82
Journal of Agricultural Research
Vol. VI. No. a
had grown in the flasks for four weeks, the bouillon was filtered through
a Chamberland filter. This filtrate contained no living organisms, as
demonstrated by transfers of platinum loopfuls to agar plates, with no
growth on these plates after three days. When at the end of three days it
was known that the filtrate was sterile, fresh grapefruit leaves were intro-
duced into the filtrate. These leaves were sterilized, prior to their intro-
duction, by immersion for half a minute in i to i ,000 bichlorid of mercury
and by rinsing them subsequently in three changes of boiled tap water.
Negative evidence of the presence of living organisms in the filtrate con-
taining the grapefruit leaves was secured by agar plates made one week
after the introduction of the leaves into the filtrate. An examination
of the leaf tissues at the end of two weeks showed no evidence of dissolu-
tion of the middle lamellae. This was true in the case of the filtrate
obtained from both sets of the six original flasks.
In another experiment Irish potatoes were cut into slices and placed
in moist chambers on moist filter paper. Pseudomonas citri was then
Fig. 6. — Drawing of a stained section of a natural canker on grapefruit.
transferred to these cut surfaces. Within a week hemispherical areas in
which the cells were easily separable one from the other had been formed
immediately beneath the colonies. That P. citri alone had caused this
condition was shown by the reisolation in pure culture of this organism
from the softened potato tissues. Because of this result, together with
the fact, previously indicated, that the cells of cankerous tissues are so
easily separable, and in spite of the negative evidence of enzym secretion
in bouillon culture, it is believed that pectinase is secreted by the parasite.
The fact of the increased size of cells of cankerous tissue in itself sup-
ports the hypothesis that there has been an increased osmotic pressure
within affected cells. Several facts contribute toward solving the question
of how this increased pressure is brought about. In the first place the cell
contents must manifestly be modified by the dissolution of the middle
lamellae, since there would be a tendency toward the establishment of equi-
librium between the solution between the cells and the cell sap. Again,
the growth of the organism between the cells with the consequent passage
Apr. lo, 1916
Citrus Canker
83
of nutritive substances through the cell walls must exert an influence
on the concentration of the cell sap. Then, too, the gelatinous material
making up the bacterial cell walls certainly possesses considerable
power of imbibition.
Further it has previously been pointed out that Pseudomonas citri
exerts a strong diastatic activity when grown on potato cylinders. The
production of this enzym has also been demonstrated by growth on
starch agar prepared according to the method described by Crabill and
Reed (4). Within a week a clear halo around the edge of the bacterial
colony is formed on this substratum, thus making a striking ocular
demonstration of dissolution of starch by the canker organism. If
diastase, secreted by this organism, is readily diffusible through the cell
walls, and it is reasonable to suppose that it is, it can convert the rela-
tively insoluble starch into more soluble carbohydrates and thus increase
the osmotic pressure of the cell sap.
It is not impossible that these several causes of increased osmotic
pressure operating conjointly or separately may so profoundly modify the
imbibitory properties of certain col-
loidal substances within the cells that
their affinity for water is in conse-
quence greatly increased.
No attempt has been made to de-
termine the isotonic coefficient of the
cell contents of the enlarged cells, but
for the reasons just mentioned it is
believed to be greater than that of
normal cells.
DISINTEGRATION OF THE TISSUES
Fig. 7. — Cross section in outline of a spongy canker
on the rind of a fruit of Citrus decumana, show-
ing ruptured epidermis and hypertrophy of the
rind tissues, the cells of which are loosely at-
tached.
An attempt has been made to gain
certain information relative to the
organisms involved in the disintegra-
tion of cankerous tissues, together with the nature of their activity on this
tissue. It was previously pointed out that a species of Phoma is com-
monly associated with Citrus canker. Two other species of fungi belong-
ing to the genera Gloeosporium and Fusarium are also sometimes
present. Since certain bacteria and fungi are known to possess the power
of hydrolyzing cellulose (13, 15), of which complex substance cell walls
are largely constituted, an effort has been made to study the action of
the organisms associated with canker upon pure cellulose. For this
purpose cellulose agar was prepared according to the following method.
Schweitzer's reagent was first made by adding ammonium chlorid and
then an excess of sodium hydrate to a solution of copper sulphate. The
blue precipitate thus formed was washed, pressed on a cloth filter, and
dissolved in ammonium hydrate (sp. gr. 0.92). In this solvent 15 gm.
84 Journal of Agricultural Research voi. vi. no. 3
of sheet filter paper were dissolved, the solution was diluted about
10 times with water, and the cellulose was precipitated with a 15
per cent solution of hydrochloric acid. After considerable dilution
the mixture was filtered, and the residue was washed repeatedly with
water to remove all copper and chlorin. This residue was added to an
agar medium consisting of agar, 10 gm. ; monopotassium phosphate,
I gm. ; magnesium sulphate, i gm.; sodium chlorid, i gm.; ammonium
sulphate, i gm.; calcium nitrate, 0.5 gm.; and the whole was made up
to 1 ,000 c. c.
Poured plates of cellulose agar were made during May, inoculated with
Pseudomonas citri, Phoma sp., Gloeosporium sp., and Fusarium sp., and
incubated at room temperature. All grew poorly and none of the fungi
fruited on this medium. There was no evidence of the production of
cellulase except by Phoma sp. Within two weeks this organism had
formed clear translucent halos as shown in Plate IX, figure 5, indicating
that the cellulose had been hydrolyzed. Even though Phoma spp.
strongly dissolve paper cellulose, they may not behave in this manner
toward cell walls of Citrus spp., since other carbohydrates present would
be more readily available than cellulose.
A further effort has been made to determine what other enzyms are
secreted by these organisms and what part they might consequently play
in the destruction of the tissues. Accordingly, Knop's mineral nutrient
solution was prepared for use as a stock solution. This stock solution
was then tubed and sterilized. To one set of these tubes of Knop's
solution starch was added, to another saccharose, and to another
maltose. They were then set aside and tested to determine whether
they were sterile. It had previously been determined that sterilization
subsequent to adding the carbohydrates resulted in a certain amount of
conversion of these carbohydrates. When it was determined that they
were sterile, four sets of four tubes each were taken of each of the nutrient
solutions. Three tubes in each set of four were inoculated with pure
cultures of one of the four organisms mentioned above and one tube in
each set was left as a check. After 10 days the solutions were tested,
with the following results : Fehling's solution showed a strong reduction
in the starch solutions in which Pseudomonas citri and Phoma sp. had
been grown, showing the production of diastase. There was no change
in the checks nor in the solutions in which the other organisms were
grown.
Inversion of saccharose, as evidenced on the reduction of Fehling's
solution, had been accomplished in the solutions in which Phoma sp. and
Fusarium sp. had been grown, indicating the presence of invertase.
Positive tests for dextrose or glucose were secured with Barfoed's reagent
and with Nylander's reagent in these inverted saccharose solutions.
Negative results were secured with the other organisms and with the
checks.
Apr. lo. 1916 Citrus Canker 85
Phoma sp. alone seemed to have any action on maltose. Inversion
into dextrose was shown by positive tests with Barfoed's reagent.
Negative tests for lipase production were secured in the case of each of
the four organisms.
From the foregoing tests it is seen that Phoma sp. secretes cellulase,
diastase, invertase, and maltase, and must therefore be regarded as very
destructive to the carbohydrate material of diseased tissues. Cellulase
very probably aids in the destruction of the cell walls ; diastase converts
the starch into maltose and dextrin and then further acts on the dextrin.
When a few drops of iodin were added to a starch solution in which
Phoma sp. had grown, blue and red colors developed, indicating amylo-
and erythro-dextrin. Maltase probably further reduces the maltose to
dextrose.
It has also been found that Phoma sp. affects the acidity of the medium
upon which it is grown. This was determined by growth in pure culture
of the fungus on leaves and fruits of Citrus trifoliata. This material was
first macerated by passing it through a meat chopper. Thirty-gm.
samples of ground leaves and of fruits were then placed in 250 c. c.
Erlenmeyer flasks and were sterilized in an autoclave. After steriliza-
tion some were inoculated with Phoma sp. from pure cultures and others
left as checks. A copious white growth occurred on those which had
been inoculated. After a month 150 c. c. of distilled water were added
to each of the flask cultures and to the checks. The flasks were then
heated on a water bath for 30 minutes, the liquid filtered through asbes-
tos, and 25 c. c. of the filtrate taken for titration, using NI20 sodium
hydroxid, with litmus paper as an indicator. The following is repre-
sentative of the results obtained: 7.1 c. c. of NI20 sodium hydroxid
neutralized 25 c. c. of the filtrate from the leaves in the check flask and
9.8 c. c. that from the fruits in the check flask. The filtrate from the
leaves upon which Phoma sp. had been growing was neutral to litmus,
and that from the fruits required 1.3 c. c. of NI20 sodium hydroxid to
neutralize it. From this it is concluded that Phoma sp. is able to utilize
the organic acids as a source of food, a condition contrary to that which
Hawkins (7) found in a study of the chemical changes produced by the
brown-rot fungus on peaches.
TAXONOMY OF THE FUNGUS
An effort has been made definitely to assign this species of Phoma to
one of the numerous species of the form genera Phoma and Phyllosticta,
which have previously been described as occurring on parts of Citrus spp.
The pycnidia of the species under consideration are globose, ostiolate,
100 to 150A1 in diameter (PL XI, fig. 5, 6) and wholly or partially em-
bedded within the cankerous tissue. The pycnidial walls are thin,
being thickest around the ostiolum, and are very similar in color to the
corky brown host cells. The conidia are elliptical or oblong in outline.
86
Journal of Agricultural Research
Vol. VI, No. 3
hyalin, and 9 to 1 2 by 3 to 4)U. They germinate within 24 hours in water
or in a variety of culture media. A white mycelial growth is produced
on bean agar. Pycnidia are readily formed on agar (PI. XI, fig. 3)
modified by the addition of a water extract from corn meal, rice, cowpeas,
orange stems, etc. (fig. 8).
The fungus, furthermore, was very probably introduced into the
United States similtaneously with Pseudomonas citri. It is impossible to
determine the position of this organism among previously described
species, since it has been found to be morphologically not unlike several
of them. Its relation to the production of Citrus canker is definitely
established as a result of this study. Then, too, no particular difficulty
would be experienced by other investigators in identifying it because of
its association with
Citrus canker. In
view of these facts it
seems well to de-
scribe it as a new
species with the fol-
lowing brief technical
diagnosis : ^
Phoma socia, n. sp.
Pycnidia irregularly
distributed, globose,
wholly or partially em-
bedded, 100 to 150M in
diameter; walls thin,
corky brown in color,
thickened only around theostiolum, which opens centrally; conidia continuous, ellip-
tical or oblong, hyalin, 9 to 12 by 3 to 4ju.
Occm-s in the cankers produced by Pseudomonas citri on living leaves and branches
of Citrus trifoliata, C. nohilis, and Fortunella sp. and on living leaves, branches,
and fruits of C. decumana and C. aurantium.
Fig. 8. — a. Cross section of a pycnidium of Phoma socia from a grape-
fruit leaf. This material was fixed in chromo-acetic acid, embedded
in parafiin, sectioned, and stained in saffranin and gentian violet.
Drawing outlined with the aid of a camera lucida. 6, Germination
of conidia of Phoma socia after 24 hours in water, c. Mycelium of this
fungus in old cultures.
ACIDITY AND RESISTANCE TO CANKER
It is generally conceded by both nurserymen and growers and has been
substantiated by the field observations of the writer that Satsuma oranges
are not as susceptible to Citrus canker as grapefruit. This difference
may be noted when both species are grown in locations where they are
equally exposed to infection. The tolerance of bacteria to acidity has
been found to be relatively low. Resistance to certain fungus diseases,
as, for example, the resistance of hard wheat to rust, has been found (3)
1 Phoma socia, sp. nov.
Pcrithedis irregulariter distributis, globosisplus minusve immersis, loo to i^oMdiam.; contextu mem-
branaceo.corticale-brunneo.cumcellulis circa ostiolum pseudoparenchymatids, centro perf oralis ; sporulis
continuis, elUptids v. oblongis, hyaUnis 9-12 X 3-4M. Hab. in foliis ramisque, vivis Citri irifoliatae,
C. nobilis et Fortunellae sp. et quoque in foliis, ramis fructibusque C. decumanae et C. aurantii. Socia
adest Pseudomonas citri Hasse.
Apr. lo, 1916
Citrus Canker
87
to be correlated with the acidity of the cell sap. Because of these several
facts, an effort has been made to determine whether the difference in
susceptibility between Satsuma oranges and grapefruit can be accounted
for on the basis of difference in acidity. Leaves collected from plants
growing in the greenhouse were used in these tests. The leaves were
finely macerated by trituration; distilled water was then added to make
a volume equaling 200 times the weight of the finely ground leaves;
phenolphthalein was added as an indicator; and the acids present in the
sample were titrated with N/io sodium hydroxid. This method is open
to criticism where absolutely accurate determinations are sought, but
is regarded as satisfactory in indicating relative differences. Consider-
able variations in acidity of the same species were noted, dependent
largely upon the cessation of photosynthetic activity at night. Greater
acidity, as would be expected, occurred in samples collected early in the
morning. Representative results of these tests, however, are shown in
Table I.
Table l.—Acidiiy of oranges and grapefruit
Variety.
Wet weight
of tissue.
Quantity of Nlio sodiiim hy-
droxid to neutralize i gm.
wet weight of tissue.
Percentage of
moisture
in sample.
Percentage of
total acidity
based upon
Actual.
Average.
content.
Gm.
C.c.
C.c.
Satsuma (old)
3.62
3.06
4. 14
4.17
2. 92
I- 0359
.9967
I. 0406
1-0431
.9760
I. 0184
60. 2
I. 691
Satsuma (young) . .
f 2. 60
2.90
I 2.55
•9423
I. 0172
. 9609
. 9735
67.8
1.465
Grapefruit (old) . . .
2.71
2-53
1.66
2. II
I 1-95
.8672
.8695
•8434
.9005
.9128
.8787
59- 0
1.490
Grapefruit (yoimg) .
f 1-97
{ 2.03
I 1-56
.8629
.8620
•8653
1 .8634
79-9
I. 080
The leaves of Satsuma oranges are consistently higher in acid content
than those of grapefruit, since the former require 1.0184 c. c. of N/io
sodium hydroxid to neutralize i gm. of wet weight of leaf tissue, young
Satsuma leaves, 0.9735 c. c, old grapefruit leaves, 0.8787 c. c, and
young grapefruit leaves, 0.8634 c. c. When the acidity of the cell sap is
computed on the basis of the total moisture content of the leaves, it is
found to be 1.691 per cent for old Satsuma leaves, 1.465 per cent in those
of young Satsumas, 1.490 per cent in those of old grapefruit, and 1.080
88 Journal of Agricultural Research volvi.no. 2
per cent in those of young grapefruit. It will be recalled that bacterial
growth occurs on artificial media rendered acid by hydrochloric or citric
acid when a sufficient amount of acid has been used to make the acidity
of the media 2 per cent. The acidity of the leaf tissue is therefore not
sufficient to inhibit the growth of the canker organism and is not regarded
as sufficient to account for the difference in susceptibility. No deter-
minations have been made of the kinds and relative amounts of the
several organic acids in the tissues of the two species. Until this is
known there still remains the possibility of a correlation between sus-
ceptibility to canker and acidity.
CHEMICAL CHANGES IN CITRUS LEAVES BROUGHT ABOUT BY
CITRUS CANKER
Little attention has been given by the biochemist to the chemical
transformations occurring in diseased plant tissues. Such studies would
no doubt throw a flood of light upon the intimate relationship of parasite
and host and would materially contribute to our knowledge of the nature
of parasitism. The literature dealing with the chemical changes induced
by plant pathogens is more or less fragmentary, mainly because of the
inexact state of our knowledge regarding the separation and quantitative
estimation of the various compounds occurring in plant tissues. An
historical resume of this literature has therefore been purposely omitted.
However, among the recent excellent papers along this line may be men-
tioned the work of Hawkins (7) upon the changes in peaches induced by
the brown-rot organism, Scleroiinia cinerea. He found in brown-rotted
tissues an increase in acid content, a decrease in certain alcohol-soluble
substances, a decrease in the total sugar content, and practically a dis-
appearance of the cane sugar. It was with the view of determining
something of the changes produced by Citrus canker that this portion of
the investigation was undertaken.
Diseased and healthy leaves were taken from grapefruit trees affected
with Citrus canker. Circles of diseased tissue and tissue from healthy
leaves were excised with a cork borer. These leaf circles were then
triturated in a mortar until the material was finely divided, their wet
weight determined, 27.25 gm. in each case, and preserved in such volume
of 95 per cent alcohol that the alcohol concentration of the mixture was
85 per cent. This concentration could not be accurately made until it
had been determined that the moisture content of normal leaves was
61.69 per cent and that of diseased leaves 61.57 P^^" cent. The material
was then set aside for two weeks and was shaken occasionally to permit
the gradual extraction of the cold alcohol-soluble portions. The method
followed subsequently was based upon those devised by Koch (9, 10,
II, 12) for use in the quantitative chemical analysis of animal tissues.
Apr. JO, 1916 Citrus Canker 89
This method consists essentially in the separation of the material into
three fractions. Fractions i and 2 consist of the soluble portion ex-
tracted by the action of alcohol, ether, and water, and fraction 3 con-
sists of the insoluble residue. Fractions i and 2 are separated by lipoid
precipitation. The former fraction contains precipitated lipoids, while
the latter contains all nonlipoid materials, soluble in alcohol, ether, and
water. Instead of the modified Wiley extraction apparatus employed
by Koch and his pupils, a rubber analysis extraction apparatus (8) has
been employed. Extractions in this apparatus, like those with the
modified Wiley apparatus, are carried out at the boiling point of the
solvent.
In making the first alcohol extraction the preserved material was
transferred to Schleicher and Schiill extraction thimbles, previously
fitted into the siphon cups of the extraction apparatus. The preserving
liquid was then filtered through these thimbles. Perforated porcelain
plates or filter paper cut to fit were then used as covers over the material
in the thimbles.
Extraction for 12 hours with redistilled 95 per cent alcohol followed.
The alcohol was changed two or three times during this extraction in
order to prevent possible decomposition of extracted materials in the
boiling alcohol. The tissue was pressed to remove the excess alcohol,
and an ether extraction was made. This extraction was continued for
1 2 hours for the purpose of facilitating the subsequent powdering of the
tissues. The material was then removed from the extraction thimbles
to a mortar and was ground to a powder. This powder was placed in a
stoppered flask with a volume of distilled water equaling twice the fresh
weight of the material and was boiled on a steam bath for two hours.
Warm absolute alcohol was added in a sufficient quantity to bring the
alcohol content of the whole up to 90 per cent. The mixture was warmed
on the bath, with repeated shaking, and set aside until the following day.
It was then filtered through the original extraction thimbles and extracted
for 12 hours with 95 per cent alcohol. At the close of this extraction
the residue in the cups (fraction 3) was transferred to previously weighed
porcelain crucibles and dried to constant weight in an oven at 100° C.
By this procedure the alcohol, ether, and water-soluble portions (frac-
tions I and 2) were separated from the insoluble portion (fraction 3).
In further preparing the soluble portions of the material for analyses
they were combined only after the ether-soluble portion had been heated
on a water bath until the odor of ether could no longer be detected. A
little alcohol was added from time to time to take up the materials left
behind by the loss of ether by evaporation. In pouring the solutions
together a precipitate appeared which was rendered soluble by the addi-
tion of sufficient hot water to bring the alcohol concentration down to 70
per cent. The solution was then made up to 2,000 c. c, 200 c. c. of
which were taken for the estimation of solids. The remainder was
90 Journal of Agricultural Research voi. vi»no. 2
evaporated at 75° C. to a sirupy consistency or until all the alcohol had
evaporated and the sirupy mass was emulsified with warm water. This
emulsion was placed in a stoppered volumetric flask, shaken with 20
c. c. of chloroform, 10 c. c. of hydrochloric acid were slowly added, and
then it was made up to a given volume by the addition of water. The
flask was then placed for 48 hours in running water under the hydrant
to facilitate the precipitation of the lipoids in the chloroform. Filtra-
tion followed, the filtrate constituting fraction 2, and the lipoid precipi-
tate on the filter paper fraction i. The precipitate was then taken up
with a large volume of hot, 95 per cent alcohol, and kept on a water bath at
75 C, until all of the chloroform was driven off. The volume was then
increased to a convenient amount, and aliquot parts taken for analyses.
The analysis of fraction 3 included (a) total phosphorus, (6) total nitro-
gen, (c) cellulose, {d) carbohydrate after hydrolysis, (e) ash, (/) total
solids; the analysis of fraction 2 included (a) dry weight and ash made
upon an aliquot part, (&) total sugars before and after hydrolysis, (c)
total nitrogen, {d) phosphorus, (e) solids ; while the analysis of fraction i
included only (a) total solids, (6) phosphorus, (c) nitrogen, since the total
weight of the lipoidal material from the two samples differed by i mgm.
only and since the amounts were too small to admit of accurate separation.
The determinations of phosphorus were made upon aliquot parts by the
Pemberton-Neuman method described by Mathews (14, p. 893-895).
The total nitrogen was determined upon all fractions by the employ-
ment of the Gunning- Arnold modification of the Kjeldahl method. No
determinations were made of fatty acids.
In fractions 2 and 3 the carbohydrate determination included reducing
sugars, total sugars, and cellulose. Prior to the determination of reducing
sugars, the solution was freed from organic acids, tannins, and other sub-
stances capable of affecting reduction by Fehling's solution. This was
accomplished by treatment with lead subacetate in excess, after which
the solution was diluted, filtered, and saturated sodium sulphate was
added to precipitate the excess of lead. The clear filtrate was then di-
luted, and an aliquot part taken for the determination of reducing sugar
by the Bertrand volumetric method. The reducing sugar was calculated
as dextrose by the Munson and Walker tables.^ Another aliquot part
of the solution, a part of which had been used for the determination of
reducing sugars, was used upon which to determine the total sugars.
This was hydrolyzed by the addition of concentrated hydrochloric acid,
following which the solution was kept on a water bath at 69° to 70° C.
for 10 minutes. It was then cooled, neutralized with 40 per cent sodium
hydroxid, and the sugar determined as invert sugar by the volumetric
permanganate method.
' Wiley, H. W., ed. Offidal and provisional methods of analysis, Association of Official Agrricultural
Chemists. As compiled by the comjtnittee on revision of methods. U. S. Dept. Agr. Bur. Chem. Bui.
107 (rev.), p. 241-251. 1908.
Apr. 10, 1916
Citrus Canker
91
Cellulose determinations in fraction 3 were made in duplicate with
accordant results by employing Schweitzer's reagent in one case and a
solution of zinc chlorid in hydrochloric acid in the other.
Polysaccharids in fraction 3 were estimated as dextrose after 2.5 hours'
hydrolysis in a reflux condenser using 2.5 per cent hydrochloric acid.
In Table II are given the fresh weights of normal and cankerous Citrus
tissue, moisture content, dry weight, and alcohol-ether soluble and in-
soluble portions.
Table II. — Analysis of normal and cankerous tissue of grapefruit leaves
Item.
Nonnal tissue.
Cankerous
tissue.
Fresh weight
Moisture
Dry weight
Total alcohol-ether
Soluble
Insoluble
Gm.
27. 250
16. 799
10. 451
4.270
6. 181
Gm.
27.250
16. 796
10. 454
4.300
6. 154
In this composite table it is strikingly significant that only slight
differences are apparent. The moisture content of normal tissue is
slightly greater than that of cankerous tissue, and there is, of course, a
corresponding decrease in dry weight. The greater amount of alcohol-
ether soluble material occurs in cankerous tissues with a lesser amount of
alcohol-ether insoluble substance. The differences represented herein
would have little or no value in themselves if it were not that they were
obtained by the use of a refined method of analysis primarily intended to
permit the discovery of changes not indicated by ordinary methods.
Studies of the intricate relation of parasite and host have proceeded far
enough to indicate that large changes in composition of the host are not
to be expected, but rather that transformations have been produced
which, though minute in amount, profoundly affect the metabolism of
both parasite and host. Table III gives in detail the results of the
several steps in this analytical procedure.
Table III. — Analyses (in grams) of normal and cankerous grapefruit leaves
Item.
Fraction i.
Fraction 2.
Fraction 3.
Totals.
Normal.
Diseased.
Normal.
Diseased.
Normal.
Diseased.
Normal.
Diseased.
Dry weight. .
Nitrogen ....
Phosphorus. .
Reducing
sugars
0. 781
. 0196
.0123
0. 780
.686
. 0114
3-479
. 1 190
. 021
2. 008
•387
3-520
. 1120
- 0225
.806
.284
6. 181
.105
. OIOI
6. 154
.0818
• 0131
10. 441
.2436
•0434
2. 008
. 960
. 2087
•859
. 041
10. 454
. 2624
.0470
806
Sugar after
acid hy-
drolysis . . .
-573
.2087
•859
. 041
-370
•1273
.843
.041
•654
.1273
.843
.041
Polysaccha-
rids (solu-
ble)
Cellulose ....
Ash
1
27470°— 16-
92 Journal of Agricultural Research voi. vi, no. 2
It should be stated with reference to the data presented in Table III
that the figures given are the weights in grams of the several constitu-
ents as determined by employing 27.25 gm., fresh weight, of healthy and
of cankerous tissue. Since the two samples differed by only 3 mgm.
in dry weight and since the figures, to be directly comparable, should
be based on dry weights in each case, a correction of 0.24 per cent should
be applied to the analyses of diseased tissue. As this is insignificant,
the data are regarded as referable, and the corrections have not been
applied.
Because of the presence of certain enzyms, of which mention has been
made earlier in this paper, it is to be expected that the changes of great-
est magnitude would occur in the carbohydrates. That such is the
case is obvious when one notes in the totals given in Table III a reduc-
tion of all classes of carbohydrate in cankerous tissue. Thus, in equal
quantities of fresh material the amounts of reducing sugar are found to
be as 5 to 2, the total sugars as 3 to 2, and the polysaccharids as 5 to 3
when normal and diseased tissues are compared. Because of the ease
with which they are available to the invading organisms, the reducing
sugars are probably the most strongly attacked. After acid hydrolysis
the normal tissue shows more reducing sugar than the diseased, both in
the alcohol-ether soluble and alcohol-ether insoluble fractions. This
means that there is also a less amount of the higher soluble carbohydrates,
disaccharids, in diseased tissues and that they too are more easily avail-
able than the polysaccharids. The ratio of disaccharids in normal and
cankerous tissue in the alcohol-ether soluble and alcohol-ether insoluble
portions is as 3 to 2 and 5 to 3, respectively.
There is also a slight but significant decrease in the amount of cellu-
lose found in diseased tissues. Although the difference in total cellulose
in the normal and diseased tissues is slight, the results given are repre-
sentative of a considerable number of determinations in which two
standard methods were employed and in which the lesser amount of
cellulose was invariably found in the diseased tissue. Experimental
error has thus been eliminated and the results indicate a slight but un-
mistakable destruction of cellulose by the invading organisms.
The polysaccharids were determined in fraction 3 after 2.5 hours acid
hydrolysis. They were found to be present in normal tissues and dis-
eased tissues in the same proportion, 5 to 3, as were the disaccharids.
There has therefore been a corresponding reduction and utilization of
both di- and poly-saccharids by the invading organisms.
In the alcohol-ether insoluble fraction the amounts of nitrogen found
for normal and diseased tissue were 0.105 and 0.0818 gm., respectively.
If the conventional factor for these figures, 6.25, is employed, 0.654 and
0.511 gm. are obtained as the protein content of normal and diseased
tissues, respectively. The protein content of diseased tissue has there-
fore been reduced 78.16 per cent. One should therefore expect to find a
Apr. lo, i9i6 Citrus Canker 93
very material increase in the nitrogen of the alcohol-ether soluble portion
of the diseased tissue. This expectation is realized, since the nitrogen
figures for the soluble portions are 0.1386 gm. for normal and 0.7806 gm.
for diseased tissue. This represents an increase in the diseased tissue
of 37.52 per cent over the healthy tissue. This increase in the soluble
portion indicates a decomposition of the complex nitrogenous compounds
resulting in the formation of peptones and amino acids soluble in alco-
hol and ether. This difference in nitrogen content of the alcohol-ether
soluble portion takes an added significance when the nitrogen content of
fraction i and that of fraction 2 are examined separately. It will be
recalled that the nitrogen of fraction 2 represents those portions of the
nitrogenous constituents extracted by alcohol and ether which are readily
soluble in water after the combined extract has been evaporated to a
paste. They are, therefore, amino acids and polypeptids. It will fur-
ther be recalled that fraction i is obtained from the watery solution of
the alcohol-ether soluble extract by chloroform precipitation and is there-
fore lipoid nitrogen. The slight decrease in nitrogen in fraction 2, when
normal and diseased tissue are compared, is accompanied by an enor-
mous increase, amounting to 250 per cent in the lipoid nitrogen of
fraction i.
These differences in nitrogen content of the several fractions lend
themselves to two possible explanations. The first and most obvious
interpretation of the results is that the changes produced by the invading
organisms in the proteins of the host result in the formation not of
amino acids and other end products of protein decomposition but in the
production of complex intermediate substances. The other explanation
is based upon the fact that the bacteria themselves derive the nitrogen
necessary for the building of their own proteins as well as for the forma-
tion of their cell walls from the proteins of the host. Concurrently
with the reduction of the protein of the host to simpler forms a series
of metabolic processes is occurring within the invading organism which
involves the synthesis of these simple nitrogenous compounds to more
complex ones. The changes in nitrogen content of the several fractions
of the diseased tissue are therefore the result of both analytic and
synthetic processes. At present it is impossible to employ any methods,
as none have been devised, which will indicate what the end products
of decomposition of the host proteins by the invading organism are,
since the formation of these products is accompanied by their con-
comitant utilization in the manufacture of new compounds peculiar to
the body of the parasite.
The total phosphorus in the diseased tissues is greater in amount in
fractions 2 and 3 than in the normal tissues. Were the changes in the
diseased tissue purely katabolic, it would be expected that there would
be a material increase in water-soluble phosphorus derived from the
94 Journal of Agricultural Research voi. vi, no. a
decomposition of neucleoproteins. On the contrary, the phosphorus of
fraction 3 shows an increase of 30 per cent, that of fraction 2 an increase
of 20 per cent, and that of fraction i a decrease of about 7 per cent.
The increase in water-soluble phosphorus in fraction 2 indicates that
decomposition processes are taking place, but the concomitant increase
in phosphorus content in fraction 3 shows that such decomposition is
accompanied by actual synthetic processes involving the use of phos-
phorus.
No difference appears between the two tissues in amounts of ash as
shown in fraction 3. The ashing of fractions i and 2 gave unsatisfactory
results and for this reason the figures are withheld.
It is evident from the foregoing statement of results that the significant
changes brought about in diseased tissues concern carbohydrate and
nitrogenous constituents. The concurrent disappearance of mono-, di-,
and poly-saccharids from diseased tissues indicates that all the sucroclas-
tic enzyms previously shown to be formed by the organisms in pure
cultures are active in the host tissues and that the reducing sugars
formed are utilized by the organisms as sources of energy. The results
with nitrogen indicate that there is not an accumulation of the products
of protein decomposition but that the destructive transformation of
protein is accompanied pari passu by a utilization of the decomposition
products in the anabolic processes of the organisms.
AGENCIES CONCERNED IN DISSEMINATION OF CITRUS CANKER
Definite experimental data are wanting on the agencies by which
Citrus canker is spread. If we judge, however, from field observations
and from a knowledge of other bacterial plant diseases, it is evident that
rain and dew are important factors in carrying the disease to unaffected
leaves, twigs, and fruits of trees in which the diesase is already present.
Man himself is a very important agent in effecting the distribution of
canker from diseased trees to healthy trees near by. When in the cul-
tural operations of budding, cultivation, picking, etc., he comes in con-
tact with diseased trees and soon afterwards touches healthy ones, infec-
tion may result. The chances of infection are greatly increased if he
comes in contact with newly formed cankers on the diseased trees, and if
a film of moisture is present on the adjacent healthy trees which he may
touch. The most plausible explanation of the introduction of Citrus
canker into two groves which have come under the writer's observation
is through the agency of man. The owners had visited groves in which
canker occurred in order to acquaint themselves with the appearance of
the disease. On returning home they examined certain of the trees in
their own groves and these trees soon afterward developed canker lesions.
Stirling (2) reports the transmission of the disease through handhng dis-
eased leaves prior to touching healthy ones. It is highly probable that
Apr. lo. 1916 Citrus Canker 95
certain birds and insects also effect this contact of diseased with healthy
parts and are therefore to be regarded as agents in dissemination of
Citrus canker.
CONTROL OF THE DISEASE
During the summer of 191 4 those who had been attempting to solve
the problem of controlling Citrus canker realized that it was an exceed-
ingly difficult undertaking. Efforts were directed along three lines:
Exclusion, protection, and eradication.
Exclusion. — Those interested in the welfare of the Citrus industry in
Florida were the first to realize the serious nature of Citrus canker and
that it had been introduced into the State from other States and from
foreign countries. For these reasons a quarantine was imposed during
the spring of 191 4 to prevent the further introduction into Florida of
Citrus trees and buds and thus of Citrus canker. Other of the Gulf
States later in the season realized the jeopardy in which their Citrus
growers' interests were placed and issued similar regulatory measures on
the importation of shipments of Citrus stock. On January i, 191 5, a
Federal quarantine was imposed to exclude the further importation of
this disease into the United States. The agitation throughout the entire
Citrus growing section of the Gulf coast attendant on the adoption of
these regulations looking toward control by exclusion have so familiar-
ized the growers with Citrus canker that it is unnecessary to advise the
exercise of care in ordering trees to be used in setting out a Citrus grove.
It is reaUzed that in no case is it safe to purchase trees from nurseries in
which this disease occurs.
Protection. — Since certain fungicides have been successfully used in
the control of various Citrus diseases a number of experiments were
undertaken during the spring of 1914 to determine the effectiveness of
these mixtures in the control of Citrus canker. A grove of badly dis-
eased grapefruit was used upon which to make applications of Bordeaux
mixture, ammoniacal copper carbonate, and soluble sulphur. Details of
these experiments are withheld, since it was reaUzed early in the summer
that the appHcation of these fungicides was without appreciable effect in
the control of canker.
Again in the spring of 191 5 another grove of grapefruit was selected
in which to test the effectiveness of several fungicides in protecting the
trees from infection by the canker organism. All visible signs of canker
were carefully removed from the trees prior to the application of the
mixtures. Bordeaux mixture, Bordeaux mixture and bichlorid of
mercury (12 tablets in 3 gallons), Bordeaux mixture and formaldehyde
(i : 100), and a Bordeaux and lead arsenate mixture were employed.
Applications were made on March 26, April 29, and May 14, and no new
infections had developed on any of the sprayed or unsprayed trees by
the last-named date. On May 27, however, new infections were apparent
96 Journal of Agricultural Research voi. vi, No. 2
and were equally numerous on sprayed and check trees. A number of
growers have used various germicidal mixtures in attempts to find a
preparation which could be successfully employed against Citrus canker.
In no case have these efforts met with a sufficient degree of success so
that their use in canker control can be recommended. When formal-
dehyde is used in sufficient strength to cause the death of the leaf tis-
sues in a considerable area surrounding the cankers no viable organisms
can be found in the cankerous tissues in many cases. They are still
viable, however, in others, and it has also been found to be impossible
to cause formaldehyde to penetrate sufficiently deep into old suberized
limb cankers to kill the canker organisms. In the light of these tests
and in the light of the ineffectiveness of sprays in the control of other
plant diseases of bacterial origin, it is believed that there is little to be
hoped for in the use of germicides for protection against Citrus canker.
Eradication. — The history of the work of eradication of Citrus
canker, Jittle of which has been published outside of the daily press,
would in itself be voluminous, and it is not the present purpose to include
it in this account. The early efforts toward the eradication of Citrus
canker were confined to the removal of diseased parts in case the trees
were only slightly diseased. When the trees were seriously affected,
however, they were severely pruned, even though this necessitated the
removal of nearly all of the branches. Pruned trees were then thoroughly
sprayed with Bordeaux mixture. It was recommended that all the
diseased parts which had been removed should be burned.
After a few months' trial it was seen that by this procedure the treated
trees were still diseased. Further than this, adjacent trees had become
diseased, although they were apparently healthy at the time efforts
had been made to remove cankered leaves and branches from the trees
near by.
Even when the work of removal had been done by skilled hands and
when the trees had received several applications of some fungicide to
protect the new growth they were still found to become cankered.
As a result of this it was decided during the summer of 1914 that only
the complete destruction of the diseased trees by burning would be
effective. As a result of this decision the eradication campaign was
organized and a concerted, heroic effort is being put forth to stamp out
Citrus canker from the Gulf States. The intelligent observance of the
strictest sanitary precautions with reference to trees adjacent to those
which are destroyed is necessary.
SUMMARY
A serious disease, commonly known as Citrus canker, which affects
species of Citrus and Fortunella, has within the past few years been in-
troduced into Alabama and other of the Gulf States. It attacks fruits,
leaves, twigs, and larger branches, producing characteristic cankerous
Apr. lo. 1916 Citrus Canker 97
lesions. The primary cause of the disease is Pseudomonas citri, first
isolated by Hasse from grapefruit and found to be pathogenic on
grapefruit seedlings. This has been confirmed, and in addition an
organism presenting the same cultural and physiological characters has
been isolated from trifoliate and Satsuma oranges and lemons. No
difficulty has been experienced in cross-inoculating the organism on
McCarty and seedUng grapefruit, Pineapple oranges, Satsuma oranges,
and seedling trifoliate oranges. It grows readily on a variety of artificial
media, and according to the studies made its group number is 221.3332513.
Infection occurs through natural openings and through wounds. The
rapid spread of the disease is favored by the simultaneous occurrence of
newly exposed cankerous cells and the presence of a film of moisture,
especially on young parts of the plant. The bacteria occur for the most
part between the cells of the host and cause them to become considerably
hypertrophied. Little, if any, hyperplasia is believed to occur. This
enlargement of the cells is caused by the dissolution of the middle
lamellae through enzym activity and by a modification of the host
protoplast so that its osmotic pressure is increased. This increased
pressure results from the presence of the parasite between the cells and
from the passage of materials through the walls of the host, occasioned
by the growth of the organism.
Besides Pseudomonas citri, fungi belonging to the genera Phoma,
Fusarium, and Gloeosporium have been isolated from Citrus cankers. Of
the fungi Phoma sp. alone was found to be notably active in the disin-
tegration of the tissues. It is able by virtue of the secretion of specific
enzyms to utiUze the carbohydrates, cellulose, starch, maltose, and
saccharose and causes also a decrease in acidity of invaded tissues. It is
regarded as heretofore undescribed and is herein given the name " Phomxi
soda, n. sp."
The difference in susceptibility to Citrus canker of Satsuma oranges
and grapefruit can not be accounted for on the basis of differences in
total organic acids in the two hosts.
Comparative analyses of grapefruit leaves aJEFected with Citrus canker
and of healthy leaves shows that there has been in diseased leaves a de-
crease in all of the soluble and insoluble carbohydrates due to their
utilization by means of sucroclastic enzyms secreted by the canker
organisms. Apparently a decomposition of the host proteins occurs
concurrently with their synthesis in the metabolism of the parasite pro-
teins, and there results a slight increase in total nitrogen in diseased
tissues. The slight increase in phosphorus in diseased tissues is ac-
counted for in the same manner as that in nitrogen, since they appear to
be correlated. No differences in ash were found in fraction 3, and the dry
weight of diseased tissues was slightly greater than that of normal.
Rain and dew are important agencies in the dissemination of Citrus
canker. Any other agencies, of which man is probably the most im-
98 Journal of Agricultural Research voi. vi, no. 2
portant, which effect a contact of diseased parts with healthy parts, are
to be recognized as carriers.
In efforts to control the disease quarantine measures have been passed,
thus preventing its further introduction from foreign localities and from
any one of the Gulf States to any other of them. The use of spray mix-
tures indicates that they are not to be regarded as remedial measures of
appreciable value in canker control; nor will their use protect healthy
trees adjacent to diseased ones from infection.
Successful eradication seems possible, but only when the work of de-
struction of diseased trees is thoroughly done, with the observance of
proper sanitary precautions.
LITERATURE CITED
(i) Berger, E. W.
1914. Citrus canker — History. In Fla. Grower, v. 11, no. 7, p. 14-15.
(2) Stevens, H. E., and Stirling, Frank.
1914. Citrus canker — II. Fla. Agr. Exp. Sta. Bui. 124, p. 27-53, fig- 7~i4-
(3) Comes, Orazio.
1913. Delia resistenza dei frumenti alle ruggini. Stato attuale della quistione
e prowedimenti. In Atti R. 1st. Incorag. Napoli, s. 6, v. 64, 1912,
p. 421-441. Letteratura e note, p. 437-440. Abstract in Intemat.
Inst. Agr. [Rome], Mo. Bui. Agr. Intel, and Plant Diseases, year 4
no. 7, p. 117-119.
(4) Crabiix, C. H., and ReEd, H. S.
1915. Convenient methods for demonstrating the biochemical activity of
microorganisms, with special reference to the production and activity
of enzymes. In Biochem. Bui., v. 4, no. 13, p. 30-44, pi. i.
(5) Edgerton, C. W.
1914. Citrus canker. La. Agr. Exp. Sta. Bui. 150, 10 p.
(6) Hasse, Clara H.
1915. Pseudomonas citri, the cause of citrus canker. A preliminary report.
In Jour. Agr. Research, v. 4, no. i, p. 97-100, pi. 9-10.
(7) Hawkins, L. A.
1915. Some effects of the brown-rot fimgus upon the composition of the peach.
In Amer. Jour. Bot., v. 2, no. 2, p. 71-81. Literature cited, p. 80-81.
(8) Joint Rubber iNSLrtATioN Committee.
1914. Tentative specifications and analytical procedure for 30^ Hevea
rubber insulating compoimd. Preliminary report of the joint rubber
insulation committee appointed by a group of manufacturers and
users of rubber compounds, 19x1-1914. /« Jour. Indus, and Engin.
Chem., V. 6, no. i, p. 75-82, illus.
(9) Koch, W.
1909. Methods for the quantitative chemical analysis of animal tissues. I.
General principles. In Jour. Amer. Chem. Soc, v. 31, no. 12, p.
1329-1335.
(10) and Mann, S. A.
1909. Methods for the quantitative chemical analysis of animal tissues. II.
Collection and preservation of material. In Jour. Amer. Chem. Soc,
v. 31, no. 12, p. 1335-1341.
Apr. lo, 1916 Citrus Canker 99
(11) Koch, W., and Carr, E. P.
1909. Methods for the quantitative chemical analysis of animal tissues. III.
Estimation of the proximate constituents. In Jour. Amer. Chem.
Soc, V. 31, no. 12, p. 1341-1355, I fig.
(12) and Upson, F. W.
1909. Methods for the quantitative chemical analysis of animal tissues. IV.
Estimation of the elements, with special reference to sulphur. In
Jour. Amer. Chem. Soc, v. 31, no. 12, p. 1355-1364, i fig.
(13) McBeth, I. G., and Scai^es. F. M.
1913. The destruction of cellulose by bacteria and filamentous fungi. U. S.
Dept. Agr. Bur. Plant Indus. Bui. 266, 52 p., 4 pi.
(14) Mathews, A. P.
1915. Physiological Chemistry * * * 1040 p., 78 fig. New York.
(15) Scales, F. M.
191 5. Some filamentous fungi tested for cellulose destroying power. In Bot.
Gaz., v. 60, no. 2, p. 149-153.
(16) Stevens, H. E.
1914. Citrus canker. A preliminary bulletin. Fla. Agr. Exp. Sta. Bui. 122,
p. 113-118, fig. 43-46.
(17) Swingle, W. T.
191 5. A new genus, Fortunella, comprising four species of kumquat oranges.
In Jour. Wash. Acad. Sci., v. 5, no. 5, p. 165-176, 5 fig.
(18) U. S. Department of Agriculture.
1915. Citrus canker in Philippines. In U. S. Dept. Agr, Dept. Circ, v. i,
no. I, p. 8.
(19) Wolf, F. A., and Massey, A. B.
1914. Citrus canker. Ala. Agr. Exp. Sta. Circ. 27, p. 97-102, illus.
PLATE VIII
Fig. 1. — Grapefruit leaf showing young Citrus cankers.
Fig. 2. — Old Citrus canker on Satsuma leaves.
Fig. 3, 4. — Seedling grapefruit branches affected with Citrus canker.
Fig. 5. — Severe canker infection of branches of Citrus trifoliata.
(100)
Citrus Canker
Plate VIII
.>£
r
^v
n
?^^'
T 1
Journal of Agricultural Research
Vol. VI, No.2
Citrus Canker
Plate IX
Journal ot Agricultural Research
Vol. VI, No. 2
PLATE IX
Fig. I. — View of lower side of leaves of seedling grapefruit artificially inoculated
with Pseudomonas citri.
Fig. 2. — Top view of plant shown in figiu-e i.
Fig. 3. — Spongy white cankers on leaf and twig of seedling grapefruit produced by
artificial inoculation. The plants were continuously kept under a bell jar in a humid
atmosphere.
Fig. 4. — Citrus canker on Satsuma leaves resulting from artificial inoculation with
Pseudomonas citri.
Fig. 5. — Photomicrograph of section of yoimg, open canker on grapefruit.
PLATE X
Fig. I. — Natural Citrus canker infection on leaves of Citrus trifoliaia.
Fig. 2. — Mature cankers on fruit of Citrus decumana (courtesy of Dr. E. W. Berger).
Fig. 3. — Canker on seedling grapefruit leaves, entrance having been effected through
abrasions made by thorns.
Fig. 4. — Yotmg spongy cankers on fruit of Citrus decumana.
Fig. 5. — Phoma soda on cellulose agar showing dissolution of cellulose.
Fig. 6. — Mature cankerous areas on leaves of Duncan grapefruit.
Citrus Canker
Plate X
J^
Journal of Agricultural Research
Vol. VI, No. 2
Citrus Canker
Plate XI
Journal of Agricultural Research
Vol. VI, No. 2
PLATE XI
Fig. I. — Cankers on old grapefruit leaves which have enlarged during the second
growing season.
Fig. 2. — Citrus canker resulting from immersion of leaves in a bacterial suspension.
Lesions involving a large part of the lower leaf surface are thus formed.
Fig. 3. — Culture of Phoma socia showing pycnidial formation in concentric rings.
Fig. 4. — Dilution poured plate of Pseudomonas citri on green-bean agar. The spots
on the colonies are the reflection of the windows of the room in which the exposure
was made. Colonies 14 days old, the last 5 of which days the plates were kept in an
ice chest at a temperature of about 55°.
Fig. 5. — Photomicrograph of pycnidium of Phoma socia taken in reflected simlight.
Fig. 6. — Photomicrograph of pycnidia of Phoma socia taken in diffuse light.
o
ADDITIONAL COPIES
OF TmS PUBLICATION MAT BE PROCUEED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
15 CENTS PER COPY
Vol. VI AF»RIL 17. 1916 No. 3
JOURNAL OP
AGRICULTURAL
RESEARCH
CONTENTS
Page
Determination of Stearic Acid in Butter Fat - - - 101
E. B. HOLLAND, J. C. REED, and J. P. BUCKLEY, Jr.
Life History and Habits of Two New Nematodes Parasitic
on Insects - - ---- . - - 115
J. H. MERRILL and A. L. FORD
Insect Injury to Cotton Seedlings ----- 129
B. R. COAD and R. W. HOWE
DEPAR'EMENX OF AGRICULXUEE
WASHINGTON , D.e
WASHINGTON : GOVERNMENT PHtNTINO OFFICE : 1916
PUBLISHED BY AUTHORITY OF THE SECRETARY
OE AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERM AN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experimetit Stalions
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Experiment
Statimt
H. P. ARMSBY
Director, Institute of Animal IVntrition, "Die
Pentisylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
All correspondence regarding articles from the Department of Agriculture
should be addressed to Karl F. Kellermau, Journal of Agricultural Research,
AVashington, D. C.
AH correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine.
JOIMAL OF AGEIOJLTIAL RESEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, April 17, 1916 No. 3
DETERMINATION OF STEARIC ACID IN BUTTER FAT*
By E. B. Holland, Associate Chemist, and J. C. Reed and J. P. Buckley, Jr.,
Assista7tt Chemists, Massachusetts Agricultural Experiment Station
INTRODUCTION
Oils and fats are composed largely of neutral glyceryl esters together
with small amounts of free fatty acids and unsaponifiable matter.
Formerly the esters were considered simple glycerids, compounds of
glycerol and three radicals of the same fatty acids. At present the oppo-
site view seems to prevail and mixed glycerids are said to predominate
in most products. The subject is controversial and difficult of solution.
The constituents would be the same, however, in either case, whether
combined as simple or complex molecules. The object of a technical
examination of oils and fats is to isolate, identify, and determine the
various fatty acids, glycerol, and unsaponifiable bodies, although, as
Lewkowitsch asserts, this is not attainable in the present state of our
knowledge. Certain progress has been made in determining different
constituents of fats by indirect methods, such as iodin absorption, acetyl
number, and molecular- weight calculations. Direct methods of fractional
distillation, crystallization, and solubility of various salts have not, as a
rule, proved sufficiently discriminative for quantitative use.
Fatty acids constitute about 95 per cent of most oils and fats and
characterize the products to a large extent. The necessity of accurate
methods for the quantitative determination of these acids has long
been recognized not only from the standpoint of pure science but espe-
cially in physiological studies having as the object the measurement of the
effect of different food groups on the production of body and milk fats.
Many methods have been proposed since the publication of the work of
Chevreul nearly 100 years ago, but few, if any, have met with general
approval. After several years' investigations of the Partheil and Ferie
method (7),^ which proved unsatisfactory in the authors' ^ hands, a study
of methods for determining stearic acid in butter fat was undertaken.
' From the Department of Chemistry, Massachtisetts Agrictdttiral Experiment Station. Printed with the
permission of the Director of the Station.
* Reference is made by number to " Literature cited," p. 113.
' Mr. Reed was associated with the senior author in the earlier stages of the work and Mr. Buckley in the
later.
Journal of Agricultiu-al Research, Vol. VT, No. 3
Dept. of Agriculture, Washington, D. C. Apr. 10, 1916
db Mass. — i
(lOl)
I02 Journal of Agricultural Research voi. vi.no. 3
EARLIER INVESTIGATIONS
For the separation of stearic from other fatty acids, David (i) recom-
mended a special alcohol and dilute acetic-acid solution saturated with
stearic acid at 15° C, in which solution oleic acid was shown to be
soluble.
The Hehner and Mitchell (3) method for isolating stearic from other
fatty acids was based on the hypothesis that a mixture of fatty acids
heated with a solvent saturated at a given temperature with the acid
under determination might be expected on cooling to that temperature
to crystallize the whole of the acid sought, provided the other constitu-
ents did not increase the solubility. The solvent employed was methy-
lated alcohol (94.4 per cent) saturated with stearic acid at 0.2° C, pre-
pared by chilling a solution of 3 gm. to i liter overnight in ice water and
siphoning off the saturated mother liquor through a small thistle tube
covered with fine calico, using suction. The tests were conducted in a
similar manner, taking from 0.5 to 5 gm. of insoluble acids (according to
content) to 100 c. c. of alcohol-stearic-acid solution. Shaking was found
to increase precipitation. Supersaturation and esterification were recog-
nized as possible sources of error. The method gave concordant results
with solid fats containing considerable stearic acid, but slight, if any, pre-
cipitate from the acids of butter fat and from mixtures of the acids of
Japan wax and pure stearic acid.
Emerson (2) noted considerable variation in the content of different
saturated solutions and found that supersaturation seemed to occur when
less than 0.7 gm. to 100 c. c. was employed in preparing the solution.
The formation of ethyl ester appeared to be a source of error and to have
increased the apparent solubility of the stearic acid.
Kreis and Hafner (5) showed that small amounts of stearic acid below
0.1 gm. to 100 c. c. of a saturated solution formed supersaturated solutions,
and that less than 0.05 gm. gave low and extremely variable results, even
upon the addition of crystals of stearic acid.
Lewkowitsch (6, p. 556-559) claimed that the method yielded capri-
cious results with mixtures of stearic, palmitic, and oleic acids, and that
in many cases the results were entirely unreliable when other acids were
present. He stated that a considerable proportion of lauric acid would
prevent the complete precipitation of stearic acid, even when super-
saturated alcohol-stearic-acid solutions were used, and that acids of
higher melting point, when present, such as arachic, behenic, etc., would
appear in the separated acids. He reported a precipitate of 0.49 per
cent from butter fat, of which a portion might be arachic and myristic
acids.
The results obtained by various investigators indicate that the solu-
bility of stearic acid increases with the strength of the alcohol, but
the figures reported are too variable to warrant further deductions
(Table I).
Apr. 17, 1916
Stearic Acid in Butter Fat
103
Table I. — Solubility of stearic acid, according to various investigators
Investigator.
Hehner and Mitchell (3, p. 323)
Emerson (2, p. 1754)
Do
Do
Kreis and Hafner (5)
Lewkowitsch (6, p. 164)
Do
Ruttan (8, p. 440)
Approximate
strength of
alcohol.
Per cent.
94.4
95-5
95-
94.
95
94.
94.
100
Stearic add to
100 c. c.
Gm.
O. 2 to O. 5
•7
■7
•7
• 5
■3
■7
Sattiration of 100 c. c.
at 0° C.
Gm.
1400 to o. 1580
• 1223
• I 139
1035
13 10
0814
1082
373
1220 to
0810 to
PRELIMINARY WORK
In view of what has been stated, the outlook for another investigation
was not promising, although Lewkowitsch's final arraignment of the
process was not published until nearly a year after the work was under-
taken. The subject was of sufficient importance, however, to warrant
additional study whatever the outcome.
Apparatus. — To insure a uniform temperature for crystallization, a
tank was constructed of ^-inch lumber (20 inches long, 10 inches
wide, and 20 inches deep), lined with galvanized iron, provided with
a tight cover, and raised by legs to a convenient working height.
For icing, a basket (i3>2 by 6 by 18 inches) of galvanized screening
of "i^-inch mesh, holding probably 30 pounds of broken ice, was found
very satisfactory. The insulation of wood, together with the large
volume of water and ice, proved inadequate to meet the requirements
of the case, and it was necessary to install in one comer of the tank a
pump run by a motor, to keep the water in continuous circulation. With
this apparatus a constant temperature of about 0.1° C. was easily main-
tained (fig. 1,2).
Several factors had to be considered in the selection of containers
in which the tests were to be conducted. They must be of a form,
size, and weight suitable for weighing the charge on analytical bal-
ances, easily held in position in the tank, and such that the alcoholic
solution could be removed while still in the tank, leaving the crystalline
residue. After numerous experiments with globe-shaped separatory
funnels and filtering tubes, 8-ounce sterilizer bottles were adopted
and have been found fairly satisfactory. The bottles are of narrow
cylindrical form (2 by 6^ inches) and are held in place in the
tank by pockets of wire screening, with only the rubber stopper and a
small portion of the neck projecting out of the water. The solution
is siphoned off by means of a small thistle tube (>^-inch bulb) having
a felt of absorbent cotton weighing 0.020 gm. supported by a glass
bead and covered with a piece of batiste.
I04
Journal of Agricultural Research
Vol. VI, No. 3
m::^^.
Reagents. — For the preparation of an alcohol-stearic-acid solution
constituents of high quahty were deemed essential for satisfactory
work. The purification of alcohol had been a subject for study for
a number of years in connection with the ordinary analysis of oils and
fats, and excellent results were finally secured by treatment with silver
nitrate and caustic lime and redistillation. A strength of 95.25 per
cent proved a satisfactory solvent for fatty acids, and greater strength
was not considered
^^^ necessary or even ad-
visable.
One lot of stearic
acid, a mixture of sev-
eral grades, was purified
by fractional distilla-
tion of the ethyl ester
in vacuo and subse-
quent repeated crystal-
lization of the separated
acids from alcohol as
previously described
(4). Another lot of
acid with a molecular
weight of 271.13 was
purified by 10 or more
crystallizations from
alcohol to a molecular
weight of 284.25, and
a second portion to
284.71, although the
resulting leaflets were
less perfect than those
obtained by the former
process.
When using separa-
tory funnels and filter-
ing tubes, alcohol-stearic-acid solutions, saturated at o. i ° C, applied to the
insoluble acids of butter at the rate of 150 c. c. to 0.5 gm. of material,
seldom yielded an appreciable amount of precipitate on standing, even with
the addition of crystals of stearic acid and thorough agitation. Solutions
testing about 0.22 and 0.24 gm. of stearic acid to 150 c. c. gave somewhat
higher results, although of erratic and untrustworthy character. In the
attempt to develop a method with this apparatus, over 140 determina-
tions were made on butter acids, stearic acid, mixtures of butter and
stearic acids, stearic and oleic acids, and stearic, myristic, and oleic acids.
The object was not attained, and most of the data will be omitted, as
Fig. I. — Exterior of constant-temperature crystallization tank.
Apr. 17, 1916
Stearic Acid in Butter Fat
105
they would serve no useful purpose, merely indicating the time and
labor involved. The results, however, with solutions of stearic acid
appear to warrant certain deductions.
Solutions containing from 0.25 to 0.29 gm. of stearic acid to 150
c. c. crystallized, leaving a mother liquor of unlike composition (satura-
tion).
The saturation varied inversely with the quantity of stearic acid
present.
Presumably, therefore, supersaturation occurred as a result of insuffi-
cient stearic acid (Table II).
The time of standing may have had some influence, but when in
excess of 24 hours it was of minor consequence.
Fig. 2. — Interior of constant-temperature crystallization tank.
The form of the container as viewed in the light of subsequent work
was a factor of some importance ; a globe-shaped vessel was less effective
than a narrow, cylindrical one of large surface.
Table II. — Crystallization of stearic acid front solutions of different content, using
separatory funnels
Alcohol-
Alcohol-
stearic-acid
Additional
Saturation
stearic-acid
Additional
Saturation
solution
stearic acid
Precipitate.
(grams in
solution
stearic acid
Precipitate.
(grams m
(grams in
taken.
100 c. c).
(grams in
taken.
100 c. c).
isoc. c).
150 c. c).
Gm.
Gm.
Gm.
Gm.
0. 2406
0. 0100
0. 0130
0. 1584
0. 2400
0. 0304
0. 0640
0. 1376
. 2406
• 0150
0254
•1535
. 2400
• 0354
•0733
• 1347
. 2406
■ 0150
0315
■ 1494
. 2400
■0475
.0872
■ 1335
. 2406
. 0400
0859
. 1298
. 2400
. 0481
. 0910
• 1314
. 2406
.0450
099 s
. 1241
. 2400
.0491
. 0910
• 1321
. 2400
. 0200
0426
■ 1449
. 2400
. 0498
. 0960
. 1292
. 2400
.0251
0544
• 1405
io6
Journal of Agricultural Research
Vol. VI, No. 3
Stearic-acid solutions were found to crystallize more readily and with
greater uniformity in sterilizer bottles than in separatory funnels, prob-
ably owing to the more rapid chilling of the narrow column of liquid and
more thorough filtration.
Table III shows the amount of stearic acid crystallized from solutions
of different content and the saturation of the mother liquor.
Table 111.— Crystallization of stearic acid from solutions of different content, using
sterilizer bottles
Alcohol.
Stearic acid
taken.
Precipitate.
Saturation
(grams in
100 c. c).
Alcohol.
Stearic acid
takep.
Precipitate.
Saturation
(grams in
100 c. c).
C.c.
Gm.
G-m.
C.c.
Gm.
Gm.
150
0. 2000
0. 0000
150
0. 3670
0. 1880
0. 1 193
150
. 2400
. 0020
0. 1587
150
.3800
. 2000
1200
150
.2705
.0485
. 1480
150
. 4000
. 2210
1 193
150
.2815
. 0700
. 1410
150
. 4080
. 2260
1213
ISO
•3055
. mo
• 1297
150
. 4200
• 243s
II77
150
•3215
. 1280
. 1290
150
.4650
. 2980
1 1 13
150
•3475
. 1680
• "97
150
. 5000
•32SS
1 163
150
.3600
.1815
. 1190
ISO
. 6000
•43 IS
II23
Table IV. — Crystallization of stearic acid from solutions of different content, using
sterilizer bottles
Alcohol-stearic-acid solution (0.3990 gm.
in 150 c. c).
100
IIO
120
130
140
150
Alcohol.
C.c.
SO
40
30
20
10
Equivalent in
stearic acid
(grams in
150 c. c).
O. 2660
. 2926
.3192
•34S8
•3724
•3990
Precipitate.
Gm.
o- 0555
. 0980
• iSoo
•174s
•20SS
•233s
Saturation
(grams in
100 c. c).
o. 1403
.1297
. II28
. 1 142
.1113
.1103
APPLICATION OF CRYSTALLIZATION METHOD
The facility with which alcohol-stearic-acid solutions crystallize in-
creased with the concentration. Solutions of 0.40 to 0.45 gm. to 150 c. c.
formed crystals readily, gave a satisfactory amount of precipitate, and
when applied to the insoluble acids of butter yielded an additional amount
from that source. This would indicate that if the stearic-acid content
of the solution is sufficient, crystallization of stearic from butter
acids is no more difficult than from other products. The results were
very concordant for a crystallization method when all details of manip-
ulation were strictly observed: The water maintained at the required
level, properly iced at all times, and the pump run continuously at good
speed. A gentle agitation of the solution after standing overnight in
the ice tank assisted in completing the precipitation, but anything in
Apr. 17, 1916 Stearic Acid in Butter Fat 107
the nature of shaking reduced the fragile crystals to a mass and ren-
dered filtration extremely difficult or impossible.
EXPERIMENTAL METHOD IN DETAIL
Five-tenths of a gram of melted insoluble acids are placed in an 8-
ounce sterilizer bottle and 150 c. c. of an alcohol-stearic-acid solution
(3 gm. to 1,000 c. c), accurately measured with a pipette at 30° C,
added. The bottle is sealed with a solid-rubber stopper, shaken at a
gradually increasing temperature until a clear solution is obtained,
placed immediately in a pocket of the ice tank, and allowed to stand
overnight. The following morning the solution is gently agitated by
inverting the bottle several times, and in the afternoon it is siphoned off
as thoroughly as possible by means of a small thistle tube and a per-
forated rubber stopper, using suction. The residue is dissolved in
ethyl ether, transferred to a tared 140 c. c. wide-mouth Erlenmeyer
flask, the ether carefully distilled off, the residue dried at 100° C, and
weighed. As saturation may vary somewhat mth the amount of
stearic acid present and as the quantity of solution retained by the
precipitate depends in a measure on the amount of precipitate, blanks
are run on a weight of stearic acid equivalent to that expected in the
test. By deducting the additional stearic acid taken from the weight
recovered the true blank for the alcohol-stearic-acid solution is obtained.
NATURE OF THE PRECIPITATE
To ascertain whether the crystalline substance obtained from butter
acids was stearic acid or a mixture, the residues from a number of tests
(one being insufficient for accurate work) were combined and the molec-
ular weight determined by saponification. Such a determination made
after securing satisfactory control of the stearic- acid method gave 284.64,
theoretically 284.288. The melting point was not determined, as it
was considered less reliable than the molecular weight.
INFLUENCE OF DIFFERENT FATTY ACIDS ON PRECIPITATION OF
STEARIC ACID
Numerous tests were made in an effort to determine whether lauric,
myristic, palmitic, and oleic acids had any effect on the crystallization of
stearic acid and, if so, the nature and extent of such action. Table V
will serve to illustrate.
According to molecular-weight determinations the lauric and palmitic
acids were of excellent quality and the myristic and oleic acids somewhat
inferior.
Lauric, myristic, and oleic acids in relatively large amounts showed no
appreciable influence on the crystallization of stearic acid. Palmitic
acid, on the other hand, noticeably increased the solubility and affected
the crystalline structure of the precipitate.
io8
Journal of Agricultural Research
Vol. VI. No. 3
Table V. — Effect of different fatty acids on precipitation of stearic acid
STEARIC ACID
Alcohol-stearic acid solution (grains
in ISO c. c).
Additional
stearic acid
, taken.
Other acids
taken.
Precipitate.
Saturation
(grams in
loo c. c).
a 3990
•3990
•3990
•3990
Gm.
o. 1000
• IO15
• 1035
. 1000
Gm.
Gm.
O. 3420
•3430
•3415
•3405
o. 1047
• 1050
• 1073
• 1057
LAURIC- ACID
o. 3990
•3990
•3990
•3990
1043
1050
1040
1033
MYRISTIC ACID
PALMITIC ACID
0.3990.
.3990.
.3990.
•3990-
•3990.
•105s
. 1000
. lOIO
. 1040
• 1050
4000
4030
2500
2500
2000
3135
, 2980
,296s
3065
3085
1273
1340
1357
I3I0
1303
OLEIC ACID
o. 3990
•3990
•3990
•3990
1030
1040
1003
1043
The addition of palmitic acid to butter acids reduced the amount of
stearic acid recovered in the test. Some of our more recent determina-
tions indicated that the solvent action of palmitic acid can be counter-
acted in a large measure, if not entirely, by increasing the relative amount
of stearic acid in solution. With butter acids of average palmitic acid
content, an alcohol-stearic-acid solution, containing at least 3 gm. of
stearic acid to the liter, is necessary and possibly 3.4 or 3.7 gm. may prove
more reliable. This, however, seems to depend to a considerable degree
upon the alcohol-stearic-acid solution employed. Some solutions made
Apr. 17, 1916
Stearic Acid in Butter Fat
109
from purified alcohol of approximately the same strength require more
stearic acid than others to insure a constant saturation, the reason for
which we have been unable as yet to determine. Some of the results
cited in Tables VI to VIII are probably low, owing to insufficient stearic
acid in solution, although the results are all calculated with reference to
blank tests conducted under precisely like conditions.
Table VI. — Amount of stearic acid in the insoluble acids of butter fat
Sample No.
Solution A.
(0.8153)0
Solution B.
(0.8135)0
Solution C. .
(0.8142)0
Solution D.
(0.8142)0
Solution E.
(0.8147) «
vSolution F . .
(0.8147)0
Solution A:
4
4
4
4
Solution B:
5
5
Insoluble
acids of
butter taken.
Gm.
5440
5170
5235
5000
5085
5190
5230
5010
5135
5230
Alcohol-
stearic-acid
solution
(grams in
150 c. c).
O. 3990
•3990
•3990
3960
,3960
,3960
.3960
,3960
4050
4050
4050
4050
4050
4050
4470
4470
4440
4440
3990
3990
3990
3990
3960
3960
3960
3960
.3960
.3960
Additional
stearic acid
taken.
Gm.
0.0525
.0500
.0500
0515
0530
0505
, 0490
0500
0575
0500
0800
0800
0820
0805
"15
1 105
1 100
Precipi-
tate.
Blank.
Satura-
tion
(grams in
100 c. c).
Gm.
0. 2900
.2880
.2895
Gm.
0. 2375
.2380
^•2395
6 . 2383
0. 1077
•1073
. 1063
.2630
. 2640
.2625
.2605
. 2610
.2115
. 2110
. 2120
• 2115
. 2110
&. 2114
. 1230
•1233
. 1227
• 1230
•1233
.2770
. 2710
.2195
. 2210
.1237
. 1227
. 2990
•3005
. 2190
.2205
b . 2200
. 1240
• 1230
.3040
•3045
. 2220
. 2240
h . 2230
. 1220
. 1207
•3765
•3765
.2650
. 2650
b . 2650
• 1213
• 1213
.3840
•3830
•273s
2730
"•2733
•1137
. 1 140
.2930
. 2900
.2915
.2905
■2383
•2383
•2383
•2383
. 2460
. 2480
. 2114
.2114
•2555
■ 2520
.2114
. 2114
■2515
• ^500
. 2114
. 2114
Stearic
add.
Per cent.
Hydrometer reading at is-50° C. of the alcohol employed.
Average.
10. 06
10. 00
10. 16
10. 44
b 10. 17
6.80
,7-05
&6.93
8.43
8. 10
68.27
7.81
7-38
6 7. 60
no
Journal of Agricultural Research
Vol. VI. No. 3
Table VI. — Amount of stearic acid in the insoluble acids of butter fat — Continued
Sample No.
Insoluble
acids of
butter taken.
Solution A:
8
8
8
Solution C:
9^
9
9
9
Solution B:
IO&
lO
Solution C:
IO&
lO
lO
Solution B:
lib
II
Solution C:
14
14
14
15
15
i6
i6
17 c
17
18 c
18
18
I9C
19
Solution D:
20
20
21
21
" Average.
Gm.
o. 5170
• 5120
•5225
5090
515s
5150
5130
5130
5050
4995
5060
5255
5065
5150
5165
5070
5015
5045
5060
5265
5205
5045
5300
5035
4990
5105
5205
5110
5180
5025
5090
5205
Alcohol-
stearic-acid
solution
(grams in
150 c. c).
O. 3990
•3990
•3990
,4050
,4050
,4050
,4050
3960
3960
4050
4050
4050
3960
3960
4050
4050
4050
4050
4050
4050
4050
4050
4050
,4050
,4050
,4050
4050
4050
4050
4050
4050
4050
Additional
stearic acid
taken.
Gm.
Precipi-
tate.
Gm.
O. 2850
. 2840
.2830
2970
3000
2970
2980
3000
3020
3070
3075
3140
2990
3000
2645
2635
2625
2705
2715
2665
2680
2965
2980
2945
2930
2940
2925
2905
2720
2715
2630
2650
Blank.
Gm.
O. 2383
•2383
•2383
2200
2200
2200
2200
2114
2114
2200
2200
2200
2114
2 1 14
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
2200
2230
2230
. 2230
■ 2230
Satura-
tion
(grams in
100 c. c).
Stearic
acid.
Per cent.
9
8,
8,
a 8,
15
15
14,
15
'15
17
17
17.42
17.29
17.89
a 17. 56
17-30
17. 20
«i7. 25
8.62
8.58
8.47
a 8. 56
10. 01
10. 18
o 10. 10
8.83
9. 22
09. 03
15. 16
14.72
a 14. 94
14. 80
14.63
14.50
« 14. 64
13-93
13.80
o 13- 87
9.46
9-65
09. 56
7.86
8.07
I a 7. Q7
'' The cows were fed beef tallow.
f The cows were fed palm oil.
Apr. 17. 1916
Stearic Acid in Butter Fat
III
Table VI. — Amount of stearic acid in the insoluble acids of butter fat — Continued
Sample No.
Insoluble
acids of
butter taken.
Alcohol-
stearic-acid
solution
(grams in
ISO c. c).
Additional
stearic acid
taken.
Precipi-
tate.
Blank.
Satura-
tion
(grams in
100 c. c).
Stearic
acid.
Solution F:
I
Gm.
0. 5070
■5225
•5205
•5215
.5090
•5140
.5060
•5015
•5035
0. 4440
.4440
•4440
.4440
.4470
.4470
.4470
.4470
.4470
Gm.
Gm.
ao. 3850
•3915
a. 3830
•3850
^^•3525
•3525
•3535
'^^ 3505
•3520
Gm.
o- 2733
•2733
•2733
•2733
■ 2650
. 2650
• 2650
.2650
.2650
Per cent.
22.03
I
22.62
II
II
Solution E:
III
Ill
Ill
IV
622.33
21.08
21.42
621. 25
17. 19
17. 02
17.49
617.23
17-05
IV
17.28
617.17
a Molecular weight of the several precipitates, 284.54-
!> Average.
c Molecular weight of the several precipitates, 284.59-
Table VII. — Amount of stearic acid in the insoluble acids of beef tallow
Sample No.
Solution A.
Do...
Solution B.
Do....
Solution A:
I
I
Solution B:
Insoluble
acids of
beef tallow
taken.
Gm,.
o. 5280
•5155
5025
5150
Alcohol-
stearic-acid
solution
(grams in
150 c. c).
O. 3990
•3990
.3960
.3960
3990
3990
3960
3960
Additional
stearic acid
taken.
Gm.
O. 1500
•1555
1520
1550
Precipi-
tate.
Gm.
o. 3870
•3930
3690
3700
3975
3960
3740
3775
Blank.
Gm.
O. 2370
•2375
"• 2373
. 2170
.2150
«. 2160
2373
2373
. 2160
. 2160
Satura-
tion
(grams in
100 c. c).
O. 1080
• 1077
"93
1207
Stearic
acid.
Per cent.
30-34
30-79
" 30- 57
31-44
a 31. 40
» Average.
112
Journal of Agricultural Research
Vol. VI, No. 3
Table VIII. — Amount of stearic acid in the insoluble acids of palm oil
Sample No.
Insoluble
acids of
palm oil
taken.
Alcohol-
stearic-acid
solution
(grams in
isoc. c).
Additional
stearic acid
taken.
Precipi-
tate.
Blank.
Satura-
tion
(grams in
100 c. c.).
Stearic
acid.
Solution C
Gm.
0. 4050
.4050
.4050
.4050
.4050
.4050
.4050
.4050
.4050
.4050
Gm.
o- 1515
• 1500
. 2000
.2030
• 1500
■ 1510
• 1540
■ 1500
• 1500
•1565
Gm.
o- 3745
•3750
•4255
•4295
b, 4040
. 4100
.4265
.4205
•4245
■4275
Gm.
0. 2230
. 2250
•2255
.2265
0. 2250
. 2250
• 2250
. 2250
.2250
• 2250
.2250
0. 1213
. 1200
.1197
. II90
Per cent.
Do
Do
Do
Solution C:
12
12
12
12
12
12
o. 3405
. 4110
•5205
. 5000
•5215
•5135
8.52
8. 27
9-13
9. 10
9.49
8.96
08.91
o Average.
* Molecular weight of the several precipitates, 284.38.
The stearic acid obtained from the insoluble acids of butter fat by the
method described ranges from 7 to 22 per cent, which is considerably in
excess of the amount generally credited to the product. The prevaiHng
opinion was supported undoubtedly by the fact that only a small amount
of precipitate is obtainable by the Hehner and Mitchell (3) method, as
shown by several investigators.
The amount of s'^earic acid appears to be affected by the feed the animal
receives. Samples 9, 10, and 11, averaging 16.67 P^^ cent, were from
cows fed beef tallow; samples 17, 18, and 19, averaging 14.48 per cent,
were from those fed palm oil; while samples 4 to 8, 14 to 16, 20 and 21,
averaging 8.70 per cent, were from those fed a ration low in fat. It is
probable that the individuality of the animal and period of lactation
also affect the composition. The entire matter of the effect of food as
well as other influences upon the chemical character of butter fat is now
being further studied.
The stearic acid (8.91 per cent) recovered from the insoluble acids of
palm oil exceeded the amount usually reported.
SUMMARY
The results of the determinations of stearic acid in the insoluble acids
of butter fat by the method proposed show a higher percentage of stearic
acid than has been generally reported. The facts that the results are
concordant and that the molecular weight determinations of the crys-
tallized product secured by the proposed method agree closely with the
theoretical molecular weight leave no doubt as to the identity and
approximate purity of the stearic acid.
Apr. 17. 1916 Stearic Acid in Butter Fat 113
LITERATURE CITED
(1) David, J.
1878. Mdthode de dosage et de separation de I'acide st^arique et de I'acide
ol^ique provenant de la saponification des suifs. In Compt. Rend.
Acad. Sci. [Paris], t. 86, no. 22, p. 1416-1418.
(2) Emerson, W. H.
1907. The solubility of stearic acid in ethyl alcohol at zero. In Jour. Amer.
Chem. Soc, v. 29, no. 12, p. 1750-1756.
(3) Hehner, Otto, and Mitchell, C. A.
1896. On the determination of stearic acid in fats. In Analyst, v. 21, no. 249,
P- 316-332, I fig.
(4) Holland, E. B.
1911. Purification of insoluble fatty acids, /n Jour. Indus, and Engin. Chem.,
V. 3, no. 3, p. 171-173.
(5) Kreis, Hans, and Hafner, August.
1903. tJber Stearinsaure-Bestimmungen. In Ztschr. Untersuch. Nahr. u.
Genussmtl., Jahrg. 6, p. 22-27.
(6) Lewkowitsch, J.
1913. Chemical Technology and Analysis of Oils, Fats, and Waxes. Ed. 5,
V. I. London.
(7) Partheil, a., and FER16, F.
1903. Zur Kenntnis der Fette. In Arch. Pharm., Bd. 241, Heft 7, p. 545-569,
illus.
(8) Ruttan, R. F.
I9i3(?) Margaric acid and its relation to palmitic & stearic acids. In Orig.
Com. 8th Intemat. Cong. Appl. Chem., v. 25, 1912, p. 431-442.
LIFE HISTORY AND HABITS OF TWO NEW NEMATODES
PARASITIC ON INSECTS^
[PRELIMINARY PAPER]
By J. H. Merrill, Assistant Entomologist in Charge of Fruit-Insect Control, and A.
L. Ford, Assistant in Life-History Studies, Kansas State Agricultural Experiment
Station
INTRODUCTION
While investigating the life history and methods of control of the elm
borer (Saperda iridentata Oliv.) and the termite {Leucotermes Itccifugus
Rossi) at the Kansas Agricultural Experiment Station, two new nematodes
were found, one parasitic on the former and the other parasitic on the
latter. One hundred and twenty-one adult beetles obtained from one tree ^
were placed in breeding cages, but in no instance were eggs deposited,
and both sexes eventually weakened and died. Examination after death
showed that the intestines were so filled with nematodes that in only one
female were eggs even developed in the body. The death rate due to
nematode parasitization was apparently loo per cent. Several colonies
of Leucotermes Itccifugus were placed in salve boxes, together with food.
Inasmuch as Saperda iridentata had shown so high a nematode para-
sitization, it was naturally suggested that nematodes might be present
in the termites. Accordingly a number of these insects were killed and
examined, with the result that nematodes were found infesting the head
in varying degrees. Of the colonies taken, 76.92 per cent were para-
sitized with nematodes. The parasitism of the individuals in single
colonies ranged from o to 100 per cent.
DIPLOGASTER LABIATA
The nematodes were submitted to Dr. N. A. Cobb, of the Bureau of
Plant Industry, United States Department of Agriculture, for identifi-
cation. He found that the nematode parasitizing Saperda iridentata
was a new species which he named "Diplogasier labiata" (fig. i ; 2,
A-H), and described as follows:
12 17 21 '59'^' 9'
Diplogaster labiata, n. sp. 2" , — 4^2 — 4I2 — 474 — 279 ^'^^ ™™" ^'^^^ formula was de-
rived from a single specimen.) The thin layers of the transparent, colorless, naked
cuticle are traversed by fine transverse striae, resolvable with high powers into rows of
dots, more particularly near the head and on the tail, those on the tail being some-
what irregularly placed. The cuticle is also longitudinally striated, and the dots of the
transverse striations are coincident with those of the longitudinal striatic ns. The longi-
• Contribution from the Entomological Laboratory, Kansas State Agricultural College, No. 17. This
paper embodies the results of some of the investigations undertaken by the authors in the prosecution of
projects Kos. 13 and loi, Kansas Agricultural Experiment Station.
2 A tent was placed around an elm tree so that all emerging insects might be secured for breeding purposes.
Journal of Agricultural Research, Vol. VI, No. 3
Dept. of Agriculture, Washington, D. C. Apr. 17, 1916
dd Kans. — 2
(115)
ii6
Journal of Agricultural Research
Vol. VI, No. 3
Fig. i.—Diplogaster lahiata: A, Mating (X 12s); B, mature female reared in water culture (X 12s). «. lip
region, 6, esophagus, c, median bulb, d, cardiac bulb, e, intestine,/, ovaries, g, egg, h, genital pore, i, rec-
tum, k, anus; C, mature male reared in water culture (X 125), a, lip region, b, esophagus, c, median bulb,
d, cardiac bulb, e, intestine, ife, anus, to, spicula; £>, at time of hatching (X 400); £. female during process
of molting (X 123); F, dead female with young nematode which hatched within her body (X 125)'
Drawings by A. L. Ford.
Apr. 17, 1916
Two New Nematodes
117
Fig. 2.—A-H, Diphgaster labiala: Development of the egg (X 500); /, Diplogaster aerivora: mature male
reared in moist soil (X 160); J, Diplogaster aerivora: mature male reared in water culture (X 125), a lip
region, b. esophagus, c, median bulb, d, cardiac bulb, e, intestine, k, anus, m, spicula; K Diplogaster
aerivora: Ac&d female with young which hatched within her body (X 12s); L, Diplogaster aerivora-
matmg (X 123)- Drawings by A. L. Ford.
27471°— 16 2
ii8 Journal of Agricultural Research voi. vi, no. 3
tudinal striae are not present on the lateral fields, this naked space being one-third to
one-half the width of the body. The slightly conoid neck becomes slightly convex-
conoid near the head, the lip region of which is set off by a very broad, almost imper-
ceptible constriction. There are six strongly developed and fairly distinct lips,
each ending in a conoid tip, from the sum.mit of which issues a very short innervated
bristle-like papilla. The lips have a more or less distinct refractive framework and
are in all probability quite mobile. Usually in specimens which have been fixed in
Flemming's solution the tips of the lips are slightly outward-pointing, leaving a
somewhat circular refractive mouth opening about two-fifths as wide as the front
of the head. The inner surface of the lips is so strongly refractive that usually the
posterior limits of the lips are distinctly visible, more particularly as the wall of the
pharynx at this point is encircled by a very delicate refractive line lying considerably
in front of the middle of the pharynx. This latter appears to be irregularly cylin-
droid, but is slightly unsymmetrical at the base. On the whole, it is about two-
fifths as wide as the head. It appears to possess at the base a rather well-developed
but blunt, slightly inward-projecting process or tooth. In the lateral view, as the
posterior part of the pharynx appears to pass around this projection, it acquires the
slightly unsymmetrical contour already mentioned. The walls of the esophagus are
rather distinctly ceratinized. The esophagus begins at the base of the pharynx as a
tube two-thirds as wide as the base of the head and continues to have this diameter,
or a slightly greater, until it reaches a point halfway back to the median bulb. Thence
onward it diminishes slightly, so that just in front of the median bulb it is only half
as wide as the middle of the neck. The median bulb is a well-developed, elongated
or ellipsoidal, radially muscular structure, with a somewhat distinct elongated but
narrow valve. This bulb is about two-thirds as wide as the middle of the neck.
Behind the median bulb the esophageal tube continues with a diameter one-third
to two-fifths as great as the corresponding portion of the neck but diminishes very
slightly, so that just in front of the ellipsoidal cardiac bulb it is less than one-third
as wide as the corresponding portion of the neck. The cardiac bulb contains a rather
distinct and rather complicated threefold valvular apparatus and is capable of open-
ing out posteriorly, so that the lumen of the posterior part of the bulb, where it
debouches into the intestine, then becomes one-fourth as wide as the corresponding
portion of the body. The lining of the esophagus is a distinct feature throughout
its length. The intestine, which is thin-walled at first, is separated from the esophagus
by a distinct constriction. It becomes at once four-fifths to five-sixths as wide as
the body and presents at the beginning a distinct cardiac cavity. There is also a
distinct cardia. The cells of the intestine, which are of such, size that probably four
are required to build a circumference, contain rather large nuclei and are packed with
granules of variable size, the largest of which have a diameter as great as the distance
between two of the longitudinal striae, the smallest of which are very much smaller.
The lining of the intestine is refractive, so that the lumen is usually quite a distinct
feature. From the slightly raised anus the narrow, refractive, ceratinized rectum,
which is one and one-half to two times as long as the anal body diameter, extends
inward and forward. The tail end begins to taper from some distance in front of
the anus but in front of the anus tapers only very slightlj-. Behind the anus it tapers
rather regularly to an acute point. Near the middle of the tail there appears to be a
lateral papilla on each side. From the slightly raised, rather broad vulva the vagina
leads inward at right angles to the ventral surface nearly halfway across the body,
where it joins the two uteri, which extend in opposite directions. The reflexed
ovaries reach more than halfway back to the vulva, at any rate in apparently young
specimens in which no eggs exist in the uterus. The ova in the ovary are arranged
more or less single file for about half its length; toward the blind end they are
arranged irregularly. Fertilized females show sperm cells in the uterus of such a
Apr. 17. i9i6 Two New Nematodes 119
size that about four to five side by side would span the bodj-^ diameter. Numerous
micro-organisms \vere seen in the intestine.
Male formula. Yi 3^1 3*5 3 9 09 ^'^^ ™"^' v^^"?'^ specimen). The tail of
the male diflfers m'aterially in form from that of the female. It begins to taper at the
anus, and it tapers rapidly in the anterior two-thirds, more particularly in the middle
third, so that at the beginning of the final third it is only about one-tenth as wide
as at the anus. Thence onward it tapers rather regularly to the exceedingly fine ter-
minus; there is, however, a pronounced ventral elevation at the beginning of the small
part of the tail, though it remains uncertain whether this elevation is innervated.
The middle portion of the tail is strongly convex-conoid , the convexity existing largely
on the dorsal side. The cuticle of the tail presents a peculiar arrangement of the dots,
such that there is an appearance of two sets of oblique fibers crossing each other, these
fibers being arranged approximately at 45° to the longitudinal lines. The two equal,
rather uniform, somewhat arcuate, blunt spicula are about one and one-fourth to one
and one-half times as long as the anal body diameter. Their proximal ends, which
are slightly narrower than the main portion, are set off by a rather broad and prominent
constriction. At their widest part, through the middle, they are about one-fifth to one-
sixth as wide as the corresponding portion of the body. The accessory piece is about
half as long as the spicula. It is very inconspicuous near the anus, but lies parallel
to the spicula. It widens out to a somewhat clavate or elongated pyriform contour,
and has its roimded proximal end toward the dorsal side of the body, and from this
blunt end muscular fibers pass obliquely backward to the ventral surface of the tail
and join the caudal wall at a distance nearly half way from the anus to the beginning
of the narrow portion. Oblique copulatory muscles are to be seen opposite the ejacu-
latory duct for a distance about one and one-half times as great as the length of the
tail. The male papillae are arranged as follows: One ventrally submedian pair a little
in front of the proximal ends of the spicula; one ventrally submedian pair a little in
front of the anus, and one ventrally sublateral pair on the same zone; another sub-
lateral pair just opposite the anus; a lateral pair slightly behind the middle of the
enlarged portion of the tail ; a submedian pair nearly halfway from that last mentioned
to the beginning of the small part of the tail; a dorsally sublateral pair a little in
front of the beginning of the narrow portion of the tail ; three subventral pairs close
together opposite that last mentioned; between the members of these three sub-
ventral pairs, possibly a single ventral papilla. The most pronounced of these
papillae can hardly be called digitate. The ejaculator}' ducc is about two-fifths as
wide as the body. The vas deferens is nearly two-thirds as wide as the body. The
testis tapers so that at the point of inflection, a short distance behind the cardiac bulb,
it is about one-fourth as wide as the body. The blind end lies about two body widths
behind the flexure.
Habitat: Manhattan, Kans., 1915, on Saperda tridentata.
The eggs of Diplogaster labiata, elliptical in shape, about twice as long
as wide, with bluntly rounded ends, when freshly deposited, were uni-
formly dark brown or gray, but after segmentation began they became
darker. Their average length was 0.0627 mm. and the average diameter
0.031 mm. They were laid singly with apparently no preference as to
the place of deposition. Occasionally segmentation began before the
eggs were deposited. From the beginning of segmentation the cell divi-
sions could be plainly followed throughout (fig. 2, A-H).
A few hours before emerging, the folded young nematodes made
slight movements within the egg. Later these movements became
I20 Journal of Agricultural Research voi. vi.No.s
more vigorous until finally they ruptured the shells and emerged, after
which the egg walls collapsed. Occasionally a young nematode hatched
within the body of a dead female. In cultures the eggs hatched in
from 30 to 32 hours from the time of deposition, and the nematodes
matured in from 7 to 10 days. The males appeared to mature slightly in
advance of the females.
At hatching, the young nematodes were about 0.2 mm. in length (fig.
I, D), very slender, and sluggish, and remained for a time in a curled
position. Later they straightened out their bodies and became very
active. The young worms were almost transparent (in water cultures),
there being no solid food in the alimentary canal. As development pro-
ceeded, the young became darker in color and more active. At the end
of 5 days the sex organs began to appear, and in from 7 to 10 days the
nematodes reached maturity.
Specimens which were isolated and kept under observation were noted
to molt at least three times, these molts occurring about three days apart.
The process of molting (fig. i , E) was as follows : The nematode first fas-
tened its posterior end to any surface upon which it might be resting.
The skin then broke at the anterior end and the nematode began to
emerge. At first the process was very slow, owing to the fact that the
opening of the molt skin was smaller in diameter than the middle part of
the body. By moving vigorously from side to side, the nematode slowly
worked its way out of the skin. After the widest portion of the body
had passed through the opening, no further resistance to emergence was
offered, as the posterior end rapidly decreased in diameter. The nema-
todes were not always able to emerge, as occasionally specimens were
found which died before completing the process. Molting lasted from
45 minutes to 6 hours.
The adults and the young were similar in form and food habits, but
differed in that the adults possessed sex organs. The mature females
were about 0.7 mm. in length and 0.03 mm. in diameter, while the males
were about 0.6 mm. in length and 0.02 mm. in diameter.
As soon as maturity was reached, mating began (fig. i, A). The male
fastened its caudal end around the middle of the female's body. During
this process the male held its body rigid, while the female moved vigor-
ously from side to side. It was not uncommon to find males in the act
of mating with their bodies wrapped twice about the females. Toward
the end of the process the female increased her activity and soon shook
the male free. Many matings were observed, the shortest of w^hich
lasted about 2 minutes and the longest 30 minutes.
Proportion of sexes. — Of 367 specimens examined, 229 were found
to be females and 138 were males. In other cultures in which counts were
not made the females were noticed to be more abundant than the males.
Apr. 17, i9i6 Two New Nematodes 121
Period of oviposition. — While in the specimens of Diplogaster
labiata under observation mating usually occurred but once, occasionally
a few individuals mated a second time. Oviposition began from two to
four hours after mating and lasted over a period of about two days, during
which time the average number of eggs deposited was seven.
Habits. — These nematodes infested the intestines of adults of Saperda
iridentata in such large numbers that they prevented these insects from
performing their natural functions. They lived in the alimentary canal
in such large numbers that they ruptured the walls of the canal and,
escaping into the body cavity of the insect, caused its death.
The examination of individuals of Saperda iridentata which had died
in this manner rarely showed eggs that had started to develop. Speci-
mens of Diplogaster labiata placed in water cultures were fed on macerated
bodies of Saperda Iridentata. They flourished on this, but since the supply
was soon exhausted, substitute foods had to be used. Different substances
were tried with varying success, but macerated beetles placed in water
seemed to be the most satisfactory. Nematodes in cultures without food
usually did not live longer than two days. The presence of food acted as
a stimulant to copulation and oviposition, but both varied directly with
the abundance and adaptability of the food.
. The nematodes seemed to show no preference to either day or night for
depositing their eggs or any other of their habits.
Length of active breeding state. — If the nematode is considered
to be mature from the time of mating, it spends an average of about two
days as a normal active breeding adult.
DIPLOGASTER AERIVORA
In 1856, Charles Lespes^ gave a meager description of a nematode
which he found parasitizing Leucotermes lucifugus. His description is
short and so indefinite that it might apply to several species of nematodes,
but the habits he discusses closely resemble those of the nematodes found
in L. lucifugus in Kansas. However, Dr. Cobb identified this nematode
as Diplogaster aerivora (fig. 2, I-L; 3) and described it as follows:
Diplogaster aerivora, n. sp. j76~^3'9 '4*9 ^5 9 ^26 '"^ ™™' The transparent, mod-
erately thin layers of the colorless naked cuticle are traversed by ftne transverse
striae, resolvable with high powers under favorable conditions. The cuticle is trav-
ersed also by 24 longitudinal strise. These longitudinal striae are sometimes resolvable
into quadrate elements, each consisting of foiu- punctations arranged in a quadrangle
whose width is equal to the width of the stria. In the majority of specimens these
quadrate elements were not to be seen. The distance between the striae varies in
different parts of the body up to about twice their width. The striations of tlie
cuticle, both transverse and longitudinal, vary within pretty wide limits, the varying
> Lespc's, Charles. Sur un ndmatoide parasite des Termites. In Ann. Sci. Nat. Zool., s. 4, t. 5, p. 335-556.
1856.
122
Journal of Agricultural Research
Vol. VI, No. 3
4b c d
Fig. 3.-i?.>Wa./.ra.nWa.- A. FormfcundintemnteCX .so): B attimeof hatcM^^^^^^^
(><7;):\E.maturefemalerearedinwaterculture(Xx.5); F-M,developmeatoftheegg(X5oo). Draw
ings by A. L. Ford.
Apr. 17, i9i6 Two New Nematodes 123
conditions evidently being a function among other things of the age or condition of
tlie cuticle. There are lateral wings, though these consist simply of a pair of slightly
modified longitudinal striae.
The conoid neck becomes convex-conoid toward the truncated head, which is not
set off in any way. There are six comparatively well amalgamated lips, each of which
bears two innervated papillae, one on the forward surface and somewhat forward point-
ing, and one on the outer surface and somewhat outward pointing. The anterior of
these two papillae is extended beyond the surface of the lip in the form of a minute
seta or innerv^ated papilla, and corresponds to the cephalic seta of other species of
Diplogaster. The contour of the lip is not much disttu-bed by the presence of the
posterior papilla, which is sometimes very difficult to see. Close behind the lateral
papillae or setae there are minute openings in the cuticle, which in character closely
simulate the amphids in some otlier species of Diplogaster, notably those of D. fictor.
No doubt these are really the outward expression of minute amphids. Distally the
lips have thin extensions which can close together over the pharynx in such a fashion
that the front of tlie head is comparatively flat, though the tips of these lips may be
recurved and point forward so as to make an exceedingly minute elevation at the
middle of the front of the head. The latter has its front surface on the whole very
slightly depressed.
The pharynx is about as deep as the front of the head is wide, and bears near its
base on the dorsal side a relatively large, rather acute movable conoid tooth or onchus,
which reaches about one-third the distance to the lips when the latter are closed,
but which is relatiA^ely farther forward when the mouth is open. In addition there
is a very much smaller sub median projection that undoubtedly may be denomi-
nated a rudimentary onchus. When the lips are closed the pharynx is a little wider
at the base than anteriorly. At the base of the lips, opposite the posterior circlet
of labial papillae, the width of the pharynx is a little more than one-third that of the
corresponding part of the head. Posteriorly, however, the width appears to be nearly
three-fifths that of the corresponding portion of tlie head, at least when the head is
viewed in profile. The walls of the pharynx are thin but refractive and fairly well
ceratinized. The surface of the dorsal onchus is more highly ceratinized than that
of other portions of the pharynx. Both the onchus and the wall of the pharynx have
a yellowish or brownish color like that of the spicula. The end of the esophagus
receives the base of the pharynx and is at once fully two-thirds as wide as the corre-
sponding portion of the head. It continues to have the same diameter for some dis-
tance, then begins to expand and continues to do so to some^vhat behind the middle
of the neck, where it rather suddenly diminishes in diameter in such a way that it
is proper to speak of a median bulb, although the anterior end of this bulb is not
very distinctly set off by constriction from the anterior esophageal tube. This bulb
contains an elongated valvular apparatus which is about one-third as wide as the bulb
itself. This latter is three-fourths as wide as the corresponding portion of the neck.
Notwithstanding the rather massive character of this median bulb, the succeeding
portion of the esophagus is only about one-fourth as wide as the corresponding portion
of tlie neck. However, it soon begins to widen and forms a somewhat pyriform
cardiac bulb three-fourths as wide as the base of the neck. This bulb does not con-
tain any very evident valvular apparatus, though in it there are faint indications of
a modification of the esophageal lining. The intestine joins the posterior surface of
the cardiac swelling, and at this point is about one-third as wide as the correspond-
ing portion of the body. There is no very distinct cardia. The intestine widens
out rather gradually and attains a width at least half as great as that of the body.
The tail end of the female begins to taper from some distance in front of the anus.
This latter is slightly raised, especially its broader posterior lip. Behind the anus
the tail diminishes somewhat more rapidly for a short distance and tliereafter tapers
regularly to the hairfine terminus. From the anus the rectum, which is about as long
124 Journal of Agricultural Research voi. vi, N0.3
as the anal body diameter, extends inward and forward. Nothing definite is known
with regard to the lateral fields.
From the well-developed, slightly depressed vulva the vagina leads inward at right
angles to tlie ventral surface halfway across the body, where it joins the two symmetri-
cally placed uteri. The internal female organs are double and reflexed, and the
ovaries, which are rather narrow and packed with small ova arranged irregularly,
reach back to the vulva or even beyond. The ellipsoidal eggs are about as long as
the body is wide and about two-thirds as wide as long. Their shells are smooth and
rather thick. Specimens have been seen in which well-developed embryos existed
in the eggs contained in the uteri. Other specimens have been found in which two
to three dozen embryos had escaped from the eggs and then devoured the whole
interior of the mother's body. The excretory pore is located opposite the cardiac
swelling.
Male formula. ^ ^54 ^l\ '"fo' ^45 ^-^ '^^' The tail of the male diminishes
suddenly in diameter from the raised anus in such fashion that at a distance from the
anus not very much greater than the anal body diameter it has a diameter only
about one-fourth to one-fifth as great as at the anus. At this point, which is immedi-
ately behind the posterior group of male papillae, the tail begins to taper rather
gradually and somewhat imiformly, and continues so to do to the hairfine terminus,
though there is at first a very slight increase in the diameter, so that the tail has the
appearance of being very slightly constricted just behind the posterior caudal group
of male papillae. There is no spinneret, and there are no caudal glands. The two
equal, rather slender, tapering, arcuate, brownish, acute spicula are about one and
one-half times as long as the anal body diameter. At their widest part, a little
distance behind the cephala, the spicula have a width about one-tenth as great
as that of the corresponding portion of the body. From this widest part they taper
gently toward the cephalated proximal ends. In the other direction the spicula
taper regularly to their acute terminals. The accessory pieces surround the spicula
at their distal extremities. The portion of the spiculum surrounded by the accessory
piece constitutes about one-sixth of the length of the former. Extending backward
from this encircling part of the accessory piece is a median arcuate portion arranged
nearly parallel to the spicula and having its proximal end somewhat cephalated. The
entire lengtli of the accessory piece, including this median dorsal portion, is about one-
third that of the spicula. Like the spicula the accessory pieces are brownish in color.
The hemispherical-conoid innervated supplementary male organs are located as fol-
lows: In front of the anus three pairs, two of which are ventrally submedian and one
sublateral ; the sublateral pair is nearly opposite the middle of the spicula, and is on
nearly the same zone as the posterior of the two ventrally submedian pairs; the
anterior submedian pair is a little in front of the proximal ends of the spicula. Behind
the anus the papilla; are arranged as follows: One pair sub ventral or ventrally sub-
median immediately behind the anus, two pairs sublateral, and three closely approx-
imated pairs of small size, subventral. This latter group of three pairs is slightly
farther behind tlie anus than the foremost preanal pair is in front of it. The three pairs
do not appear to be uniform in structure, the two anterior appearing to be mere inner-
vations, while the posterior one is a distinctly raised innervated papilla like the
preanal ones. The posterior of the two pairs of sublateral postanal papillae is a trifle
in front of the group of three just mentioned, while the anterior is about halfway
between the group of three and the anus. The anterior border of the anus consti-
tutes a sort of rudimentary flap with an innervation. The testis is single and rather
broad and tubular. It extends forward and is reflexed a short distance behind the
base of the neck. The reflexed narrower part of the testis is about twice as long as
the corresponding body diameter.
Habitat: Manhattan, Kans. Found feeding on grasshopper eggs after the eggs had
been deposited in the ground.
Apr. 17, 1916 Two New Nematodes 125
The eggs of Diplogasier aerivora, which are elliptical in shape, averaged
about 0.062 mm. in length and 0.0335 "im. in diameter. When freshly
deposited, they were dark brown in color, but became transparent as the
embryo developed. Segmentation often began before the eggs were de-
posited and the succeeding cell divisions could (fig. 3, F-M) be readily
followed throughout. The eggs were numerous and could be found lying
close together in groups of from about 6 to 30. The eggs hatched in
about 18 hours from the time segmentation was first noticed. Toward
the end of the egg stage the living worm (fig. 3, M) could be plainly seen
moving about within the egg wall. These movements became more
active until the worm finally ruptured the wall and escaped.
At the time of hatching, the young nematodes (fig. 3, -B) of this species
averaged 0.2145 mm. in length. At this stage the sex organs could not
be distinguished, because of their poor development. In water cultures
the worms grew very rapidly and reached maturity in three to four days.
The females matured slightly in advance of the males (fig. 2, /). D.
aerivora never exceeded 0.5 mm. in length nor completed its life cycle
while within the termite (fig. 3, A). The nematodes remained in the
termite in this form for an indefinite length of time, but upon emerging
into moist soil they matured in about two days.
Although molting occurred in this species as in D. lahiata, it was much
more difficult to observe; and, while it was not observed more than once
in any individual, it is probable that more molts did occur. Molting
required less time in D. aerivora than in D. lahiata, and the posterior
end of the nematode remained free throughout the process.
In the older water cultures the adults became so numerous that they
appeared as a living mass to the naked eye. The females, which were
much larger than the males, averaged 0.99 mm. in length and 0.067 ni"^-
in diameter, while the males averaged 0.75 mm. in length and 0.046 mm.
in diameter. When free in moist soil, the worms became even larger;
the females (fig. 3, D) averaged 1.632 mm. in length and 0.1192 mm, in
diameter, and the males (fig. 3, E) averaged 1.1425 mm. in length and
0.0724 mm, in diameter.
When reared in water cultures, the females appeared darker than the
males, but when found in the soil both sexes appeared pearly white.
The alimentary canal of the female, like that of D. lahiata, was spiral,
while that of the male was straight. The posterior end of the female's
body tapered into a long, threadlike process, but in the male th:>s process
was shorter and its body ended in an abrupt hook.
Process op mating. — The process of mating in D. aerivora (fig, 2, L)
was much the same as in D. lahiata. The male clasped the female
slightly back of the middle of the body, so that its anal opening was in
direct apposition to the genital pore of the female. In mating, the pos-
terior end of the male usually completely circled the body of the female,
although exceptions occurred. Although many instances of mating
126 Journal of Agricultural Research voi. vi, No. 3
were observed, none lasted over 4^ minutes. As the mating neared com-
pletion, the female became more active and broke free.
Relation and economy of the sexes- — Both males and females
mated repeatedly with different individuals. A single female was ob-
served to mate with 7 different males, and during this time laid a total
of 317 fertile and 14 infertile eggs. The length of time from the first
to the last mating was 13 days. The greatest number of fertile eggs
produced from a single mating by any individual under observation was
125, but the average number was 52.63. A single male was successfully
mated with 10 different females, the latter depositing 624 fertile eggs.
The total time which elapsed during these 10 matings was 19 days.
Time and method of oviposition. — A single instance was observed
of a female depositing a fertile egg 30 minutes after mating, although
from one to two hours are usually required. The eggs developed in the
ovaries in large numbers and were rapidly discharged through the
genital pore. With age the females became very sluggish and did not
appear to be able to discharge their eggs; consequently these eggs
hatched within the body of their parent, where they fed on her internal
organs. Usually they were unable to escape, although instances were
observed where they escaped through the genital pore of the mother
(fig. 2, K).
Proportion of sexes. — ^Three hundred specimens were examined,
and of these 138 were males and 162 were females. In all cultures the
females seemed to be more abundant.
Habits. — These nematodes were found parasitic in the heads of
Leiicoiermes lucifugus, where under natural conditions the number varied
from o to about 75. Where heavy infestation occurred, the termites
became sluggish and often died. These worms were usually more numer-
ous in the immediate region of the mouth parts of Leucoiermes lucifugus,
although it was not uncommon to find them in the upper part of the cav-
ity of the head. A great many termites were dissected, and in no case
were nematodes found in the abdomen. In infested colonies nematodes
were often seen in the surrounding soil. These usually were found in
masses, feeding upon the bodies of dead termites or other available
decaying matter. Specimens of D. aerivora placed in water cultures
were found to flourish in the same food that was used for D. lahiaia.
It was necessary to feed these nematodes each day, for without food
they died in a very short time. As in D. lahiata, the presence of food
appeared to stimulate copulation and consequently caused an increase in
oviposition.
So far as could be determined, these nematodes showed no preference
to either day or night in mating, oviposition, or other habits.
Length of active breeding stage. — The active breeding life of the
female extended over a period of about 13 days, while that of the male
was about 19 days. The complete life cycle of D. aerivora required
from four to five days. As the individuals of this species which were
Apr. 17. i9i6 Two New Nematodes 127
examined had no hibernation stage, their hfe cycle was continually
repeated under favorable conditions. Insufficient moisture and lack of
suitable food seriously interfered with the development of these
nematodes.
A series of experiments was carried on to ascertain whether it is
possible to introduce these parasites into Letccotermes lucijugus. Good
cultures of nematodes were obtained in moist soil, into which specimens
of L. lucijugus were placed. After two days a number of these termites
were dissected, and it was found that there was an average of 22.9 nema-
todes in each head. In three days this average rose to 32.9 and in four
days it was 46.6. In each instance the check count remained the same,
being about 3 nematodes per head. After remaining in a similar culture
for 1 2 days, all the termites died and the bodies were found to be literally
alive with nematodes.
SUMMARY
(i) The eggs of Diplogaster labiaia hatched in from 30 to 32 hours,
while those of D. aerivora hatched in about 18 hours.
(2) The eggs of D. lahiata were deposited singly, while those of D.
aerivora were deposited in groups.
(3) More cases of eggs hatching in the body were found in D. aerivora
than in D. lahiata.
(4) The eggs of both species developed similarly.
(5) Both species, when reared in water cultures, used the same food,
but in nature they had different hosts.
(6) Both species molted, but the process dififered in that D. lahiata fas-
tened its posterior end, while D. aerivora did not.
(7) The adults of D. aerivora were larger than those of D. lahiata and
required much less time to mature.
(8) In water cultures, the females of both species were more numerous
than the males.
(9) Although mating was similar in both species, D. lahiata required
more time for the process.
(10) Individuals of D. lahiata usually mated but once, while those of
D. aerivora mated repeatedly.
(11) Neither species in their habits showed any preference to day or
night.
(12) The females of D. aerivora had a period of oviposition of about
13 days, while in D. lahiata this period lasted only about 2 days.
(13) In both species adaptable and plentiful food acted as a stimulant
to reproduction.
(14) Both species attacked insects, but in different regions of the body,
as D. aerivora was found in the head while D. lahiata was found in the
intestines.
(15) The life cycle of D. lahiata required more than twice as much
time as did that of D. aerivora.
(16) D. aerivora was successfully introduced into the termites.
INSECT INJURY TO COTTON SEEDLINGS ^
By B. R. CoAD and R. W. Howe, Entomological Assistants, Southern Field Crop
Insect Investigations, Bureau of Entomology
INTRODUCTION
The present work deals with leaf mutilation of cotton seedlings
{Gossypium spp.) caused b}^ insects. The observations were made in
the vicinity of Tallulah, La., during the spring of 191 5. Such injury to
cotton seedlings is probably found throughout the entire area of cotton
cultivation in the United States. The senior author has noted it in
many parts of Texas, both the drier and more humid portions, in Louisi-
ana, and in Arizona on irrigated cotton. Since these localities approxi-
mate the extremes of rainfall, temperature, and sunshine under which
cotton is cultivated, it is reasonable to expect the injury at almost any
place.
CHARACTER OF INJURY
The injury varies much in appearance and intensity, but all of the
examples which have come to the attention of the authors have certain
more or less constant characteristics. This is frequently noticed as soon
as the seedUngs appear above the ground, although it may not appear
until later. The time of the cessation is also variable, but it does not
seem to continue after the plants reach a height of 10 to 12 inches and
usually stops much earlier. In the vicinity of Tallulah this injury is
seen from the first sprouting of the plants until the latter part of May.
The first appearance is characterized by irregular holes appearing in
the cotyledons. These vary from small holes through the leaf or small
marginal incisions to almost complete loss of the leaf. Following this
the later leaves are attacked in the same manner, with all possible varia-
tions in the type and degree of the injury. In some cases the terminal
bud may be lost.
LABORATORY STUDIES
Efforts were made to secure growing plants at the earliest possible
date. For this purpose cotton seed was planted in boxes and pots in the
laboratory during the very early spring, but lighting facilities were so
poor at this season that the plants failed to thrive. The first healthy
seedlings which were secured sprouted in the laboratory hotbed March
16 from seed planted in the middle of February. Seed planted in another
part of this hotbed on March 5 sprouted well a little later. This hotbed
' The investigations upon which this paper is based were conducted under the direction of Mr. W. D.
Hunter, in Charge of Southern Field Crop Insect Investigations. Bureau of Entomology.
Journal of Agricultural Research, Vol. VI, No. 3
Dept. of Agriculture, Washington, D. C. Apr. 17, 1916
de K— 29
(129)
I30
Journal of Agriciiliural Research
Vol. VI, No. 3
was covered with glass during the night and was only opened during
the warmer part of the day. The plants appeared perfectly healthy at
all times and grew well.
Other plantings were made in the laboratory yard at intervals during
March for studies under outside conditions. Later, seeds were germi-
nated between layers of moist absorbent cotton and placed in pots con-
taining soil sterilized by baking. These pots were then placed in large
screen cages and the plants were allowed to grow under this protection.
The first injury was noted in the hotbed on March 31. These seedlings
had sprouted March 1 6 and at this time were about 3 inches tall. They
had been protected from cold by the glass covers, and the soil had been
well manured. On this first morning a number of plants were found to
have been injured.
Following this the progress of the injury was noted carefully. All
plants were examined daily and those showing injury were tagged. In
this manner a record of the number of plants injured each day was
secured. On the morning of April 14 nine new seedlings were injured;
on April 1 5 five, on April 1 6 six, on April 1 7 two, and on April 1 8 three.
In order to determine the period in which the injury was incurred,
both morning and evening counts were started. These showed the num-
ber of seedlings injured during the night and during the day. These
observations were started April 22 and continued until May 4. The
results are presented in Table I.
Table I. — Comparison of day and night injury to cotton seedlings
Date of examination .
Apr. 22.
23-
26.
27.
28.
29.
30-
Number
of seed-
lings
injured
during
day.
Number
of seed-
lings
injured
during
night.
Date of examination.
May I
2
3
4
Total
Number
of seed-
lings
injured
during
day.
26
Number
of seed-
lings
injured
during
night.
49
From this table it is seen that 66 per cent of the injury appeared
during the night and 34 per cent during the day.
On April 14 this same type of injury appeared upon seedlings which
had just sprouted in the laboratory garden, and from that time it appeared
about as abundantly here as in the hotbed.
The rapidity with which the injury was produced was quite striking,
and special studies were made upon this point. A number of appar-
ently healthy and entire seedlings were examined morning and evening,
and in that way the amount of injury produced in a single night was
Apr. 17. J9i6 Insect Injury to Cotton Seedlings 131
determined. This was done in both the hotbed and the garden, and
the results were the same in both cases. I^eaves which were entire and
uninjured at nightfall would show large holes often occupying one-half
of their area on the following morning. Later observations have shown
practically entire leaves disappearing in the same manner during the
night.
During the first few days when the injury was appearing in the hotbed
a number of examinations were made during the daytime in the attempt
to find some insect producing the injury, but not a single individual
which could be suspected of being the cause was noted. However,
on April 6, 50 square inches of the hotbed soil were examined to a depth
of 3 inches, and 12 cutwonns were found. If this was a fair sample
of the hotbed, the soil there certainly contained hundreds of the worms.
Eleven of these larvae were very small, while one was about an inch in
length.
The presence of these larvae in the hotbed and the fact that they
were known to feed upon plant leaves made it seem quite possible that
they were responsible for more or less of the injury. Consequently
several examinations were made at night, and a number of cutworms
were found feeding on the leaves of the plants. At this time the same
injury was noted on clover and weed leaves in the hotbed.
Several half-grown cutworm larvae collected on cotton in the garden
and hotbed were placed on the surface of the soil in a pot containing
a number of seedlings. This pot was placed in a screen cage and the
larvas attacked the seedlings at once. Plate XII, and Plate XIII,
figure I, show several seedlings injured by these larv^.
STUDIES OF CLIMATIC FACTORS
A number of tests were conducted to determine whether any of the
injury could be due to the exposure to low temperatures during the
night or to the hot sunlight in the morning before the plants had time
to become warm. In the first test a wooden frame was erected over a
cotton planting in the laboratory garden just prior to the sprouting of
the plants. This frame was 2}4 feet in height and was covered with
8-ounce duck. This cloth was placed over the frame at sundown each
day and allowed to remain until about 10 o'clock the following morning.
In this manner the radiation was reduced under this cover during the
night and the plants were protected from sudden exposure to the sun-
light in the early morning. A minimum thermometer was suspended
under the cover in the center of the bed about 1 5 inches from the ground
and another was suspended at the same height in the open garden a few
feet away. Records continued for a few nights showed only a slightly
higher temperature under the shelter, so the frame was lowered to within
i^ feet of the ground and the thermometers were lowered to 6 inches.
Following this the minimum temperatures under the cover usually ranged
a few degrees higher than in the open.
132 Journal of Agricultural Research voi. vi, no. 3
This frame was first erected April 26 and on April 30 the first seedlings
appeared above the ground. Of the 18 which sprouted this first day, 5
showed injur3^ On May i, 9 of the 45 seedlings showing above the
ground were injured, while on May 3, 10 out of 50 were injured. On
May 4, 16 out of 70 and on May 5, 22 out of 70 were injured. These
observations were continued until May 8 and new seedlings were injured
practically every day.
On May 8 a second test of the same sort was started. In this case,
however, the cotton row was covered just before sprouting with heavy
pasteboard boxes, i foot square and 8 feet long. These boxes were cov-
ered with several layers of 8-ounce duck and were only removed from
over the plants during the hotter part of the day. Minimum thermom-
eters were arranged under the boxes and in the open in the same manner
as that just described in the preceding test. In this case considerable
differences in the nightly minimum temperatures were noted. It was
usually from 3 to 6 degrees warmer under the box than in the open. On
May 12 the first seedlings appeared, and of the 39 in sight, 3 showed
injury to the leaves. This test was continued six days longer and the
injury continued to appear.
For comparison with the seedlings growing in the garden and hotbed,
a number of seeds were planted at intervals in pots and crocks containing
soil sterilized by baking. Part of these were allowed to remain exposed
in the open, while others were placed in screen, cages. In the hundred
or more seedlings grown in this manner not a single sign of injury was
found, whereas the injury was appearing abundantly on plants growing
in the garden and hotbed at this same time. From this it seemed quite
evident that the cause of the injury was located in the soil which had
not been baked.
FIELD OBSERVATIONS
As the injury was appearing in the various fields at this same time,
efforts were made to learn its extent and to discover any insects which
might cause the lesions. In these studies all insects which were known
to be leaf feeders were noted and an attempt was made to secure positive
samples of their injury to cotton. On April 19 four small lepidopterous
larvae were found feeding upon the leaves of cotton seedlings at a plan-
tation near Tallulah. The injury which they were producing was
apparently identical with that already noted. These larvae belong to
the family Liparidae and are commonly known as "tussock moths"
(H enter ocampa leucostigma Smith and Abbot) . On this same date three
larvae of the same species were found feeding on the seedlings in the
hotbed and one was found in the laboratory garden. Following this
the field examinations showed a considerable number of these larvae to
be present around Tallulah, and associated with them were found several
species of cutworms and "measuring worms." All produced nearly the
same type of injury to the seedlings.
Apr. 17, 1916
Insect Injury to Cotton Seedlings
133
In order to determine definitely the amount of injury present in the
various cotton fields around Tallulah and also the prevalence of the
worms, a considerable number of examinations were made during the
latter part of April. In these observations only the worms found on
the cotton seedlings were noted. In order to make the figures more
accurately represent the condition of the field, the plants were examined
in groups of 100 each in all parts of the field. The results are summarized
in Table II.
Table II. — Records of examinations for insect injury to cotton seedlings infields around
Tallulah, La.
Date.
Niunber
of seed-
lings ex-
amined.
Number
of seed-
lings
injured.
Percent ^umber
Type of soil.
Remarks.
1,000
200
2,300
1,000
800
800
2,300
1, 200
1,000
400
266
30
534
84
188
54
207
380
227
43
26. 6 \ 25
Sandy
do
All tussock larvas; very small.
Seedlings just above the
ground.
Eleven cutworms and 14 tus-
sock larva.
22 and 23
32.2 25
8.4 ! I
23-5 ; I
6.7
13-4 9
31-7 4
22. 7 I
...do
Buckshot .
Sandy
Buckshot .
Sandy
...do
...do
..do
Do
six cutworms and 3 tussock
larvae.
Two geometrid larvae and a
tussock larvae.
Total
11,000
2,013
66
Forty-five tussock larvae, ig
cutworms, and 2 geometrids.
Weighted av-
erage. . .
18.3
1 1
From this it is seen that the percentage of plants injured at the various
plantations visited ranged from 6.7 to 32, with an average of 18.3 per
cent for the 11,000 seedlings examined. In the course of these obser-
vations 66 lepidopterous larvae in all were found. By far the greater
part of these were the "tussock" larvae and the remainder were either
cutworms or "measuring worms."
The possibility of the soils having some influence upon the extent of
damage was considered, but the writers were unable to secure sufficient
information to allow definite conclusions. Soils in the vicinity of Tallulah
may be roughly classed as either "sandy" or "buckshot." The former is
the light, sandy land found on the bayou fronts, while the "buckshot" is
the dark, heavy, stiff" "back land." Under boll-weevil conditions "buck-
shot" land is not adapted to cotton culture; hence, only two fields of this
type of soil were located for study. The percentage of injured seedhngs
in these two fields was 6.7 and 8.4. These were the lowest records made
and are considerably below the average of sandy fields near by. Whether
or not this lesser degree of injury was due to the soil is open to doubt.
Owing to the "coldness" of " buckshot" land in the spring, the cottonseed
germinates slowly and consequently the plants were considerably smaller
27471°— 16 3
134
Journal of Agricultural Research
Vol. VI, No. 3
than those on sandy land. This may have caused the difference in the
percentage of injury. However, only one suspected larva (a cutworm)
was found in the two fields.
The different lepidopterous larvae noted were all observed to be feeding
upon the leaves. The tussock larvae were much the more abundant and
evidently produced a great deal of the injury. During the earlier exami-
nations nearly all of these tussock larvae were quite small. The injury
produced varied somewhat with the size of the larva. The very small
individuals fed only upon the epithelium of the lower side of the leaf and
the injury was not visible from above. With a slight increase in size
the larvae started to feed through the leaf and at this stage produced the
peculiar type of injury shown in Plate XIII, figure 2. I^ater the older
larvae (one-half to full grown) ate large holes in the leaves, and the injury
could no longer be distinguished from that of the other species concerned.
Plate XIII, figure 3, shows the injury produced by one nearly full-grown
tussock larva when confined in a large screen cage with cotton seedlings
growing in a pot.
About May i nearly all cotton fields under observation suddenly began
to show, greatly increased injury until within a few days many fields had
practically every plant more or less mutilated. This proved to be due
to an invasion of grasshopper nymphs. These speedily became very
abundant and swarmed over the young cotton, feeding principally upon
the leaves. This is shown in Plates XIV and XV. These cotton leaves
were collected in the field when the young grasshoppers were feeding
upon them.
A little later in May the 12-spotted cucumber beetle, or adult of the
southern corn rootworm (Diabrotica 12-punctata Olivier) , became abundant
locally and added to the injury. The work of these beetles closely
resembled that of the worms and grasshoppers, though the holes made
were usually not very large. At this' same time woolly-bear larvae began
to appear in the fields and produced the same injury.
Following this great increase in injury to the plants caused by the
grasshoppers, counts were made to determine the percentage of injured
seedlings in four average fields near Tallulah. The information secured
from these examinations is shown in Table III.
Table III. — Abundance of injured cotton seedlings after the grasshopper invasion
May 14.
IS-
17-
17-
Number of
seedlings
examined.
800
3>5oo
2, 000
I, 000
Total
Weighted average.
7.300
Number of
seedlings
injured.
792
3,446
I, 920
I, 000
7,158
Percentage,
injured.
99.0
98.5
96. o
100. o
98.0
Apr. 17, 1916
Insect Injury to Cotton Seedlings
135
Here it is seen that 98 per cent of the 7,300 seedHngs examined had
been injured by some of the various agencies operating prior to that
time. High as they are, these figures are representative of average condi-
tions in the fields near Tallulah.
ACTIVE PERIOD OF LARV/E
On April 14 continuous examinations of cotton seedlings were made
from 8 a. m. until noon and from i to 5 p. m. on two plantations
near Tallulah. The day's records of worm collections were divided into
hourly periods and in this manner the active time of the various larvae
was noted. The results of these studies are shown in Table IV. From
this it is seen that the tussock larvae were much the more abundant
throughout the day and there seemed to be no time at which they were
especially abundant on the plants. The same seems to be true of the
other larvae.
Table IV. — Records of field examinations for larvce by hourly periods on two plantations
near Tallulah, La.
Period.
Number and kinds of larvae found.
First plantation.
Second plantation.
8a
m. to 9 a. m
2 tussock larvae, i cutworm
1 tussock larva.
9a.
m. to loa. m
7 tussock larvae
4 cutworms, 2 yellow
"woolly-bear" larvae.
10 a. m. to II a. m . .
9 tussock larvae
9 tussock larvae, 2 cut-
worms, 3 yellow " wool-
ly-bear "larvae.
iia
. m. to 12 noon. . . .
3 tussock larvae, 2 small cutworms .
No examinations.
I p.
m. to 2 p. m
7 tussock larvae, i small cutworm.
No worms.
? n.
m. to3p. m
I yellow "woolly-bear " larva.
6 tussock larvae
2 unknown larvae.
^P-
m. to4p. m
6 tussock larvae, i small cutworm. .
3 unknown larvae.
4 p.
m. to 5 p. m
Summary
7 tussock larvae, 2 yellow "woolly-
bear " larvae.
I tussock larva, x geome-
trid, 6 unknown larvae.
57 tussock larvae, 5 cutworms, and
14 tussock larv£e, 8 cut-
3 yellow ■ ' woolly-bear ' ' larvae.
worms, 5 yellow
"woolly bears," i ge-
ometrid, and II un-
known larvae.
Total, both
71 tussock larvae, 13 cutworms, 8
plantations.
yellow "woolly-bear" larvae, i
geometrid, and 11 unknown
larvae.
INJURY TO TERMINAL BUDS
The greater part of the feeding of the insects just mentioned is confined
to the leaves. However, a considerable number of plants were found
with the terminal buds either partially or completely destroyed. Plate
XVI, figure I, shows the usual location of this injury. This seedUng was
found in the field with a lepidopterous larva embedded at the base of
the bud (a). The small cavity where the larva was feeding is shown in
the photograph. From this the injury progresses until often all the buds
and small leaves above point a are eaten out.
136
Journal of Agricultural Research
Vol. VI, No. 3
ULTIMATE EFFECT OF INJURY UPON THE PLANTS
The preceding pages have shown the different insects contributing to
the mutilation of cotton seedlings, but it is the ultimate effect upon the
cotton production of the plants which determines the economic impor-
tance of the injury. This is a point upon which it is difficult to secure
accurate data, but a certain amount of information has been gathered by
the writers.
A number of plants are evidently killed outright by the feeding of the
insects; but this number appears to be so small, even in fields very heavily
infested, that it is of no practical importance.
The leaf feeding is also of very doubtful importance. In severe cases
it retards the growth of the plants somewhat and occasionally dwarfs
them permanently, but usually they recover very rapidly, and there is no
visible effect other than the slight retardation.
Apparently it is the injury to the terminal buds which produces the
most important economic effect. When this bud is injured or destroyed,
the development of the plant is greatly changed. Instead of having a
single main stem extending to the top of the plant, two or more large
branches develop just below the injured bud and serve as stalks to pro-
duce the fruiting branches. Usually several very abnormal clusters of
leaves form around the stalk near the injury. In Plate XVI, figure 2,
the result of similar injury is shown in comparison with a normal plant.
These two plants were collected in the garden at the laboratory and
were stripped of their leaves before being photographed. Plant B shows
a normally developed stalk and its branches, while plant A shows the
deformity caused by the destruction of the terminal bud.
About the middle of June a number of examinations were made in the
fields near Tallulah in order to determine the abundance of these deformed
plants. The results of these examinations are given in Table V.
Table V. — Records of field examinations for deformed cotton plants at Tallulah, La.
June 8.
9
9-
10.
II.
II
16.
17-
Total
Weighted average.
Number of
plants
examined.
4, 000
100
100
1, 000
100
500
400
600
6,800
Number of
plants
deformed.
3
7
87
4
63
42
553
Percentage
deformed.
7.8
3-0
7.0
8.7
4.0
12. 6
8.2
7.0
Location.
Plantation.
Hotbed.i
Laboratory garden.^
Plantation.
Do.
Do.
Do.
Do.
' Just prior to this examination the plants in the garden and hotbed had been hand thinned; and as the
poorest plants were removed, the percentage of deformed plants was evidently greatly lowered.
Apr. 17, 1916
Insect Injury to Cotton Seedlings
137
From this it is seen that the percentage of deformed plants ranged
from 3 to 10.6, with an average of 8.1. As these same fields furnished
the records given in Tables II and III, and were shown in the latter to
have practically every plant more or less mutilated, it seems evident that
only a comparatively small amount of the injury produces final deformity.
However, an injury which deforms only 8 per cent of the plants in a field
still is of considerable importance.
When this deformity was first observed it was at once noted that the
injured plants were not forming as many squares as normal plants of the
same age and height. Further studies showed this effect to be so pro-
nounced that counts were made in the fields to determine the relative
squaring of deformed and normal plants. In these observations, every
time a deformed plant was found its squares were counted, and likewise
those on the nearest eight normal plants of the same size. The average
of these normal plants was compared with the number upon the deformed
one. In 40 cases out of the 229 recorded the squares on the injured
plants exceeded the average of the nearby normal plants, but in all others
the average of the normal ones was considerably higher than the number
on the injured plants. A summary of these observations is given in
Table VI.
Table VI. — Effect of deformity upon fruiting of cotton plants
Deformed plants.
Normal plants.
Date.
Number
observed.
Total
squares.
Average
squares
per plant.
Maxi-
mum
squares
per plant.
Number
observed.
Total
squares.
Average
squares
per plant.
Maxi-
mum
squares
per plant.
June 10
11
II
16
17
87
4
63
33
42
248
23
52
405
559
2.8
5-0
0.8
12.3
13- I
10
9
6
26
34
700
32
502
264
336
3,804
267
1,105
3,931
6, 122
5-4
8.3
2. 2
14.9
18.2
16
13
10
34
53
Total
229
1,287
1,834
15,229
Weighted
averages
5-6
8. 2
The 229 deformed plants averaged 5.6 squares per plant, while the 1,834
normal ones averaged 8.2 squares. This gives a difference of 2.6 squares
per plant in favor of the normal plants at the time of these observations.
From these figures it is evident that the necessity for the additional
vegetative development before squaring retards the fruiting of the plants
considerably. This is a point of great importance in cotton culture
under boll-weevil conditions. The primary requisite for a successful
crop in the presence of the boll weevil is early, rapid, and prolific fruiting.
This allows the safe "setting" of a crop before the weevils multiply
138 Journal of Agricultural Research voi. vi. No. 3
sufficiently to infest all the squares. Hence, any agency which retards
the formation of the squares in the early spring does a very serious injury
to the crop. While the deformed plants may overtake the normal plants
later in the quantity of fruit, this fruit will be produced too late to insure
safe maturing.
Another effect of the deformity which may be of considerable import-
ance is the ease with which the plants are split when the two or more
branches fork at the same point. This gives a very weak stalk, and a
comparatively slight jar will split it. In fact, the weight of a crop of
bolls will break many of the plants.
SUMMARY AND CONCLUSIONS
From the various observ^ations discussed in this paper it seems that
mutilation of cotton seedlings may be produced by any of several insect
pests. These consist of a number of species of lepidopterous larvae
(cutworms, measuring worms, "woolly-bear" larvse, tussock-moth
larvae, etc.), grasshoppers, and leaf beetles. In all fields several species
of these pests were present, and in many fields all of them were found.
During the spring of 191 5 at Tallulah, La., the tussock larvae were re-
sponsible for most of the damage early in the season and then were sup-
planted by the grasshopper nymphs. However, the relative importance
of the various species undoubtedly varies with the locality and season.
Tests made with plants protected from low temperatures during the
night and from bright sunshine in the early morning demonstrated that
the injury would appear about as abundantly on these plants as on the
unsheltered plants in the garden and field. Seedlings in large number,
raised through this period in pots and crocks containing baked soil,
failed to show the slightest trace of injury, although they were fully
exposed to the weather.
Injury to cotton by cutworms has been known for many years, but
usually has been considered to consist only of the cutting of the plant
stem near the ground. In 1897 Howard * published a brief review of the
information then available concerning these larvae, but did not mention
them as leaf feeders. In 1 905 Sanderson ^ mentioned the injury due to
these worms and also discussed the work of Prodenia ornithogalli. This
species he recorded as being diurnal in habits and feeding upon the
leaves, but he considered the damage to the squares and bolls as its
most important injury. Sanderson also mentioned the "woolly-bears"
as occasionally damaging cotton by feeding upon the leaves.
In actual effect upon the plants it seems that the injury of the various
species may result in death of the plant, dwarfing of growth, or defoimity
iHo-ward, L. O. Insects affecting the cotton plant. U. S. Dept. Agr. Farmers' Bui. 47, 32 p., 18 fig.
1897.
2 Sanderson, E. D. Miscellaneous cotton insects in Texas. U. S. Dept. Agr. Farmers' Bui. 223, 24 p.,
39 fig. 1905.
Apr. 17, 1916 Insect Injury to Cotton Seedlings 139
of the stem, producing retardation of the fruiting. Of these the deform-
ing of the stalk is evidently much the more important. Field examina-
tions have shown that an average of 8 per cent of the plants in the
fields under observation were deformed and that these abnormal plants
averaged 2.6 squares per plant less than the normal ones about the
middle of June. As the cotton in these fields averages about 4 feet
between the rows and is spaced about 18 inches in the drill, this would
mean a loss of over i ,500 squares per acre at the critical period in cotton
production in the presence of boll weevils.
The "woolly-bear" larvse mentioned in this paper were reared and
proved to be Estigmene acraea Drury. Two of the cutworms have been
identified by Mr. S. H. Crumb, of the Bureau of Entomology, as Prodenia
ornithogalli Guenee and Peridroma margaritosa Haworth, var. saucia
Hiibner.
PLATE XII
Fig. I. — Cutworm injury to cotton seedlings; produced in breeding cages.
Fig. 2, 3. — Cutworm injury to cotton seedling.
(140)
Insect Iniury to Cotton Seedlings
Plate XII
Journal of Agricultural Research
Vol. VI, No. 3
Insect Injury to Cotton Seedlings
Plate XIII
Journal of Agricultural Researcli
Vol. VI, No. 3
PLATE XIII
Fig. I. — Cutworm injury to cotton seedling.
Fig. 2. — Tussock larva feeding upon cotton leaf. The ragged injury shown here is
usually produced by the smaller larvae.
Fig. 3. — Injury produced by a nearly full-grown tussock larva when confined in a
screen cage containing jxjtted cotton plants.
PLATE XIV
Cotton leaves showing grasshopper injury.
Insect Injury to Cotton Seedlings
Plate XIV
Journal of Agricultural Research
Vol. VI, No. 3
Insect Injury to Cotton Seedlings
Plate XV
Journal of Agricultural Research
Vol. VI, No. 3
PLATE XV
Fig. I. — Underside of cotton leaf showing grasshopper injury. This shows a num-
ber of places where the very small nymphs ate only the epithelium and did not pene-
trate the leaf.
Fig. 2. — Cotton leaf showing grasshopper injury.
PLATE XVI
Fig. I. — Injury to terminal bud of cotton by lepidopterous larva. This worm was
embedded at point a.
Fig. 2. — Two cotton plants from laboratory garden with leaves removed. Plant A
shows the abnormal forking caused by injury to the terminal bud, while B is a normal
stalk. The absence of fruit on plant A is due to the deformity.
Insect I njury to Cotton Seedlings
Plate XVI
Journal of Agricultural Research
Vol. VI. No. 3
ADDITIONAL COPIES
OF THIS PtTBUCATlON MAY BE PEOCtTEED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRrNTING OFFICE
WASHINQTON, D. C.
AT
15 CENTS PER COPY
Subscription Price, 83 Per Yeas
Vol. VI
APRIL 24, 1916
No. 4
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS
A Sex-Limited Color in Ayrshire Cattle
EDWARD N. WENTWORTH
Watermelon Stem-End Rot . . . .
F. C. MEffiR
Effect of Pasteurization on Mold Spores .
CHARLES THOM and S. HENRY AYERS
Crowngall Studies, Showing Changes in Plant Structures
Due to a Changed Stimulus
ERWIN F. SMITH
Page
141
149
153
Effect of Water in the Ration on the Composition of Milk 167
W. F. TURNER, R. H. SHAW, R. P. NORTON
and P. A. WRIGHT
179
DEPARTMENT OF AGRICUITUEE
WASHINGTON, D.C.
WAeHINGTON : GOVERNMENT PfllNTINQ OFFICE : 191«
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALIvEN
Chief, Office pf Experi»:ent Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Ento^wlosy
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Patlwlogist, and Assistant
Dean, Agricultural Experiment Stationof
the University of Minnesota
All correspondence regarding articles from the Department of Agriculture
should be addressed to Karl F. Kellerman, Joiu^al of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultiu-al Research, Orono, Maine.
JOHNEOFAGRICDLTIIALffiSEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, April 24, 191 6 No. 4
A SEX-LIMITED COLOR IN AYRSHIRE CATTLE*
By Edward N. Wentworth,
Professor of Animal Breeding, Kansas Agricultural Experiment Station
TYPES OF INHERITANCE AS RELATED TO SEX
Two general types of inheritance as related to sex exist, aside from the
ordinary secondary sex characters. Sex-linked inheritance depends on
the great mass of hereditary factors that have been shown to be linked in
transmission to the sex-determining factors; while sex-limited factors
follow the simple Mendelian scheme of inheritance, but show a reversal
of dominance in the two sexes. Frequently these two latter terms are
used synonymously, but since there is a distinction between the two
classes of transmission, and since the term "sex linked" is so much
more descriptive of the hereditary phenomena to which it has been
applied than is the term "sex limited," the foregoing terminology is used.
HISTORICAL REVIEW
The classical case of sex-limited inheritance was reported by Wood (7) ,
who made reciprocal crosses of the Dorset sheep, a breed horned in
both sexes, with the Suffolk, a breed polled in both sexes. All Fj
individuals were the same, so far as the type of cross was concerned, the
males being homed and the females polled. In the Fj generation the
fact that dominance differed in the two sexes resulted in three males
being horned to one being polled, and three females being polled to one
being homed.
Similarly in 191 2 the writer reported a pair of rudimentary teats in
swine, located on the lower part of the scrotum of the male and on the
inner thighs of the female, behind the inguinal pair, which presented the
same phenomenon in transmission, the character being dominant in the
male and recessive in the female.
Gerould (2)^ reported in 1911 on the inheritance of yellow and white in
the common clover butterfly (Colias philodice). White is dominant to
1 Paper No. 3 from the Laboratory of Animal Technology, Kansas Agricultural Experiment Station.
2 Reference is made by number to "Literature cited," p. 147.
Journal of Agricultural Research, Vol. VI, No. 4
Department of Agriculture, Washington, D. C. Apr. 34, 1916
df Kans. — 3
(141)
142 Journal of Agricultural Research voi. vi. No. 4
yellow in the female, but it is recessive in the male. Something lethal
seems to be connected with homozygosis for white; hence, white as a
somatic character appears only in the female. The yellow female is YY,
the white female YW. Males are either YY or YW, but are always yellow.
Jacobson (3) made some observations on Papilio mennon L., which
were studied from a Mendelian standpoint by De Meijere (5) in 1910.
There are three varieties of females in this species known as Achates,
Agenor, and Laomedon, respectively, in the order of their dominance.
The males corresponding to these three forms are all alike, although
each of the female patterns may be carried in a recessive manner. Fur-
thermore, De Meijere believes that the female carries the male pattern
homozygously ; but, owing to the reversal of dominance, the male
character never becomes somatic. The I^aomedon probably represents
the female expression of the male condition. The principal difference
between this and the previous cases is that the changes in dominance
affect the homozygotes as well as the heterozygotes.
AYRSHIRE BLACK-AND-WHITE
A case which seems to fall under this genera^, sex-limited group is found
in the inheritance of black-and-white as alternative to red-and-white in
Ayrshire cattle. While the general breed color is red-and-white, black-
and-white animals have been known for some time, as shown by Kuhl-
man (4). Practically no attention has been paid to the mode of inherit-
ance of this color, since in America it has been considered undesirable
and selection against it has been practiced. It is difficult to state
whether the black is due to a true black pigment or whether it is simply
a very dense red. Under the microscope typically black granules seem
to be present, but no chemical solutions of the pigments have yet been
attempted.
SOURCE OF THE DATA ^
The Ayrshire herd bull at the Kansas Experiment Station, Melrose
Good Gift, is a very deep mahogany-and-white ; in fact, the black-and-
white previously referred to. It is through the study of his ancestry
and breeding performance, the ancestry and breeding performances of
the cows in the herd, including the black-and-white animals, and the
records of some of the former herd bulls that the present data were secured.
In all, 63 individuals were included. Much larger numbers might have
been obtained by adding the progeny of red-and-white males and females
to the table; but since they demonstrated no facts different from those
here included, their records are not presented.
* Acknowledgments are hereby niade to Prof. O. E. Reed, of the Department of Dairy Husbandry,
Kansas Experiment Station, for facilities extended in obtaining the data.
Apr. 24. 1916 A Sex-Limited Color in Ayrshire Cattle 143
PROGENY OF MELROSE GOOD GIFT FROM RED-AND-WHITE COWS
Fifteen red-and-white cows in the herd were mated to Melrose Good
Gift to produce 20 calves, of which 10 were black-and-white bulls and 10
were red-and-white heifers. All of the bulls were as red as the heifers
at birth, but at 2 to 4 months of age the blackish tinge began to develop,
and within 4 months the youngsters became distinctly black-and-white.
The heterozygous male progeny of Melrose Good Gift differed from the
homozygous male progeny in that the black tinge developed more slowly
and also became much less intense on maturity. While in the mature
homozygous bull the black is very distinct throughout the pigmented
areas, in the mature heterozygous bull the black may appear only as a
streaked border where the pigmented spots adjoin the white, or at the
limbs, muzzle, ears, and tail. The main portions of the colored parts of
the animal are usually a very dark red which blends gradually, although
in a particulate manner, into the blacker borders. The heterozygous
heifers are red-and-white, and while occasional dark hairs are found, no
regular means whereby the heterozygous red-and-white females could be
distinguished from the homozygous red-and-white females was discov-
ered. It should be further noted that the black color of the homozyg-
ous female is by no means as intense as that of the male, although the
black is indisputably present.
HETEROZYGOUS BLACK BULLS TO HOMOZYGOUS RED COWS
Johanna Croft King, College Marquis, Sir Croft of Spring City, Wool-
ford's Good Gift, and Lessnessock Oyama's Good Gift were bulls which
by their breeding performance and somatic description must have been
heterozygous for the black factor. The last two bulls were found in the
pedigree of Melrose Good Gift, while the first three were used at one time
or another at the college as herd bulls. Records of these in matings to
homozygous red-and-white cows were available for all except Woolford's
Good Gift, and the result showed four red-and-white heifers, four black-
and-white bulls, and 5 red-and-white bulls. This is the most probable
distribution of colors in both the males and females and is perfectly in
alignment with the interpretation of the method of inheritance as given.
The reciprocal cross of red-and-white bulls to black-and-white cows
gave two black bulls to one red bull and two white heifers, also the most
probable expectation.
BLACK-AND-WHITE COWS MATED TO RED-AND-WHITE BULLS
Only three calves were available from this type of mating, all red-and-
white daughters of Bangora, the original black-and-white cow in the
herd. While the numbers are too small to be conclusive, yet they
conform to the expectation.
144
Journal of Agricultural Research
Vol. VI, No. 4
RESULTS OF THE DIFFERENT CROSSES
If the factor for the black-and-white color is represented by B, the
hereditary constitutions are as follows: BB is always black-and-white;
bb is always red-and-white; Bb is always black-and-white in the male and
red-and-white in the female. All of the nine possible matings were
discovered, as shown in Table I.
Table I. — Results of nine possible matings of Ayrshire cattle
Sires.
Dams.
Male offspring.
Female offspring.
Black-and-
white.
Red-and-
white.
Black-and-
white.
Red-and-
white.
BB
BB
Bb
I
o
lO
3
I
4
o
2
O
0
0
0
0
0
5
0
I
7
3
0
0
2
I
0
0
0
0
0
BB
I
BB . .
bb
10
Bb
Bb
BB
Bb
I
0
Bb
bb
4
bb
BB
3
2
bb* . .
Bb
bb
bb
9
Total
21
20.75
13
13-25
6
5-25
30
Expected .
30-75
The expectations here presented are based on the most probable result
of each of the matings, considered on an individual basis with reference
to the number of animals produced by each type of mating, but without
figuring the proportions of the sexes as equal. From these data it would
appear that the black-and-white color of Ayrshire cattle behaves in an
ordinary sex-limited manner similar to the horns in sheep as discussed by
Wood (7) and the rudimentary mammae in swine as reported by the
writer (6).
DISCUSSION
Arkell and Davenport (i) have reported on the inheritance of horns in
sheep and have attempted to bring it under the ordinary sex-linked
scheme of inheritance by an ingenious system of inhibitors and horn
factors. Such an explanation was doubtless justified when horns in
sheep were the only character known in which the reversal of dominance
in the two sexes existed, but now that at least two other characters are
known in which an exactly similar system of inheritance occurs, it seems
unnecessary to assume the complexities hypothesized by these investi-
gators. Instead, the much simpler and probably more perfectly descrip-
tive explanation adopted by Wood (7) in his original paper seems more
logical.
Apr. 24. 1916 A Sex-Limited Color in Ayrshire Cattle 145
COLOR RECORD OF PROGENY IN AYRSHIRE CATTLE
The following record presents the data considered in this paper. The
term "red" refers to red-and-white and the term "black" refers to black-
and-white. The hereditary constitution assigned the breeding animals
retained in the herd or found in the pedigrees of animals in the herd is
also given.
Johanna Croft King, Bb (described as fvSir Croft of Spring City, Bb (black),
dark). 1 Johanna of Juneau, bb (red).
College Marquis, Bb (described as dark).{^^'^!^^^ f.^'^f'f' ^^ ^'''^^•
^ ' iMaggie of Woodruff, Bb (red).
Woolford's Good Gift, Bb (described as f^^"'"^''°'=^' Oyama's Good Gift, Bb (de-
mahogany), scribed as dark).
I Pearl 3d of Woolford, bb (red cow).
Melrose Good Gift, BB (black.and-white).|!!;°°^°''^'^,^°«^' ^^^S^l^-
(.Florence Melrose, Bb (red cow).
College Maud, bb (red) {^^''^^/t °;,W°°d™«' ^^ (red).
^ ' V ^ IStarof Hill view, Bb (red).
(White Prince, Bb (described as mahogany in
pigmented areas).
Star of Hillview, Bb (red).
College Marquis 2d, bb-t. ,, /College Marquis, Bb (dark).
College Marquis 3d, bb/^ ' ICoUege Maud, bb (red).
(See progeny of College Maud.)
Progeny of College Maud 31350 (red), bb:
One red heifer by unknown red bull, bb.
One red heifer by College Marquis, Bb.
Three red bulls by College Marquis, Bb.
One red heifer by Johanna Croft King, Bb.
One red heifer by Sir Croft of Spring City, Bb.
One red heifer by Melrose Good Gift, BB.
Progeny of College Maud 2d (red), bb (daughter of College Maud by College Marquis):
One red heifer by College Marquis, Bb.
Progeny of College Maud 2d's heifer (red), bb (daughter of College Maud 2d by College
Marquis) :
One black bull by Sir Croft of Spring City, Bb.
One red biill by College Marquis 3d, bb.
One black bull by Melrose Good Gift, BB.
One red heifer by Melrose Good Gift, BB.
Progeny of Kansas Croft Maud (red), Bb (daughter of College Maud by Sir Croft of
Spring City):
One red heifer by Melrose Good Gift, BB.
One black bull by Cavalier's College Master, bb.
Progeny of Johanna Croft Maud (red), bb (daughter of College Maud by Johanna Croft
King):
One red heifer by Melrose Good Gift, BB.
146 Journal of Agricultural Research voi. vi, no. 4
Progeny of Georgie Em 25749 (red), bb:
One red heifer by Sir Croft of Spring City, Bb.
One red heifer by College Marquis 3d, bb.
One black bull by Melrose Good Gift, BB.
One red heifer by Melrose Good Gift, BB.
One red heifer by College Marquis 2d, bb.
Progeny of Georgie Croft (red ) , bb (daughter of Georgie Em by Sir Croft of Spring City) :
Three black bulls by Melrose Good Gift, BB.
Progeny of Marquis Em (red), bb (daughter of Georgie Em by College Marquis 3d):
One red heifer by Melrose Good Gift, BB.
One black bull by Melrose Good Gift. BB.
Progeny of Johanna of Juneau 26290 (red), bb:
One black bull by Sir Croft of Spring City, Bb.
One red heifer by College Marquis 3d, bb.
Twins (one black bull and one red heifer) by Melrose Good Gift, BB.
One red heifer by College Marquis 2d, bb.
Progeny of Elizabeth of Juneau 26292 (red), bb:
One red bull by Sir Croft of Spring City, Bb.
One red bull by College Marquis 3d, bb.
Two black bulls by Melrose Good Gift, BB.
Progeny of Rose of Oakdale 26291 (red), bb:
Two red bulls by College Marquis 2d, bb.
One red bull by College Marquis 3d, bb.
One red heifer by Melrose Good Gift, BB.
One red heifer by Cavalier's College Master, bb.
Progeny of Rosa Lee Melrose (red), bb (daughter of Rose of Oakdale by Melrose Good
Gift, BB):
One red bull by Cavalier's College Master, bb.
Progeny of Canary Belle 25748 (red), bb:
One red bull by Sir Croft of Spring City, Bb.
One red bull by College Marquis 3d, bb.
One red heifer by Melrose Good Gift, BB.
One black bull by Melrose Good Gift, BB.
One red heifer by Cavalier's College Master, bb.
Progeny of Melrose Canary Belle, (red), Bb (daughter of Canary Belle by Melrose Good
Gift, BB):
One red heifer by Cavalier's College Master, bb.
Progeny of Feamot of Oakdale 26289 (red), bb:
One black bull by Sir Croft of Spring City, Bb.
One red heifer by College Marquis 3d, bb.
One red heifer by Melrose Good Gift, BB.
One red bull by College Marquis 2d, bb.
Progeny of Lady Marquis Feamot (red), bb (daughter of Feamot of Oakdale by
College Marquis 3d):
One red heifer by Melrose Good Gift, BB.
Apr. 24. i9i6 A Sex-Limited Color in Ayrshire Cattle 147
Progeny of Bangora 29700 (black), BB:
One red heifer by Marquis of Woodruff, bb.
One red heifer by College Marquis, Bb.
Two black heifers by College Marquis, Bb.
One black bull by Sir Croft of Spring City, Bb.
One black bull by Johanna Croft King, Bb.
One black heifer by Melrose Good Gift, BB.
One black bull by Melrose Good Gift, BB.
One red heifer by Cavalier's College Master, bb.
Progeny of Bangora 2d (black), BB (daughter of Bangora by College Marquis):
One black bull by Johanna Croft King, Bb.
Two black heifers by Melrose Good Gift, BB.
Progeny of Bangora 's Melrose (black), BB (daughter of Bangora by Melrose Good
Gift, BB):
One red heifer by Cavalier's College Master, bb.
CONCLUSIONS
(i) Black-and-white color is a simple allelomorph of red-and-white
color in Ayrshire cattle.
(2) In the male the black-and-white character is dominant and in the
female the red-and-white character is dominant.
(3) Males heterozygous for the two characters are black-and-white,
while females heterozygous for the two characters are red-and-white.
LITERATURE CITED
(i) Arkell, T. R., and Davenport, C. B.
1912. Horns in sheep as a typical sex-limited character. In Science, n. s. v.
35, no. 897, p. 375-377-
(2) Gerould, J. H.
1911. The inheritance of polymorphism and sex in Colias philodice. In Amer.
Nat., V. 45, no. 533, p. 257-283, 5 fig. References to literature, p. 283.
(3) Jacobson, Edward.
1909. Beobachtungen tiber den Polymorphismus von Papilio memnon L. In
Tijdschr. Ent., deel 52, afl. 3/4, p. 125-157, 2 fig., tab.
(4) KUHLMAN, A. H.
1915. Black and white Ayrshires. In Jour. Heredity, v. 6, uo. 7, p. 314-322,
illus.
(5) MeijERE, J. C. H. de.
19 10. tJber Jacobsons Ziichtungsversuche beziiglich des Polymorphismus von
Papilio Memnon L. 9 und iiber die Vererbung sekundarer Ge-
schlechtsmerkmale. In Ztschr. Indukt. Abst. u. Vererbimgslehre,
Bd. 3, Heft 3, p. 161-181, pi. 3.
(6) Wentworth, E. N.
1912. Another sex-liraited character. In Science, n. s. v. 35, no. 913, p. 986.
(7) Wood, T. B.
1905. Note on the inheritance of horns and face colour in sheep. In Jour. Agr.
Sci., V. I, pt. 3, p. 364-365, pi. 4.
WATERMELON STEM-END ROT
[preliminary paper]
By F. C. Meier,
Sftident Assistant, Cotton and Truck Disease Investigations,
Bureau of Plant Industry
During the last few years in certain parts of the United States shippers
have been seriously troubled by a decay which attacks watermelons
{Citrullus vulgaris) in transit and may sometimes destroy or render
unsalable a large percentage of a shipment before it reaches its destina-
tion. Owing to this fact, in the season of 1915 the Department of
Agriculture began a careful investigation of shipping conditions, in the
course of which the present writer had an opportunity to make a labora-
tory study of some decayed material.
This material was taken from a shipment received in Washington,
D. C, on July 24, 1915. The shipment consisted of five carloads of
approximately 900 watermelons each, no one car of which yielded more
than 300 salable melons, owing to the prevalence among them of the
disease. The decayed watermelons were distributed through the car
entirely without reference to position, a fact which made it seem mani-
festly impossible that the trouble could have originated from mechanical
or chemical injury received from contact with the walls or the floor of
the car.
This examination indicated, moreover, that, as has been reported in
the case of other shipments, the injury of these watermelons had occurred
in a very uniform manner. In its early stages the presence of the decay
was indicated by a watery discoloration of the rind in an area closely sur-
rounding and apparently extending from the stem. Beginning in this
way there were all stages of decay up to those where about half or three-
quarters of the melon were involved. In such cases the rind of this por-
tion had become soft and wrinkled, so that in cross section it appeared
much like that of the watermelons shown in the lower row of Plate XVII,
figure I . The meat below this part of the rind was slimy and blackened,
while that at the opposite end of the melon remained sound, not having
as yet become included in the decay. Owing to the warm, moist con-
ditions at this season, the portion involved was covered by a gray or some-
what black mold, so that the origin of the trouble could not be readily
ascertained.
An abundance of material being available at this time, an attempt
was made to find out whether the injury was due to the action of some
fungus, and, if this proved to be the case, to obtain the specific organism
Journal of Agricultural Research, Vol. VI, No. 4
Dept. of Agriculture, Washington, D. C. Apr. 24, 1916
dg G— 76
(149)
150 Journal of Agricultural Research voi. vi. No. 4
in pure culture. In endeavoring to obtain such cultures, the following
procedure was adopted. Several watermelons were selected in which
the decay was just beginning to be apparent. A razor was flamed;
and with this, a funnel-shaped section, which included a portion of
both diseased and healthy tissue, the two being separated by a more or
less distinct line of demarcation, was cut from the melon. After the
razor had been flamed again, the section was divided along the line of
demarcation which distinguished the advancing edge of the decay, the
plug being cut from the inside toward the outer surface. This gave
access to a portion of the rind to which the fungus filaments were proba-
bly just advancing and which would be unlikely to contain concomitant
forms. From this region, using a sterile platinum needle, small portions
were removed from just below the surface and placed directly on syn-
thetic agar in sterile Petri dishes. After two days, during which the
plates were kept at a temperature of 27° C, an abundant mycelial
growth of a gray color appeared in every instance. A number of trans-
fers of the mycelium thus obtained were made to potato cylinders, and in
all cases a fungus developed which possessed the characteristics of the
genus Diplodia. In order to test the capacity of this organism for pro-
ducing the decay, the pure culture was inoculated into a sound water-
melon at three widely separated points, at each of which the character-
istic rot was reproduced.
The direct connection between this fungus and the disease having
been thus indicated, 16 healthy watermelons were obtained for more
inoculations. They were bought at the wharf in Washington, D. C,
and came from the Pyankatank River district in Virginia, a region free
from the disease, so far as is known. It may be well to mention in this
connection that the decay has usually been reported as occurring on
the variety known as " Tom Watson." This is probably due to the
fact that in the last few years this melon has been grown somewhat to
the exclusion of other varieties. Of the melons chosen for inoculation,
three were "Excel" melons; the remainder were of the "Tom Watson"
variety.
These melons were placed on a table near a large window which was
kept open the greater part of the day, and were protected from the
direct light of the sun by a cardboard screen. For a period of nine
days, during which time the melons were under observation, the average
temperature was 26.5° C. Of these 16 watermelons, 8, two of which were
of the "Excel" variety, were inoculated with the fungus, the cultures
used in this case having been derived from the original subculture.
This was accomplished by making with a sterile knife at a single point
near the stem an incision, into which a bit of the growing fungus mycelium
was introduced. A similar wound was made in the remaining 8 melons,
including the third "Excel" variety, but no infectious matter was
introduced. Within 36 hours the 8 inoculated melons began to show
Apr. 24, 1916 Watermelon Stem-End Rot 151
signs of decay, while the 8 checks remained perfectly sound through-
out the course of the experiment. There was no decay present on
the inoculated melons except that which originated at the point of
inoculation.
The decay is first noticeable as a somewhat circular discolored area
surrounding and extending from the point of inoculation. On the
watermelons observed in the laboratory this area gradually increased in
size until at the end of six days about half of the melon was involved.
At this time the advance of the decay seemed to become less rapid and
the area which was first decayed began to show a blackening due to the
formation of pycnidia by the fruiting fungus. This area spread daily,
and at the close of nine days the stem end of the melon presented a
withered, charred appearance. Plate XVII, figure i , is a reproduction of a
photograph of nine of these melons. The four in the upper row are
checks; the five below were inoculated.
The fructification of the fungus may be briefly described as follows:
Pycnidia separate or confluent, smooth or, under moist conditions,
covered with loose olivaceous hyphae, 180 to 250^1 in diameter. Spores
24 to 2>on by 10 to 14/X, oval, uniseptate, dark brown. On the material
taken from the watermelons inoculated in Washington no paraphyses
could be detected. They are present, however, when the organism is
grown upon potato cylinders, a fact which would tend to support the
conclusions reached by Taubenhaus,^ to whose work reference will be
made in the following paragraph.
It has long been known that those members of the Sphaeropsideae
which produce brown uniseptate spores are extremely variable. The
distinctions between the genera Diplodia, Botryodiplodia, Chaetodiplodia,
Lasiodiplodia, and Diplodiella have been based on slight structural vari-
ations in the pycnidia. The points of separation are the relation of the
pycnidia to one another, whether scattered or cespitose; their relation to
the host, whether subcutaneous, erumpent, or superficial; the presence
or absence of bristles and of paraphyses. These are all characteristics
which one might expect to vary somewhat with the characteristics or the
condition of the host. This variation probably occurs; and for this rea-
son there has been some uncertainty as to the proper position certain
species should occupy in classification. Botryodiplodia theobromae Pat.,
which causes a dieback of Hevea braziliensis in Ceylon, southern India,
and the Malay States, is an example; and in his account of this fungus
Petch ^ remarks that —
Among the names which are known to refer to this species are Macrophoma vestita,
Diplodia cacaoicola, Lasiodiplodia theobromae, Diplodia rapax, and there are probably
others. Botryodiplodia theobromae is its earliest name, as far as is known, but some
prefer to call it Lasiodiplodia theobromae.
* Taubeahaus, J. J. The probable non-validity of the genera Botryodiplodia, Diplodiella, Chaetodip-
lodia, and Lasiodiplodia. In Amer. Jour. Bot., v. 2, no. 7, p. 324-331, pi. 12-14. iQiS-
'Fetch, Thomas. Physiology & Diseases of Hevea braziliensis . . . 268 p., 16 pi. Loudon, 19x1.
152 Journal of Agricultural Research voi. vi. No. 4
Taubenhaus, as a result of his inoculations upon sweet potato (Ipomoea
batatas) with Diplodia tubericola E- and E-, Diplodia gossypii Zim.,
Diplodia natalensis Pole Evans, and Lasiodiplodia theobromae (Pat.)
Griff, and Maubl., suggests that the characteristics of the genus Diplodia
be so extended that it may include all of the five genera.
This genus, although it is not thought to include forms which are
absolute parasites, is nevertheless a source of serious trouble among some
of our cultivated plants. The injury is usually confined to a fruit rot
or to a dieback of the younger branches or shoots as in the Citrus disease
prevalent in Florida and the Isle of Pines.^ In both cases the fungus
has been described as following an injury which has been previously
inflicted either by mechanical means or as the result of the action of some
other fungus. In the United States the more important crops which
hitherto have been known to be affected are sweet potato, Citrus fruits,
corn (Zea mays), and cotton (Gossypium spp.) In our Southern States
the Diplodia injury is of considerable consequence in connection with
these products. As one enters the Tropics the number of plants which
are attacked increases. Among the list of hosts found here are Citrus
spp., Hevea spp., Theobroma cacao, and Thea spp. In certain cases where
the growing plant is attacked, the injury produced is sufficient to cause
the death of the host, as is the case with Diplodia vasinfecta Petch,
which causes an internal rootrot of tea.
Since the cotton, sweet-potato, and watermelon fields of the South are
not widely separated, it is of some interest from the economic standpoint
to know whether a species found on one host will grow equally well upon
another. Plate XVII, figure 2, shows a watermelon nine days after it
had been inoculated with a culture of Diplodia tubericola E. and E.
obtained from Mr. L. L. Harter, of the Bureau of Plant Industry. The
decay took the same course in this melon as has been described for the
other inoculated material, which is shown in Plate XVII, figure i. The
pycnidia which were produced, however, retained the paraphyses.
While the Diplodia injury is apparently the cause of serious loss in the
watermelon industry, there are other ways in which the crop suffers. Dr.
W. A. Orton, Pathologist in Charge of Cotton and Truck Disease Investi-
gations, Bureau of Plant Industry, who has made a careful study of ship-
ping conditions, is inclined to believe that the injury is confined to
certain districts. In other sections, anthracnose, due to Colletotrichum
lagenarium, is the source of considerable trouble. To the losses thus
caused by fungi must be added a small percentage of melons which have
been damaged by rough treatment and by the use of cars which have
been employed for the transportation of fertilizer or chemicals to the
fields.
' Earle, F. S., and Rogers, J. M. Citrus pests and diseases at San Pedro in 1915. In San Pedro Citrus
Path. I,ab. ist Ann. Rpt. 1915, p. 5-41, 19 fig. [1915.]
PLATE XVII
Watermelons, showing the effect of inoculation with species of Diplodia:
Fig. I. — The upper four melons were held as checks; the lower five are melons
nine days after having been inoculated with a culttu-e of Diplodia sp. which had been
isolated from a decaying watermelon obtained from a freight car at Washington, D. C.
Fig. 2. — A watermelon nine days after having been inoculated with a culture of
Diplodia tuhericola E. and E.
Watermelon Stem
-End Rot
Plate XVII
J^^
^^Btt^^gf^^t^^k
Nv«|
li.
Br^^BSI^^^^Hr
HI^hK
^m^f^K
^ f ^ ^^^^^^jBH^I
B«^
1
^
Journal of Agricultural Research
Vol. VI, No. 4
EFFECT OF PASTEURIZATION ON MOLD SPORES
By Charles Thom, Mycologist, Bureau of Cfiemtstry, and S. Henry Ayers, Bacteri-
ologist, Bureau of Animal Industry
INTRODUCTION
Definite experiments to determine whether spores of the common sapro-
phytic molds sur\'ive the temperatures used for the pasteurization of
milk have not been reported. These spores are certainly present and are
frequently abundant in ordinary market milk. Vague and general state-
ments that such organisms do or do not survive are not uncommon, but
are not supported by reference to actual work. To obtain such data
studies were made with spores from pure cultures of a series of molds in-
cluding several species of Penicillium, Aspergillus, and of the mucors,
with, in some experiments, the addition of Oidium (Oospora) lactis and
one strain of Fusarium. These sets of experiments were made to test,
as carefully as laboratory conditions would permit, the temperatures used
in pasteurization by the "holder" process, those used in the "flash"
process, and the effects of dry heat.
EXPERIMENTS WITH THE HOLDER PROCESS OF PASTEURIZATION
Bacteriological studies of milk treated by the holder process have fixed
the temperatures between 140° and 145° F. (60° to 62.8° C), main-
tained for 30 minutes, as the minimum heating for the destruction of
pathogenic organisms which may be found in milk. Although certain
bacteria survive this heating it has been found that milk so treated is free
from the ordinary disease-producing organisms, safe for consumption,
unchanged in taste, and low enough in acid organisms to be handled with-
without souring too quickly.
To study the effect of this process of pasteurization on mold spores,
conidia from pure cultures of molds were first transferred to tubes of
sterile water to obtain a suspension of spores. Transfers from such a
suspension reduce the danger of such spores being blown by air currents
into the cotton plugs and upon the walls of the test tubes used, where
they might escape the full temperature applied to the milk. In the first
series the inoculations were made by transferring i c. c. of this suspension
in sterile pipettes into duplicate tubes of sterile milk. In a later series a
platinum loop was used, since the tendency of the conidia to float thickly
upon the surface of the water made this a quick and effective method of
handling them. For most species it was thus possible to transfer spores
enough to make a visible film over a part of the surface of the milk. None
Journal of Agricultural Research, Vol. VI, No. 4
Dept. of Agriculture, Washington, D. C. Apr. 24, 1916
dh A — 20
(153)
154 Journal of Agricultural Research voi. vi. no. 4
of the species used produced visible growth except upon or near the sur-
face of the milk. Observations of growth must include, therefore, the
surface of the milk and especially the glass from the surface of the milk
upward for a few millimeters, since most molds begin to grow first upon
the glass. When no spores occurred upon the glass a free-fioating colony
in one case escaped observation until it fruited.
The inoculated milk tubes, with the exception of the control tubes,
were heated in a water bath in which the water was agitated and the tem-
perature of the milk was recorded in a control tube by a thermometer
placed in the milk. The temperature in the tubes was not allowed to
vary more than half a degree in either direction. The results of the exper-
iments with the holder process are shown in Table I. In preparing this
table the records of the checks, or unheated tubes, of successive exper-
iments were found sufficiently uniform to permit them to be averaged
and appear but once. Experimental tubes were made in duplicate; and
when the results were not reasonably harmonious the work was repeated.
Table I summarizes the tabulated data from a series of experiments
extending over a period of several months.
Apr. 24, 1916
"^
:f;
•i^-
'^
•^
L)
<
Effect of Pasteurization on Mold Spores
155
o o
feO
feO
•sAcp 9
■SABp t"
■sAsp I
•sAcp 9
•SABp t'
•sABp Z
OOCOOOOOOOO -o -oooooo 5
• ■ ■ &
OOwiOOOOOOO 'O -oooooo
•sAep 9
'I- ;OOOOOOOOOOO0OO00OO •0OOOOOO0OO\OOO
•SABp ■^
oooc>oooooooooo>oo-!t- -OOOOOOOOr-inOOO °4;
■sAcp z
t<CJ
•sABp 9
OOWOi'lOOOOWOttoOOO'O -OOONOOOOOOOOOt^O
§5
•sAep t-
+ I -V
•sXep z
fcO
fed
too
O Ov
-a.
•sABp 9
•sAep *■
r-00 CO -OOOOOt^OOOOOOOO -O • ■('lOt^-OOOO • '0000 3a
S3
+ : I + : : : I : . . ,
W)i^iri •OOvOOO'O-iot^.roi'lOO -O • -CiON -OOOO ■ -O^OO
•sAbp e
•sABp 9
)»n0000^oo^-0 -o
> t^oo 00000
■sAvp P
■sA-ep z
■sAep 9
■sABp t"
■SABp i
•sABp 9
+ I I
: Ui : : : .' •5.0
<^O\O0O\0t^0000OOOOOO»00OiOO'«t^OO0OOOOO>t^OO rt'*'
_ gg
)^-^(^>rOPO^N w rO^N rOcorO-^fOt^ *^>0 (•^ (*) w to
O 00 00 VO MX
770 « P
el
4)
■o ..
m" a>
d S
;3S'
v?3
: s-3
I O. 'S. O
I ^ to
^^
■saassssaas ■ • •« • • • • s.2 » » S t*
Qt&fet&fefefcSs&QQQS^QQQo^iSaC'G
b
0
i-t
he
0)
■0
0
&
U
«
ai
■£
■Ti
>.
J3
09
&
27472°— le
156
Journal of Agricultural Research
Vol. VI, No 4
Y
"1^
•<A,
too
O <3
too
too
o o
too
tod
too
too
•sAcp 9
•SA'Bp't'
•sAvp i
•sAep 9
•sAcp t'
•sXep e
■sAcp 9
•sA'Ep *
•sAep I
•sABp 9
■SABp t"
■sA^p Z
•sA'Ep 9
•SA'Ep fr
■sAcp C
•sABp 9
•SABp fr
•sAep i
•sABp 9
•sAcp fr
•sABp r
to o o o o o
o o o o o
00'>tOOOOOOOOOOOOOOOOO
oonooooooooooooooooo
OOOOOO •OOOMOOOOOOOOOOOOOOOO'OO'OOm
o o o o o o
O O « O r) o
O Oi'<^- O 01 O
r*50 00 ^*0 O
<0 00 o o • t
OOOi^^-OOOOOOOOOOOOOOOONOtoO'
to ■ ix-O r> o r^ O O O m -OOOOOOOWOO
>0 lovO "* O
M 't **5 *0 O
000»n^O«OfO
looOvO '«t^')'00«OOOW O -vOOiOiO
«^co«*3fO«OOMmOO«0
00 -OtOOOOOOifOOfO
Tj- O '^ fOO^-N O to
++
:°iid
O S4J 5
te o o O
O aa^
•sA'Ep 9
)t»O00Q000>O t^^OOO t^O 00000 t^t^li^OO ^t^O»0000 viOkO O«00 o t*
•sAep ^
f0'tJ't'*»n^iO^io 1/lsO ^iDUiiowjtOlOTj-toW r^^O ^ VI fo^O 00 w^ »/l 1^ ^
•sA'Ep C
fO « CO w ro ^ (
^
«W«^POH« mO wm^^
I TT rfl M 00 « -O O
fO CO f*3 PO
SsS^s
3
^ 9 ^ S
3 • §
u u u u "J u^a la^ O.^ S •Oi'<\ w k. ■-)■-) F> F»
ssssssssssssssssssss
•S -2 -S -S -S -s -S •§ -S -2. S-SS-S ■§■§■§ ■§•§■§
8888S8888a8S8SS88g8g C
raoooooooo
Apr. 24, 1916
Effect of Pasteurization on Mold Spores
157
A study of Table I shows that very few mold spores survive exposure
to 140° F. (60° C.) in milk for 30 minutes and that at 145° F. (62.8° C.)
still fewer are found. With reference to significant organisms, among
the mucors the Mucor racemosus group (3513, 3523.6, 3560) and
Rhizopus nigricans, which are found more frequently than all others of
this group combined, were destroyed at 130° F. (54.5° C). The common
green species of Penicillium are mostly dead at 130° F. (54.5° C); a
few stand 135° F. (57.2° C), but two, one of them an undescribed soil
organism, survived 140° F. (60° C.) for 30 minutes. Among species of
Aspergillus, however, the strains of A. flavus, A. fumigaius, and A. repens
all survived 145° F. (62.8° C.) for 30 minutes; A. repens and A. fumigatus
both survived 150*^ F. (65.6° C). These three species are always found in
forage and feeding stuffs; hence, milk is more or less subject to contamina-
tion with them. A. repens grows
very poorly in milk, however, and
the examination of a great many
cultures of milk and its products has
shown that the actual development
of A . flavus and A . fumigatus is com-
paratively rare. Although these
organisms grow at blood heat and
have demonstrated their pathogen-
icity even to human beings at rare
intervals as causes of disease in the
lungs, there is no report of their
growth in the alimentary canal.
The destruction of mold spores by
the holder process of pasteurization
is shown more clearly in figure i,
where the results have been plotted.
Pasteurization of milk at 145° F. (62.8° C.) may therefore be regarded
as destroying mold spores completely enough to render them a negligible
factor in the further changes found in the milk.
EXPERIMENTS WITH THE FLASH PROCESS OF PASTEURIZATION
In working with continuous pasteurizers, temperatures of 165° to 175°
F. (73-9° to 79.5° C.) are reached by heating within a period of approxi-
mately 30 seconds and maintained about 30 seconds. This is followed
by quick cooling. Tower temperatures have not been deemed satisfactory.
A series of experiments was therefore planned to subject the freshly inocu-
lated spores of species of Penicillium, Aspergillus, and of themucors to these
temperatures and to determine their relative ability to survive such heat-
ing. For this purpose glass tubing about 3 mm. in diameter was drawn
into capillary form so that each tube had 3 01 4 inches of the original tub-
'• ^ 5^'^ ;P^ S^- ISJ^ S^ IS'*^. ?^
Pig. I. — Curve of the number of species of molds
surviving pasteurization of milk for 30 minutes
at a series of temperatures.
158
Journal of Agricultural Research
Vol. VI, No. 4
ing with 2 to 4 inches of capillary tube approximately 0.5 mm. in diame-
ter. The open end of each tube was plugged with cotton. The tubes were
packed into a copper case and dry-sterilized. F'or each experiment a few
drops of sterile milk were transferred to the conidial surface of a colony
and the conidia stirred into the milk. A column of milk 15 to 30 mm.
long, bearing numerous conidia, was then drawn into the capillary tube
and the end sealed in the flame. Experiments had shown that alcohol
boiling at 172.4° F. (78° C.) when so treated would boil in 20 to 30 sec-
onds when the tubes were thrust into water at 1 74.4° F. (79. i ° C.) . This
showed that milk containing mold spores could be heated in from 20 to
30 seconds in capillary tubes to any given temperature when immersed
in water 2 degrees Fahrenheit above the desired pasteurizing temperature.
In our experiments, therefore, it was possible to duplicate flash pasteuri-
zation on a laboratory scale; for example, to pasteurize at 165° F.
(73.9O C.) the capillary tubes containing milk and mold spores were held
in water at 167° F. (75° C.) for i minute. During this period about 30
seconds were required to heat the milk and it was held at the pasteurizing
temperature the other half minute. This is approximately the heating
period of milk in commercial flash pasteurization. After heating for the
required time, the tubes were cooled by thrusting them into cold water.
The tip of the capillary was then broken off and the contents streaked
upon slanted Czapek's solution agar. The slants were incubated, observed
occasionally, and the results of the various experiments were tabulated
separately and then brought together in Table II.
Tabi.E II. — Comparative effect of heating mold spores in milk to temperatures of from
145° to 175° F. {62.8° to 79.5° C.)for JO seconds 1
Serial
No.
Growth of spores.
Name of mold.
Not
heated
(control).
Heated
to
145° F-
(62.8°
C).
Not
heated
(check).
Heated
to
'^5°F-
(68.3°
C).
Not
heated
(check).
Heated
to
165° F.
(73-9°
C).
Heated
to
175° F.
(79-5
C).
•a
a
T3
0
"0
0
0.8
.8
.6
.6
. 0
•9
.0
CD
•a
0
I.O
I.O
I.O
I.O
.0
I.O
.0
i
•a
0-3
•4
•4
•S
•S
-3
-3
•3
-3
•3
•5
■5
•9
•9
i
0.7
.8
.8
.8
•9
.8
■7
.6
.8
-7
.8
I.O
I. 0
I. 0
•0
"5
as
a
•a
>o
T3
1
0
•a
>•
Aspergillus candidus.
Asperaillusflavus series
Do
106
108
3538. 108
Rgi36
SC171
118
2705
3512
3555-21
no
III
3534-a
3534-b
3534-c
112
113
0.4
■9
■9
.8
•7
•9
.8
0.7
I. 0
I. 0
I. 0
•9
I.O
I.O
0.0
.0
.0
.0
.0
. 0
. 0
. 0
. 0
.0
.0
?
.0
. 0
. 0
. 0
0. 0
.0
.0
.0
. 0
.0
. 0
. 0
.0
.0
.0
1'
0.6
I.O
0.0
0.0
0.0
0.0
Aspergillus fumigatus
.6
-4
I.O
.8
.0
•5?
.0
I.O?
.0
.0
.0
.0
Do
Do
.8
•3
•9
•9
.8
.8
•9
.8
• 9
1.0
I.O
I.O
I. 0
I. 0
I. 0
I. 0
. 0
•3
.0
.6
.6
-7
. 0
. 0
. 0
I.O
.0
I.O
I.O
•9
. 0
. 0
Aspergillus nidulans
•5
.6
•9
I.O
.0
.0
.0
.0
.0
.0
.0
.0
Aspergillus niger, var. altipes
Aspergillus cinnamomeus . . .
.0
. 0
.0
.0
•5
•7
I.O
I.O
.0
.0
.0
. 0
. 0
. 0
.0
.0
Aspergillus ochraceus
Aspergillus oryzae
•5
.8
. 0
.0
.0
.0
1 I.O, a typical colony; decimals, proportionate growth; o.o. no growth; ?. inharmonious results.
Apr. 24, 1916
Effect of Pasteurization on Mold Spores
159
Table II. — Comparative effect of heating mold spores in milk to temperatures of from
145° to lyf F. {62.8° to y9.S° C.)for jo seconds — Continued
Serial
No.
Growth of spores.
Name of mold.
Not
heated
'control).
Heated
to
145° F-
(62.8°
C).
Not
heated
(check).
Heated
to
(68.3°
C).
Not
heated
[check).
Heated
to
165° F.
(73-9°
C).
Heated
to
175° F.
•0
•0
1
•a
0
-0'
>>
a
0
>.
M
■a
■3
•0
"d
1
■d
-O
d
"d
i
•d
VO
i
•d
1
•d
0.8
•S
.8
•9
1. 0
I.O
I.O
I.O
0.8
?
.0
.0
I.O
•5?
116
Ra42
3522.30
3522.36
3556
3509
3565
3514. CI
3513
3523-6
3560
3Rn.
Syn.
.8
•7
•9
I.O
I.O
1.0
.0
.6
.0
.0
I.O
.0
Aspergillus sp
. 0
.0
•S
.8
.0
.0
.0
• 0
Do . . .
Do
Do
■9
•9
•9
.8
.8
I.O
I.O«
I.O
I.O
.8
. 0
.6
•5
.8
.8
.0
I.O
I.O
I.O
I. 0
•S
•S
•7
•S
■9
•9
I. 0
I.O
■9
I. 0
I.O
I. 0
I.O
.0
?
.0
.0
.0
.0
.0
.0
■ 0
?
.0
.0
.0
.0
.0
. 0
.0
.0
. 0
.6?
.0
.0
.0
.0
.0
.0
.0
•7
.0
.0
.0
. 0
. 0
.0
.0
.0
Aspergillus parasiticus "
Do
•7
I.O
.0
.0
. 0
.0
Circinella umbellata . .
.8
•7
I.O
I.O
.0
.0
.0
.0
.0
.0
.0
Mucor racemosus (group) . . .
Do . .
.0
Do
CO
.8
•9
.8
.8
•9
•3
-9
.8
.8
•7
.8
.8
.6
-9
.8
I.O
I.O
I.O
I.O
I.O
I.O
bact.
I. 0
I.O
I.O
I.O
I.O
I.O
1.0
I.O
I.O
I.O
.0
.8
•9
9.0
• 0
. 0
.0
. 0
. 0
.6
.6
•4
. 0
•9
•4
I.O
.0
I. 0
I.O
I. 0
.0
.0
.0
.0
. 0
.8
.8
I. 0
•5
■9
•7
I.O
I.O
.0
.0
.0
.0
.0
.0
.0
Rhizopus nigricans
.0
Fusariutn sp. . .
•5
•3
•3
■9
-4
•4
•9
-5
I.O
'.'6'
.6
I. 0
.6
.6
I.O
.6
Penicillium atramentosum. . .
Penicillium biforme
38
39
2
5
6
26
IS
23
2543-a
16
34
20
14
■ 3523-4
9
10
II
102
103
I
2683
17
2670
18
46
2546
66
45
27
3551
2552
2643
3028
J5i4-a
28
63
3525.61
,3553
3555- 18
3555- 19
•5
■S
• s
•3
-4
. 0
-9
•7
•5
I.O
I.O
.6
•7
• 9
.0
I. 0
I.O
I.O
.0
.0
.0
. 0
.0
. 0
.0
.6?
.0
. 0
. 0
. 0
.0
.0
.0
.0
I.O
.0
. 0
.0
.0
. 0
.0
.0
.0
.0
. 0
.0
.0
PenicMium brevicaule
Penicillium camembertt
Penicillium camemberii, var.
rogeri
.0
. 0
.0
Penicillium ckrysogenuni . . . .
Penicillium citrinum
Penicillium comm-une
Penicillium cvclopium
.0
.0
.0
.0
Penicillium di'oaricatum
-4
-5
■5
•5
•S
•3
.4.
•3
•5
■4
•4
•3
•3
•4
-3
•5
•4
•3
•3
•3
•3
•4
•3
•4
•4
•5
.6
-9
•9
•9
•9
.6
.8
.8
.8
•9
-9
•7
.6
.8
•9
•9
I.O
•9
■7
.8
.6
.6
.8
•9
.0
.0
.0
. 0
.0
• 0
Very
.0
•4
.0
. 0
.0
.0
.0
. 0
.0
.0
?
.0
.0
.0
.0
.0
.0
.0
.0
. 0
.0
. 0
. 0
.0
. 0
slow.
.0
.8
. 0
.0
.0
.0
.0
.0
.0
.0
•7
.0
.0
.0
.0
.0
.0
.0
.0
.8
•9
.0
.0
.0
.0
.6
•4
•4
.8
.8
I.O
.0
.0
.0
. 0
.0
.0
.2?
.0
.0
.5?
Penicillium. (Ciiromyces) sp.
Penicillium granulaium
•9
-9
•4
.8
.6
.8
•4
.8
•9
I.O
I.O
•9
•9
.7
•9
.8
I. 0
I. 0
. 0
.0
•S
•4
. I
?
•4
. 0
. 0
I. 0
.0
I.O
I.O
•5?
•7?
.8
.0
. 0
.0
.0
Penicilliu m luteiun
•5
•3
•9
•3
•9
■7
. 0
•7
?
.0
.0
. 0
?
.0
.0
.0
. 0
. 0
.0
.0
.0
Penicillium notaium
Penicillium. oxalicum
Penicillium pinophilum
Penicillium puberulum?
Penicillium purpurogenum. .
Penicilliu m purpurogenum ,
.0
.0
.0
.6
• s
•5
•3
.8
•5
•5
■5
.8
•9
I.O
•9
•4
I.O
I.O
.8
.8
I.O
. 0
•3
.0
.0
.0
.0
.0
.0
.0
. 0
•S
. 0
.0
.0
.0
.0
.0
.0
. 0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.6?
Penicillium rotiueforti
Penicillium solituin
.0
• 4
.8
•9
.8
■9
•9
•9
•9
•7
.8
.9
.8
.8
I.O
I.O
I. 0
I.O
I.O
I.O
I.O
I. 0
I. 0
I. 0
I. 0
•4
.0
.6
?
•4?
.0
?
•7
.0
•4?
.6
. 0
?
.8
.0
I.O
.0
.6
.8
.0
.6?
.8
.0
■4
.0
.0
.0
Penicillium spinulosutn
Penicillium variabile
Penicillium viridicatum
Pemcillium ■viridicatum ,var. ?
Do
.0
.0
.0
•5
•7
■5
.4
•5
•7
I.O
I.O
•9
I.O
.0
.8?
.0
.0
.0
.0
•9?
.0
.0
■ 0
. 0
.0
.0
.0
.0
?
.0
Do
.0
Penicillium {Ciiromyces) sp.
Do
.0
.0
Do
.6
•4
■9
■7
.0
.0
.0
.0
■ 0
.0
.0
Do .
.0
Do
.
■9
•4
•7
.6
•9
.0
.0
i6o
Journal of Agricultural Research
Vol. VI, No. 4
From Table II it is seen that very few of the forms are killed in 30
seconds at 145° F. (62.8° C); nearly all, however, are destroyed at 155°
F. (68.3° C). None of the colonies found at 165° F. (73.9° C.) and
175° F. (79.5° C.) were produced in both tubes. The chance of error
is not fully eliminated in these cases. The consistent character of the
whole table and the innocuous character of the few organisms in which
occasional colonies occurred after heating show that temperatures of
165° to 175° F. (73.9° to 79.5° C.) for 30 seconds do practically destroy
the spores of these molds as they may be found in milk, although a few
conidia in some species may occa-
sionally survive.
Figure 2 shows graphically the
effect of the flash process of pasteur-
ization on mold spores.
DESTRUCTION OF MOLD SPORES
BY DRY HEAT
The third series of experiments
was planned to find the relative
ability of the spores of approxi-
mately the same organisms to endure
heating in dry air for the same
period as used for heating in milk.
After some experimentation the fol-
lowing method was used: Strips of
heavy filter paper were cut wide enough so that only the edges would come
into contact with the glass when dropped into test tubes. A drop of sterile
water carrying a suspension of the spores under experiment was depos-
ited in the middle of the paper strip and allowed to evaporate overnight.
The tubes were then immersed in liquid heated to the desired tempera-
ture and held 30 minutes after check tubes carrpng thermometers indi-
cated that the air in the tubes had reached the same degree. The tubes
were then removed and cooled. Melted agar was allowed to run into
each tube to form a slant and the cultures were set away at room tem-
perature. Observations of growth were made as in the previous experi-
ments and the results tabulated in the same manner in Table III.
0^?^
0^^
•o<^
^(t>
^'1
^/^^// /='AST£:ey/?/Z^770A/
Fig. 2. — Curve of the number of species of molds
surviving flash pasteurization at a series of
temperatures.
Apr. 34, 1916
Effect of Pasteurization on Mold Spores
161
•t?.
•^
<>
-§
'?^
U
1
.g
a
0
i
.g
5
a
1
5i
Heated
to 250° F.
(121.1°
C).
•sABp 8
0
6
0 0
0
0
0
0
0
0
0
0
0
0
0
0
0
TfOO
0 0
H
0
0
•SABp 1'
0 0
6 •"*
00000
000
0
0
000 00
0
^A^^^
A
•SABp 8
■*CJO 00
6 • •
0> t^ 0»vO 00
00 t^ t*
■*
0
CO r^ t^ 1^ t^oo 0 0
•0
•sAbp t'
n.. ^
>o t^ t fO -t
•* •* ■*
'^
CO
■* ■<»■ ■* CO •>«■ -^^O CO
•*
•sABp 9
0
0
0
000 0^_^ 00 000 0^ 0
0 CO
»oo 000 000 O^Ot »o*o
■sAEp e
'^ 6
0 t^ 0 N \r,sO^ \r, 0
0
CO CO CO 10 4^ O^ OsO
•sABp 9
6 • •
C> C»00 Ot 0^ Oi C\ 0 <?^C^0^^■
00 00
O^O^O^OlO^O\O00 O»00
•sAbp e
"" 6
'T ":; "^ ": t?': ': " ??
N «
cococoio-U-rtiocON «
S
"2
S
s
V
.g
1
1
1
1
s
1
1
•sABp L
t^ 10
6
00 t^
•sAep t
6
<^
"t
CO
00 to
■SABp i
6
00 0
0 0
0
0 0
0
0
0 00 t~ OV OlOOO
•SABp ^
0""
n 0
0 0
0.°
0 0
0
0
.. t. — ?- •?
"0 1 •
•SABp i
o» o» o\ t^
6 ' ' '
•SABp t'
6
Tf 1000
000
00 t~
00 ■*
t^
CO
■O rf^ 00 CO ■
r^ •
•sABp 8
w
6
0 •<l-0 NOO-*0OK-i'*^ot
0 0 "1
0
't
0 ^ «o -Jj-oo
•SABp t'
« 0 0 0
0 0 t
^
^
to ^ CO •<1-
•sABp 8
6
00
•sAep fr
6
Tf-flON TffON t rr,^\nf m
?" «
•^
CO
CO CO CO « >o
■sABp L
0
6
?
0 0
0
0 to
0
•SABp E
•sABp i
-noo) p^^^
(lOii
aq ^o^
6
00
00 CO
00
t^ 10
"?
•sAbp S (•;
6
»t>oN 0 tn'<l-«'0
U1
0 tntn
•sABp S
-noo) pa^?
(10J:>
sq lojst
10 t- </>!» >"oo >o CO >o >^ t^oo t^co t^oo -a- 1^« 00 10 0.
6
\0 t^O »o t^
d
•c
vOcoooo MOO mn u-io '-' ■>t^'«r« fO
00 ^
0 « M <NO C^WI-^fOfOO
WTfNM iooOmh wo
2
"o
S
"5
i
■c
'c
•c
"o
(11
a
a
'A
>
s
» c
C
s
/v.
k
i
>
S
j
.■c
■5:
1
J
k
1
s
k
k
k
5
j
P
c
p
C
P
c
S
e
" !
■* k
c
c
c
P
a
•0
ft:
c
j
^<
V.
S
;:•
I/-
1 :
d
S
3
l62
Journal of Agricultural Research
Vol. VI. No. 4
3
a
a
o
u.
o
1
3
-M
a!
a
S
■M
o
-^J
•d
S
t*
a
■>
CS
i-.
V
■u
-3
5
f
o
•g
o
a
1
1
"o
o
O
Heated
to 250° F.
(121.1°
C).
•sAep 8
0
0
0
0
0
■sAcp t'
0
0
0
0
0
•sAep 8
o\
0
r*
00
00
•SABp t'
'?
•*
T
•*
00
Do 0 .
W 0^
•sABp 9
0
6
0
M
rv.
0
0
n..
0
•sAep e
0
-*
^.
i^
0
n-.
-0
•0 , .
•sABp 9
00
6
o>
0
0 0\
0
o>
0
•sAbp C
^
?
'^
« «
?
'^
CO
•dfeo
•sAep i
0
•sABp t'
0
tM 0
•sABp I
0
6
0
0
to t^
00 00
T
0
•SABp >
0
6
■St
0
n- ■*
Tf ■*
t
0
•d 1 .
•SABp i 1
•SABp t-
6
00
CO
00 in
00 00
t^
7
•SABP 8 ^°°°°'^°°'^
^O N
0 0
Tj- 0 ''J- N
0
0 0 0 0 CO 0 fO
•sAbp t
0000 0^ 0 0^
6
ri-n-
0 0
fv. rv. A- f»-
0
0000 0
lilt
■sABp 8
•SABp t'
•>j->o t in t^ r~ io>o
6
to to
f?
fC^ t^ «
?
^ PO »o\o to t^ 10
•a too
^c °cO^
■SABp i
c5
0 0 tH
0 0
m n 0
fO **
"O ^ 0. 0 -^ 0
0 \
•sAep e
to
6
°°-.
0 0
rv-rv- 0
fv-f^
M •* 0^ 0
0 \
-noo) p3:>-caq ^o^si
6
<^00 00
a\ r^
t^ 0\ (>
fO t*
00 t^ \0 \0 00 CO
00 1
•sAepS (-D^s-zs;
6
0 fO rooO 0 ^
•0 m
VO
'^\0 t- Tj-
to
•^ CO T^
to
f
•sAtp S (joj^
-no3) poiEcq 50^
6
00 r~o 000 OD 00
r^^o
0, t-
u^OO 00 VI
t*
l> 10 !>.
?
»r
1
T
00 r^ 0 w-nO vO »n ro rosO ** O ■'J- tO 0\
mOO com t^MTt- '^O -"t « 10
M M so 0 to \r
i
i
i
1
1
1
1
'i
1
1
s
i
1
1
g
.;
a
s
■1
a:
s
S
a
■c
s
1
-«
s
a;
s
s
s
c
S
.s
a
1
c
1
it
0,
1
1
1
s
s
i
ii
a!
1
s
.;
'?
a!
s
:"t
s
a
1
s
a;
s
!
s
•0
s
S
.;
a!
s
a
>
g
1
s
,«,
■£.
II
II
a,
•5
0
.;
a!
s
§
•1
s
a!
g
s
a;
i
1
S
.;
1^
■1
■S
g
.:
■;
1
a,
S
s
.?
i
a!
1
»^
a
s
•1
1
Apr. 24, 1916 Effect of Pasteurization on Mold Spores
163
0
0
00
-r
0 0
0 0
Oi 0
M fO
vO
00
"C'
0
^
»-)
•*
•^
00
00
>o
00
"^ 0 0 0 t 0 0
nor)
000 00
0
wo tovo w n- 1-
-0 mco
« 0 "^ ^ 0 0 -
?
^ 0 ^ >/-. 0 0 ^
^
00 t- Ov
« 0.00
CO
* « « m
" T
TT
^00 00 0^
>o 0\
0 00
" H 00 0\
M r^.cc TfOOO t*^ \n r*l Vi Vi
»no 0 »/^ 0 irt to »/i vo
a a m f. ^ m n "i rn
\
?
■J
■1
1
Q
i
- c
c
: i
c
i
: c
if;
ft
'p
' c
J 64
Journal of Agricultural Research
Vol. VT. No. 4
A study of Table III shows that mold spores possess much greater
ability to withstand dry heat than heating in milk. Very few forms
were destroyed at i8o° F. (82.2° C), but they include Penicillium
hrevicaule, which has a thick-walled spore and in laboratory cultures
has remained viable at least 7 years. Only a few species of Penicillium
survived heating to 200° F. (93.3° C.) for 30 minutes. All these are
forms which grew at 98.6° F. (37° C), and some of them are widely
distributed.
Aside from A. wentii, all the species of Aspergillus survived heating
at 200^ F. (93.3° C). Several of them survived at 230° F. (110° C),
but after 250° F. (121.1° C.) for 30 minutes no species showed growth
after 6 days' incubation. Three of six mucors, however, survived the
heating to 250° F. (121.1° C.) for 30 minutes. These species were killed
quickly by both forms of heating
in milk. The results of these
experiinents are plotted in
figure 3.
The destruction of mold spores
by dry heat has no relation to the
subject of pasteurization of milk,
but it is of scientific interest.
DISCUSSION OF RESULTS
These results with mold spores
agree in general with bacterio-
logical studies of pasteurization.
Fig. 3.-Curve of the number of spedes of molds sur- Very few of thcSC Orgauisms
viving dry heat for 30 minutes at a series of temper- found in milk SUrvive af tCf 30
minutes' heating to 145° F. (62.8°
C). Certain molds, notably Aspergillus fumigatus and A. flavus, do
survive, but they occur only occasionally in milk. Oidium lactis and the
mucors, which are probably more important as milk-borne organisms
than all the rest, are destroyed at the low temperatures used in the holder
process of pasteurization. In the flash process very few mold spores
survived at 165° F. (73.9° C). Occasionally some spores seem to have
escaped destruction at 175° F. (79.5° C), but the organisms surviving
in these cases were of minor importance in the decomposition of dairy
products. In confirmation of these results the writers have had access
to unpublished data of Mr. R. O. Webster, of the Bureau of Chemistry,
giving cultural analysis of butter made from flash-pasteurized cream on
a commercial basis. Cultures from this butter showed no mold spores,
while cultures made at the same time from country butter showed 20,000
to 60,000 per gram.
Apr. 24. i9i6 Effect of Pasteurization on Mold Spores 165
Mold spores in milk seem, therefore, to be destroyed completely or
reduced to negligible numbers by both of the standard pasteurization
processes.
Careful study of the cultures showed that the first efifect of heating was
to delay germination. This is indicated in the tables by the reports of
successive examinations of the same culture. In Table I three reports are
given; later only two reports. The third and fourth observations, how-
ever, were usually made. At times heating to a degree just under the
death point delayed germination almost the full length of the usual
growing period of the species. The number of possible sources of error
was so great that the results of observations have been tabulated and
compared. When essential harmony of results was not obtained, the
work was repeated. In a few cases the continued lack of consistent
results for particular organisms is indicated by the interrogation point in
the tables. Even with these precautions the data obtained can be said
to apply only to the strains used. This is indicated by comparing the
results given for the Aspergillus flavus group or for the four members of
the A. niger group. These results do not prove that other strains of
these groups would respond exactly as here tabulated. In fact, more
extended studies (as yet unpublished) of these two groups indicate that
organisms otherwise undistinguishable may differ greatly if we measure a
single physiological reaction. Such quantitative differences may persist in
continued cultures,, but are hardly comparable to differences in the kind
of reaction as a basis for separating species. Inside the race or strain,
conidia transferred from the same culture respond very differently.
There is frequently a survival of a few spores where a majority of the
spores die. There may be, therefore, a difference of as much as 20° F.
(11. 1 C.) between the temperature at which an occasional culture is
completely killed and that at which cultures of that species are uniformly
killed. These results resemble those obtained in determining the thermal
death point of bacteria.
The applicability of these results to the occurrence of mold spores in
substances other than milk has not been tested. The variation in com-
position of the substratum together with the heating may at times
introduce a considerable variation. In general, however, it is clear that
mold spores are easily killed by heat when suspended in fluid. The
tables have been studied in an attempt to correlate resistance with size
of spore or thickness of spore wall. No such correlation has been found.
There is, therefore, no suggestion as to the nature of the difference in
these organisms which affects their resistance to heat.
SUMMARY
(i) The holder process of pasteurization, in which milk was heated to
145° F. (62.8° C.) and maintained at that temperature for 30 minutes,
killed the conidia of every species investigated, except those of Asper-
1 66 Journal of Agricultural Research voi. vi.no. 4
gillus repens, A. flavus, and A. Jumigatus. The molds which sundve are
found only occasionally in milk.
(2) The flash process of pasteurization, where milk was heated to 165°
F. (73.9° C.) for a period of 30 seconds, destroyed the spores of all the
molds tested with the exception of many spores of one form and occa-
sional spores of three more forms. At 175° F, (79.5° C.) only occasional
spores of two forms developed.
(3) When the heating process was performed in dry air for a period
of 30 seconds at 200° F. (93.3° C), 31 out of 42 forms of Penicillium and
7 out of 24 forms of Aspergillus were destroyed, but none of the cultures
of the mucors. A temperature of 250° F. (121.1° C.) over a period of
30 minutes killed all the forms of Penicillium spp. tried, but left an
occasional living spore in one species of Aspergillus and three out of six
mucors.
EFFECT OF WATER IN THE RATION ON THE
COMPOSITION OF MILK
By W. F. Turner, R. H. Shaw, R. P. Norton, and P. A. Wright, of the Dairy
Division, Bureau of Animal Industry
INTRODUCTION
Experiments conducted at Brownsville, Tex., by the Dairy Division
of the Bureau of Animal Industry indicate that the feeding of prickly-
pear (Opuntia spp.) lowers the percentage of fat in milk. In comparison
with other feeds prickly-pear contains a large amount of water and
mineral matter. It was thought by the writers that one or both of these
constituents might be responsible for the reduction in fat percentage;
consequently experiments were conducted at Beltsville, Md., to deter-
mine the influence of the water. Work with the mineral matter is now
in progress.
The literature dealing with the effects of watery feeds or water in the
ration upon the quantity and the quality of milk produced contains
many conflicting statements. No doubt the difficulty of eliminating all
factors except the watery character of the ration is largely responsible
for the conflicting nature of these statements.
Gilchrist (i)^ reports very little difference, if any, in quantity and
quality between the milk produced by cows either on pasture only or
on a daily ration of mangels in varying amounts up to 86 pounds per
cow and that produced by the same cows on a ration of hay and grain.
Tangl and Zaitschek (12) state, as the result of extensive experiments
to determine the influence of watery feeds on milk secretion, that there
is no difference between the composition of the milk from cows fed on a
watery ration and that from cows fed on a dry one. They state that it is
not true that watery feeds cause the production of thinner milk than
dry feeds.
Lauder and Fagan (10, p. 9) reached the following conclusions from
experiments extending over a 3-year period, using 60 cows and feeding a
large ration of turnips {Brassica rapa) to compare with a dry or concen-
trated ration :
The feeding of a ration containing a large quantity of water does not increase the
percentage of water in the milk or reduce the percentage of fat.
The greater yield of milk was obtained from the cows on the concentrated ration.
On the other hand, the milk from the cows on the turnip ration contained a higher
percentage of fat, and a greater total weight of fat was secreted in the milk.
' Reference is made by number to " Literature cited," p. 177-178.
Journal of Agricultural Research, Vol. VI. No. 4
Dept. of Agriculture, Washington, D. C. Apr. 24, 1916
di A— 21
(167)
1 68 Journal of Agricultural Research voi. vi. No. 4
Holtsmark (6) reports that there is no decrease in the fat content of
the milk of cows on a liberal daily ration of concentrated feed and cut
straw, with as much as jj pounds of turnips per head, after this ration is
substituted for one consisting of hay, straw, concentrates, and a small
quantity of roots.
A writer in the Journal of the Board of Agriculture (3), JUondon, Eng-
land, concludes from a study of the work of various investigators that,
although many feeds have a specific effect on the yield and quality of milk,
it may be attributed to stimulating substances in the feeds rather than
to water content. These substances have a physiological rather than a
nutritive effect and are present in feeds in small quantities only.
As the result of a number of experiments conducted and a review of
previous work of the same character, Jordan (8, p. 69) states that, "Con-
trary to a notion held by many, it is not possible to water a cow's milk
through her drink or through the ingesting of watery feed."
The Journal of the Board of Agriculture, London (2), reports that a
dairyman was convicted in the French courts for selling adulterated milk.
The conviction was based upon the assumption that it is possible to
water milk either by feeding cows on watery feeds, by causing them to
drink water in large quantities, or by making them drink immediately
before milking. To prove the fallacy of this assumption, the Board con-
ducted experiments with a number of cows. After feeding them an
excess of common salt (sodium chlorid), or limiting the water drunk
after free access to it, or permitting them to drink only immediately
before milking, it was found that no change is produced in the composi-
tion of the milk.
At Offerton Hall, Durham, England, a series of experiments was con-
ducted to determine how the composition of milk is affected by feeding
wet brewers' grains. The first of these experiments (7, p. 35) indicates
that the feeding of these grains to cows whose milk is habitually low in
butter fat is not to be recommended, especially during the earlier stages
of the lactation period, when the grains tend slightly to reduce the yield
of fat. The writer advises dairymen to use such grains sparingly. Later
experiments (13, p. 19-20) indicate that the grains may be fed safely if the
ration contains other feeds also, and that there is no appreciable lowering
of the butter fat when the grains are fed in moderate quantities.
In a general article upon the effect of difterent feeds upon the quality
of milk, McConnell (11) says:
It is a matter of common knowledge that the lush grass of spring, an excess of man-
golds, or too many brewers' grains will promote a great flow of milk, but that that milk
will be poor, and farmers who do not do anything to modify such feeding will find their
milk coming dangerously near the "standard."
Hansson (4), of the Stockholm Agricultural Experiment Station, in a
review of the work of various investigators concerning the effect of dif-
ferent feeds upon the fat content of milk, concludes that there are on
Apr. 24, 1916 Effect of Water in Ration on Milk 169
this point distinct differences among different feeds, but that the effect
of any feed depends upon the composition of the other components of
the ration. He states that roots have a favorable effect upon milk
secretion, but tend slightly to lower the fat content.
Koch (9) reports extensive feeding experiments at Rosenhof in which
cows were fed beet roots (Beta vulgaris), and gives the following conclu-
sions :
An increase in fat units (total fat) with beet-root feed, an increase of the amount of
milk combined with a decrease in the fat content. However, the increase in quantity
exceeded the decrease in quality so much that the cows gave 6 per cent more total fat
on the beet-root ration.'
PLAN OF INVESTIGATION
The experimental work to determine the effect of water in the ration
upon the composition of milk was conducted at the Dairy Division farm,
Beltsville, Md., and included parts of three different lactation periods.
The four following methods for supplying rations of widely different
water content were tried:
1. A full allowance of drinking water as compared with a limited
supply, the ration otherwise being alike in both cases.
2. A heavy ration of turnips as compared with a dry-roughage one.
3. Wet beet pulp as compared with dry beet pulp.
4. Green crimson clover ^Trifolium iiricarnaium) as compared with the
cured hay.
As the change in the fat content of the milk noted during the prickly-
pear experiments took place within a few days after the change in the
character of the ration and continued throughout the 80-day period, it
was decided that for this work two lo-day periods of feeding any one
ration, with a lo-day transition period intervening, and equal periods of
feeding the comparative ration, would give time enough for any change
in the composition of the milk to take place. In each series of experi-
ments the milk from each cow was weighed at each milking, and lo-day
composite samples were taken for analysis. The data obtained from
each series of experiments are given separately.
FULL VERSUS LIMITED ALLOWANCE OF WATER
In this series of experiments eight cows were used and all received the
same general treatment. For the first two lo-day periods the animals
were given water ad libitum twice daily. Then a definite quantity of
water, not more than 75 per cent of the full allowance, and in some cases
less than 65 per cent, was given for two lo-day periods following a lo-day
transition period. The quantity of water given in the limited water ration
was so reduced that, when watered once a day, all cows drank the quantity
allowed. After a second lo-day transition period, a full allowance of
water was again given for two lo-day periods. This completed the work
1 Authors' translation.
I70
Journal of Agricultural Research
Vol. VI. No. 4
with all but two cows, which were given a still more reduced allowance
of water following the second full-allowance period. Table I gives the
results for each cow.
Table I. — Comparison of the effect of a full and a limited allowance of water on the
composition of milk
cow loo
Water allowance.
Full
Do.
Transition. .
Limited . . . .
Do
Transition. .
Full
Do
Transition. .
Limited . . . .
Do
Average :
Full....
Limited
Total
milk.
Pouiids.
2 20. 6
240. 6
205-3
198.8
199-3
197.6
172. 2
167. o
149.8
135-8
138. o
200. I
168.0
Total
water.
Pounds,
412. 5
502,
340
340
340
434
378.0
358.0
200. o
205.0
215.0
Fat.
Per cent.
412. 5
275.0
4-65
4.80
Pounds.
9-93
10.83
9.44
9-54
9-37
8.60
8.44
7-85
7- 19
5-65
6.62
9. 26
7-79
Specific
gravity.
1-033
1.032
1-033
1.034
1-033
1.032
1.032
I-OJ3
1-033
1.032
1.032
Solids
not fat.
Per cent.
9. II
9. 19
9-27
9-36
9. 18
9-
9-
9-
9-
9-
9-
9. 12
9.24
Mois-
ture.
Per cent.
86.39
86. 23
85.96
Ash.
Per cent.
o. 720
. 710
. 720
. 710
•70s
. 700
•745
•747
•755
•770
•750
730
734
Total
prote-in.
Percent.
3- 35
3^35
3-3(>
3-52
3^53
3^41
3.68
3.66
3.80
3^72
3.61
3^51
3-59
Full
Do
Transition. .
Limited . . . .
Do
Transition. .
Full
Do
Transition. .
Limited . . . .
Do
Average:
Full....
Limited
iqi.
8
188.
0
179.
0
175-
2
179.
0
181.
6
184.
5
180.
4
172.
9
157-
8
151-
5
186
2
165
9
500.5
502.0
320. o
320.0
320. o
496. o
462. o
473-0
250.0
255-0
280. o
484. o
294. o
6. 00
II. 51
1.036
9-98
84. 02
6. 00
10.88
1-035
10. 14
83.86
6.25
II. 19
1.036
10. 18
83-57
6. 10
10. 69
1-035
10. II
83-79
6.03
10.79
1-035
10. 21
83.76
5-50
9-99
1.034
9.91
84-59
5- 73
10-57
1-035
10. 00
84.27
6. 00
10.82
1-035
9.81
84.19
5-95
10. 29
^•035
9-95
84-07
6. 00
9-47
1.034
10. 10
83.90
5- 80
8-79
1-033
9.82
84.38
5-93
10.94
9.98
84.08
5-98
9-94
10. 06
83.96
770
770
750
740
740
730
720
740
755
750
730
750
740
3-94
3.88
cow 19
Full
Do
Transition ....
Limited
Do
Transition ....
Full
220. 5
228.6
213.0
213. I
202.5
203.3
198.5
193.2
520. 0
492.0
300.0
345- 0
305-0
622. 0
520.0
520.0
5-30
5-20
5.18
5-42
5-33
5.60
5^50
5^25
II. 69
11.89
11.03
11-55
10.79
11.38
10. 92
10. 14
1.036
1-035
1.036
1-035
1-035
1.034
1.034
1.034
9-83
9.82
10. 09
9.66
9.98
9.66
9-74
9.71
84.87
84.98
84.73
84.92
84.69
84. 74
84.76
85.04
770
745
735
750
780
765
775
76s
3-73
3.86
3^73
3^77
3.82
3-83
3-88
Do
3-74
Average :
Full
Limited . .
205. 2
207.8
513-0
325-0
5^31
5-37
II. 16
II. 17
9-77
9.82
84.91
84.80
.764
.765
3.80
3-79
Apr. 24, 1916
Effect of Water in Ration on Milk
171
Table I. — Comparison of the effect of a full and a limited allowance of water on the
composition of milk — Continued
cow 8
Water allowance.
Full
Do
Transition. .
Limited . . . .
Do
Transition. .
Full
Do
Average :
Full....
Limited
Total
milk.
Pounds.
252. 2
253-0
234- 9
220. 7
208. o
217. 6
205.7
209. 4
230. I
214.3
Total
water.
Pounds.
573-0
553- o
346.0
350
350
509
500,
536
540.
350.
Fat.
Per cent.
4-55
15
20
30
65
30
30
50
4-37
4-47
10. 06
9-58
Specific
gravity.
1-033
1-033
1.034
1.034
1.034
1-033
1.034
1-033
Solids
not fat.
Per cent.
9-30
9. 10
9-39
9-44
9-35
8.98
9-25
9. 18
9. 21
9-39
Mois-
ture.
Per cent.
86.15
86.88
86.13
Ash.
Total
protein.
Per cent.
0.730
•723
•707
•731
.714
.727
.724
•732
727
722
Per cent
3-17
3-13
3. 12
cow 17
Full
194.9
206.8
173-5
188.2
174.0
184.7
164.9
156.4
465-0
437-0
267. 0
310.0
300.0
443-0
499.0
504-0
5-30
4-95
5-28
5-15
5.20
5- 18
5.20
5.18
10.33
10. 24
9. 02
9.69
9-05
9-57
8-57
8. 10
1.034
1-035
1-035
1.034
1-035
1-034
1-033
1-033
9-73
9.88
9-73
9-53
9.92
9-44
9.69
9-70
84.97
85-17
84.99
85-32
84.88
85-38
85. II
85.12
74
72
715
720
755
740
770
755
3^70
3^77
3.62
3-68
3.68
3.81
3.86
3^76
Do
Transition ....
Limited
Do
Transition ....
Full
Do
Average :
Full
Limited. .
180. 7
181. I
476.0
305-0
5-16
5-17
9-31
9-37
9-75
9-72
85.09
85.10
•744
•737
3^77
3.68
cow 9
Full
Do
Transition . .
Limited ....
Do
Transition. .
Full
Do
Average :
Full
Limited .
182. 2
153-2
199.7
410. 0
193-7
432.0
181. 0
386.0
163.5
300.0
142.9
300.0
165.7
526.0
170.4
556-0
164.9
541.0
485.0
300.0
4.40
4-15
4. 20
4-05
4-15
4. 10
4-15
4-30
8.79
8.04
7. 60
6.62
5-93
6-79
7.07
7-09
7-75
5-77
031
030
031
032
031
030
031
031
?.83
i. 40
^•53
?. 79
3.64
?. 27
i-53
?-65
60
71
-77
-45
.27
. 16
. 21
-63
-32
-05
87-15
87. 18
744
709
711
724
703
698
712
704
717
713
2.78
2.65
2. 76
2. 72
2-54
2. 69
2-73
2.84
2-75
2.63
cow 14
Full
Do
Transition . .
Limited . . . .
Do
Transition . .
Full
Do
Average :
Full....
Limited
27472°— 16-
269. I
429.0
5.10
13-72
1-033
9.17
85-73
723
263.9
470.0
4.80
12. 67
1-032
9. II
86.09
723
236.5
338.0
5-40
12.77
1-033
9. 10
85^50
754
232.9
305-0
5-00
II. 64
1-033
8.97
86.03
756
222. 6
300.0
5.10
11.36
1-033
8.91
85^99
739
235-2
566.0
4.90
II. 52
1.032
8.84
86.26
727
227. 6
494.0
5-05
II. 50
1-032
9. 06
85.89
737
211. I
484. 0
4.70
9.92
1-031
8.99
86.31
724
242.9
469. 0
4.91
11-95
9-09
86.00
727
227.7
302.0
5-05
II. 50
8.94
86.51
747
3^04
3- 10
3.18
3.22
3-25
3. 16
3.21
3.06
3- 10
3-23
172
Journal of Agricultural Research
Vol. VI, No. 4
Table I. — Comparison of the effect of a full and a limited allowance of water on the
composition of milk — Continued
cow 2
Water allowance.
Total
milk.
Total
water.
Fat.
Specific
gravity.
Solids
not fat.
Mois-
ture.
Ash.
Total
protein.
Full
Pounds.
238.6
236-5
201. 6
181. 9
188.5
172.7
189.9
202. 9
Pounds.
517-0
595-0
388.0
350-0
350.0
589.0
630.0
639.0
Per cent.
5- 60
4-85
5. 60
5-50
5-55
5-30
4-95
4-70
Pounds.
13-37
11.47
II. 29
10. 00
10.47
9-15
9.40
9-54
1-033
1.034
1-035
1-035
1-035
1.032
1.034
1-033
Per cent.
9-55
9.68
9.92
9-57
9-57
9-32
9-54
9. 62
Per cent.
84-85
85-47
84.48
84-93
84.88
85-38
85-51
85.68
Per cent.
0. 72A
Per cent
^. 50
Do.
Transition ....
Limited
Do.
Transition ....
Full
737
775
769
746
738
754
747
3-67
3-85
3.82
3-77
3-48
3-56
Do
3-89
Average:
Full
Limited. .
217. 0
185. 2
595-0
350-0
5. 02
5-52
10.94
10.23
9. 60
9-57
85-38
84. 90
.740
•757
3-65
3-79
In studying the data obtained in these trials it will be noted that all
the milk constituents except the fat show very little variation during the
different periods, and that these differences are attributable more to the
individual animals than to the character of the ration. Taking the
average figures for the two classes of rations, it will be seen that the full
water allowance ration tended to increase the quantity of milk produced
and to cause a slight reduction of the fat content of the milk. A study
of the data for individual cows by separate periods, however, will show
that this average effect of the different rations is caused more by the
order in which the rations are fed than by their character. Of the data
obtained from the eight cows used in this test those from only one (No. 2)
show indication of any effect of the ration upon the composition of the
milk, and the data from the seven other cows are so negative that this
variation is probably caused more by the individual than by the ration.
Two of the cows, Nos. 17 and 19, show practically no variation in either
quantity or quality of the milk produced; one other, No. 100, decreased
gradually in the quantity of milk produced and increased gradually in
quality, regardless of the ration; while the remaining four, Nos. 8, 9, 14,
and 2 1 , gave milk the fat content of whifch varied considerably from nor-
mal in different periods, even on the same ration. These variations were
independent of the character of the ration — that is, the abnormal per-
centage of fat was in some cases found when the full allowance of water
was given and in other cases when the quantity was reduced. A sum-
ming up of all the data obtained shows that the feeding of rations whose
water content is varied by controlling the quantity of water drunk has
no influence upon the composition of the milk produced.
Apr. 24, igi6
Effect of Water in Ration on Milk
173
TURNIPS VERSUS DRY-ROUGHAGE RATION
In this series of experiments four cows were used, the experimental
period consisting of six test periods and two transition periods.
Figure i shows the grouping of the cows and the character of the ration
fed during each period.
As much as 90 pounds of turnips a day was fed to the cows on the wet-
roughage ration, with the addition of 4 pounds of clover hay. The
roughage ration of the dry-roughage group consisted entirely of clover
hay. The grain ration was the same for both groups. In Table II
COI^^ A/p
/^£r£'£>
/^£r£ro
TXi4MS/T/OA/
/«2£Z29
23^A^£>2^
2S^M> 2P
7Z//?A//A>S
TC//?A//^iS
^^K^£^^^d!^.
Oj^l^KVA^^^
Fig. I.— Grouping of cows and kind of ration fed cows 23, 24, 2s, and 27
both the quantity of water drunk and the total water content of the tur-
nips are given, turnips being considered as having 90 per cent of water,
as shown by Henry and Morrison (5, p. 645).
Table II. — CoTnparison of the effect of turnips and a dry-roughage ration on the com-
position of milk
cow 23
Ration.
Wet
Do...
Transition
Dry
Do...
Transition
Wet
Do...
Average :
Wet. .
Dry..
Wet
Do...
Transition
Dry
Do...
Transition
Wet
Do...
Average :
Wet. .
Dry..
Total
milk.
Lb.
234.7
236.9
225.2
212. 4
214. 6
204. 2
198.6
192.9
215.8
213-5
Water
in ra-
tion.
Lb.
94
123
446
712
734
190
62
92
723
Tur-
nips.
Lb.
774
810
261
630
810
810
801
Fat.
'. ct.
. 10
•30
•03
. 00
■03
.90
• 13
•05
4. 14
4. 01
Lb.
9. 62
10. 19
9.08
8. 50
8.65
7.96
8.20
7.81
8-95
8-57
Specific
gravity.
1.032
1.030
1-031
1.030
1.030
I. 030
1.030
I. 031
Solids
not fat.
P.ct.
64
50
52
44
40
55
47
69
57
42
Mois-
ture.
P.ct.
87.26
87. 20
87-45
87.56
87-57
87-55
87.40
87.26
87. 28
87.56
Ash.
P.ct.
0.750
. 720
•735
•715
. 700
•725
•725
• 740
732
707
Total
pro-
tein.
P.ct.
3- 18
3-09
3-17
3-09
3- 13
3- II
3-24
3.28
3.20
3-"
cow 24
255-1
251-3
234- 5
226. 4
226. 7
234.0
237.2
227. I
242. 7
226. 5
72
389
607
658
119
56
144
774
810
261
630
810
810
632
8or
4. 10
4. 10
4-3°
3.80
4-
3-
4-
4-
00
4. 10
3-90
035
035
035
034
033
033
033
034
9.96
8.83
9-
9-
9-
9-
9-
9-
9-31
9-51
9-53
9. 26
86.37
86.83
710
690
690
645
635
695
680
715
699
640
3-64
3-52
3-42
174
Journal of Agricultural Research
Vol. VI, No. 4
Table II. — Comparison of the effect of turnips and a dry-roughage ration on the com-
position of milk — Continued
cow 25
Dry
Do...
Transition
Wet
Do. . .
Transition
Dry
Do. . .
Average :
Dry..
Wet. .
Total
milk.
Lb.
199.8
195-5
218. 2
203.0
198. O
177-3
175-3
160. 4
182.7
200. o
Water
in ra-
tion.
Lb.
604
541
158
304
505
532
Tur-
nips.
Lb.
85
810
810
180
810
Fat.
P.d.
4-23
4-39
3-99
Lb.
8.45
8.41
9-27
8.08
7.92
7-50
7-89
7-3°
8.01
8.00
Specific
gravity.
-033
-033
.032
.032
-033
.032
.032
.031
Solids
not fat.
P.ct.
9. II
07
8.87
9. 01
9. 06
Mois-
ture.
P.ct.
86.66
86.63
86.86
86.99
86.90
86.60
86.69
86.58
86.64
86.94
Ash.
P.ct.
•750
•730
.745
•715
•725
•755
•715
•730
731
720
Total
pro-
tein.
P.ct.
3-14
37
cow 27
Dry
Do...
Transition
Wet
Do...
Transition
Dry
Do...
Average :
Dry..
Wet. ,
223. 2
207.7
223.3
237.8
240. o
214. o
199.0
175-4
201.3
238.9
591
578
132
407
548
576
585
810
810
180
810
■30
. 10
. 20
. 00
.04
•03
■05
-30
4. 19
4. 02
8.43
9. 60
1-033
t-033
1-032
1-033
1-033
[.034
1.032
[.031
9. 01
8.91
86.80
87.06
•730
. 710
•735
•695
•765
•745
•750
•765
741
730
3- 12
3-14
2-99
3-29
3-24
3-28
3-17
3-14
In this series of experiments the data show conflicting results. All
the cows gave more milk when fed the turnip ration, and they also ate
that ration much more readily than they did the entire dry-roughage
one. The two cows that were fed the ration in the order wet-dry-wet
gave milk of a higher fat content on the wet ration, while those fed
in the dry-wet-dry order gave the higher percentage of fat when the
dry ration alone was fed. None of the other constituents of the milk
were appreciably affected, and in the case of the fat content the data
are so conflicting that they seem to have been caused by some factor
other than the ration.
DRY VERSUS WET BEET PULP
Two cows were used in this trial, one being fed wet, dry, and wet beet
pulp in successive periods, with a transition period after each change in
ration, and the ration of the second cow being just the reverse. While
being fed dry beet pulp each cow received 10 pounds daily. The wet
ration consisted of 40 pounds of the wet beet pulp, or 10 pounds of the
Apr. 24, 1916
Effect of Water in Ration on Milk
175
dry, with 30 pounds of water added, the pulp used having been found to
absorb three times its weight of water. In all conditions except as to
the pulp the two rations were alike for each cow in the different periods.
In Table III the quantity of water in the beet pulp, as well as the quan-
tity of water drunk, is given:
Table III. — Comparison of the effect of dry beet pulp and wet beet pulp on the compo-
sition of milk
cow 22
Ration.
Total
nailk.
Water
in
ration.
Pulp.
Fat.
Specific
gravity.
Solids
not fat.
Moisture.
Ash.
Total
pro-
tein.
Dry
Lb.
209. 6
201.8
199-3
189.5
185.4
180. 9
167.7
Lb.
590
540
273
306
487
479
472
Lb.
300
300
49
Per ct.
4.80
4-85
4- 65
4- 65
4.80
4.80
4. 90
Lb.
10. 06
9-79
9.27
8.81
8. 90
8.68
8.22
1.034
1.036
1-035
1-035
I- 035
1.036
1-035
Per ct.
9-93
10. 08
10. 00
10. 15
9-85
9.76
9-59
Per ct.
85.27
85.07
85-35
85. 20
85-35
85-44
85-51
Per ct.
0. 760
Per ct.
3-65
3-70
3-89
3-88
3-92
3.88
3.82
Do
797
789
796
795
797
790
Wet
Do
Transition
Dry
Do
Average :
Dry
Wet
190. 0
194.4
520
290
300
4.84
4-65
9.19
9.04
9.84
10. 07
85-32
85.27
.788
•792
3-76
3-88
cow 18
Wet
Do. . .
Dry
Do. . .
Transition
Wet
Do. . .
Average :
Wet..
Dry. .
193-7
185.0
196. 6
176.9
183.2
166. 9
159.6
176.3
186.7
340
369
472
511
348
327
383
355
492
300
300
228
300
300
300
5.10
5.20
5. 00
5- 40
5.20
5.60
5.60
5-37
5.20
9.88
Q. 62
9-83
9-55
9-53
9-35
8.94
9-45
9.69
1.032
1-033
1.034
1-033
1.034
1-035
1.034
9-23
9- 19
9. 60
9.48
9-45
9.40
9-30
9.28
9-54
85-34
85.27
740
747
730
733
760
766
754
752
731
The data from these two cows give negative results so far as the efifect
of the water in the ration upon the composition of the milk is concerned.
One cow, No. 22, gave milk slightly lower in fat content when the wet
beet pulp was fed; but the other gave opposite results, the milk testing
higher than that produced when the preceding dry ration was fed. The
quantity of milk produced by both cows decreased at a normal rate.
GREEN VERSUS CURED CRIMSON CLOVER
In this series of experiments four cows were used. For a period of 10
days they were each fed all the fresh-cut green crimson clover that they
would consume, and composite samples were taken during the period.
176
Journal of Agricultural Research
Vol. VI, No. 4
Later, when the clover had been harvested and had become well cured,
the same four cows were fed all the cured product that they would con-
sume, and composite samples again taken. No weights of water drunk
were taken, but as the green clover contained 71.23 per cent of water
and the cured hay but 8.33 per cent, there was an appreciable difference
in the quantity of water in the rations of the two test periods. Table IV
gives the results for each cow. The figures in parentheses following the
class of ration show the total number of pounds of the cured or green
clover fed.
Table IV. — Comparison of the effect of green and cured crimson clover on the composi-
tion of m,ilk
cow 23
Ration.
Green (405)
Cured (180)
Green (415)
Cured (180)
Green (400^
Cured (165)
Green (505)
Cured (220)
Milk.
Lb.
132.0
107. I
Total
water in
rough-
age.
Lh.
15
Fat.
Per ct.
5.81
4.53
Lb.
4.40
4- 23
Specific
gravity.
I. 029
I. 031
Mois-
ture.
Per ct.
86.94
86.97
Ash.
Per ct.
0.723
• 744
Total
protein.
Per ct.
3.18
3-38
COW 25
163.2
167-3
296
15
4-05
3.60
6.61
6. 02
1.030
1.032
87.26
87-45
.724
-742
3-17
3-19
cow 27
161. I
128. o
28s
14
3-75
3.60
6. 04
4. 61
030
032
87.58
87-35
738
783
3-05
3-17
333-5
297.2
360
18
3-65
3.20
12. 17
9-51
1.028
1.030
88.36
88.85
.696
•725
2. 78
2.77
The length of time covered by this series of experiments, 10 days on
each ration, was too short to give more than an indication of the results
which a complete investigation would give. The data obtained, how-
ever, show that the water in the ration supplied by a green roughage, as
compared with the cured product, does not lower the fat content of the
milk. The results of these experiments would even indicate an opposite
effect, for in all cases the cows gave higher testing milk and three of them
produced more milk on the green feed.
Apr. 24. 1916 Effect of Water in Ration on Milk 177
SUMMARY
Four different methods of varying the water content of the ration were
used in this work.
(i) A full versus a limited allowance of drinking water.
(2) Turnips versus a dry-roughage ration.
(3) Wet versus dry beet pulp.
(4) Green versus dry crimson clover.
Eight cows were used in the experiments conducted by the first
method, four in the second, two in the third, and four in the fourth.
In every case except when the crimson clover was fed the amount of
water drunk by the different animals, as well as the difference in the
water content of the roughages under comparison, was determined.
With all except one cow. No. 22 in the wet versus dry beet-pulp group,
the amount of water in the dry ration did not exceed 75 per cent of that
supplied by the wet ration, and with some cows that were given a limited
allowance of water the dry ration contained less than 60 per cent of the
water content of the full-allowance ration.
Cow 22 in the wet versus dry beet-pulp group received, when the dry
ration was fed, 88 per cent of the water content of the wet ration.
In the green versus cured crimson-clover group, the former contained
71.23 per cent water and the latter 8.33 per cent. The daily ration of
green clover varied from 40 to 50 pounds per head, and of the cured hay
from 16 to 22 pounds per head.
Certain individual cows at times produced milk having an abnormal
fat content. This effect was apparently independent of the ration, as it
occurred not only with the high water-content ration but -with the dry as
well. A study of the data obtained in the four series shows that the
watery character of the ration has no effect upon the fat content of the
milk.
There was even less variation in the other milk constituents than in the
fat. This indicates that rations of varying water content have no effect
upon the composition of milk.
LITERATURE CITED
(1) Gilchrist, D. A.
1909. Variations in the composition of milk and their probable causes. In Dur-
ham Co. Coun. Educ. Com. Rpts. Dairy Invest. Offerton Hall, 1909,
p. 7-27.
(2) Great Britain Board of Agriculture.
1911. Effect on milk of water or watery food given to cows. In Jour. Bd. Agr.
[London], v. 17, no. 11, p. 910-911.
(3)
1913. The effect of watery foods on milk. In Jour. Bd. Agr. [London], v. 20,
no. 5, p. 385-392.
1 78 Journal of Agricultural Research voLvi, no. 4
(4) Hansson, Nils.
1912. Der verschiedenartige Einfluss der Futtermittel auf die Milchabsonderung
und Fettproduktion der Kiihe. In Fiihling's Landw. Ztg., Jahrg. 61,
Heft 10, p. 337-351-
(5) Henry, W. A., and Morrison, F. B.
1915. Feeds and Feeding . . . ed. 15, 691 p. Madison, Wis.
(6) HOLTSMARK, B.
1897. Nogle lagttagelser over Melkens Fedtinhold ved Tvimipsfodring. In
Tidsskr. Norske Landbr., Aarg. 4, p. 161-169.
(7) Jones, C. B.
1907. The effect of brewers' grain on milk, hi Durham Co. Coun. Educ. Com.
Offerton Bui. 2, p. 23-36. Also in Durham Co. Coun. Educ. Com.
Rpts. Dairy Invest. Offerton Hall, p. 81-92. 1909.
(8) Jordan, W. H.
1908. Investigations in animal nutrition. In 26th Ann. Rpt. N. Y. State Agr.
Exp. Sta., pt. 3 (25th Anniv. Rpt. 1882/1907), p. 66-71.
(9) Koch, B.
1901. Untersuchungen iiber den Einfluss der Menge des aufgenommenen
Wassers auf die Milchsekretion des Rindes. In Jour. Landw., Bd. 49,
Heft I, p. 61-88.
(10) Lauder, Alexander, and Fagan, T. W.
1912. On the effect of heavy root feeding on the yield and composition of
milk. Edinb. and East of Scot. Col. Agr. Rpt. 26, 56 p., tab.
(11) McConnELL, Primrose.
1911. Watering milk. In Dairy, v. 23, no. 268, p. 102.
(12) Tangl, F., and Zaitschek, A.
1911. tjber den Einfluss verschiedener wassiger Futtermittel auf die Menge
und Zusammensetzung der Milch. In Landw. Vers. Stat., Bd. 74,
Heft 3/5, p. 183-249.
(13) Walker, F. P.
1909. A second experiment to test the effect of brewers' grains on the quantity
and quality of milk. In Durham Co. Coun. Educ. Com. Offerton
Bid. 3, p. 5-20. Also in Durham Co. Coun. Educ. Com. Rpts. Dairy
Invest. Offerton Hall, p. 93-107. 1909.
CROWNGALL STUDIES SHOWING CHANGES IN PLANT
STRUCTURES DUE TO A CHANGED STIMULUS
[PRELIMINARY PAPER]
By Erwin F. Smith,
Pathologist in Charge, Laboratory of Plant Pathology,
Bureau of Plant Industry
Some recent experiments with crowngall have led to a number of
discoveries which may be summarized as follows :
First, as everyone knows, the tendency of cambium is not simply to go
on indefinitely producing more cambium but to elaborate out of its
embryonic elements formed structures, tracheids, wood vessels, wood
fibers, ray cells, sieve tubes, etc., all having a definite arrangement and
a well-defined polarity, but when internodal stem cambium is inoculated
with the crowngall organism (Bacterium iumejaciens) the ordinary physio-
logical tendencies are upset, as already shown in 1911 and 191 2,* and
several very interesting new phenomena make their appearance: (i) The
elements of the formed or mature tissues are produced in less numbers
than ordinarily, and these elements have lost the whole or a considerable
part of their polarity, so that the most bizarre complexes of twisted and
distorted tissues arise; (2) the parenchymatous elements are greatly
increased in number and reduced in size, since under the bacterial stimu-
lus many of the cambium cells appear to have lost all power to produce
mature tissues and at the same time have acquired a new growth impetus,
a tendency to an uncontrolled, pathologically embryonic, cell multipli-
cation, the result of which is a tumor of indefinite extension — the ordinary
naked crowngall, containing the distorted formed elements above referred
to and in addition exhibiting a marked hyperplasia of the parenchyma;
(3) correlative with these changes, over which the plant has no control,
is a tendency to open wounds and to early decay and also to the formation
of daughter tumors.
Second, when, by means of very shallow needle pricks, similar inocu-
lations are made into the internodal cortex of young stems (the so-called
fundamental tissue, which is still in a growing condition) a similar cell
proliferation occurs, the elements of which are very small in comparison
with those from which they have developed, because under the changed
stimulus they are kept embryonic and are compelled to divide soon after
previous divisions, so that they can never reach maturity either in size
1 Smith, Erwin F., Brown, Nellie A., and Townsend.C.O. Crown-gallof plants: its cause and remedy.
U. S. Dept. Agr. Bur. Plant Indus. Bui. 213, 215 p., 36 pi. 1911.
Smith, Erwin F., Brown, Nellie A., and McCulloch, Lucia. The structure and development of crown-
gall: a plant cancer. U. S. Dept. Agr. Bur. Plant Indus. Bui. 255, 60 p., 109 pi. 1912.
Journal of Agricultural Research, Vol. VI, No. 4
Dept. of Agriculture, Washington, D. C. Apr. 24, 1916
ds G — 77
(179)
i8o Journal of Agricultural Research voi. vi. no. 4
or function as long as the stimulus lasts. These inoculations (on the
Paris daisy) have brought out another interesting fact. As the tendency
of young fundamental tissue (the growing point) is to form a stele in its
center, so when the mature tissues of the stem cortex are brought under
the new stimulus and begin to proliferate, in the manner of embry-
onic tissues, primitive but imperfect stele-forming tendencies are devel-
oped in the tumor. I have not seen an actual shoot produced by such
a tumor; but sieve tubes and trachei are formed in it (out of descend-
ants of cortex cells, be it remembered); and cross sections of some of
these small tumors show that these stelar elements have a tendency to
be arranged in the form of a closed structure (primitive stele), although
often this is not pronounced. These superficial tumors have no connec-
tion with the xylem or phloem of the true stele, for in no case did the
needle punctures enter as far as the phloem, much less the cambium, and
serial sections show clearly that all of their structures (blastomous cells,
trachei, and sieve tubes) have been developed wholly, out of cortex cells
(probably cortex mother cells). After a few weeks such shallow timiors
cease to grow, or develop very slowly, owing to imperfect nutrition (lack
of all connection with the xylem and phloem of the plant).
Third, when the crowngall organism (hop strain) is inoculated into
the leaf axils of young growing plants (species of Pelargonium, Nicotiana,
Lycopersicum, Citrus, Ricinus, etc.) the buds of which are in a dormant
state and which under ordinary conditions will continue dormant —
namely, unless the top of the plant is removed — a new type of tumor
develops, one hitherto not seen in crowngall. Inoculating in this way
I have obtained tumors covered all over with diminutive, abortive leafy
shoots, or flower shoots, if flower anlage have been disturbed. The
shoots may be variously twisted, fused, and fasciated, as in the common
house geranium {Pelargonium spp.) shown in Plate XVIII. This appar-
ently is what happens : The growth of the tumor distorts the tissues, tear-
ing the anlage into small fragments which are variously distributed and
develop on or in the tumor into organs of a size proportional to the size
of the included fragment — here as part of an ovary or anther, there as
a shoot. These pathological shoots live but a short time and are quite
unable to carry on the normal activities of the plant when the other
leaves are removed. I have believed for a long time that fasciation
must be due to a bacterial infection ; but this is, I believe, the first time
that anyone has obtained it by means of a pure-culture inoculation.
The results obtained by inoculating the upper leaf axils of young
growing plants of the castor-oil plant {Ricinus communis) are prompt
and quite as striking (PI. XIX).
On tobacco plants {Nicotiana tabacum) these teratoid tumors, devel-
oped in leaf axils <Pls. XX and XXII), have also produced secondary
tumors repeating the structure of the parent tumor. Such tumors have
been obtained both in stems and leaves, especially when inoculations were
Apr. 24. 1916 Crowngall Studies 181
made early; and they contain, along with the proliferating tumor cells
(blastomous cells), the same teratoid elements as the primary tumor.
These are true daughter tumors, being connected back to the primary
tumor by a tumor strand which is quite different both in structure and in
location (PL XXI) from that occurring in the Paris daisy. The latter, it
will be remembered, follows the line of the spiral vessels in the inner wood,
and is parenchymatous in its structure, containing only here and there
some vessels (scattered trachei) . This tobacco tumor strand occurs in the
cortex, consists almost entirely of vessels, and is a true stem (stele),
although developed under a pathological stimulus, and in a part of the
plant where no stele was ever seen before — namely, in the outer cortex,
through which it can be traced (parallel to the long axis of the stem) for
long distances and from which at intervals leafy tumors are sent to the
surface of the plant. I^rom its frequent proliferation in the form of
tumors it is evident that parenchymatous (blastomous) elements must
also occur in the strand, but they are not abundant. In fact, in the
parts I have examined they are almost as infrequent as are trachei in
the daisy strand. Cross sections and longitudinal sections of this re-
markable tumor strand show it to have spiral vessels in its center, sur-
rounded by trachei cut by ray cells, beyond which is a cylinder of
cambium surrounded by a cylinder of phloem, containing well-developed
sieve tubes. This tiny stele has no cortex or epidermis because it does
not need any, being surrounded and sufficiently protected by the normal
cortex of the tobacco stem. This is a phenomenon due apparently to
my new manner of inoculation (into shoot anlage), because some years
ago by inoculating intemodally on tobacco stems I obtained and figured*
tumors and a tumor strand in cortex corresponding to those found in the
Paris daisy — that is, composed chiefly of small-celled parenchjona.
The difference in results must therefore be due to difference in the
kind of tissue inoculated, each developing pathologically according to
its own growth tendencies.
Fourth, on some plants (which were tobaccos) I have also obtained
leafy tumors by making my bacterial inoculations in places where no
bud anlage are known to exist — for example, in the middle of leaves.
Ordinarily when leaf tissue in tobacco grows, it only produces more leaf
tissue;^ but when the crowngall organism (hop strain) is pricked into
midribs or side veins, tumors arise and a portion of them are leafy — that
is, bear shoots. I have obtained 27 such leafy tumors on a single plant
and several on a single leaf, all within a period of a few weeks (PI. XXIII).
It is easy to obtain them. The young leaves yield a larger proportion
of such tumors than the older ones, and I have observed no shoot-bearing
tumors on leaves which were fairly well developed when inoculated.
• Smith, Erwin F., Brown, Nellie A., and McCulloch, Lucia. The structure and development of crown-
gall: a plant cancer. U. S. Dept. Agr. Bur. Plant Indus. Bui. 253, pl. 102-103. 1912.
2 I have never got any leaf cuttings of it to take root.
1 82 Journal of Agricultural Research voi. vi, no. 4
Rapidly developing young tissues seem to be necessary. Here again,
a changed stimulus has produced a more embryonic and primitive
condition, as shown by the appearance of these shoots. It is a patholog-
ical phenomenon, but one of more than passing interest, for, unless I am
much mistaken, it has wide physiological and pathological bearings.
It is another proof that the immature cell wherever it is located carries
the inheritance of the whole organism, and that what it will finally
become, as it matures, depends on the stimuli withheld from it or applied
to it. In other words,' it is so much evidence that any young cell may
become a totipotent cell if it is subjected to the proper stimulus, and this
stimulus may be either physiological, resulting in a normal structure,
as when the top of a plant is removed and a new top grows in its place
out of so-called adventitious buds (regeneration phenomena), or patho-
logical, resulting in an embryonic teratoma, as when a tumor-producing
schizomycete is introduced into sensitive growing tissues.
PLATE XVIII
Teratoid crowngalls produced in Pelargonium spp. by inoculating Bacterium
tumefaciens (hop organism through sunflower) into upper leaf axils on January 13,
1916. Photographed at the end of 74 days. At X the top of the shoot bearing five
or six leaves was removed to show the tumor more distinctly. All of the leafy shoots
here shown and many others too small to be seen distinctly in the photograph are
outgrowths from the tumor, which also bears red abortive flower anlage. The upper
shoot (L) was also flattened and fasciated (several shoots fused together) and the front
leaves (P P) were turning yellow and dying.
Crowngall Studies
Plate XVIII
Journal of Agricultural Research
Vol. VI, No. 4
Crowngall Studies
Plate XIX
Journal of Agricultural Research
Vol. VI, No. 4
PLATE XIX
Teratoid crowngalls produced in castor-oil plant (Ricinus communis) by inoculating
Bacterium, tumefaciens (hop strain), the inoculations being made in the upper leaf
axils of young, vigorous, unbranched plants.
Fig. A. — A red-stem variety. Leaves refiexed; axis distorted; and feeble shoots
developing out of the axillary tumors. There are on the tumors other smaller shoots
not shown here distinctly. Time, 13 days.
Fig. B. — A green-stem glaucous variety. As in figure A, but time 17 days. Here
also internal growths (root anlage) are pushing up the tissues of the stem below the
lower leaf. A few days later these roots appeared on the surface, both of this intemode
and of the one above it. This phenomenon has been recorded previously by the
writer as sometimes occtu^ring on inoculated stems of the Paris daisy and other plants
in the vicinity of developing tumors (Smith, E. F. Bacteria in Relation to Plant
Diseases, vol. 2, fig. 26. 1911).
PLATE XX
Teratoid crowngalls produced in tobacco by inoculating Bacterium tumefaciens
(isolated from a hop tumor several years ago and passed through a sunflower in 1915).
The inoculations were made by needle pricks in the axils of the lower leaves (under
the arrows), at which places small leafy tumors developed. These sent tumor strands
into the midribs of both leaves (L L) and later secondary teratoid tumors {S T T)
burst through and covered the top of the midrib. From the upper leaf axil also a
tumor strand developed, passing upward through 5 intemodes and then out into the
midrib of a leaf for several inches, giving rise at frequent intervals to tumors bearing
leafy shoots (teratoid elements) and to others free from them. This tumor strand
{T St) was not on the surface of the stem, as might appear from the photograph, but
was near enough to show through as a translucent band about 2 mm. wide. Time,
26 days.
Crowngall Studies
Plate XX
^i
s
STT
/
^ m^
»^.
\
Journal of Agricultural Research
Vol. VI, No. 4
Crowngall Studies
Journal of Agricultural Research
Vol. VI, No. 4
PLATE XXI
The teratoid tumor strand of Plate XX, which gives rise during its course to
more than 30 small tumors.
Top. Cross section of outer part of right side of stem of tobacco plant shown on
Plate XX. P, outer edge of the phloem; E, epidermis; T St, tumor strand, which
is bedded in the normal cortex of the stem.
Bottom. Longitudinal section from upper part of the above tumor strand, more
highly magnified, showing it to be a true stele. The coarse-celled tissue at top and
bottom is the normal cortex of the stem. The pathological tissues are 5 T, sieve
tubes; C, cambium; Tr, trachei; Sp, spiral vessels.
PLATE XXII
Teratoid crowngalls produced in a tobacco plant by inoculating Bacterium tume-
faciens (hop strain through sunflower) into the leaf axils. Small tumors soon appeared
where inoculated and these are now covered with pale leafy shoots which have swollen
(tumefied) bases and are beginning to die. The top was cut away on the 26th day,
and the plant was unable to make a new one out of these pathological shoots, but has
grown it (X) from an uninoculated lower leaf axil. Time, 73 days.
Crowngall Studies
Plate XXII
Journal of Agricultural Research
Vol. VI, No. 4
Crowngall Studies
Plate XXIII
Journal of Agricultural Research
Vol. VI, No.
PLATE XXIII
Teratoid crowngalls produced in tobacco leaves with the hop strain of Bacterium
tumefaciens by local (leaf) inoculations — that is, inoculation in places where shoot
anlage are not known to exist.
Fig. A. — Portion of an upper leaf showing fotu- shoot-bearing tumors growing from
upper surface of the inoculated midrib. Leaf inoculated February i6, 1916. Pho-
tographed on April i.
Fig. B . — Same as A , but the leaf reversed and the midrib stripped of its blade to
show two other shoot-bearing tumors which have developed from its under surface.
Actual height of the tallest shoot, 1.5 cm.
Fig. C— From middle of another leaf on the same plant as A , but fiuther magnified
and photo made on an orthochromatic plate to show the pale green character of the
shoot as contrasted with the dark green of the stirrounding leaf (which is also in
shadow). This tumor and its shoot arise from a branch of the midrib, the latter
in cross section being shown at X. A smaller teratoid tumor bearing two shoots (at
either side of C) developed on the upper surface of the leaf and the one bearing the
longer shoot on its lower surface. The actual length of this shoot was 1.5 cm. The
leaf was curved downward and the shoot was growing out horizontally. Time, 45 days.
27472°— 16 4
ADDITIONAL COPIES
OF THIS PXTBLICATION MAT BE PROCTTEED FEOM
THE SUPERINTENDENT OF DOCT7MENT3
GOVERNMENT PRXNTING OPTICE
■WASHINGTON, D. C.
AT
15 CENTS PER COPY
SiTBSCRiPTioN Price, $3.00 Per Ykab
A
Vol VI NIAY 1, 1916 No. 5
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS
Bffect of Certain Species of Fusaritun on the Compositioii
of the Potato Tuber ------- 183
LON A. HAWKINS
Hyperaspis binotata, a Predatory Enemy of the Terrapin
Scale 197
F. L. SIMANTON
DEPARTMENT OF AGRICULTURE
WAS HINGTON , D.C
WASHtHSTON : eOVEqNMENT PBINTIM3 OfTKJE ; t»M
PUBI<ISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
ffOK TBS DEPARTMBin
VOR THB ASSOCIATION
KARL F. KELLBRMAN, Chairmak RAYMOND PEARL
Phyiiologist and Assistant Chief, Buraatt
*f PUnt Industrr
EDWIN W. ALLEN
Chiet. Office of Experiment Stations
CHARLES L. MARLATT
Assistant Chief, Snreau of Entomology
Bioloeist, Main* Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition, Tke
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station *t
the University of Minnesota
All oorre^)ondence regarding articles from the Department of Agriculture
should be addressed to Karl F. Kellerman, Joxmial of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, JoiutuI of Agricultiual Research, Orono, Maine.
JOURNAL OF AGEICETIAL ISEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, May i, 1916 No. 5
EFFECT OF CERTAIN SPECIES OF FUSARIUM ON THE
COMPOSITION OF THE POTATO TUBER* .
By LoN A. Hawkins,
Plant Physiologist, Plant Physiological and Fermentation Investigations,
Bureau of Plant Industry
INTRODUCTION
Potato tubers (Solanum tuberosum) are subject to attack by various
parasitic fungi. Some of these organisms invade the tuber, kill the
cells, break down the cell walls, and cause, directly or indirectly, a
more or less complete disorganization of the host tissue. What con-
stituents of the potato are most easily destroyed by the fungus and
what compounds can not be utilized by it either in respiration or to
build up its own tissue are of considerable interest in the study of the
physiology of parasitism. It was to obtain information on the effect
of some potato tuber rot fungi upon the tissues of the host plant that
the present study was planned and carried out. In this investigation
the effect of Fusarium oxyspormn Schlecht. and F. radicicola Wollenw.
on the sucrose, reducing-sugar, starch, pentosan, galactan, and crude-
fiber content of the potato was studied. Some experiments were dupli-
cated also with F. coeruleum (Lib.) Sacc.
The three species of Fusarium just mentioned are all parasites on
the potato tuber. Smith and Swingle (9) ^ considered F. oxysporum
to be the cause of a serious rot of potato tubers. Wollenweber (10)
did not agree with these writers, and contended that this fungus, while
the cause of a wilt disease of the potato plant, was not able to rot the
tubers. This conclusion of Wollenweber's has recently been disproved
by Carpenter (4), who corroborates the findings of Smith and Swingle
on this point. With this species and with F. radicicola, the latter con-
sidered by Wollenweber and by Carpenter to be the cause of a tuber-rot
of considerable importance, the writer experienced no difficulty in
obtaining well-rotted tubers in two to three weeks after inoculation.
' The work described in this paper was carried out in cooperation with the Office of Cotton and Truck-
Crop Diseases. The writer thanks Mr. C. W. Carpenter, of that office, for cultures of the fungi used.
The writer's thanks are also due Mr. A. A. Riley, of the Office of Plant Physiological and Fermentation
Investigations, for assistance in the experimental part of this study.
^ Reference is made by number to " Literature cited," p. 196.
Journal of Agricultural Research, Vol. VI, No. s
Dept. of Agriculture, Washington, D. C. May i. 1916
dl G— 78
(183)
184 Journal of Agricultural Research voi. vi, No. s
EXPERIMENTAL METHODS
As the methods for sterilizing, sampUng, and inoculating followed in
this study were similar to those outlined in a study of the brownrot
of the peach (7), they will be discussed here only briefly. The sterile
tubers were sliced longitudinally into four parts with a flamed knife.
Particular attention was given to obtaining portions of approximately
the same weight and same proportionate amount of cortex and pulp.
Each quarter was placed in a small wide-mouthed flask or large test
tube which had been stoppered with cotton, sterilized, and weighed.
The containers with the portions of potatoes were weighed again and
the samples were ready for inoculation. Two of the quarters from each
potato were inoculated from stock cultures of some one of the fungi
used in these experiments and a small quantity of sterile water was
added to each of the four containers. The four samples, two inoculated
and the two corresponding control samples, were placed side by side at
room temperature until the inoculated portions were well rotted. They
were then prepared and analyzed. The difference between the sound
and the rotted portions in the content of the compounds determined
was considered to be due to the action of the fungus. All control por-
tions infected at the time the samples were prepared for analysis and
all inoculated portions infected with organisms other than those used in
the inoculations were discarded.
All samples were prepared for analysis by cutting them into very thin
slices with a sharp knife and washing them into the proper vessel. Pre-
cautions were observed, of course, that none of the juice or pulp should
be lost. The methods of analysis for agricultural chemists ^ were
usually followed in the determination of the various compounds. The
sugars were extracted from the tissue with alcohol and determined as
in the work on the brownrot of the peach. The method of extraction
is the alcohol method of Bryan, Given, and Straughn (3), somewhat
modified to suit the conditions of the experiment. The amount of cane
sugar was in all cases calculated from the reducing power of the extract
before and after inversion with acid.
The starch determinations in the preliminary experiments were made
only by the direct acid-hydrolysis method using the finely ground potato
which- had been extracted with alcohol. In the work with the sound
and the rotted portions of the tubers, series of analyses were also made
by the diastase method with subsequent acid hydrolysis.^ Tollen's
phloroglucid method ^ was followed in all cases in the determination of
the pentosans. The methyl pentosans were determined according to
the method of Ellett and Tollens (6), by extracting the precipitated
phloroglucid with alcohol. The galactans and the crude fiber were de-
• Wiley, H. W., ed. Official and provisional methods of analysis, Association of Official Agricultural
'Chemists. As compiled by the committee on revision of methods. U. S. Dept. Agr., Bur. Chem. Bui. 107
■(rev.), 272 p., 13 fig. 1908.
May 1. 1916 Effect of FusaYium spp. on Potato Tubers
185
termined by the usual methods ^ in dry ether-extracted samples which
had been ground. For the percentage of dry matter the sliced-up
samples were placed in glass-stoppered weighing bottles and covered
with alcohol. The alcohol was then driven off and the samples dried
to constant weight. All data were calculated to the original wet weight
of the potato used. The potatoes used in the experiment were smooth
white potatoes usually purchased at the local market. The cultures
of fungi used in the experiments were subcultures from Carpenter's cul-
tures of F. oxysporum 2,Z95 and 3315; F. mdicicola 31 13 and 3319, and
F. coeruleum 3501.
EXPERIMENTATION
To determine the amount of variation in content of the different com-
pounds in the four quarters of the potato, series of preliminary analyses
were carried out. In these the potato was sampled in the usual way,
except that the portions were sliced immediately and prepared for
analysis. The results of these analyses are shown in Tables I to VI.
Table I. — Reducing sugar and sucrose content of quarters of sound potatoes
[Expressed as percentage of wet weight]
Potato No.
Reducang sugar.
Quarter
A.
Quarter
B.
Quarter
C.
Quarter
D.
Sucrose.
Quarter
A.
Quarter
B.
Quarter
C.
Quarter
43
44
46
49
87
88
105
o. 10
.06
•17
. 02
o
O. II
.06
• 14
. 02
09
07
14
03
O. II
.06
.19
. 02
o
o
o
o. 04
.04
. 02
•03
.07
.06
• 05
,04
03
03
07
OS
08
o. 04
• 03
.04
• 03
.07
.06
.07
o. 02
• 03
.04
.04
.06
• 05
.06
Table II. — Starch content of quarters of sound potatoes determined by the direct acid-
hydrolysis method
[Expressed as percentage of starch, wet weight]
Potato No.
Quarter A.
Quarter B.
Quarter C.
Quarter D.
70.
47-
50-
17.88
16.43
15.04
18.56
19.52
15-35
16. 04
17. 16
17.07
15.64
16. 00
16.54
17. 26
16. 04
14. 50
17.27
Table III. — Pentosan content of quarters of sound potatoes
[Expressed as percentage of wet weight]
Potato No.
Quarter A.
Quarter B.
Quarter C.
Quarter D.
19.
28.
47-
207.
0.51
•37
.41
•51
0.43
•35
.40
.48
o. 46
•35
•37
.48
48
37
40
51
> Wiley, H. W., ed. Op. cit.
i86
Journal of Agricultural Research
Vol. VI. No. s
Table IV. — Galactan content of quarters of sound potatoes
[Expressed as percentage of wet weight]
Potato No.
Quarter A.
Quarter B.
Quarter C.
Quarter D.
ii8
0.025
.028
.034
C.030
. 020
. 024
0.034
. 027
.030
0. 027
. 029
• 025
Table V. — Crude fiber content of quarters of sound potatoes
[Expressed as percentage of wet weight]
Potato No.
Quarter A.
Quarter B.
Quarter C.
Quarter D.
102
210
211
0.50
•47
.40
0.59
•47
.41
0.48
•47
•39
0.47
.44
•36
Table VI. — Dry matter in the quarters of sound potatoes
[Expressed as percentage of wet weight]
Potato No.
Quarter A.
Quarter B.
Quarter C.
Quarter D.
118
119
120
138
143
20.86
20. 70
19-73
24. 12
24. 19
19.96
20. 96
20. 23
24. 58
22.37
21.95
19. 16
21. 42
25-36
24. 81
17.78
19.77
20.73
24. 80
22. 82
Tables I to VI show that there is considerable variation in the percent-
age of some of the compounds in different quarters of the same tuber,
though usually the actual difference is not great. It is noticeable that
two portions of the same tuber are more nearly alike in composition than
samples from different potatoes. The method, therefore, which involves
the comparison of the content of two quarters of the same potato is
more accurate than one based on a comparison of the composition of
two different potatoes. The experiments in which sound and rotted
quarters were analyzed to determine the effect of the fungi upon the
potato show that data from which definite conclusions may be drawn
can be obtained by this method.
Inasmuch as the mycelium of the fungi was present in the rotted por-
tions of the potatoes, it was of interest to determine what influence the
compounds elaborated by these fungi would have on the apparent com-
position of the tuber. Quantities of mycelia of the two fungi F. radicicola
and F. oxysporum were accordihgly grown on potato extract. This
medium was prepared by boiling sliced potatoes until they were soft,
filtering the extract through cotton, and sterilizing it in suitable flasks.
May 1. 1916 Effect of FusaHum spp. on Potato Tubers
187
The flasks of this medium were inoculated and the fungi allowed to grow
for two or three weeks. The mat of mycelium was then removed, washed,
dried, ground, and analyzed. The data obtained from these analyses,
calculated as percentage of the dry weight, are given in Table VII.
Table VII. — Amount of akohol-insoluble substance reducing Fehling's solution when
hydrolyzed with dilute hydrochloric acid, pentosans, methyl pentosans, galactans, and
crude fiber in mycelium of Fusarium oxysporum and Fusarium radicicola
[Expressed as percentage of dry weight]
Species.
Alcohol - insoluble
substance reduc-
ing Fehling's solu-
tion when hydro-
lyzed with dilute
hydrochloric acid
(as dextrose).
Pentosans.
Methyl
pentosans.
Galactans.
Crude fiber.
Fusarium oxysporum . .
Fusarium radicicola
1 34. 58
i 31- 90
I 31- 63
J 31- 48
2-53
2. 60
I. 20
I. 20
0-73
.68
1.50
1.50
0.86
.66
.72
.64
21.8
18.4
20.3
17.6
It is apparent from Table VII that the fungi growing on the culture
media prepared from potatoes produce pentosans, methyl pentosans,
galactans, and a considerable quantity of substance which is insoluble
in alcohol and reduces Fehling's solution when hydrolyzed with dilute
hydrochloric acid. That this last-mentioned substance can not result
from the hydrolysis of the pentosans is evident from the relatively small
pentosan content of the mycelium. The amount of substance which is
considered as crude fiber in the table is also quite marked. It is evident,
then, that both fungi build compounds which may be expected to raise
the content of pentosans, galactans, and other substances in the tissue
of the potato when the fungi and host are analyzed together. It must
be remembered, however, that the percentages given in Table VII are
related to dry weight of washed fungus mycelium and that the content
of mycelium in 25 gm. of wet weight of the potato rotted with either of
these fungi would be small.
The general appearance of the rotted portion of potato was typical
for tubers rotted with these fungi at laboratory temperatures (from 20°
to 25° C.) in a saturated atmosphere — that is, it was a wetrot (4, p. 187).
The skin apparently was uninjured and could have been removed entire
in most cases. The inner portion was soft and generally disorganized.
Microscopic examination showed that the cells of the interior were appar-
ently free from each other, as if the middle lamellae had been dissolved.
The starch grains did not appear to have been eroded in the time allowed
for the experiment. The method of preparing the quarters of potato
for analysis has been described.
The starch and sugar determinations were usually made on the same
portion by extracting the pulp with alcohol, the extract being used for
Journal of Agricultural Research
Vol. VI, No. s
the sugar and the soHd residue for the starch determinations. The
effect of three species of Fusarium, F. oxysporum, F. radicicola, and
F. coeruleum, on the starch and sugar content of sound and rotted
quarters of the same tubers was studied. The data obtained from the
determination of the sugars are shown in Table VIII.
Table VIII. — Reducing sugar and sucrose content of the sourid and rotted quarters of
potatoes
[Expressed as percentage of the original wet weight]
Species of Fusariiun and potato No.
Reducing sugar.
Rotted quarter. Sound quarter.
Sucrose.
Rotted quarter. Sound quarter,
Infected with Fusarium oxy-
sporum:
160
159
158.... ,
Infected with Fusarium coeru-
leum:
149
150
151
Infected with Fusarium radi-
cicola:
32
26
34
41
0. 04
0.31
0. 10
.04
.28
0
0
.44
0
• 13
.40
. 12
.04
•47
.24
• 17
•37
0
0
•03
.04
0
. 02
.04
0
•03
. 02
0.66
.67
1.03
39
66
24
19
09
42
In Table VIII it may be seen that all three species of Fusarium used
the sugars. In most cases practically all the sugar had disappeared from
the rotted portion, the cane sugar being utilized almost if not quite as
completely as the reducing sugars. That the fungi could use disaccharids
directly — that is, without breaking them down to their constituent
monosaccahrids — seemed unlikely. It was therefore probable that the
fungi secreted enzyms which were capable of hydrolyzing cane sugar,
and possibly maltose also. To determine this point, tests were made
for sucrase and maltase in extracts of the mycelium of F. oxysporum and
F. radicicola. The fungi were grown for about three weeks or until a
thick mat of myceUum was formed on potato extract. The felt was then
separated from the liquid, ground up in a mortar and digested for 48
hours under toluol. The extract was filtered off and portions of it added
to solutions of the sugars of known concentration. Controls of the boiled
extract were also prepared. After the preparations had been allowed
to stand overnight at laboratory temperature the amount of reducing
sugar was determined. It w^as found that in the preparations of un-
boiled extract the sugars, both sucrose and maltose, were inverted almost
quantitatively. The boiled extracts were practically without effect. It
is evident then that the two fungi secrete both sucrase and maltase.
May 1. 1916 Effect of FusaYium spp. on Potato Tubers
189
The starch determinations in the sound and rotted portions of the
same tuber were made by two methods, as has been said. The data
obtained by the direct acid hydrolysis method are given in Table IX,
while the results of the determinations by the diastase method with sub-
sequent acid hydrolysis are shown in Table X.
Table IX. — Starch content of sound and rotted quarters of potatoes infected with different
species of Fusarium, as found by direct acid hydrolysis
[Expressed as percentage of starch of the original wet weight]
Fusarium oxysporum.
Potato No.
Fusarium coeruleum.
Potato No.
Fusarium radicicola.
Potato No.
Rotted
quarter.
Sound
quarter.
Rotted
quarter.
Sound
quarter.
Rotted Sound
quarter. } quarter.
165
168
180
14.40
14. 08
16. 16
13-53
14.67
14. 62
149
150
151
19.19
18. 72
22. 06
18. 69
17.48
22. 22
34
26
32
41
16. 18
15.24
15-15
16.83
15- 79
16. 60
16. 01
16.85
Table X. — Starch content of sound and rotted quarters of potatoes infected with different
species of Fusarium asdetermined bythediastase method with subsequent acid hydrolysis
[Expressed as percentage of starch, wet weight]
Potato No.
25
33
27
Fusarium oxysporum.
Rotted
quarter.
17.77
12. 50
15-32
Sound
quarter.
16.85
II. 32
14.05
Potato No.
47-
Fusarium. radicicola.
Rotted
quarter.
17-50
16.66
Sound
quarter.
16.66
15. 16
The effect of the fungi upon the starch in the potatoes is in marked
contrast to their action on the sugars. In Table IX, which gives the
results of starch determinations by the direct acid-hydrolysis method, it
may be seen that the starch content of the rotted portion appears to
be higher in many cases than that of the corresponding sound quarter.
In the determinations by the diastase method followed by acid hydrolysis
the apparent starch content of the rotted portion is always higher, as
shown in Table X. The fact that the fungi build up substances which
are insoluble in alcohol and reduce Fehling's solution when hydrolyzed
with dilute hydrochloric aicd, as shown in Table VIII, would account
for any apparent increase in starch content in the rotted portion when
the starch is determined by the direct acid-hydrolysis method. If the
substances are also either soluble in hot water originally or made so by
the diastase treatment, the apparent increase in starch content when
the starch is determined by this method would be explained. In the
diastase method the starch paste is liquefied by the action of the diastase,
190 Journal of Agricultural Research voi. vi. No. s
then filtered, and the filtrate hydrolyzed with dilute hydrochloric acid.
Some of the mycelium of these fungi was extracted with alcohol, and
then dried and extracted with hot water. The extract was then filtered
off, treated with hydrochloric acid exactly as in the acid hydrolysis of
starch, neutralized and tested for reducing substances. A considerable
quantity was found. The filtrate did not give a qualitative test for
pentosans. The apparent increase in starch content in the rotted por-
tions of the potatoes, then, is due to compounds laid down by the fungi.
From the fact that only a small amount of mycelium of these fungi
could be present in the rotted potato it would seem probable that it
the starch were attacked to any extent the apparent starch content as
obtained by acid hydrolysis would be lowered in all cases. To obtain
further information on this point experiments were carried out to ascer-
tain whether these fungi secreted diastase and if so, whether this enzym
could break down the starch grains of the potato.
Extracts of the undried, ground mycelium of the two fungi, F. oxy-
sporum and F. radicicola, were made with 50 per cent glycerin. These
extracts were filtered after 24 hours through absorbent cotton and por-
tions added to a 2 per cent solution of "soluble starch." Suitable con-
trols were prepared and all preparations allowed to stand in an incubator
under toluol at 30° C. for 48 hours. At the end of this time the starch
was practically all broken down by the extracts of both fungi. Similar
experiments were carried out with starch paste made from potato starch
with positive results. The fungi then secrete diastatic enzyms. The
experiments, however, did not prove that the diastases were able to
attack the starch grains before they were broken down. Brown and Mor-
ris (2) have shown that malt diastase can not act on ungelatinized potato
starch, though the starch grains of barley are readily eroded by it.
Whether the enzyms in the extracts of the myceUum could erode the
starch grains of the potato at room temperature was determined by
placing some well-washed potato starch in extracts and allowing the
preparations to stand under toluol. They were shaken up and examined
from time to time, but no sign of erosion of the starch grains was evi-
dent even at the end of a week. The extracts used were tested on starch
paste or "soluble starch" with positive results. Smith and Swingle (9)
mention that the starch in the potatoes rotted with F. oxysporum was
apparently not eroded. It is, of course, possible that the potato starch
grains are very slowly attacked by the diastases of these fungi or that
some inhibitor is present which prevents the action of the enzym on
the starch in this condition at the temperature at which these studies
were made. These points should be investigated. At present, however,
the conclusion seems warranted in view of the evidence that the starch
of the potato is not appreciably affected by the fungi.
From the fact that these fungi penetrate the cell walls or parts of the
cell walls of the potato in living parasitically upon their host, their effect
May 1. 19x6 Efject of Fusariuw, spp. on Potato Tubers
191
on the constituents of the cell wall was considered of especial interest.
The substances studied in this investigation which may be considered
to be, in part at least, components of the cell walls are the pentosans,
crude fiber, and galactans (5). Inasmuch as the fungi apparently do not
affect the skin in rotting the potato, it was considered of interest to find
out the relative distribution of the pentosans and crude fiber in the skin
and inner portion of the potato. For these analyses the potatoes were
peeled as thinly as convenient and determinations made on the weighed
peeling and inner portion separately. The results of the pentosan
determinations are given in Table XI.
Tabi^E XL — Pentosan content of the peeling and inner portion of potatoes
[Expressed as percentage of pentosans, wet weight]
Potato No.
116
133
134
Skin.
O. 62
.88
.80
Inner
portion.
0.28
•39
•47
140
163
164
Skin.
02
72
07
Inner
portion.
©•59
• 36
• 50
When the pentosan content is calculated as wet weight, it is about
half as great in the inner portion of the tuber as in the skin. There is,
nevertheless, a considerable amount of the furfurol-yielding compounds
in the fleshy part of the potato. Inasmuch as the fungus has practically
no effect on the skin, it is to be considered that practically all changes
in the pentosan content that take place during rotting are in the inner
portion of the tuber. The results obtained from the pentosan determina-
tions on the sound and the rotted portions of the potato tubers are shown
in Table XII.
Table XII. — Pentosan and tnethyl-pentosan content of sound and rotted quarters of
potatoes
[Expressed as percentage of pentosans, wet weight]
Infected with F. oxysporum
29
30
35
40
Infected with F. radicicola:
171
174
176
Sound quarter.
Total
pento-
sans.
53
53
45
52
28
37
25
Pen-
tosans.
0.47
.41
•36
.42
32
19
Methyl
pen-
tosans.
O. 06
. 12
.09
. 10
•05
•05
.06
Rotted quarter.
Total
pento-
sans.
o. 50
.46
•44
•37
•25
.29
.26
Pen-
tosans.
0-35
•35
•35
.26
. 20
. 24
Methyl
pen-
tosans.
0.15
. II
.09
. II
•05
•05
•05
192
Journal of Agricultural Research
Vol. VI, No. s
Table XII shows that the total pentosan content, which includes all
furfurol-yielding matter, and the pentosan content, which is the total
pentosan content after the methyl pentosans have been extracted, are
higher in all but one instance in the sound portions of the tuber. There
is slightly more variation in methyl pentosan content; it is the same or
greater in the rotted as in the sound portion in all but two cases. The
fungi evidently use the pentosans, but do not affect the methyl pentosans
to any extent. It is to be remembered that these fungi build up both
pentosans and methyl pentosans when growing on potato extract. The
content of these substances, then, in the rotted portions given in Table
XII is undoubtedly the difference between the amount of pentosans
broken down by the fungi in the interior of the potato and the amount
built up by the fungi. The destructive processes evidently proceed
more rapidly than the constructive, and some of the pentosans of the
potato are used either in respiration or in the building up of other
compounds.
From the effect of the fungi on pentosans it was considered probable
that enzyms which could hydrolyze these compounds were present in
the mycelium. Experiments were undertaken to determine this point.
The experiments were carried out as described in a previous paper (8) ,
except that the fungi were grown on potato extract instead of a synthetic
medium with gum arabic as a source of carbon. Xylan from rye straw
was used as a substrate. The results of these experiments are given in
Table XIII.
Table XIII. — Effect of boiled and unboiled extract of mycelium upon xylan from rye
straw, as shown by alcohol-soluble furfurol-yielding m.aterial and substance reducing
Fehling's solution. (0.2 gm. of xylan in each preparation was maintained at jo° C.
for one week.)
Species of Fusarium.
Fusarium radicicola . .
Fusarium. oxysporum
Quantity of cuprous oxid
derived from material
reducing Fehling's so-
lution.
Unboiled.
Mgm.
45-2
44.8
14. I
16. 4
Boiled.
Mgm.
15-4
14.8
6-5
6-5
Quantity of alcohol-solu-
ble furfuroi - yielding
substances as pento-
sans.
Unboiled.
Mgm.
13- I
13- I
18.6
14. 6
Boiled.
Mgm.
5-7
5-7
6.8
6.8
It is evident from Table XIII that the extracts of the fungi are able
to break down xylan prepared from rye straw to an alcohol-soluble
compound which reduces Fehling's solution and which forms furfuroi
when boiled with hydrochloric acid. The fungi then secrete an enzym
or enzyms which can break down xylan probably to xylose.
The crude fiber of the potato is undoubtedly a mixture of compounds,
among which are some of the cell wall constituents, including whatever
May 1, 1916 Effect of Fusariufn spp. on Potato Tubers
193
cellulose may be present. The distribution of the crude fiber through-
out the tuber is not as uniform as that of the pentosans, as is shown by
a comparison of Tables XI and XIV.
Table XIV. — Crude fiber content of the skin and inner part of the potato tuber
[Expressed as percentage on both a wet weight and dry weight basis]
Potato No.
Percentage of crude fiber, wet
weight.
Percentage of crude fiber, dry
weight.
Skin.
Inner portion.
Skin.
Inner portion.
210
1-54
I' 33
I. 20
0. 25
•25
.36
II. II
6.61
7.89
I 16
211
212
I 82
From Table XIV it may be seen that the crude-fiber content of the
peeling is ^}4 to 6 times greater than that of the inner portion calculated
on a wet weight basis and from 4 to 10 times greater on the basis of dry
weight. The inner portion of the potato contains usually a lower per-
centage of crude fiber than of pentosans.
The determinations of crude fiber on the sound and rotted portions of
the potato tubers are given in Table XV.
Table XV. — Crude-fiber content in sound and rotted quarters of potatoes
[Expressed as percentage of wet weight]
Rotted with Fusarium radicicola.
Rotted with Fusarium oxysporum.
Potato No.
37-
39-
115
Rotted
quarter.
56
57
40
Sound
quarter.
54
56
37
Potato No.
177
178
179
Rotted
quarter.
O. 71
• 73
Sound
quarter.
0.58
.62
.62
The crude-fiber content is always higher in the rotted quarter of the
tuber than in the corresponding sound portion, though the difference is
not great. As has been mentioned earlier in this paper, the fungus builds
up a considerable quantity of substance which is not dissolved in either
the acid or alkali used in the crude-fiber determination ; to this is due the
rise in the crude-fiber content of the potato during rotting. It is possible,
of course, that the fungi may break down the crude fiber of the host plant
and build up some similar substance with greater rapidity. From the
evidence brought out in these experiments, then, it is impossible to draw
definite conclusions.
The substances in the potato which give mucic acid when boiled with
proper concentration of nitric acid are considered in this study as galac-
tans. They are present in small quantities in the potato, and the com-
194
Journal of Agricultural Research
Vol. VI. No. s
bination in which they occur in the tuber was not investigated. Galactose
might occur in combination with raffinose, in a glucoside or combined in
the cell walls. It probably occurs in plants most commonly in the last-
mentioned combination. The effect of the fungi upon the galactan con-
tent of the potato is shown in Table XVI.
Table XVI. — Galactan content of sound and rotted quarters of potatoes
[Expressed as percentage of wet weight]
Rotted with Fusarium radicicola.
Rotted with Fusarium oxysporum.
Potato No,
Rotted
quarter.
Sound
quarter.
Potato No.
Rotted
quarter.
Sound
quarter.
27
42
0.039
•033
. 029
o. 062
. 060
.030
166
167
172
o. 069
.068
.081
o. 071
. 076
.083
It is evident from the table that the fungi lower the galactan content
of the potato. The fungi produce galactans when growing upon potato
extract and the data in Table XVI show that the breaking down process
proceeded faster than the building up.
The amount of dry matter of the sound and rotted quarters determined
as mentioned earlier in this paper is shown in Table XVII.
Table XVII. — Amount of dry matter in sound and rotted quarters of potatoes
[Expressed as percentage of wet weight]
Rotted with Fusarium radicicola.
Rotted with Fusarium oxysporum.
Potato No.
27
42
Rotted
quarter.
20. 83
19.88
20. q8
Sound
quarter.
21. 19
22. 59
22. 13
166
167
Rotted
quarter.
17-73
18.93
18. 17
Sound
quarter.
18.91
20.45
19.36
As was to be expected, the rotting of the potato by the fungi lowered the
percentage of dry weight as calculated to the original weight of the portion
of the potato used in the experiment. This is probably due to an increased
respiration — that is, a respiration of the quarter of the potato plus the
respiration of the fungus which in a given time is greater than a portion
of the same potato alone.
DISCUSSION
From the foregoing pages it is evident that the tuber-rot fungi used
in this study considerably alter the composition of the potato. That
they should be able to utilize the sugars of the potato was to be expected.
Most fungi use glucose readily as a source of carbon. Behrens (i) has
shown that Sclerotinia fructigenia lowers the sugar content of apples in
May 1, 19:6 Effect of FusaHum spp. on Potato Tubers 195
rotting them. The brownrot fungus of peaches reduces the sugar
content of that fruit. The presence of the enzyms sucrase and maltase
in fungi has frequently been recorded.
The starch content of the potato makes up the greater part of its dry
weight and may be regarded as stored food material. That the fungi
which so efficiently utilize the monosaccharids and disaccharids of the
potato tuber are unable, apparently, to affect this polysaccahrid is of
considerable interest. The fungi grow for the most part in the cell walls
and thus are not closely in contact with the starch grains. This might
retard the action because of the low rate of diffusion of the diastase but
could hardly inhibit it entirely. The fact that the diastases of these fungi
had no apparent effect on unbroken starch grains in vitro during the
time of the experiment, while potato starch when gelatinized was readily
hydrolyzed by these enzyms, indicates that the rate of action under
what are usually favorable conditions for such reaction is to say the
least very low. The experiments seem to show that enzymic studies
are of doubtful value in determining the effect of the parasite on the host
plant unless corroborated in a study of the physiological relations existing
between the two organisms. The effect of the fungi on the pentosan and
galactan content of the potato shows that they can break down at least
some of the constituents of the cell wall. Now, when a parasitic fungus
such as those used in this study enters a cell of its host plant, it must
either force its way in mechanically by exerting sufficient pressure to
rupture the cell wall or a portion of the cell wall must be dissolved.
Likewise, in growing between the cells of the host plant where no appreci-
able intercellular spaces exist, the cells must be forced apart mechanically
or some parts of the cell walls dissolved. It is evident from their effect
on the pentosans that these fungi are able to dissolve at least some portions
of the cell wall. That they secrete enzyms which can hydrolyze xylan is
more evidence on this point. The crude-fiber content of the potato was
increased in rotting owing to the formation in the fungi of some sub-
stances which were not broken down by the acid or alkali treatment in
the crude-fiber determinations. Therefore it was impossible to obtain
evidence as to the effect of the fungi upon the crude fiber. As shown in
the tables the crude-fiber content of the inner portion of the potato is
not high. It is noticeable that throughout this study the different
species of Fusariura had practically the same effect on the potato.
CONCLUSION
In conclusion, it has been shown in this study that the fungi in the
potato reduced the content of sugar, both sucrose and reducing sugar,
pentosans, galactans, and dry matter. The starch and methyl pentosans
are apparently not affected appreciably, and the crude-fiber content was
not reduced. It was shown that these two species of fungi secrete
sucrase, maltase, xylanase, and diastase; the last-mentioned enzyra is
apparently incapable of acting on the ungelatinized potato starch.
196 Journal of Agricultural Research voi. vi. No. s
LITERATURE CITED
i) BehrENS, Johannes.
1898. Beitrage zur Kenntnis der Obstfaulnis. In Centbl. Bakt. [etc.], Abt.
2, Bd. 4, No. 17/18, p. 700-706.
2) Brown, H. T., and Morris, G. H.
1890. Researches on the germination of some of the Gramineae. In Jour.
Chem. Soc. [London], v. 57, Trans., p. 458-528, 2 pi.
3) Bryan, A. H., Given, A., and Straughn, M. N.
19 II. Extraction of grains and cattle foods for the determination of sugars:
a comparison of the alcohol and the sodium carbonate digestions.
U. S. Dept. Agr. Bur. Chem. Circ. 71, 14 p.
4) Carpenter, C. W.
1915. Some potato tuber-rots caused by species of Fusarium. In Jour. Agr.
Research, v. 5, no. 5, p. 183-210, pi. A-B (col.), 14-19.
5) CzAPEK, Friedrich.
1905. Biochemie der Pflanzen. 2 Bd. Jena.
6) EllETT, W. B., and ToLLENS, B.
1905. Ueber die Bestimmung der Methyl -Pentosans neben den Pentosanen.
In Ber. Deut. Chem. Ges., Bd. 38, No. 2, p. 492-499.
7) Hawkins, L. A.
1915. Some effects of the brown-rot fungus upon the composition of the peach.
In Amer. Jour. Bot., v. 2, no. 2, p. 71-81. Literature cited, p. 80-81.
8)
1915. The utilization of certain pentoses and compounds of pentoses by Glom-
erellacingulata. /n Amer. Jour. Bot., v. 2, no. 8, p. 375-388.
9) Smith, Erwin F., and Sv/inglE, D. B.
1904. The dry rot of potatoes due to Fusarium oyxsporum. U. S. Dept. Agr.
Bur. Plant Indus. Bui. 55, 64 p., i fig., 8 pi.
(10) WollEnwEbER, H. W.
1913. Studies on the Fusarium problem. In Phytopathology, v. 3, no. i,
p. 24-50, I fig., pi. 5. Literature, p. 46-48.
HYPERASPIS BINOTATA, A PREDATORY ENEMY OF
THE TERRAPIN SCALE
By F. L. SiMANTON,
Entomological Assistant, Deciduous Fruit Insect Investigations, Bureau of Ento?nology
INTRODUCTION
One of the most effective enemies of lecanium scales is the coccinellid
beetle Hyperaspis hinotata Say. Its economic importance was impressed
on the writer during the seasons of 191 2 and 191 3, when he was studying
the life history and control of the terrapin scale (Eulecanium nigrofasciatum
Pergande). Throughout the spring and early summer the larvae, con-
spicuous by their flocculent covering, could be found in large numbers
feeding upon the immature scales and overturning the adult scales. The
adult beetles do not feed upon the mature scales, but they destroy the
young and also attack aphides, or plant lice, and other soft-bodied insects.
In view of the economic importance of this beetle a study of its life
history was undertaken at the suggestion of Dr. A. L. Quaintance, in
charge of Deciduous Fruit Insect Investigations, Bureau of Entomology.
The work was begun in the summer of 191 2 and completed in 1913.
HISTORICAL SUMMARY
Very little has been written about Hyperaspis binoiata. Say (i, p. 303),*
in 1826, described the male under the present name, and the female as
Coccinella normaia. G. R. Crotch (2, p. 380) considered the form with
the subapical red spot as a variety of H. signata Olivier, and gave as
synonyms H. hinotata Say, H. normata Say, and H. lettcopsis Melsheimer,
T. Iv. Casey (3, p. 124), in 1899, considered H. hinotata Say as a
distinct species and gave the following synonymy: H. signata Le
Conte, H. normata Say, H. affinis Randall, and H. leucopsis Melsheimer.
J. G. Sanders (4, p. 3), in 1905, mentions H. hinotata as a valuable
predatory enemy of Pulvinaria spp. J. B. Smith (5, p. 606; 6, p. 570),
in the same year, reported the same species as reducing an infestation
of Pulvinaria spp. at Montclair, N. J., from 500 to 1,000 scales to a leaf
to about one dozen scales to a leaf.
S. A. Forbes (7), in his annual report for 1908, mentions the species as
one of the principal enemies of Pulvinaria spp. in Illinois. In 1910, W. S.
Blatchley (8, p. 523), gives a key to the species of Hyperaspis found in
Indiana and remarks that H. hinotata Say is " a variety of H. signata Oliv.,
having the subapical spot lacking, color and structure otherwise exactly
as in that species." W. E. Britton (9, 8), in 1914, treats this species,
' Reference is made by number to " I,iterature cited," p. loj.
Journal of Agricultural Research, Vol. VI, No. s
Dept. of Agriculture, Washington, D. C. May i, 1916
do K— 30
(197)
198
Journal of Agricultural Research
Vol. VI. No. s
mentioning it as a great destroyer of the cottony maple scale {Pul-
vinaria vitis Linnaeus) and stating that it feeds upon both the woolly
maple-leaf scale (Phenacoccus acericola King) and the tulip scale (Eule-
canium tulipiferce Cook).
These references bring the history of the species down to the date of
the present paper, which deals with the life history and habits of the
species when feeding upon the terrapin scale.
DISTRIBUTION
H. hinotata occurs in most of the territory east of the Mississippi
River and extends west of this river in some States to the semi-
arid region. It is most abundant in the Atlantic States from Connecticut
to Maryland, but is common from New Jersey to Illinois. All localities
known to the writer are indicated upon the map (fig. i).
F^Tp^— ■^— ^~
/~j\ T \
^^
^^^ r-A/
('7-~X_7 L
k f
(|Q^]~^_^a
( (/S — t
\ M
tr^^y
\ \J-~-4— _ \ \J
■llsj--^
^a^M^
i^
^w"^
m^i
y~^ -\ \
Fig. I. — ^Map showing the distribution La the United States of Hyperaspis hinotata:
0=doubtful record.
HOSTS
►= definite record;
H. hinotata feeds upon honeydew, aphides, aphis eggs, and mealy
bugs and other soft-bodied scales. The larvse, so far as observed,
feed upon scale larvae and young scales. They seem to have preyed
originally upon species of Pulvinaria, to the egg masses of which the
larvae have a superficial resemblance. The species thrives upon the
terrapin scale and seems to be rather more abundant where it preys exclu-
sively upon this scale.
May I, i9i6 Hyperaspis hinotata 199
DESCRIPTION OF LIFE STAGES
IMAGINAL vSTAGE
The adult (PI. XXIV, fig. i, 2) is a small hemispherical beetle which passes the
winter in rubbish or under bark. It was described by Say (i) in 1826 from the
male as follows:
" Black, lateral margin of the thorax and head yellow; each elytron with a rufous
spot; body rounded-oval, convex, punctured, black, polished; head pale yellow,
labrum and transverse line on the vertex piceous; thorax with a yellow margin
extending for a short distance on the anterior margin; anterior margin with an obsolete
yellowish line interrupted in the middle; el}-tron each with a rufous, orbicular,
central spot."
EGG STAGE
The egg (PI. XXIV, fig. 3), which was first obtained by the writer in 1913, isoblong-
elliptical and somewhat depressed; 10 specimens measured from 0.6 to 0.775 ™ni-
in length (average, 0.704 mm.) and from 0.218 to 0.4 mm. in width (average 0.312 mm.).
In color it is light salmon, changing ultimately to ash-gray; the shell is membranous,
becoming indented with age. Hatching takes place through a longitudinal slit on
the upper surface.
IvARVAI^ STAGE ^
The first instar has characteristic markings, and represents a rather primitive type
of coccinellid larva. The other instars are similar to the first, but they are covered
by a white fleece of wax filaments which masks their characters.
First instar (PI. XXIV, fig. 4). — Length 1.22 mm. (1.125 to 1.275 mm.), width
0.478 mm. (0.450 to 0.575 nim.); body grayish white, semiopaque, cylindrical, and
tapering caudad. Head black, with a white trident spot over the epicranial and
frontal sutures; three pairs of. ocelli present; length 0.125 mm., width 0.225 mm.
Thorax sparsely pilose, the segments each with a pair of black dots; prothorax with
two black clouded areas surrounding, but mainly cephalad of the dots. Abdominal
segments each with a row of eight hairs and a pair of long lateral setae; ninth segment
black above; tenth segment, the so-called anal lobe, retractile.
Second instar (PI. XXIV, fig. 3, a). — Length 2.5 mm. (1.3 to 2.75 mm.), width 1.08
mm. ; body yellowish white, pubescent and covered with a white fleece. Head black
with the trident spot mildly obscured; length 0.175 mm., width 0.325 mm. Thorax
white, immaculate; legs gray, marked with black. Abdomen devoid of conspicuous
lateral setae.
Third instar. — Length 2 to 3.38 mm., mostly 2.5 mm.; width 0.9 to 1.75 mm.,
mostly 1. 125 mm. Head black, pigmentation on the posterior part of labium con-
fluent; length 0.275 to 0.3 mm., width 0.45 to 0.5 mm., mostly 0.475 ™m- Abdomen
with eight pairs of conspicuous blood pores. Otherwise as in the second instar.
Fourth instar (PI. XXV, fig. i, 2). — Length 2.5 to 6.25 mm., mostly 5.5 mm.;
width 1. 125 to 2.5 mm., mostly 2.25 mm. Body subglobose, yellowish gray. Head
glabrous, white, flecked with black, pigmentation on the posterior part of labium not
confluent on the median line; length 0.3 to 0.375 mm., mostly 0.35 mm.; width 0.575
to 0.65 mm., mostly 0.6 mm. Otherwise as in the third instar.
PUPAL STAGE
Pupa (PI. XXV, fig. 3, 4) inclosed within the larval skin; length 2.03 to 4.19 mm.,
mostly 3.9 mm.; width 1.77 to 1.86 mm.; color uniform chestnut-bro%vn ; ovate, with
a depressed segmented area on the dorsum; dorsal surface hispid; ventral surface
mildly pilose.
' A detailed morphological study of this larva by Dr. Adam Boving is in course of preparation.
36286°— IG 2
200 Journal of Agricultural Research voi. vi. No. s
HABITS AND SEASONAL HISTORY
THE BEETLES
The beetles emerge from hibernation at Mont Alto, Pa., about the
middle of April and commence mating about the 20th of that month.
When the species is feeding upon the terrapin scale, the beetles hibernate
for the most part at the bases of scale-infested peach {Amygdalus persica)
trees. After emerging from hibernation they soon depart in search of
food and do not return to the peach until the adult scale, which the
beetle is unable to destroy, begins to deposit honeydew — about the
middle of May. For the rest of the season the species remains upon the
peach, feeding upon the scale and its honeydew. The overwintering
beetles are nearly all dead by the middle of July, while the new brood
of beetles escapes from pupse for the most part during the first half of
that month.
There is some indication of a second brood, but there is not enough
evidence at hand to establish it.
THE EGGS
A very typical group of four eggs just as they were deposited is shown
in Plate XXIV, figure 3. It will be noticed that the eggs are not clustered,
but are placed more or less at random in the irregularities of the bark
adjacent to the host. The terrapin scale upon which the species was
feeding is found only upon young wood, the growth rings of which
supply a convenient shelter for the eggs of the beetle. It is not unusual,
however, to find eggs in crevices at the base of fruit spurs or even upon
smooth bark. It is worthy of note in this connection that the eggs
are not placed under the scales. It was found that the membranous
shell became dry and shriveled in from three to six days, and that the
egg changed to an ash-gray near the end of the incubation period.
The first eggs of the season were laid upon the twigs of scale-infested
peach trees at Mont Alto, Pa., on May 3, 1913, but were immediately
consumed by the beetles, as were all later eggs, until the food supply
became abundant. It was not until May 26 that eggs were permitted
to hatch. Oviposition reached its maximum about June 5, and con-
tinued in a small way until September i . Owing to the tendency of the
beetles to devour their eggs, it was not possible to determine definitely
the beginning of oviposition or the total number of eggs; 36 was the
largest number obtained from a single female, but there were indications
that several times that number had been deposited. Incubation lasts
from six to eight days; the average for 18 eggs deposited between June
27 and 30, 191 3, was seven days.
May I, 1916
Hyperaspis hinotata
201
THE LARV^
The larvae at the time they escape- from the &gg have the pigment
lacking from the head, legs, and ninth abdominal segment. They begin
searching at once for the terrapin scales ; and when one is found, a larva
enters the brood chamber through the anal cleft, where it remains during
the first and second instars. The first noticeable appearance of the
coccinellid larvae in the orchard, which occurs about June 18, coincides
with the beginning of reproduction of the terrapin scale. Once within
the brood chamber of a scale the coccinellid larva (PI. XXIV, fig. 4) preys
upon the new-born young of that particular scale until the end of the
second instar, by which time the rapidly growing coccinellid displaces
the scale.
The second molt is made in the open, mostly at the base of a fruit
spur. In the third and fourth instars many mature scales are destroyed,
being displaced (PI. XXIV, fig. 5) by the coccinellid larvae as these thrust
their heads into the brood chambers to secure the young scales. When
all the old scales have been destroyed, the ladybird larvae, which now
have a superficial resemblance to mealy bugs, migrate to the leaves and
there continue to feed upon such of the scale larvae as were able to reach
the leaves. It is estimated that a single coccinellid larva will destroy
90 mature scales and 3,000 larvae.
The length of the larval instars, together with the number of specimens
used in their determination, is shown in Table I.
Table I. — Length of the larval instars of Hyperaspis binotata
First. ..
Second
Third..
Fourth.
Number of
specimens.
17
Length of instar.
Average.
Minimum.
Days.
2.98
Days.
Maximum.
Days.
4
3
4
12
The dates at which the respective instars occur in the field are given
in Table II. The first and second dates show the time of greatest abun-
dance; the first and last dates show the total time of occurrence for each
instar.
Table II. — Sequence of the seasonal appearance of the larval instars of Hyperaspis binotata
in the field
Instar.
Date present iu field.
First
June 17 to 20 to Sept. 15.
June 20 to 22 to Sept. 20.
June 22 to 25 to Sept. 25.
June 25 to July 7 to Sept. 30.
Second
Third
Fourth
202
Journal of Agricultural Research
Vol. VI. No. s
The author has depended upon head measurements in distinguishing
the instars; a key for this purpose (Table III) has proved satisfactory.
As will be seen from the table, it is only necessary to consider the width
of the head.
Table III. — Key for detennining the larval instars of Hyperaspis binotata according to
width of head
Instar.
Width of head.
First
Mm.
0.225
•325
•475
. 600
Second
Third
Fourth
THE PUPA
The pupal period lasts for from 10 to 13 days, averaging 12 days.
Pupae appear in the field early in July and are most abundant from
the 7th to the 20th of the month. They are found, surrounded by
the last larval skin, attached to leaves or concealed in clusters under
bark. An occasional one may be found as late as October.
NATURAL ENEMIES
There seem to be very few enemies of this ladybird. No parasites
were obtained, and no birds were obser\'^ed to feed upon it. Aphis
lions were found preying upon the eggs, and a common plant bug,
Brochymena sp., was taken upon two occasions with this coccinellid
impaled upon its beak.
SUMMARY
Hyperaspis binotata Say is found
westward to the semiarid region,
bodied scales and is very effective
scale and the terrapin scale. The
deposited singly on twigs adjacent to
39 days and is as follows: Incubation,
instar, 2 days; third instar, 3 days;
days.
in the eastern United States and
It feeds upon aphides and soft-
in controlling the cottony maple
eggs are salmon-colored and are
the hosts. The life cycle requires
7 days; first instar, 3 days; second
fourth instar, 12 days; pupa, 12
May 1, 1916 Hyperaspis hinotata 203
LITERATURE CITED
i) Say, Thomas.
1826. Descriptions of new species of coleopterous insects inhabiting the United
States. In Jour. Acad. Nat. Sci., v. 5, p. 293-304.
2) Crotch, G. R.
1873. Revision of the Coccinellidae of the United States. In Trans. Amer.
Ent. Soc, V. 4, p. 363-382.
3) Casey, T. L.
1899. A revision of the American Coccinellidae. In Jour. N. Y. Ent. Soc,
V. 7, p. 71-163.
4) Sanders, J. G.
1905. The cottony maple scale. U. S. Dept. Agr. Bur. Ent. Circ. 64, 6 p.,
4%-
5) Smith, J. B.
1906. The cottony maple scale. Enemies of the insect. In N. J. Agr. Exp.
Sta. Ann. Rpt. [i904]-5, p. 605-607, illus!
6)
1907. The signate lady-bird beetle. In N. J. Agr. Exp. Sta. 27th Ann. Rpt.
[19051-6, p. 570-572, illus.
7) Forbes, S. A.
1908. Cottony maple scale in Illinois. Insect enemies. In 24th Ann. Rpt.
State Ent. 111., p. 114-115, illus.
8) Blatchley, W. S.
1910. On the Coleoptera known to occur in Indiana. 1,386 p., illus. Indian-
apolis. (Ind. Dept. Geol. and Nat. Resources Bui. i.)
q) Britton, W. E.
1914. Some common lady beetles of Connecticut. Conn. Agr. Expt. Sta.
Bui. 181, 24 p. (p. 8), 24 fig.
PLATE XXIV
Hyperaspis binotata:
Fig. I. — Male, showing the characteristic markings. Much enlarged.
Fig. 2. — Female, showing the dorsal view. Much enlarged.
Fig. 3. — Eggs and a second-instar larva, a, Second-instar larva as disclosed by
displacing the host; b, larvae of the terrapin scale, Eulecanium nigrofasciatum; c, a
displaced scale; d, eggs "in situ "; e, egg somewhat enlarged.
Fig. 4. — First-instar larva.
Fig. 5. — Method of attacking the mature scales during the third and fourth instars.
(204)
Hyperaspis binotata
Plate XXIV
Journal of Agricultural Research
V^l. VI, No. 5
Hyperaspis binotata
Plate XXV
" I
Journal of Agricultural Research
Vol. VI, No. 5
PLATE XXV
Hyperaspis hinotata:
Fig. I. — Mature larva as it appears when attacking the leaf -attached larvae of the
terrapin scale, Eulecanium nigrofasciatum.
Fig. 2. — Ventral view of mature larva.
Fig. 3. — Dorsal view of pupa, showing the last larval molt skin and the depressed
segmented area.
Fig. 4. — Ventral view of pupa.
ADDITIONAL COPIES
OF THIS PUBLICATION MAT BE PKOCUEED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT
10 CENTS PER COPY
SUBSCRIPTION PRICE, S3.00 PER YEAR
V
Vol. VI JvIAY 8, 1916 No. 6 *
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS
Page
Test of Three Large-Sized Reinforced-Concrete Slabs
under Concentrated Loading - - - - - 205
A. T. GOLDBECK and E. B. SMITH
Occurrence of Sterile Spikelets in Wheat - - - 235
A. E. GRANTHAM and FRAZIER GROFF
DEPARTMENT OF AGRICULTURE
WASHINGTON, D.C.
wA«HiNoTON ! ooveiNMEKT pniNTiNa omct t ist*
PUBLISHED BY AUTHORITY 01^ THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Physiologist and Assistant Chief. Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment StatioTis
CHARLES L. MARLATT
Attestant Chief, Bureau of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL
Biologist, Maine Agricultural Experiment Station
H. P. ARMSBY
Director, Institute of Animal Nutritian, The Penn-
syhatiia State College
E. M. FREEMAN
Botanist, Plant Pathplogist, and Assistant Dean,
Agricultural Experiment Station of the Univer-
sity of Minnesota
All correspondence regariding 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 Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine,
JOURNAL OF AGRICDLTffiALlSEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, May 8, 1916 No. 6
TESTS OF THRKE LARGE -SIZED REINFORCED -CON-
CRETE SLABS UNDER CONCENTRATED LOADING
By A. T. GoLDBECK, Engineer of Tests, and E. B. Smith, Associate Mechanical Engi-
neer, Office of Public Roads and Rural Efigineeri^ig
INTRODUCTION
Numerous instances occur in reinforced-concrete design in which the
use of slabs supported at two ends only is required, and in many such
cases the critical loading is concentrated at one or more points. Such
a condition may exist on slab-bridge floors, box culverts, on floors of
buildings where heavy machinery is housed, and in other constructions
where loads are concentrated.
If a slab, supported at two ends and carrying a single concentrated load,
is imagined to be divided into narrow strips extending from support to
support, it would seem reasonable to assume that the strip immediately
under the load carries a ver}^ large part of it and that the adjacent
strips receive a smaller amount, depending upon their distances from
the load. The most remote strips, those at the edges of the slab, would
then probably receive very little load. The question which concerns
the designer of such a slab is that of the relative magnitude of the stresses
at different distances from the load.
Up to a few years ago the technical literature on this subject was prac-
tically nonexistent, and the result was that engineers relied largely on
their judgment when called upon to design slabs subjected to concen-
trated loads. Very naturally, large variations in load-distribution
assumptions were made, and as a consequence there were great differ-
ences in the design even when the span and load to be carried were
practically identical.
The necessity for definite knowledge on this subject was very forcibly
brought to the attention of the engineers of the Office of Public Roads
and Rural Engineering a few years ago, and a set of tests was made by
one of the authors on slabs of 3-foot and 6-foot span length.^ These
tests gave some useful and rather surprising results that have since been
' Goldbeck, A. T. Tests of reinforced-concrete slabs under concentrated loading. In Amer. See. Testing
Materials, Proc. i6th Ann. Meeting 1913, v. 13, p. SsS-.'iyj, 10 fig. 1913. Discussion, p. 874-883, 4 fig.
Journal of Agricultural Research, Vol. VI, No. 6
Dept. of Agriculture, Washington, D. C. May 8, 1916
dj D— 8
(205)
2o6 Journal of Agricultural Research voi. vi. no.6
verified ; and in order to carry the investigation farther, with slabs of longer
span than those previously investigated, the present series of tests was
undertaken at the Arlington Experimental Farm of the United States
Department of Agriculture.
OBJECT OF INVESTIGATIONS
The theory applied to the design of narrow rectangular reinforced-
concrete beams involves the assumption that the stress is constant
throughout the width of the beam. In a wide slab the stress distribu-
tion varies from a maximum at the point of application of the load to a
minimum at the extreme edges. Obviously then, if the rectangular-
beam theory were applied to the design of slabs under concentrated
loads, the width h used in the design formulas can not be taken as the
entire width of the slab. The rectangular-beam theory, however, could
be utilized in wide-slab design if it were known what width h should be
substituted in the design formulas, and it is the object of this paper to
explain tests for determining this width and to demonstrate the appli-
cation of the theory of narrow rectangular beams to the design of wide
slabs supported at two ends and subjected to concentrated loads.
EFFECTIVE WIDTH
The width of the slab that should be used in the rectangular-beam
formulas when applied to slab design will be termed the "effective width"
of the slab. It is that width over which, if the stress were constant and
equal to the maximum stress under actual conditions, the resisting
moment would equal the resisting moment of a slab of the same depth
and full width, but having varying stress distribution. If the straight-
line theory of stress distribution from neutral axis to upper fibers is
assumed to be applicable to slabs, the resisting moment of a given slab
is dependent on the total stress in the concrete or steel at the dangerous
section. The total stress in the concrete, however, is governed by the
stresses in the top fibers, and these stresses are proportional to the unit
deformations. If, then, there are two slabs of equal depth, one having
uniform distribution of deformations and the other a varying distribution,
but with their maximum deformations identical, they will likewise have
equal resisting moments if the summations of the deformations over their
respective widths are identical.
In figure i , which represents a slab in position on two supports with a
concentrated load P, is illustrated the method of obtaining "effective
width." Strain-gauge readings are taken of the fiber deformations per-
pendicular to the supports, as indicated at eg. These concrete deforma-
tion values are plotted to scale, as, for instance, at fh, giving the deforma-
tion curve JHF, inclosing the area AJHFE. This curve shows the varia-
tion of stress from the center to each of the two free edges of the slab, and
the area under the curve is a function of the total concrete-resisting
May 8, 1916
Tests of Reinjorced-Concrete Slabs
207
moment of the slab. The area BDGI, obtained by dividing the area
AJHFE by its maximum ordinate CH, has the same total concrete-
resisting moment with the stress uniformly distributed as the whole slab,
and its width BD is that which may be effective in furnishing sufficient
resistance under these conditions to carry the load. The width BD,
obtained in this manner, is the "effective width."
DESCRIPTION OF APPARATUS
Load-applying apparatus. — The slabs tested were 32 feet. wide, with
a span length of 16 feet, and in order to accommodate such extraordinarily
large test specimens it was necessary to build special apparatus. Two
supports 32 feet long were constructed of reinforced concrete, and em-
bedded in each of them at the center were two loop-welded eyes car-
rying four 24-inch 80-pound I beams 6 feet above the level of the supports
Fig. I . — Diagram illustrating the method of obtaining ' ' effective width ' ' in reinforced-concrete slab tests.
(PI. XXVI). Loads were applied by means of a hand-operated hydraulic
jack mounted between the slab and the overhead I beams.
For weighing the loads a specially calibrated chrome-nickel beam
(PI. XXVI) was mounted between the jack and the load-applying I beams,
and its deflection at the center was a measure of the load applied. This
chrome-nickel beam was 7 inches wide, 5 inches deep, and 27 inches in
span, and its deflection was measured with an Ames dial reading to o.oooi
inch. The dial was fastened to the beam and its plunger rested on a
_J/^-inch square steel rod mounted on the side of the beam at the neutral
axis. It was found that by fastening an electric buzzer on this rod
more consistent readings could be obtained with the dial. The entire
load-applying device was calibrated in a 200,000-pound universal testing
machine, and the beam deflections corresponding to known loads were
obtained. A deflection of approximately 0.0001 inch occurred for each
500 pounds of load applied. A number of calibrations were made and a
calibration curve was plotted. When used for measuring loads, it was
only necessary to read the central deflection on the Ames dial and the
corresponding load could be read from the curve. '
2o8 Journal of Agricultural Research voi. vi, no. e
Deformation-measuring apparatus. — Deformations of the top of
the slab were measured at right angles to the supports, and also, in the
case of one slab, parallel to the supports, with a Berry strain gauge of
2o-inch gauge length. The degree of accuracy attained was probably
within 0.0002 inch in that gauge length. Short brass plugs drilled at one
end with a No. 55 drill were embedded in the concrete, or in some cases
cemented in holes drilled for the purpose; and the movements of these
plugs as measured with the strain gauge were considered the fiber defor-
mations. •
In the last slab tested (No. 934) deformation readings were also taken
of the steel reinforcement, and for this purpose holes were drilled in the
steel bars 20 inches apart to accommodate the points of the strain
gauge. Although readings were not taken on all of the bars, a sufficient
number were measured to determine the distribution of the steel stresses
throughout the slab. The layout of strain-gauge points between which
readings were made is shown in figures 2, 3, and 4. The arrowheads
mark the position of the points on the top of the slab and in the case
of slab 934 (fig. 4) the gauge points in the steel are marked by small
circles.
Deflection-measuring apparatus. — The deflection measurements
were made in somewhat different ways during these tests, and the ap-
paratus was improved as the tests progressed. In its final form in slab
934, the deflection-measuring equipment consisted of a network of piano
wires stretched tightly at a fixed distance above the concrete supports,
and being entirely independent of the slab. At the points where measure-
ments were taken, steel plates were set in plaster of Paris on top of the
slab. Readings were then made between these plates and the wires by
means of a specially designed instrument consisting of a brass stand
carrying a bell-crank lever, one end of which touched on the piano wire
above and the other end bore on the plunger of an Ames dial. By means of
a slow-motion screw the end of the bell-crank lever was adjusted to touch
the wire as indicated by an electric buzzer. The dial readings taken at
different loads then indicated the deflections at the various points on
the slab. This instrument is probably a more convenient form of measur-
ing device than the ordinary inside micrometer and is accurate to
0,004 inch.
DESCRIPTION OF SPECIMENS
All three specimens were 32 feet wide, 16 feet span, and were made
of machine-mixed concrete in the proportions i to 2 to 4. Potomac
River sand and gravel were used as the aggregates, mixed with Port-
land cement. A rather wet mix was used, and the work of molding was
done by laborers at the ArHngton Farm who were experienced in work
of this character. There was no attempt to make the concrete any
better than it would ordinarily be made in the field, but efforts were
May 8, 1916
Tests of Reinjorced-Concrete Slabs
209
directed to secure work thoroughly representative of that obtained
under field conditions. The sand was a good grade for use in concrete,
and the gravel was clean, well graded, and free from weak pebbles.
The steel reinforcing consisted of ^-inch plain square bars in slabs 835
and 930, and the bars in slab 934 were ^-inch square. The yield point of
this material is. about 39,000 pounds, and the ultimate strength 60,000
pounds per square inch.
The slabs were necessarily built in place on their supports, and the
forms were struck at the end of about two weeks. The concrete was
sprinkled daily for several weeks during the earlier stages of hardening
and was allowed to cure protected from the weather until the destruc-
tion of the slab.
Table I contains the essential data concerning the slabs tested.
Table I. — Description of reinforced-concrete slabs used in tests^
Serial No.
Thickness.
Reinforcing.
Modulus of
elasticity
of concrete.
Central
Total.
Effective.
Size.
Spacing.
Per cent.
load
of slab.
835
930
934
Inches.
12
10
7
Inches.
6
Inches.
K (plain
square).
K (plain
square).
K (plain
square).
Inches.
10.5
8.87
5-56
0-75
•75
• 75
2, 900, 000
4, 000, 000
3, 000, 000
Pounds.
119, 000
80,000
40, 000
' The slabs were not reinforced transversely.
At the time the slab specimens were made, 8 by 16 inch concrete
cylinders were molded from the same mixture and were allowed to cure
under the same conditions as the slabs. These were tested later for their
crushing strength and modulus of elasticity.
METHOD OF TESTING SLABS
At the age of 28 days the initial strain-gauge and deflection readings
were taken with no load on the slab. The first load was then applied
through an 8-inch cylindrical bearing block set in plaster of Paris at the
center of the slab. Strain-gauge and deflection observations were made
again over the entire slab. Due account was taken of the air and concrete
temperatures in order to make corrections for any appreciable change
occurring during the progress of the tests. The increments of load
applied to the difi"erent specimens were varied in the diff"erent slabs,
depending on their thickness, and the aim was to stress neither the steel
nor the concrete beyond working limits, also to obtain about five incre-
ments of load within the working load.
2IO Journal of Agricultural Research voi. vi, no. 6
After readings over the entire slab had been taken, check readings were
made at various points; and invariably it was found that these check
readings showed an increased deformation in the concrete even though its
temperature remained constant. Moreover, upon releasing the load en-
tirely it was found that considerable permanent deformation remained in
the concrete. This phenomenon can be attributed only to the "flow" or
gradual change in length of the concrete even when under small stresses
and is significant, for it shows the importance of the time effect on the
relation of stresses and strains in concrete. If the strain readings on
the top of the slab, loaded for five or six hours, be used to estimate the
stresses in the concrete, based on the initial modulus of elasticity of the
concrete, this estimated stress will be greatly in excess of the true stress
conditions.
In view of the fact that the deformations which take place in the
concrete under a sustained load are continually increasing and remain
partially permanent, and that the only deformations of value are those
indicative of the stress, all of the final calculations and deductions are
based upon results obtained by taking zero deformation readings just
before applying the load. Deformations thus obtained by taking the
difference between the strain-gauge readings at the zero load and the test-
ing load (all within an hour or so), represent more accurately the elastic
deformations and are a better indication of the stress existing in the
concrete than those obtained from any initial or previous zero readings.
GRAPHICAL REPRESENTATION OF DATA AND RESULTS
A great amount of numerical data has been taken during the tests of
these three concrete slabs. Some of these data were preliminary and
served only to indicate methods and limits. Those data which have a
direct bearing upon the problem are shown graphically in the accom-
panying curves (fig. 2-28).
Figures 2,3, and 4. — The layout of the points in the concrete and the
steel over which the strain-gauge readings were taken are shown in figures
2, 3, and 4. In a few cases readings were m.ade between all points, but
in general only the readings along a center line (5-6) parallel to the
supports were taken, as this gives sufficient data for determining the
effective width. In all mention of strain-gauge or deformation readings
it should be understood that they are measured between points on a line
perpendicular to the supports, unless expressly stated to be otherwise.
Figure 5. — Figure 5 shows the variation of the concrete deformations
for different concentrated center loads, along the center line of the slab.
The ordinates of these curves are influenced slightly by the time factor
or "flow" in the concrete; hence, the values for the effective width b are
somewhat erratic in their relation to the load.
Tests of Reinforced-Concrete Slabs
211
L
'«'S,0;9I-
»
■
-H
1 1
1 — ;0e—
--'-^02-'
<■- — oi—'
■-—02 — '
\ct.
lO— — —
1 4-
1 C
{g ■
i
1
°-
1 3
j(0
1 01
0-
z ■
IZ'
•
1
1^ .
1'^
/-
"X
b
n
\,
SW.Corner*
jx— — —
jo- .
IQ. .
I(J, ,.
{(O ■
1 "bi 6"
e A
9
9 f C
Z
1
1
212
Journal of Agricultural Research
Vol. VI. No. 6
iJodS 0-9I -
—pZ-
Zf- ^—
9 1'?
-.02-
V
May S, 1916
Tests of Rein forced-Concrete Slabs
213
1-
— .'©.91 —
—
—
—
—
1
H
f-<««iS ,o»i -j
1
o'X
2 i
-c?'
6
e
^
"-cf
S
^ -
«
j
■1
lb
1
- 4
i?5-
--:OE- -
— -oa— -
— :
w —
-— .oz ^
-oa^ — H
— ;Cr2-^
— .OE - -
—^03- —
f— .02 —
1
1-.
,
-1 1
i£
i«
■
A
O
a
'0
-■,
'-5
*"*
'~>
x-i-
">
' "^
1I
->'
-
[~
a
ji
w
1
01
5
,•-
,^
,?
,•=
,1
'"t
:
1
1
?
9
t
..
r:
♦ .
..
. ti
.;
j
ir
r ]
,-1 1
ll
■-B
-i
J
._J
f- — s^
^
">i4=^
4
4
*ml
^
s,
I-l
' *•
--».
s.
S)
5.t
P
IL.
1
u.'
1
»
.
c
J,
»
s
Sl
s
1^
J
uj]
1
I
it i
5
s
I
I
si
u
'id 1
■l
«
-^
-s
.^
^
■^
-^
-S
■'t
.1
[0
Ql
...L'^
\l
-s
e
.
^
.
..
^
^1
•0
(11
10
6
a
a
cp
-
[0
0]
Id
IDI
«&.
e
e
/.
>0
s
. «J
z
;+
s
1
1
.„,
i 1
'■?
t
214
Journal of Agricultural Research
Vol. VI, No. 6
FiGUREC 6. — Two curves, A and B, are shown here to indicate the def-
ormations which resulted from the removal of the forms. The flow, or
increase in the deformations, is about 8o per cent in three days. The
curves C-D, K-F, G-H, I-J, and K-L show the large difference in the
deformation and effective width values between those obtained by the
use of a zero strain-gauge reading taken several weeks before, with sev-
eral intervening loadings, and those obtained from a zero reading taken
just before the loading. The data and results of curves C, E, G, I, and K
are the only ones of value.
A
20. OC
0 lb:
C I
■> E
F
e
H
I ,
K
L M
M
0 1
■ 0
R
5
~~-
^■■
wv.
FT -7 2-55% SPA
(
?5 0(
10 \.B
B
-..^
\>=
11-50
T-T
19«fS
SPAN
-^
30, oc;
0 LB
>.
—
^
~^
C
1
/•
^
b
llOl
FT.= e
8-8/,
SPAN
3?,>iC
OLfS
—
■
0
^
/^
^
b=
12-11 F
T-=72
Tf,i
PAN
-^
SCA
\
>001 !
<• UN
TPE!
OPMA'
ION
Fig. 5. — Concrete deformation curves for concentrated center load on slab 835.
Figure; 7. — The difference between these curves shows the magnitude
of the set, or permanent deformation, which may occur between two
applications of the load, each loading having been applied immediately
after a zero reading of the strain-gauge points, with 24 hours intervening
between the loadings. The second application shows a smaller deforma-
tion than the first. This is true for both the concrete and the steel
deformations. The effective widths are based upon the first application
of the load.
Figure 8. — These curves are shown to emphasize the importance of
considering the time factor and its effect upon the deformations in con-
crete structures. Curve i shows the immediate effect of the load. After
about 5 hours the load was removed, then again applied 20 hours later
May 8, 1916
Tests of Reinforced-Concrete Slabs
215
and allowed to remain on for two days, giving curves 2 and 3. The load
was then removed, and curve 4 shows the amount of set about two hours
later. This set is somewhat reduced after a few days' rest. The values
of the effective widths shown in this figure differ very largely and are
also indicative of the fact that the time factor is very important.
A
0 LE
5.SH(pWINC
D
CFFE
L
CT 0
F
10VAL
s +
OF F
1 I
3RMS
»
•^
L
M
^
■0
3-
MOUf
IS Ar
"ER R
EMOV
VL
19 t
AYS
3
DAYS
AFT
:!? RE
MOVA
L
22 t
AYS
B
20.0c
)0 LI
IS.
33 [
AYS
'
^rt?<
)M NE
WZD
— b
!0 TA
LOAC
I3.04
\EN V
(NG.
FT'S
lUST
25^
BEFOI
E
r
Ffro
M OR
GtNAl
- ZCR
D
p
76.0
)0 LI
SS
^
-y'
-H
r"*
AKEN
AT T
IE AG
tor:
100 1
JOAY
JAYS
s
— c
■FP?0
1 NtV
V ZEI
0
b.|2
97 F-
: •an
%OF
SPAN
E
F
'
""
—
^
/
•^RC
M OF
iGlUt
L ZEI
L
\
^
^
/"
b-i;
62 Fl
•98.
9%C
F SP/
N
35.0
>o U
>3.
100 I
AYS
■
—
r
FROM
NEW
ZER
>
b»u
il FT
•90.'
%o(
SPA
4
0
M
\
^
^
^
^
\
/^
-^RO^
b'lO
ORK
11 FT.
;iNAL
•100
ZERO
r<j^oi
SP/>
■•1
-^
15,0c
0 LI
5S
46 C
AYS
ROM
\EW ZERO
b"li,
OF Sf
10 FT
■716'
M
J
' 1
^
—
-^
-FRO
,1 OR
.79 r
3INAL
:-l23
ZERC
" SPA
SI
22.5
DO Li
13.
•42 q
AYS
■
FIJOM
NEW
:ero
b-i:
1.71 n
•03.7
%OF
SPAN
L
___
_^
^^
raot.
ORIC
INAL
:cRO
SCAL
:00c
i
01 IN
UNIT
OtFQ
BMAT
ON
Fig. 6. — Concrete defonnation curves for slab 930.
Figure 9. — Concrete deformations under 2-point loadings are shown
for two-load values. The 40,000-pound load was applied immediately
after taking the zero reading, and the deformations taken at once. The
load was then increased to the 80,000-pound value and deformations
again taken. The whole operation required not over two hours. The
local effect at the load points is very pronounced for the larger load.
2l6
Journal of Agricultural Research
Vol. VI, No. 6
A
6
C
0
E
F
L H
I
J
K
L
M
/^
L^SC
r— '
00 L
BS,
,NOV
J
■
ul
:^--_
__^.
-—
'S^
— 1
-J
'NOV
■AN
2
— ^^^
b=i
►7 FT
=7S>S
•/. I7J
UJ
L»10C
00 L
B5.
0
O
==^
F^~'
_,.
___.
"~
^^
C"
.
P
b=n
5 FT
= 109
.-?;?.(
AH
^'
r
L'5<
OO I
BS.
'■
— >B=
^
^-.-
—
;^
:z^
^^
~ — ■
"--
■''.
(7=14
IFT =
881
£ 5F>
.N
J
5CAL
;=oo
501 Ul
(IT Of
FORMfflON
u
L=10
300 V
BS.
r.^^
i.
:^
^^
\,
^^
■'^-
•~-^
„^
--■'•
'"^
..^
^
^
-^
b«i7
3 FT..
111.2;
iSR
kN
-^
FIPS
6Eca
-NO
• APPl
NPA
ICATIO
•PLIC
N
(AFTE X REM IVW^ 1
-
1
Fig 7. — Deformation curves for slab 934.
T
■
A
i
i
C
i
i
E
F
1 ■
' 1
I
t
K
\
1
r
2
1
£8
—■:!
~ —
— :::
^
^
^-'
;;;::::^
•^^^^
^_^^-.
- —
"HI
EFFKT
lYEbl
ox NO.
2<1V»
FT»05
FT-in
?^
^v
^^
--"'
""^
§
\
FT.-I14
FT.-107
^7-
\
^v
_.-^
.//
SCA
.E-O.OC
31 IN.
INIT C
EFORl
1.
■
^5=
—~^
4
_____
•
z
9
"^
J7^
■^
-^
iz
"^■"
P^
--'^
"^'
\
S?
EFFECT
vt bf
ENO.I
■ 2
.«i)8F
19J&F
>1».7'
f ^
^
"^
;.-^
•0
•5
'A,
'»07F
r-.iiaz
(
Fig. 8. — Deformation curves for slab 934, computed from first zero reading.
May 8, 1916 Tests of ReinfoTced-ConcYete Slabs 217
TW
> POl
r-jT u:
AD
A
A0,0<
s
0 I.B
c c
C
r
6
H
I J
K
L
M
N
0 P
«
R
s
— ■
.
■ —
k-— '
b. i<
>3e F
T = 10
)-4f.
SPAN
80, OC
0 LR
I.
^
N,
-/■
^
y
^
■
b-2
:52 r
r=i4c
.7^5
PAN
1
POS
TION
OF
L
sl
LOAC
TWC
INQ
POl
IN
5
>.LE
aooo
1 IN. I
INIT C
EFOR
M.
Fig. 9. — Concrete deformation curves for slab 835 with 2-point loading.
P,o
OOIJB
A
B
c
0
E
it
r G'
0, e
1 1
I
J
M
1
A
■
■^-
---
-—
b-I
rser-
r-io7
»%o
-aw>
<<
^
-— ■
-—
b-
yrrz 1
T.-oe
.2%<
)rsp«
kN
''T'
—
-—
---
---
^
b-
18.73
-r-io
-^5-!?
or s
>AN
■Xil
jQOif
i
^^
^
,y
' '
V.
y'
X
-^
b-I
75 n
•.OC
-%
or s
PAM
3C
aix*
i300<
.....
^N.T C
crew
tUTK)
-
I'lG. 10. — Concrete deformation curves for slab 934 with a-point loading.
2Ii
Journal of Agricultural Research
Vol. VI, No. 6
The effective width is not materially affected for the 40,000-pound load;
but for the 80,000-pound load, which produces the working fiber stress,
the effective width is very largely increased. .
Figures 10 and ii. — The curv^es on these figures show a more pro-
nounced local effect in the concrete at the load points than the same
character of loading on the thicker slab. It should be noted that for the
working load of 20,000 pounds the effective width for this 2-point loading
is the same as for the single-point center loading.
Figures 12 and 13. — The results for 4-point loading under different
loads are shown in these cur^^es for slabs 835 and 934. The effective
TV
■0 PC
INT L
OADl
•K^
1
c
\
E
\
a
I
* 1
K
J
A
50C
0 LB
!
,^
■
-
R
IC50C
10 La.
'•^~
—
1
— ■
\5,00
> lb£
■-
-
-^
C
-^
^
■^
■^
20, 0(
0 lb:
:
P
^
^
"^
■^
A
^
— ^
"^
CALE.
iooo
IN. U
NIT C
EFORI
wriol
(
Fig. II. — Steel deformation curves for slab 934 with 2-point loading.
width is materially affected by the width between the load points; it
seems to be increased by not less than 56 per cent of the span length for
slab 835, and 93 per cent for slab 934.
Figures 14, 15, and 16. — The deflection data are shown on these
figures. The curves are plotted to show the deflection values along
a center strip parallel to the supports. In figure 14 cur%"es have been
plotted showing the flow^ and set in the slab under a sustained load and
as effected by two applications. Two values for effective widths are
shown, which have been obtained from the deflection curves in the
same manner as from the concrete deformation curves described above;
but these values should not be used in the design of slabs.
May 8, 1916 Tests of Reinjorced-Concvcte Slabs 219
FOUC
POlN
T U3A,01NS
-woe
0 L&
c t
■ z
F
S
H
I
1 K
L
M
N
0 f
0
R
a
A
'
b=l
IA4.F
T = 9
'•T%
)PAN
1
&O,0
00 L
is.
,
b-i£
.7©f
•« in-
17,6
PAN
100,0
^OLE
s.
—
,
■04 F
■
--
'-'
b=2
t:=I3I
5f»S
PAN
?
J^
posn
LOAW
lOMOF
N6 0^
.
'r-
1J
■2
1
FOUB
POINT
i
5CAL
.= 0.0(
01 IN.
UNVT
OEFO
'MAT!
OH
s- *""
1
Fig. 12. — Concrete deformation curves for slab 835 with 4-point loading.
i
1
1
1
4 pqiNT I
OADi
-IS
A i B
c
D
E >
F <
H
i I
I
K
L
M
A
0,00
5 uas
-
, —
b
= 24^
n: =
,33.a
7. C
F &R
VN
i
B
10,00
) UK
1 — i — — \ — '
.
. 1
!
b=30.^ FT =
1
168-S
% J
F SP/
N
1
1
1
1 I
;J3
■
POSl
r.oN '
IF
3?' i—
FOUR
POl^
TS
SCALf
«0.0<
01 IN
UNV
r OE -ORM
^,T10^
I^G. 13. — Concrete deformation curves for slab 934 with 4-point loading.
220
Journal of Agricultural Research
Vol. VI, No. 6
Figures 17, 18, and 19. — After each slab was broken the cracks in
the top and bottom were drawn to scale. The heavy full lines forming an
approximate circle or ellipse around the load point are the tension cracks
on the top of the slab caused by the overhang of the ends, after a large
center deflection, at about breaking load. The remarkable symmetry of
Fig. 14. — Deflection curves for slab 934 on first application of load.
these cracks is worthy of notice. There seems to be no definite relation
between the effective width at working loads and the width over which
the cracks extended at failure; in fact, it is hardly reasonable that there
should be any definite relation, for one case is dealing with safe working
stresses within the limit of elasticity, and the other with breaking loads.
May 8, 1916
Tests of Reinjorced-Concrete Slabs
221
Table II shows the breaking loads and their relation to the depth of the
slab. Note that the breaking loads are almost directly proportional to
the squares of the depths.
'
, ^
1
■c
f-
/
.^
/
^
i
^
f^^^a^
85
1
I
0/
i
1
1^
o7
0
in'
— r-
y
T
\
■c ^
•10,0c
-5,00
10 LE.
0 lb;
5.
^ 1
--■
^
"-^
~
^
\
\
k
1
>^
i-
— h-
M'i^
\
1
~^
\
\\
0
„
-**==
,L-..
: o.'oe
IN
1
t
6-9-
-'
Fig. 15. — Deflection curves for slab 934 on second application of load.
Table II. — Breaking loads of reinforced-concrete slabs and their relation to the depth of slab
Serial No.
835
930
934
36287°— 16 2
Effective
thickness, d.
8K
6
no. 25
72.25
36. 00
Breaking
load.
119, 000
80, 000
40, 000
Relations.
3.06
2. 01
I. 00
Loads.
2. 00
I. 00
222
Journal of Agricultural Research
Vol. VI, No. 6
STRESS DISTRIBUTION OVER THE WHOLE SLAB
For the purpose of determining the distribution of stress over the top
of the whole slab, deformation readings at right angles to each other
were taken on slab 934 for a working load of 10,000 pounds concentrated
at the center.
^^
^_
1
^
^
^
^
^
^
w
jl
1
J
^
y
^
Hi
f
::^
k
^^
—f —
T
^
/
/
„-/
/ J
/
'
ll
A
n
N,
vt
^
W
^,
^
^
'^
—
:r:i
^
^
1 —
«
_jd
~~~j
- — ^
in
in
<a
_l
r'
0
J
0
0
\
0
0
0
0
0
\\-
t
-^
^
Jr
■1 ■
^
^
^
"\
\
— *"H'
\ "\
" r
J^l
^^
\
\V
\
\\
Itt
;::>-
SCAI
.E.
.0
oeir
.
^
^
^
4^
'^==:;
^
^
Fig. 16. — Deflection curves for slab 934 vrith 2-point loading.
Figures 20 and 21. — The deformations measured perpendicular to the
supports and plotted on base lines parallel to the supports are shown in
figure 20. These curves show the variation of deformations along lines
parallel to the supports. The same deformations plotted on bkse lines
perpendicular to the supports, to show the variation in that direction,
May 8, 1916
Tests of Rein forced-Concrete Slabs
223
are plotted on figure 21. Each curve as shown is an average of the
plotted points. The light vertical lines serve only to locate each curve
with its base line.
•
32
•-0-
f
^"1
~^r~""
1
>
>
}
Pl' C ' 1 \
-^
J'-o'-H ^ •'
\
/ -^
\ /
"X
\i
V;\i 1
6'-2'-
6'-0'-
h
Fig. 17. — Diagram showing effect of breaking load on slab 835.
The variation of the distribution along lines parallel to the supports is
is somewhat gradual and does not show any sudden changes; but the
-32'-o'
Fig. 18. — Diagram showing effect of breaking load on slab 930.
variation across the span near the center of the slab becomes somewhat
critical at and near the load point, and this was more pronounced in the
concrete than in the steel. (The steel data are not shown.)
224
Journal of Agricultural Research
Vol. VI, No. 6
Figure 22. — Lateral strain-gauge readings were taken on points paral-
lel to the supports over the middle third of the slab, and these are plot-
ted on base lines both parallel and perpendicular to the supports. The
groups of closely drawn parallel lines serve only to connect each curve
with its base line. Compression values of the deformations are plotted
either to the left or below the base lines, and to the right or above, for
values of tension in the concrete. The variations in these lateral de-
formations are the reverse of those of the longitudinal deformations
shown in figures 20 and 21 ; they are more critical along lines parallel to
the supports.
Figure) 23. — The data of the last three figures have been collected and
plotted as " iso-deformation lines," giving a series of lines or contours
32 FT.
i _
Fig. 19. — Diagram showing efiect of breaking load on slab 934.
which represent equal deformations in the concrete on the top of the
slab. The lines, as drawn, are averages of the plotted points. Figure
23 (also fig. 26) is more for academic interest and should be of service in
the theoretical consideration of stress distribution.
Figures 24 and 25. — These figures are similar to figures 20 and 21, and
are plotted in the same manner, except that they represent the distri-
bution of deformations under a working load of 40,000 pounds applied
at four points. No lateral deformation readings are shown. The load
points are indicated in figure 25. The local effect at the loading points
is very pronounced.
Figure 26.— The data of the last two figures mentioned have been here
collected and show the "iso-deformation lines" for the 4-point loading
of 40,000 pounds, total. (See description of figure 23.)
May 8, 1916
Tests of Reinforced-Concrete Slabs
225
226
Journal of Agricultural Research
Vol. VI, No. 6
May 8, 1916
Tests of Rein forced-Concrete Slabs
227
228
Journal of Agricultural Research
Vol. VI, No. 6
May 8, 1916
Tests of Reinforced-Concrete Slabs
229
1 ~ J
230
Journal of Agricultural Research
Vol. VI, No. 6
May 8, 1916
Tests of Reinforced-Concrete Slabs
231
CONCLUSION
If figure 27 is referred to, the influence on the effective width of the
magnitude of the load and the manner of interpreting the results may be
Fig. 26. — Iso-defortnation lines for slab 934 under 40,000-pound 4-point loading. Deformations measured
perpendicular to supports.
seen. It has been pointed out that the correct method of obtaining
deformations is to base all calculations on zero readings taken just before
ISO
»20
/
\
no
NO. 9
5.*-e
:n.
/
\
100
A
Vi
NC
I930
-8J,
IN.
00
/
0
NC
©30
L ZE
-eii
JO
^
rtO
/
/
^
-^
^
--
■—^'
NCW
zcpo
^
^
-NO e
135 -l<
>ilN.
>
/o
60
no
'^O
^0
70
10
0
c
c
c
1
)
>
c
c
c
>■
c
(
>
S
[
1
•
>
1
J
!
'
LOAD CONCCMTHATCD AT CCNTCR
Fig. 27. — Curves showing efifective width versus load (concentrated center load).
the load has been applied (designated on the curve as "new zero"). In
the case of slab 930, figure 8, note the difference in effective width obtained
232
Journal of Agricultural Research
Vol. VI, No. 6
Z lOO
depending on the manner of considering the zero readings. The more
conservative values are obtained by basing the calculations on the "new
zero" readings, as was done in the case of slabs 930 and 934. Note that
with an increase in load, the effective width seems to increase slightly.
Values for effective width were obtained from the steel deformations, as
well as from the concrete deformations, but it was found that the
concrete deformations
gave the most conserv-
ative widths, and these
were therefore plotted.
In figure 28 the effect
of variation in thick-
ness of slab on effective
width may be seen.
Note that as the thick-
ness increases, the ef-
fective width decreases,
varying from 109 per
cent of the span length
for a 6-inch slab to 75
per cent of the span
for a loj^-inch slab.
The least value for ef-
fective width shown by
these tests is roughly,
then, about 0.7 of the
span length. Judging from the curve of variation, it would seem that
under extremely heavy loads, requiring very thick slabs, the effective
width might be decreased as low, possibly, as 0.6 of the span length.
However, 0.7 of the span will always be safe, and in general is a sufficiently
conservative figure to use.
Table III. — Effective widths of reinforced-concrete. slabs, 16-foot span by 32 feet zvide,
for center loading
\
HO. 9
3A
\
\
\,
\
N^
).93<:
^JSI
xaas
EFFCCTIV-E TMICKNCSS . IN MSCMES
Fig. 28. — Curve showing effective width versus thickness.
Center load.
Pounds.
15,000. . .
20,000
25,000.
32.500 ■
35,000.
Safe load . .
Slab 83 s (ioJ4 inches effect-
ive thickness).
ii.6feet=72.3 percent
of span.
11.5 feet=7i.9 per cent
of span.
12. 1 feet=75.7 percent
of span.
12. 1 feet=75.7 percent
of span.
Slab 930 (85-^ inches effect-
ive thickness).
II. 4 feet=7i.6 per
cent of span.
13.0 feet=8i.2 per cent
of span.
12.9 feet=8i . I per cent
of span.
i4.5feet=9o.7 percent
of span.
Slab 934 (6 inches effective
thickness).
12.7 feet=79.5 percent
of span.
17.5 feet=io9.3 P^*"
cent of span.
i2.9feet=8i.i per cent 17.5 feet=io9.3 per
of span. cent of span.
May 8, 1916 Tcsts of Rezfiforced-Concrete Slabs 233
APPLICATION OF RECTANGULAR-BEAM THEORY TO DESIGN OF
SLABS UNDER CONCENTRATED LOADS
The usual rectangular-beam design formulas may be applied to the
design of slabs by merely substituting for b its value as determined by
these investigations, b^o.yL. The corresponding formulas then be-
come—
FOR SLABS UNDER CENTRAL
FOR RECTANGULAR BEAMS CONCENTRATED LOADS
(i) M,=y,fjtjbd^ M,=y,fj^jU.d^
(2) M=pfjbd' M=pfjy-Ld^
(3) p=h
joa.
bd "jLd
(4) P=f /f — \ p=
(5) k=^-\ 2pn-^{pn)' — pn f^='\/2pn-\-{pny—pn
It is interesting to note that in substituting for M^ and Mg in formulas
PL
I and 2 their value ^ — , the L's cancel, showmg that the safe load-carrying
capacity of the slab is independent of the span; thus —
PL 7 7
1 becomes — = ^fJij—Ld^ or P=—fJijd'^
2 becomes ?^=pfjlLd'' or p=plAf jd^
The above investigations were made on slabs the width of which was
twice the span length, so that the stress at the extreme edges was very
small. The conclusions must therefore be applied to such cases only.
When the ratio of width of slab to span length is less than 2, these conclu-
sions may or may not apply, and additional investigations are now being
made to determine the proper value of effective width to use under such
conditions.
PLATE XXVI
Fig. I. — Load-applying and load-measuring apparatus for testing reinforced-con-
crete slabs, showing set-up for 4-point loading.
Fig. 2. — Load-measuring apparatus and hydraulic jack for testing reinforced-con-
crete slabs.
(234)
Tests or Reinforced Concrete Slabs
Plate XXVI
Journal of Agricultural Research
Vol. VI, No. 6
OCCURRENCE OF STERILE SPIKELETS IN WHEAT
By A. E. Grantham, Agronomist, Delaware Agricultural Experiment Station, and
Frazier Groff, Student, Delaware College
INTRODUCTION
The average spike of wheat (Triticum spp.) contains from 15 to 20
spikelets, each of which under favorable conditions is capable of pro-
ducing two or more kernels. Ordinarily, however, the lower two or
three spikelets on the spike do not develop. The only indications of
their absence are the joints or nodes of the rachis which are thus exposed
(PI. XXVII). Hunt states that often in the cultivated varieties and
always in the wild species the lower one to four are sterile. In this paper
the term "sterile spikelet" is used to designate those spikelets at the base
of the spike which for some reason fail to develop and produce seed. No
account was taken of the sterile florets which might occasionally occur
within the spikelet. The absent spikelets, as shown by the naked rachis,
were the only ones estimated as sterile.
MATERIAL AND METHODS
During the summer of 191 5 the writer had the opportunity of making
a detailed study of the occurrence of sterile spikelets in a large number
of varieties of wheat under test by the Department of Agronomy at the
Delaware Agricultural Experiment Station. These varieties and strains
of wheat, 188 in number, had been sown the previous autumn by two
methods: First, by a grain drill as under ordinary field conditions, at the
rate of 7 pecks per acre; second, by the centgener or hill method, leaving
the individual plants 6 inches apart each way. By the former method
the plants were very close in the rows, which were 8 inches apart. This
gave an opportunity to determine to what degree the closeness of the
plants or rate of seeding influenced the frequency of sterile spikelets.
The data for each variety were secured in the following manner: The
total number of fertile and sterile spikelets were counted on 25 represen-
tative spikes of each variety. The means of the fertile spikelets and the
sterile spikelets were taken separately and the percentage of sterile spike-
lets was determined for each variety of wheat. Where the varieties were
planted in hills 6 inches apart each way, five plants of five culms each con-
stituted the 25 spikes, the spikelets of which were counted. In this man-
ner the actual number of sterile spikelets and the percentage of the total
number of spikelets were determined for the 188 varieties and strains
under the two methods of planting.
Journal of Agricultural Research, Vol. VI, No. 6
Dept. of Agriculture, Washington, D. C. May 8, 1916
dk Del.— 2
(235)
236
Journal of Agricultural Research
Vol. VI, No. 6
EFFECT OF RATE OF SEEDING ON STERILITY OF SPIKELETS
It was found (see Table I) that the actual number of sterile spikelets
per spike (average of 25 spikes) ranged from 1.84, the lowest, to 5.52,
the highest, for varieties in drills; and in hills, from 0.28 sterile
spikelets, the lowest, to 3.76, the highest. The percentage of sterile
spikelets per average spike in drill rows ranged from 11.5 per cent, the
lowest, to 36 per cent, the highest. In hills the percentage of sterile
spikelets among the varieties ranged from 1.5 per cent to 23.5 per cent.
The mean number of sterile spikelets for all varieties in drill rows was
3.47; in hills, 1.73. The mean percentage of sterile spikelets for all
varieties in drills was 21.8 per cent; in hills, 10 per cent.
The data indicate that the spacing of the wheat plants has a direct
bearing on the number of sterile spikelets. Wljieat planted in hills has
more space in which to develop and invariably sends up a greater number
of tillers than wheat sown in drills. It has also been observed that the
period of maturation is prolonged where the wheat plant has more space.
Under these conditions the vegetative activity of the plant is more pro-
nounced, as shown by an increased number of culms, broader leaves, and
heavier straw.
Table I. — Number and percentage of sterile spikelets on 25 spikes of bearded and smooth
varieties of wheat in IQ15
Variety.
Bearded
or
smooth.
Total number
of spikelets.
Number of
sterile spikelets.
Drill. Hill. Drill. Hill.
Percentage of
sterile spikelets.
Drill. Hill
Acme
Acme Bred (Maryland)
Acme Improved (Maryland)
Ahrens (Indiana)
American Banner
American Bronze
Babcock (Michigan 07664)
Bearded Purple Straw
Bearded Winter (Michigan 9850) .
Bearded Winter Fife
Beech wood Hybrid
Beloglina
Berkeley
Berkeley Awnless
Blue Stem
Broughton
Buda Dawson (Michigan 310717).
Buda Pest
Canadian Hybrid
Century
China
Clawsons Longberry
Cooks Brookmont
Councilman
Craigs Favorite
Currells Prolific
Crimean
14.92
13. 12
16. 92
16. 04
17.88
14. 64
IS-
16. 28
18.28
15-50
15-78
16. 12
14. 12
15-44
15. 20
14.36
15.48
17.40
15.84
17.04
17.64
13-36
15-3
15.48
15-5
14.64
15.80
15. 08
15.68
17.92
16. 92
19.96
14. 72
15.04
16.28
17.48
16. 16
13-52
16. 20
17. 08
15-44
15.2
14. 80
14.
15.80
16. 24
15-
18. 24
15. 16
15-76
17.44
16.
17.28
3-56
2. 96
3- 04
2.36
2. 76
72
60
44
76
80
2.84
3-76
2. 00
2. 84
2.68
3-72
3- 04
4. 12
3- 00
3-48
3-
3-
2.
3-
3-
3-
3-72
2.84
2.80
1.56
2. 20
2.88
2. 96
2. 12
2. 56
2.48
2.32
I. 24
I. 64
1.80
1.48
I. 20
1. 12
2.08
2. 00
2. 00
1.36
I. 76
1. 12
2. 04
23. 69
19.84
23-17
13-94
17-
20. 80
24- 59
22. 22
23.09
20. 79
16. 00
17.99
23-32
14. 16
18.38
17.63
25.90
19-63
23.67
18.93
20. 42
19.27
24.25
17.23
22. 99
22.39
23-77
23-54
18.83
17-83
8.70
13-53
14.48
20. 10
14.09
15-72
14. 18
10.39
13.90
13.09
13-58
8.03
10.73
12. 16
10. 22
7-59
6.89
13.26
10. 96
13- 19
8.26
10. 09
6.86
11.80
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
237
Table I. — Number and percentage of sterile spikelets on 25 spikes of bearded and smooth
varieties of wheat in 191 5 — Continued
Variety.
Bearded
or
smooth.
Total number
of spikelets.
Drill. Hill.
Number of
sterile spikelets.
Drill. HiU
Percentage of
sterile spikelets.
Drill. Hill.
Dawsons Golden Chaff
Defiance
Diamond Grit
Dietz
Dietz Longberry
Doub
Dunlap
Early Harvest
Early Red Chief
Early Red Clawson
Early Windsor
Eclipse
Egyptian Amber
Enterprise
European Century
Farmers Trust
Fish Head
Fulcaster
Four Row Fultz
Jersey Fultz
Fultz
Fultz Mediterranean
Genessee Giant
Giant Square Head
Goens
Goens Awnless *
Gill
Glace
Gold Coin
Golden Bronze
Greening (Michigan 126)
Gypsy
Hedges Prolific
Hercules
Harvest King
Hickman
Himgarian (Michigan 913802)
Hybrid Sel. 13
Hyde Michigan 6
Imperial Amber
International 6 (Michigan 61)
Jones Early Red Chaff
Jones Longberry
Jones Mammoth Amber
Jones Paris Prize
Jones Winter Fife
Kansas Mortgage Lifter
K. B. 2
Kharkov ■. . . .
Klondike
Lancaster- Fulcaster
Lancaster Red
Lebanon
Mammoth Red
Martins Amber
Malakoff
Massey
36287°— 16 .3
16. 36
13. 60
18.76
15.28
15.04
15. 00
15. 24
15.60
16. 24
16.76
16.56
17.24
17.64
15-36
16. 76
18.08
15.8c
15-56
16.56
14.84
16.32
16. 40
18. 04
17.40
15-36
15-44
15-44
19. 00
17.24
16. 24
16. 04
17.76
15-32
15. 16
15-72
15-45
14.92
18. 00
16.84
15-92
15- 63
15.80
18.32
18.96
16.88
19. 60
14.64
18.44
13.96
16.84
14.32
15.84
14.88
15.80
18.24
14. 24
16.88
17.32
13.76
18.76
15-44
16.55
13-52
15. 00
15.08
17-56
16. 92
17.80
19.00
17.08
16.44
18.48
17.40
18.24
15.28
17.76
15.64
17. 60
16. 92
19.44
18.36
15. 00]
15-32
15. 12
19. 80
17.36
18.32
17.92
18.84
16.80
17.08
16. 20
15.68
17.96
22. 08
19.88
18.68
18. 00
17.76
17-
22.68
18. 12
20. 20
16. 20
20. 12
15-76
17.92
15.24
16.36
15.84
16.84
21. 00
14.96
20. 04
3.00
3-24
5. 28
3-24
2.86
3. 12
3-92
2. 60
2. 64
3-56
3-
3-
4-
3-
3-
4-
3.00
2. 96
2. 40
2. 72
2.36
1. 96
4. 20
4.28
3- 24
2.48
2. 52
4.68
3. 60
3- 20
3- 52
4. 16
2.80
3- 20
2. 64
2.44
3-92
24
4.04
3- 20
3
3
3
3
4
3
2.88
4.08
2'. 48
2.44
52
2.56
1. 60
3-52
1.68
2. 16
I. 64
.40
1. 16
1.28
2. 40
2.80
1.44
2.08
2. 24
2.32
2. 12
.48
.60
•56
.28
1.28
.64
1. 04
.76
•36
2.08
1.44
2. 04
1.80
2. 24
• 44
.60
I.
I.
2. 56
2. 12
2.56
.92
.48
2. 36
1.48
I. 04
1.44
1.68
I. 16
1. 32
1.44
1.52
2. 16
2. 00
2.32
1. 64
2. 24
23.82
28. 14
21. 20
19. 01
20. 80
25.72
16.66
16.25
21. 24
21. 02
22.73
26.
/5
19. 80
24.77
18.98
19. 02
14.49
18.32
14.46
11-95
23.28
24-59
21. 09
16. 06
16. 32
24.63
20.88
19.70
21.94
23.42
18. 27
21. 10
16.79
15.84
26. 27
21-33
19.23
25-37
20.47
20.75
23-36
26.37
17-77
15-71
17.76
20. 82
22. 63
18.76
22. 06
26. 01
23-65
18. 22
22. 36
17.41
14-45
14-77
11. 62
18. 76
10. 87
13- 19
12. 13
12. 26
5-57
2. 27
6.85
7. 19
12. 62
11.63
8-75
11. 25
12.87
12.71
13-87
2. 76
3-84
3-18
1-59
6.58
3-48
6. 92
4. 96
2.38
10. 50
8.29
11. 13
10. 04
11.88
2. 16
11. 00
3-70
9-43
10. 02
11-59
10. 66
13-70
10.44
5.18
2.68
10. 40
8. 22
5-14
8.88
8-34
7-36
7.42
9-44
9.29
13- 63
11.87
11. 04
10. 96
11. 12
238
Journal of Agricultural Research
Vol. VI, No. 6
Table I. — Number and percentage of sterile spikelets on 2j spikes of bearded and smooth
varieties of wheat in IQ15 — Continued
Variety.
Meally
Mediterranean
Michigan Amber
Millers Pride
Miracle
Missing Link
Morse
New Amber Longberry
New Soules
Nigger
Nixon
Ohio 5507
Ontario Wonder
Orange
Pesterboden
Perfection
Plymouth Rock
Poole
Pride of Genessee
Prosperity
Purple Straw
Red Cross
Red Hussar
Red Rock
Red Wave
Reiti
Reliable
Rochester Red
Rocky Mountain
Royal Red Clawson . . .
Rudy
Rudy Hard
Ruperts Giant
Rural New Yorker ....
Russian Amber
Shepherds Perfection. .
Silver Sheath
Silver Wave
Smiths Rustproof
Soumans Champion . . .
Spayde
St. Louis Grand Prize.
Stone
Swamp
Theiss
Turkey Red
Turkish Amber
Velvet Chaff
Valley
Wayside Wonder
Whedling
White Eldorado
Wyandotte Red
Tennessee 3608
Tennessee 3609
Tennessee 3611
Tennessee 3614
Bearded
or
smooth.
Total number Number of
of spikelets. sterilesptkelets.
s
B
s
B
B
B
S
B
S
B
S
S
s
s
B
S
S
s
B
s
B
S
B
B
S
B
B
S
B
S
B
B
S
S
B
B
B
B
S
B
B
S
B
B
B
B
B
B
Drill. Hill.
17.72
15.20
15-92
14. 20
15-7
17. 76
14.48
18. 80
17. 16
12. 80
15.96
17. 28
17.80
15.08
15-1'
14. 92
16. 72
15.64
18. 76
16.84
15.28
16. 60
12. 76
13-32
18. 16
17. 20
15- "■"
15.96
15. 00
14.96
12.52
13.84
17.24
16. 92
15. 84
16. 64
13-52
17-32
16. 72
16.80
16. 52
18. 00
13. 00
14.88
13.28
14.44
15.48
16.80
16.68
14-32
13.80
15.96
14.64
18. 12
17. 60
15. 16
15.28
20. 00
16. 00
17-52
16.84
16.84
19.84
16.08
20. 16
17.80
14.40
15. 60
17.92
20. 32
15.96
15.96
16.44
18.88
18. 20
21. 48
19.56
20.44
20. 32
18. 12
15.68
19.52
21. 60
18.68
18.80
15.68
16. 52
14. 20
16. 12
19. 80
19.56
19-52
20. 56
17-52
19.48
20.68
19. 28
19.76
20. 40
16. 96
17-
17-52
16. 60
16.
18. 04
17. 76
17.08
15. 60
18.36
16. 96
22. 48
19. 68
20. 16
18. 40
Drill. Hill.
3.68
3.00
2.64
4.40
4.64
2. 24
4-
3-58
2.68
3-50
3-72
4. 20
2. 72
2.80
2. 92
3-52
2. 72
5-32
3-24
2. 72
3-44
3-36
4.24
4.44
3-84
3-96
3-96
3- 24
2. 72
2. 92
4.40
3-72
4.80
4-56
3-68
4- 76
4.28
3
4
3
3
3-68
3-
2.
76
„ 28
4.04
4. 60
3-92
2. 40
3-
2.80
5-52
4.40
4-32
3.80
2. 16
I. 96
1. 64
2. 20
1.80
2.
.56
I. 40
I. 24
1.56
.84
I. 04
1. 76
1.68
1.80
.60
2. 00
2.28
3-76
I. 60
2.
I. 24
.72
1. 92
1.56
2. 20
2. 24
1.68
I. 92
.48
I. 04
1.32
1. 60
1.44
2. 76
2.80
2. 40
2.
2. 72
1.88
1.56
1.24
2.32
.84
I. 60
I. 72
1. 24
1.88
2. 32
I.
.92
I. 24
1. 16
3- 12
2. 20
2. 76
2. 40
Percentage of
sterile spikelets.
Drill. Hill
16. 02
24. 12
18.84
18.59
27.98
26. 12
15-47
25-73
20.86
20.93
21. 92
21.52
23-59
18.03
18.46
19-57
21. 05
17-39
28.35
19.23
17.80
20. 72
22.57
25. 22
23-34
25-81
24.
24. 81
26. 40
21.65
20. 92
21. 09
25-51
21. 98
30-30
27.40
27. 14
27.48
25-59
22. 38
25.18
19. II
29-53
24-73
28. 91
19. II
21. 18
24. 04
27-57
27-37
17-39
21. 80
19. 12
30.46
25.00
28. 49
24.86
10. 80
12. 25
9-36
13. 06
10.68
12. 50
3-48
6-99
6.96
10.83
5-38
5.80
8.66
10.53
11. 90
3-64
10.59
12. 52
17-50
8.17
"•35
6. 10
3-97
12. 24
7-83
10. 18
11.99
8-93
12. 24
2. 90
7-32
8. 18
8.08
7-36
14- 13
13.61
13.69
14-79
13- 15
9-74
7-89
6. 07
13.68
4.69
9.70
10.37
7-34
10. 42
13.06
9-83
5-89
6-75
6.82
13-87
11. 17
13.69
13.04
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
239
Table I. — Number and percentage of sterile spikelets on 25 spikes of bearded and smooth
varieties of wheat in igi^ — Continued
Variety.
Bearded
or
smooth.
Total number
of spikelets.
Drill. Hill.
Number of
sterile spikelets.
Drill. Hill
Percentage of
sterile spikelets.
Drill. Hill.
u. s
u. s
u. s
u. s
Tennessee 3617
Tennessee 3277
U. S. 2980
3608
3609
3610
3612
U. S. 3613
U.S. 3614
Abundance
Auburn Red
Australian Red
Banat
Bulgarian
California Red
Davidson
Deitz Amber
Deitz Mediterranean .
Early Pearl
Early Ripe
Economy
Egyptian
Farmers Friend
Ghirka Winter
Goings
Grand Prize
Invincible
Jones Red Wave ....
Kentucky Bluestem.
Lancaster
Lehigh
Petigree Giant
Red May
Reiti
Sibleys New Golden .
Texas Red
Tread well
Tuscan Island
Ulta
Winter Chief
Winter King
Wisconsin 13
Leaps Prolific
17. 16
18.36
14.84
17.92
17. 12
18. 16
17-52
15.00
16. 40
16.44
16.08
14.52
13-72
14.88
14.96
15-95
14. 52
14.44
13.04
15.08
14.48
14. 08
13. 00
15-76
15. 16
17.76
17.92
18. 20
14.84
14.36
13.96
18. 24
14-32
14. 72
14.44
13.04
15-24
15-32
13-32
15-44
14. 16
13. 12
17. 06
20. 44
21. 60
17.08
21. 24
19. 68
21. 56
21.36
20. 40
19. 20
17.76
16.88
15.84
15- 52
16.68
15-76
17.44
15.88
15.64
13.92
16.08
15-36
17. 20
14.40
18.56
14. 80
19.72
20. 60
20. 21
17.28
15. 16
17. 16
19.40
16.84
17-
17.84
17.28
18. 00
16. 72
16. 04
17- 56
15.04
16.36
4.68
5- 40
2. 92
4.84
4-36
3-52
4.48
3-92
3-40
2. 60
3-40
3-24
3-48
4. 16
2.88
1.84
3.80
3-50
2. 56
2.80
2. 72
2.36
2. 64
2. 00
2. 72
2. 20
1.68
2.52
2. 56
1.80
2. 00
2.32
2. 12
2. 64
2. 04
•32
.80
I. 24
1.32
•72
.68
1. 16
1.80
1.68
2. 12
1. 04
2. 00
2.32
I. 64
I. 40
I. 96
2.88
I. 40
•36
1.68
1-52
I. 76
1. 72
2. 00
I. 24
•44
1.28
I. 24
.68
27. 27
29.41
19.67
27. 06
25.46
19.38
25-57
26. 13
20.73
15.81
21. 15
22.31
25-36
27-95
19.25
11.52
26. 17
24.23
19.64
18.56
18.78
36.07
22. 76
28.88
19.78
18.91
24-33
23.29
21. 02
24.23
25. 21
20. 83
18.43
25. 14
24.65
29- 14
30.18
25-58
28. 52
15-54
22.88
28.65
13-36
11-54
12. 22
11. 70
12. 80
II. 17
7-79
79
54
36
26
74
13-38
17. 01
12.23
2.03
4-58
7.80
8.43
5-17
4. 22
7-55
10. 46
11. 69
11.36
7. 02
10. 14
11. 26
8.15
8. 10
12. 92
16.78
7. 21
2.13
9-39
8.54
16.18
9-55
II. 96
7-73
2. 50
8.51
7-58
3.60
Average .
15-85
17-13
3-47
73
10. 09
Of the 188 varieties and strains of wheat under observation, io8 were
beardless and 80 bearded. To determine whether the presence or absence
of awns as a morphological character was in any way correlated with the
occurrence of sterile spikelets, the varieties were tabulated so as to show
the distribution of bearded and of beardless varieties with reference to
the percentage of spikelets (see Table II). The data in this case were
taken from the varieties sown in drills.
240
Journal of Agricultural Research
Vol. VI, No. 6
Table II. — Arrangement of bearded and beardless varieties of wheat with reference to the
percentage of sterile spikelets
Percentage of barren spikelets.
II to 15
i5to 17
17 to 19
19 to 21
21 to 23
23 to 25
25 to 27
27 to 29
29 to 31
Total
Total
number of
varieties.
12
27
32
30
31
25
17
5
Number of
beardless
varieties.
12
19
18
14
7
80
Number of
bearded
varieties.
o
8
14
16
24
23
17
5
108
Percentage of each class
to total number of —
Beardless
varieties.
10. O
15-0
23-7
22. S
17-5
8.7
2-5
o
o
Bearded
varieties.
7-4
12. 9
14.8
22. 2
21. 2
15-7
5-5
Table II shows that the bearded varieties as a class have a higher
percentage of sterile spikelets than the beardless wheats. There are 20
of the 80 varieties of beardless wheat which have more than 15 per cent
of sterile spikelets, while not a single variety of bearded wheat has less
than 17 per cent of sterile spikelets. Of the 108 bearded varieties 45
have not less than 25 per cent of sterile spikelets. Only two of the 80
beardless varieties have 25 per cent of sterile spikelets. The average
percentage of sterile spikelets for all the beardless varieties is 17.8; for
the bearded, 24.1; a difference of 6.1 per cent in favor of the beardless
varieties. The individual variety having the lowest percentage, 11.5,
was beardless, while the variety having the highest percentage of sterile
spikelets, 36.7, was bearded. All of the varieties Vv^hich are mentioned
above were sown under like conditions of soil preparation and fertilization
and planted at the same time.
EFFECT OF TIME OF SEEDING ON STERILITY
The next step was to determine the effect of time of seeding and of soil
treatment on the frequency of sterile spikelets. As it happened, an
experiment was already under way on different dates of sowing wheat,
including two varieties, one bearded and the other beardless, on both
fertilized and unfertilized soil. These plants were in hills 6 inches apart
each way. In the manner followed above, the total number of spikelets
and that of sterile spikelets per spike were combined, and the average
was determined for the two varieties under different dates of planting
on both treated and untreated soil (Table III).
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
241
Table III. — Effect of date of planting on the number of sterile spikelets in 25 spikes of
two varieties of wheat on fertilized and on unfertilized soil
RED WAVE (BEARDLESS)
Date of planting.
Total number of
spikelets.
Fertilizer.
No ferti-
lizer.
Number of sterile
spikelets.
Fertilizer,
No ferti-
lizer.
Percentage of sterile
spikelets.
Fertilizer.
No ferti-
lizer.
Sept. 17.
24.
Oct. I . .
IS.
22 .
Average .
21. 4
20. 5
20. o
21-5
21. 3
19.7
17.7
18.4
20.3
18.8
19.9
20. 9
2.8
2. 2
2. 2
2. I
2. 2
1-5
1.6
20. 7
19.
1-7
13-4
II. I
II. I
10. I
10. 6
5-4
10.3
12. I
9-7
10.8
8.4
8.4
5-7
9-3
MIRACLE (BEARDED)
Sept.
Oct.
24.
IS-
Averao;e.
16.7
15.0
2-3
1-5
13.8
16. 2
14.9
2.6
I. 2
16. 4
18.9
15-5
2.6
1.8
14. I
16. 0
15-4
3-1
1.8
19.7
15-7
16.8
1.6
1.8
10. I
16. I
15-4
•9
.4
S-7
16.6
iS-S
2. 2
1.4
13-3
10. 4
8.5
II- 5
12. 1
10. 9
2.8
9.4
Table III shows that the number of sterile spikelets per spike varies
considerably from the earliest seeding, September 17, to the latest,
October 22, but in no regular manner. The latest seeding in every case
shows the smallest number of sterile spikelets. This holds true for both
varieties and under both soil conditions. If the average is taken of the
number of sterile spikelets under the six different dates of seeding, it is
found that there are more sterile spikelets where fertilizer was used than
where no application was made. This also holds true for both varieties.
Expressed as a percentage, the average of sterile spikelets for the different
rates of seeding with the beardless variety is 10.3 per cent where ferti-
lizer was used and 9.3 per cent on untreated soil. That of the bearded
variety was 13.3 per cent of sterile spikelets as an average for the different
dates of seeding on treated soil and 9.4 per cent on the untreated. It
will be noted that the latest seeding of each variety has as many spikelets
as the earliest, and that there are more than twice as many sterile spike-
lets in the latter than in the former. This may be partially accounted
for b)' the fact that the later plantings did not have a full stand of plants,
thus giving the individual wheat plant more space. This explanation
is in accord with results obtained under the different methods of seeding
(see Table I) — that is, that fewer sterile spikelets were found in the
thinner plantings.
242
Journal of Agricultural Research
Vol. VI, No. 6
The tillering in the early plantings was nearly 100 per cent greater
than in the later plantings. The tillering for each variety on fertilized
soil for a given date was 50 per cent greater than where no fertilizer was
used. The general efifect of the date of seeding seems to indicate a
tendency toward a smaller percentage of sterile spikelets in the later
seedings. The relation of the number of sterile spikelets to yield does not
seem to affect the yield seriously, since the fertilized wheats produced
two or three times as much grain per spike as the unfertilized. The
difference in yield per spike seems to be due largely to quality (size) of
kernel.
Table IV. — Relation of the effect of different fertilizers and combinations of fertilizers
to the occurrence of sterile spikelets
Treatment.
Dawsons Golden Chafi
(smooth).
Total
num-
ber of
spike-
lets. 1
Num-
ber of
sterile
spike-
lets.'
Per-
cent-
age of
sterile
spike-
lets.
Lehigh (bearded).
Total
num-
ber of
spike-
lets.i
Num-
ber of
sterile
spike-
lets.i
Per-
cent-
age of
sterile
spike-
lets.
Nitrogen, phosphorus, and potassium
Nitrogen and phosphorus
Phosphorus and potassium
Nitrogen and potassium
None
Nitrogen
Phosphorus
Potassium
16.8
18.2
18. 2
18. 2
17.0
16. 9
15-9
17. o
1.36
1.68
I. 92
I. 08
1.05
.92
1.56
I. 32
8.0
9.2
10. 2
5-9
6.1
S-4
9-7
17.9
18.2
17. 2
17-3
16. 7
18. o
15.0
16.8
1.32
2.08
1.80
.92
I. 01
1.36
I. 40
I. 24
7-3
II. 4
10. 4
5-2
6.0
7-5
9.2
7-3
' Average of 25 spikes.
EFFECT OF FERTILIZERS ON STERIEITY
The effect of different elements of plant food, singly and in combina-
tion, on the number of sterile spikelets is seen in Table IV. The wheat
was planted by the centgener method, the individual plants being 6
inches apart each way. On each of the plots sufficient fertilizer of
each mineral ingredient was supplied to produce a 50-bushel crop of
wheat, provided that it were all used. The nitrogen was applied for
a 25-bushel crop, it being assumed that the soil carried a fair reserve
of this element. The nitrogen was applied in equal parts by weight
of nitrate of soda and dried blood; the phosphoric acid was carried as
acid phosphate and the potash as muriate of potash. It will be noted
that where the fertilizers were applied singly nitrogen gave the lowest
percentage — 6.4 — of sterile spikelets as an average for the two varieties.
Potash came next with 7.5 per cent, and phosphoric acid stood highest,
with 9.4 per cent of sterile spikelets. Where two elements were used
in combination, phosphoric acid and potash led, with an average of
May 8. 1916 OccuYYence of Sterile Spikelets in Wheat 243
10.4 per cent for the two varieties; phosphoric acid and nitrogen com-
bined gave 10.3 per cent of sterile spikelets, while nitrogen and potash
gave 5.4 per cent. Since phosphoric acid gave the highest percentage
of sterile spikelets when used alone, it, would seem that this element of
plant food is largely responsible for the sterile spikelets, as in every
combination in which it is used the number of sterile spikelets is greater
than where nitrogen and potash are used singly or in combination. The
untreated plot gave 6 per cent of sterile spikelets, the lowest for the
series except where nitrogen and potash were used in combination,
which gave 5.5 per cent. The complete fertilizer gave an average of
7.6 per cent of sterile spikelets. From these data it would seem that
there is a tendency for phosphoric acid to produce a larger percentage
of sterile spikelets than either potash or nitrogen. However, the fairly
high percentage of sterile spikelets in the case of the wheat treated
with phosphoric acid did not afFect the yield per plant or spike. Under
this treatment the yield and quality of the grain surpassed that under
either nitrogen or potash.
CORRELATIONS
In order to determine what relation might exist between the total
number of spikelets per spike and the number of sterile spikelets, the
readings constituting the averages for the 25 spikes of each variety were
arranged in correlation tables. The beardless varieties form one table
and the bearded the other. Thus, the readings were the average of
each variety and the array or distribution in the table was made up of
varieties. The data were secured from the plants in hills. Since the
number of spikelets per spike in a large measure determines the length
of spike, the relation found will be closely associated with the length of
the spike. In Table V, which includes the beardless varieties, the coef-
ficient of correlation between the number of sterile spikelets and the
total number of spikelets is 0.543 ±0.054. The bearded varieties show a
correlation which is expressed as r = 0.598 ±0.041. It appears that the
number of sterile spikelets per variety bears a direct positive correla-
tion to the total number of spikelets or the length of head. The varie-
ties with the shorter spikes have decidedly fewer sterile spikelets. The
relation betvv'een the number of spikelets and the length of spike may
not be close, inasmuch as there may be more or less range among varie-
ties as to the condensation or closeness of the spikelets on the spike.
However, the long spikes are made up of a relatively larger number of
spikelets than the short ones, and the actual percentage of sterile spike-
lets may be smaller in the long spikes, as will be pointed out later.
244
Journal of Agricultural Research
Vol. VI, No. 6
Table V. — Correlation between the number of sterile spikelets and the total number of
spikelets in beardless and bearded varieties of wheat
BEARDLESS VARIETIES »
Number of sterile spikelets.
12 to
13-
13 to
14.
14 to
IS-
15 to
16.
16 to
17-
17 to
18.
iS to
19.
19 to
20.
Total.
I
14
10
I
7
12
I
I
I
7
4
3
32
35
9
2 to ?
2
8
2
■I to 4
3
3
I
I
4 to t;
Total
2
10
25
21
13
6
2
79
BEARDED VARIETIES 2
I to 2
0
2 to 3
^ to A
4
2
16
6
15
I
I
6
18
7
18
■■■'6
4
II
3
4
5
58
27
6
A to i;
c to 6
Total
4
18
2^1 TT
10
II
12
109
"
lr=o.S43±o.os4-
' r=o.598±o.04i.
CORRELATION BETWEEN THE) PERCENTAGE OF STERILE SPIKELETS AND
OTHER CHARACTERS OF THE WHEAT PLANT
For the purpose of studying the relationship between the percentage
of sterile spikelets per plant and other characters, 300 plants of the
variety Velvet Chaff were pulled, dried, and later carefully measured.
The plants had been grown by the centgener method, 6 inches apart
each way. The percentage of sterile spikelets was used rather than the
actual number, for the reason that the length of spikes, which deter-
mines the number of spikelets, varies so greatly. The measurements of
length were taken in -centimeters and those of weight in milligrams.
Biometrical data were secured for the statistical relationship between
the percentage of sterile spikelets per plant and (i) the number of culms
per plant; (2) the yield of grain per plant; (3) the yield of grain per
spike; (4) the length of the culm; (5) the length of the spike; (6) the
average weight of the kernel; and (7) the number of spikelets. In the
above determinations the plant was used as a unit, the value for each
character being determined by taking the average of the respective
readings.
CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS PER
PLANT AND THE NUMBER OP CULMS PER PLANT
An inspection of Table VI shows only a slight degree of correlation
between the percentage of sterile spikelets and the number of culms,
which is negative. The coefficient of correlation is — 0.076 ± 0.039. Evi-
dently there exists no appreciable relationship between the percentage of
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
245
sterile spikelets and the number of tillers per plant. The less vigorous
plants, indicated by the smaller number of tillers per plant, do not show
a higher percentage of sterile spikelets than the more thrifty plants.
Table VI. — Correlation between the percentage of sterile spikelets per plant and ike
number of tillers per plant in wheat^
Percentage of sterile
Number of tillers per plant.
Total.
spikelets per plant.
I
2
3
4
s
6
7
8
9
10
II
12
13
0 to 3
I
2
2
2
2
3
10
2
I
I
I
13
4
14
7
4
4
I
4
15
20
10
I
2
7
19
II
II
2
I
5
16
II
4
I
8
14
14
5
I
2
7
6
2
6
^ to 7
I
2
3
I
I
4
I
I
I
I
2
46
85
95
48
7 to 1 1
I
I
I
I
I
1 1 to 15
11; to IQ
I
2
19 to 23
23 to 2 7
7
27 to 31
31 to 35
I
Total
8
22
46
53
50
38
42
18
7
6
5
3
2
300
' r=— o.o756±o.0387-
CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS AND
THE YIELD OF GRAIN PER PLANT
Between the percentage of sterile spikelets and the yield of grain per
plant (Table VII) the coefficient of correlation is negative, — 0.306 ±0.035.
This correlation is fairly high and though expressed negatively indicates
that the higher yielding plants have a smaller percentage of sterile spike-
lets than those of low yield.
Table VII. — Correlation betiveen the percentage of sterile spikelets per plant and the yield
of grain per plant in wheal ^
yield of grain per
plant (in milligrams).
Percentage of sterile
spikelets per plant.
8
S
0
0
8
8
0
0
8
0
■♦-'
8
0
0
3
§
8
0
i
8
0
0
8
8
o_
0
8
8
0
0
8
8
0
I
I
0"
0
8
8
•6
0
•a
0
8
0
i
0
1
0 to •?
I
5
3
16
8
3
5
2
7
12
20
12
I
2
6
IS
16
4
I
6
■2 to 7
I
2
• 6
7
5
2
I
6
23
9
4
I
7
10
13
7
2
10
5
5
4
3
2
4
I
3
2
I
I
I
I
46
85
95
48
II
7 to II
I
I
I
II to 1 5
I ^ to 10
10 to 2'?
2 ^ to 2 7
7
I
27 to ■?!
? I to ■? t;
I
I
Total
34
42
54
44
43
37
19
12
7
8
3
3
I
3
300
-o.30S7±o.03S3.
246
Journal of Agricultural Research
Vol. VI. No. 6
CORRELATION BETWEEN THE PERCENTAGE OE STERILE SPIKELETS AND THE
AVERAGE YIELD OE GRAIN PER SPIKE
The average yield of grain per spike (Table VIII) was determined by
dividing the total weight of grain per plant by the number of spikes per
plant. The coeflficient of correlation between this yield and the percentage
of sterile spikelets is again negative, —0.589 ±0.02 5, which indicates a
much closer relationship between the low percentage of sterile spikelets
and yield of grain per spike than is shown between the same character
and the yield per plant. There is a rather high correlation existing be-
tween the percentage of sterile spikelets and the yield of grain per spike.
Table VIII. — Correlation between the percentage of sterile spikelets per plant and the
average yield of grain per spike in wheat ^
Yield of grain per spike (in milligrams).
Percentage of
sterile spikelets
per plant.
8
0
0
0
0
§
8
0
0
0
0
8
I
0
0
0
0
0
0
8
t
0
0
0
8
8
0
0
0
p
0
0
8
0
0
0
0
8
•0
8
0
0
0
0
8
1
2
0
0
CO
0
0 to 1
I
2
7
17
8
I
I
8
14
8
I
7
13
3
I
I
I
6
I
I
I
6
•? to 7 . . . .
I
I
4
II
2
4
2
9
8
3
I
3
17
16
9
I
7
12
13
2
6
6
4
I
2
I
/!6
85
95
18
2
3
I
6
2
I
2
IS to 19
19 to 23 ... .
2
2
II
7
I
2 7 to 11
I
I
? I to ■? C
I
Total . .
4
8
12
ZZ
36
36
46
34
32
26
16
6
2
7
I
I
300
ir= — o. 5S8S±o.o2S4.
CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS PER
PLANT AND THE AVERAGE LENGTH OF CULM PER PLANT
In this case the average length of culm per plant (see Table IX) was
found by taking the sum of the lengths of the culms of a plant in centi-
meters and dividing it by the number of culms. The correlation coeffi-
cient is —0.448 ±0.03 1. This is a rather high degree of correlation and
is expressed as negative, although with reference to the relation of the
two characters compared it means that the longer culms tend to form a
lower percentage of sterile spikelets. This is what might be expected,
since the yield of grain per spike is generally closely associated with the
length of spike, and that in turn with the length of culm.
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
247
Table IX. — Correlation between the percentage of sterile spiklets per plant and the
average length of culm in wheat ^
Length of culm (in centimeters).
Percentage of sterile spike-
lets per plant.
0
0
0
0
2
•0
0
0
0
0
2
0
0
0
I
6
13
21
9
2
3
0
0
7
17
21
8
2
8
0
C\
3
9
15
15
7
0
0
8
0
2
5
0
0
0
0 to 3
6
•J to 7
5
II
16
13
4
I
12
19
9
5
6
3
2
I
I
I
I
46
85
95
48
y to 1 1
3
6
7
I
I
1 1 to 1 5
I
2
I
I
2
I
I
11; to 10
19 to 23
I
21, to 27
.
7
27 to •?!
^ I to •? ?
I
Total
I
5
4
19
50
55
57
49
40
16
3
I
300
' r=— o.4482±o.03i6.
CORRELATION BETWEEN THE PERCENTAGE OE STERILE SPIKELETS AND THE
LENGTH OF SPIKE PER PLANT
The average length of spike per plant was determined by dividing
the sum of the lengths of the spikes per plant by the number of spikes.
The calculations were expressed in centimeters. The coefficient of
correlation between these two characters is —0.451 ±0.031 (see Table X).
Since the longest spikes usually occupy the longest culms, we should
expect the same relationship between the length of spike and percentage
of sterile spikelets as was found between the latter character and the
length of culm (see Table IX). There is a very close relation, the
coefficient of correlation with the culm being 0.448 ±0.031, a difference of
0.003 between the two coefficients.
Table X. — Correlation between the percentage of sterile spikelets per plant and the
average length of spike in wheat ^
Length of spike (in centimeters) .
Percentage of
sterile
spikelets
per plant.
00
0
-0
0
CO
0
0
0
•0
0
00
0
06
0
^0
00 1 o>
06 i oi
1-
0
00
0
t
6
0
CO
0
0
0
0
vO
0
0
0
00
0
i
0
5
j2
0 to 7,. . . .
i
I I
3 : 5
6 19
12 J17
8 1 4
I ' . . .
2
6
5
3
I
4
2
I
I
6
•? to 7 . . . .
I
I
4
II
12
4
5
14
15
8
I
12
17
10
5
8
14
4
2
2
46
85
95
48
7 to II. . . .
3
II
3
2
I
I
I
I
II to 15....
ic; to IQ. . . .
I
I
I
2
3
2
3
2
I
I
I
10 to 2^. . . .
I
I
I
I
2 •? to 2 7 . . . .
I
7
27 to ^I. . . .
^i to -i^. . . .
1
I
5
Total. .. .
I
I
3
8
21
32
31 46
43
45
28 17
10
3
4
0
I
I
300
r=-o.45i5±o.03io.
248
Journal of Agricultural Research
Vol. VI, No. 6
CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS AND
THE AVERAGE WEIGHT OF KERNEL
To get the average weight of kernel per plant the total weight of ker-
nels per plant was divided by the number of kernels and the result
expressed in milligrams. The coefficient of correlation is —0.421 ±0.032
(see Table XI). This indicates a decided tendency for the heavier kernels
to be associated with a low percentage of sterile spikelets. This is in
accord with the relations found to exist between the length of culm and
spike and the percentage of sterile spikelets. The more vigorous plants,
as indicated iDy an increased length of culm and spike , generally bear kernels
of a larger size. Hence, the correlation between the percentage of
sterile spikelets and the weight of kernel — in other words, the quality of
the grain — is in the same direction and approximates the other coefficients
very closely .
Table XI. — Correlation between the percentage of sterile spikelets per plant and the average
weight of the kernel in wheat ^
Percentage of sterile spike-
lets per plant.
Weight of kernel (in milligrams).
2 to
4
4 to
6
6 to
8
8 to
10
10 to
12
lato
14
14 to
16
I
4
IS
29
10
I
2
16 to
18
3
II
26
13
5
2
18 to
20
I
7
13
13
I
20 to
22
'"s'
4
5
I
22 to
24
24 to
26
Total.
0 to •?
I
2
I
2
2
I
6
0 to 7
2
I
6
6
I
2
6
13
7
3
2
8
17
II
15
I
3
46
85
95
48
II
^ to II
1 1 to 1 5
I
I
3
I
3
!■; to 10
I
10 to 2"?
2 •? to 2 7
7
I
2 7 to "? I
I
■? I to^ C
I
I
Total
I
2
8
17
33
55
62
60
35
18
4
5
300
' r = — o.4209±o.0320.
CORRELATION BETWEEN THE PERCENTAGE OF STERILE SPIKELETS PER
PLANT AND THE AVERAGE NUMBER OF SPIKELETS PER SPIKE PER PLANT
A relationship is shown below between the percentage of sterile spikelets
per plant and the total number of spikelets per plant. The coefficient of
correlation is low, —0.152 ±0.037 (see Table XII). There is only a
slight tendency for plants with a low percentage of sterile spikelets to
be associated with a large number of spikelets per plant. As the number
of spikelets determines to a large extent the length of the spike, it would
be supposed that a greater correlation would exist between the number
of spikelets and the percentage of sterile spikelets. This may be ex-
plained by the fact that the total number of spikelets includes both fertile
and sterile spikelets. Also, there may be more or less variation in the
condensation of the spikelets which go to make up the spike.
May 8, 1916
Occurrence of Sterile Spikelets in Wheat
249
Table XII. — Correlation between the percentage of sterile spikelets per plant and the average
number of spikelets per spike per plant in wheat ^
Percentage of sterile
Number of spikelets per spike.
spikelets per plant.
12
13
14
IS
16
17
18
19
20
21
22
23
24
0 to T,
I
I
4
5
I
3
7
7
20
4
I
I
5
4
20
12
3
2
I
I
14
26
20
10
I
2
3
10
27
18
12
I
9
15
7
I
I
6
3 to 7
46
85
95
48
7 to 1 1
2
I
I
I
II to 15
I
3
I? to 10
I
I
I
I
10 to 2 ^
I
23 to 27
7
27 to ^I
31 to ^i;
Total
I
2
s
15
39
49
73
71
38
3
3
0
I
300
' r — — o.iS24±o-0375'
Table XIII. — Variation constants in wheat
Plant as the unit.
Mean.
Standard deviation.
Coefficient of
variation.
Sterile spikelets. . . .per cent. .
11.73 ± 0
198
5. 105 ± 0. 141
43- 51 ±1-406
Number of tillers per plant. . . .
s- 193 ±
091
2. 336± . 064
44. 98 ± I. 468
Yield of grain per plant. mgm. .
2,048. 333 ±51
12.=;
1,312. 82I±36. 157
64. 09 ±2. 381
Yield of grain per spike. mgm. .
379- 833 ± 5
475
140. 599 ± 3- 872
37.02 ±1. 151
Length of culm cm . .
91. 4i7±
366
9. 4ii± .259
10. 29 ± . 286
Length of spike cm. .
9. oi9±
043
I. ii3± .031
12.34 ± -344
Weight of kernel mgm. .
15- 033 ±
151
3. 88i± .107
25.82 ± .756
Number of spikelets per spike .
i7-857±
065
I. 688 ± . 046
9-335± -257
Coefficient of correla-
tion.
Sterile spikelets and number of tillers per plant
Sterile spikelets and yield of grain per plant mgm .
Sterile spikelets and average yield of grain per spike mgm. ,
Sterile spikelets and average length of culm per plant cm.
Sterile spikelets and average length of spike cm. ,
Sterile spikelets and weight of kernel per plant mgm. .
Sterile spikelets and the average number of spikelets per spike,
per plant
-o. o75±o. 038
- .3o6± .035
- .S89±
- .448±
- •45i±
- . 421 ±
024
031
031
032
- . i52± .037
SUMMARY
(i) The number of sterile spikelets per spike in wheat is directly
affected by the rate of seeding or the spacing of the plants. The more
space allowed each plant the smaller the number of sterile spikelets on
each spike.
(2) The bearded varieties of wheat as a class have a higher percentage
of sterile spikelets than the beardless varieties. Of the 188 varieties
250 Journal of Agricultural Research voi. vi, no. &
examined the smallest number of sterile spikelets was found on a beard-
less variety and the largest number on a bearded variety.
(3) Early seeding seems to increase the percentage of sterile spike-
lets on each spike. Wheat seeded very late had the smallest percentage
of sterile spikelets.
(4) The application of nitrogen alone as a fertilizer produced the lowest
percentage of sterile spikelets. Phosphoric acid singly gave the highest
percentage of sterile spikelets, while potash was intermediate as to the
percentage of sterile spikelets. Where two elements of fertilizers were
combined, phosphoric acid and potash gave the highest percentage of
sterile spikelets, with nitrogen and phosphoric acid next and nitrogen
and potash last. In every instance the check or untreated plots
gave a lower percentage of sterile spikelets than those treated with a
complete fertilizer.
(5) There is a distinct correlation between the length of spike as
expressed by the number of spikelets and the number of sterile spikelets.
As the number of spikelets per spike increases (in other words, the length
of spike), the number of sterile spikelets becomes greater. That is,
varieties with the shorter spikes tend toward a smaller number of sterile
spikelets than the varieties with the longer spikes. However, the per-
centage of sterile spikelets per spike may be greater among the varieties
with the shorter spikes, as was shown to be the case where spikes of vary-
ing lengths within a single variety were examined.
(6) There is only a very slight correlation between the percentage of
sterile spikelets and the number of tillers to each plant.
(7) The yield of grain per plant is correlated to a fair degree with a
low percentage of sterile spikelets.
(8) The weight of the kernel or quality of grain is correlated to a con-
siderable degree with a low percentage of sterile spikelets.
(9) The yield of grain per spike, the length of spike, and the length of
culm are strongly correlated with a low percentage of sterile spikelets.
(10) There is a slight correlation between the average number of
spikelets per spike and a low percentage of sterile spikelets.
ADDITIONAL COPIES
OP THIS PUBLICATION MAY BE PROCURED FROM
THE SUPEKINTEXDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
15 CENTS PER COPY
SUBSCRIPTION PRICE, S3.00 PEB YeaS
V
PLATE XXVII
Comparison of the number of sterile spikelets on bearded and beardless varieties
of wheat :
On the left two heads of a bearded variety of wheat showing a large number of
sterile spikelets. On the right two heads of a beardless variety showing compara-
tively few sterile spikelets. Both varieties were gro-mi the same year under like
conditions of soil and treatment.
Occurrence of Sterile Spikelets in Wheat
Plate XXVI
Journal of Agricultural Research
Vul. Yl, No.
VoL VI ^dAY 15, 1916 No. 7
JOURNAL OP
AGRICULTURAL
RESEARCH
CONTENTS
Effect of Cold-Storage Temperatures upon the Pupae of the
Mediterranean Fruit Fly - - - - - - 251
E. A. BACK and C. E. PEMBERTON
Effect of Climatic Factors on the Hydrocyanic-Acid Content
of Sorghum --------- 261
J. J. WILLAMAN and R. M. WEST
Egg and Manner of Oviposition of Lyctus planicollis - 273
THOMAS E. SNYDER
DEPARTMENT OF AGRICULTURE
WASHINGTON, D.C.
WA8HIN<3TDN : OOVERNKENT PRINDNa OIDCE : 1i1»
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATIOn
KARL P. KELLERMAN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Ittdustry
EDWIN W. ALLEN
Chief, Of f ice of Experiment Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Mx^«rrm*ni
Station
H. P. ARMSBY
Director, Institute of Ani?nal Nutritiwi, Th*
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
All correspondence regarding articles from the Department of Agiiculture
should be addressed to Karl F. Kellerman, Journal of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine.
JOURNAL OF AGRICETIAL RESEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, May 15, 1916 No. 7
EFFECT OF COLD-STORAGE TEMPERATURES UPON THE
PUP^e OF THE MEDITERRANEAN FRUIT FLY ^
By E. A. Back, Entomologist, and C. E. Pemberton, Scientific Assistant, Mediter-
anean and other Fruit-Fly Investigations, Bureau of Entomology
INTRODUCTION
The use to which cold-storage temperatures may be put as an aid in
offsetting the disastrous results of attack by the Mediterranean fruit fly,
Ceraiiiis capitata Wied., has already been made the subject of discussion
by the writers.^ In their paper, however, data on the effect of various
ranges of temperatures used in commercial cold-storage plants upon the
eggs and larval instars only are given. So far as the writers have been
able to determine, fruits of almost any variety commonly held in storage
are held at temperatures varying from 32° to 45° F., with preference
shown to a range of 32° to 36°. The effect upon over 26,000 eggs and
60,000 larvae of different temperatures, including 32°, 32° to 33°, 33° to
34°, 34° to 36°, 36°, 36° to 40°, 38° to 40°, and 40° to 45°, indicate that
no eggs or larvae survive refrigeration for seven weeks at 40° to 45°, for
three weeks at 33° to 40°, or for two weeks at 32° to 33°.
While the greatest danger in the spread of this pest from one country
to another lies in the transportation of the larvae within fruits, there are
certain data on record which prove that this pest may be carried long
distances in the pupal stage and arrive at its destination in a condition to
produce infestation. A fruit-fly pupa (species unknown) was found at
Auckland, New Zealand, in soil about the roots of plants imported from
Australia.^ In 1914, Sasscer'* records the discovery in Washington,
D. C, of living pupae of the papaya fruit fly (Toxotrypana curvicavda
Gerst.) in a package containing an unknown vine from Mexico. In
1 The writers wish to acknowledge the assistance given them by Mr. H. F. Willard in obtaining the data
recorded in this and in their previous paper. To obtain these data has necessitated much prolonged tedious
work extending over three years. In securing the data during 1915, Mr. Willard has not only greatly
assisted, but on several occasions during the absence of the writers has been entirely responsible not only
for the completion of experiments already started, but for the starting of others.
' Back, E. A., and Pemberton, C. E. Effect of cold-storage temperatures upon the Mediterranean fruit
fly. /n Jour. Agr. Research, v. 5, no. 15, p. 657-666. 1916.
3Kirk,T.W. Fruit flies. New Zeal. Dept. Agr. Div. Biol. Bui. 22, p. 9. 1909.
* Sasscer, E. R. Important insect pests collected on imported nursery stock in 1914. In Jour. Econ.
Ent,, V. 8, no. 2, p. 268-270. 1915.
Journal of Agricultural Research, Vol. VI, No. r
Dept. of Agriculture, Washington, D. C. May 15, 1916
dp K-3.
(251)
252 Journal of Agricultural Research voi. vi, no. 7
another instance the same investigator records finding a living adult of
the olive fruit fly (Dacus oleae Rossi) and a dead adult of another species
of fruit fly, apparently Dacus semispharens Becker. Both of these species
were in a small package containing olive seed from Cape Town, South
Africa, after having been en route 28 days. Sasscer states that according
to Silvestri it requires from 47 to 49 days in Italy for the pupae of the
olive fruit fly to yield adults; hence, it is possible for this ruinous pest to
enter the United States through the eastern ports as pupae and reach the
olive-growing sections of California before adults have emerged.
Such facts as these indicate that the Mediterranean fruit fly may be
similarly transported, and emphasize the desirability of recorded data
on the effect of cold-storage temperatures upon the pupal stages. Aside
from the practical application in the future to quarantines regulating the
shipment of fruits, the results given below throw considerable light on
conditions governing the distribution of the pest, and help explain the
varying severity of its ravages in countries having both semitropical and
temperate fruit-growing regions.
HISTORICAL REVIEW
Practically nothing has been published on the effect of cold-storage
temperatures upon the pupae of Ceratitis capitata. In 1908 Lounsbury *
in South Africa reports that in removing fruit infested with C. capitata
from refrigeration at 38° to 40° F. at the end of 21 and 27 days he found
in each instance a single pupa, but that both proved to be dead. The
experiments of the writers have demonstrated that these two pupae were
produced by larvae which formed their puparia before the fruit was placed
in storage, as larvae do not form puparia at temperatures lower than
45° to 48° F.
In 1914 Newman,^ in Western Australia, placed one box containing 50
newly formed puparia in each of four rooms held, respectively, at 32°,
36°, 45°, and 55° F. At the end of 34 days of refrigeration 25 pupae were
taken from each box held at 32° and 36°, and at the end of 70 days of
refrigeration the remaining pupae held at 32° and 36° and all held at 45°
and 55° were removed to the laboratory. None of the pupae removed
yielded adults.
EXPERIMENTAL WORK
Nearly all the experimental work with temperatures lower than 45° F.
was carried on in a thoroughly modern three-story cold-storage plant.
The temperatures of the rooms in this plant were held quite definitely
within certain fixed ranges by hourly inspections made by the storage
employees. One experiment was carried on in a second plant where, as
indicated in the text, the temperature was subject to considerable fluc-
' Lounsbury, C. P. Report of the Government Entomologist, Cape of Good Hope, 1907, p. 56. 1908.
^ Newman, L. J. Annual report of the officer in charge of the insectary for the year ended June 30, 1914.
In Ann. Rpt. Dept. Agr. West. Aust. 1914, p. 61. 1915.
May IS. 1916 Cold Storage and Pupce of Mediterranean Fruit Fly 253
tuation. The temperatures 49° to 51°, 52° to 56°, and 54° to 57° were not
obtainable in the Honolulu cold-storage plants, hence in experiments at
these temperatures ordinary refrigerators were used, as indicated. Usu-
ally pupae of all ages from i to 9 or 10 days were obtained for each experi-
ment, in order that varying effects upon pupae in different stages of
development might be noted. The pupae were sifted from sand beneath
host fruits and placed in storage either in bulk of several thousand in
large jars or, as was more usual, in smaller lots of from one to several
hundreds in vials about i inch in diameter and stoppered with cotton.
Pupae were not placed in or on damp sand or soil, as early experimental
work indicated no advantage from this treatment when pupae are sub-
jected to cold-storage temperatures. The humidity of the storage rooms
varied between 80° and 91°. After refrigeration the pupae were removed
to the laboratory, where they were daily observed for emergence records.
The term "pupa" is used to designate that period in the life history
between the formation of the puparium by the larva and the emergence
of the adult.
Temperature, 32° F. — Of the 13,900 pupae of all ages subjected to
refrigeration at a temperature varying less than half a degree either
above or below 32° F. during the experiment, none survived more than
10 days. In Table I are recorded the results of observations on pupae
refrigerated from 2 to 10 days.
Table I. — Effect upon Mediterranean fruit-fly pupa of refrigeration at J2° F. for from
2 to 10 days
Age of pupse on entering
storage
1 day.
2 days
3 days
4 days
5 days
6 days
7 days
8 days
9 days
Number of pupae yielding adults after removal to normal temperature after
refrigeration for —
2 days. 3 days. 4 days. 5 days. 6 days. 7 days. 8 days. 9 days. 10 days
15
20
32
18
28
29
48
52
51
6
28
21
17
18
27
33
39
33
3
20
Each lot removed after from 2 to 8 days of refrigeration contained 100
pupae; hence, the number of pupae yielding adults represents also the
percentage of survival. Very few pupae survived refrigeration at this
temperature for longer than one week. Thus only 3 three-day-old pupae
out of 900 pupae of all ages survived refrigeration for 8 days, and only i
three-day-old pupa survived refrigeration for 9 days. While the data
in Table I do not show it, the one surviving 9 days of refrigeration was
one out of 300 of like age, and one out of i ,900 of all ages. Not one of
4,500 pupae refrigerated for 10 days survived.
254
Journal of Agricultural Research
Vol. VI, No. 7
Temperatures of from 33° to 34° F. — Only 3 out of 207 pupae held
at 33° to 34° F. for 4 days and at 43° to 45° F. for 8 additional days
yielded adults.
Temperatures of from 33° to 36° F., averaging 34°. — A total of
over 27,097 pupae were used in experiments to determine the effect of a
temperature averaging about 34° F. but varying between 33° and 36° F.
Only I seven-day-old pupa out of 228 of like age or 1,239 of all ages
refrigerated for 16 days yielded an adult; 1,228, 1,164, i>694, and 1,931
refrigerated for 18, 20, 22, and 25 days were dead on removal from
storage. Only 3 out of 272 seven-day-old pupae, or 1,472 pupae of all
ages, produced adults after refrigeration for 15 days, while only 8 out
of 210 eight-day-old pupae and 3 out of 220 seven-day-old pupae, or but
1 1 out of 1 ,630 pupae of all ages from one to eight days old when placed
in storage, produced adults after refrigeration for 14 days. After refrig-
eration for 12 days, 12 eight-day-old pupae, 11 seven-day -old pupae, 2
six-day-old pupae, and 8 one-day-old pupae out of a total of 1,580 pupae
of all ages produced adults. From i to 30 adults emerged from lots of
all ages of pupae, totaling 1,519 forms, except from 126 five-day-old
pupae, after refrigeration for 11 days, but from i to 3 adults emerged
from all lots yielding adults, except from the seven-day-old pupae, which
yielded 30 adults from a total of 265 pupae.
Refrigeration of i ,685 pupae of all ages for 9 days did not prove totally
fatal to any age. Thus 85 out of 340 eight-day-old pupae, and 88 out
of 390 seven-day-old pupae produced adults as compared with 3 four-
day-old pupae, 7 three-day-old pupae, and 2 one-day-old pupae out of a
total of 475 pupae.
Some adults emerged from lots of pupae representing all ages on removal
from storage after 2, 3, 4, 5, 6, 7, and 8 days of refrigeration. On these
days an average of about 1,479 pupae were removed from storage. The
number of pupae surviving is indicated by the data in Table II.
Table II. — Effect upon pupcB of the Mediterranean fruit fly of refrigeration for from i to 8
days at jj° to 36° F.
Age of pupae on entering storage.
Number of pupse yielding adults after removal to normal
temperature after refrigeration for —
2 days. 3 days. 4 days, s days. 6 days. 7 days. 8 days
1 day.
2 days
3 days
4 days
5 days
6 days
7 days
8 days
87
41
49
27
12
10
190
129
59
12
76
47
21
15
405
153
56
25
44
26
14
17
228
434
59
6
II
10
10
221
146
16
8
12
4
10
5
188
216
15
2
14
5
229
150
9
6
6
3
4
2
98
91
The data in Table II are introduced to prove that refrigeration for from
I to 8 days at this temperature is not fatal, and can not be depended
May 15. 1916 Cold Storage and PupcE of Mediterranean Fruit Fly 255
upon to kill all pupae. While an average of about 1,479 pupae were re-
moved each day, the number of pupae of each age is not known; hence no
conclusion can be drawn regarding the relative effect of this refrigeration
upon pupae of different ages. The data given above for 9, 11, 12, 14, 15,
and 16 days of refrigeration seem to indicate that the older pupae with-
stand the effects of cold for a relatively longer period.
In a second experiment 50 out of 200 pupae of all ages yielded adults
after 4 days' refrigeration, and but 15 out of 207 pupae held at 33° to
34° F. for 4 days and then at 43° to 45° F. for 3 additional days.
Temperatures of from 28° to 40° F., averaging 36°. — A total of
8,500 pupae were placed in a cold-storage room the temperature of which
was subject to far greater changes than are usual in commercial plants.
While the temperature averaged about 36° F. a large portion of the time,
for short periods during the night it dropped to freezing or even 28°, and
during the heat of the day when supplies were being removed frequently
rose to 38° to 40°. As each lot of the various ages from K to 9 days re-
moved consisted of 100 pupae, the numbers of pupae yielding adults after
the various numbers of days of refrigeration represent the percentages of
survival. In Table III are recorded the effects of from i to 24 days of
refrigeration on 6,800 pupae.
Table III. — Effect upon pupce of refrigeration at temperatures varying between 28° and
40° F., but averaging about ^6° F. *
Age of pupae on entering
storage.
Number of pupae yielding adults on removal to normal temperature after
refrigeration for—
I day.
3 days.
6 days.
8 days.
days.
days.
i6
days.
i8
days.
days.
1 day.
2 days
3 days
4 days
5 days
6 days
7 days
9 days
8i
92
62
55
70
94
69
35
5
32
21
60
a 27
14
51
a 27,
" Of these three lots of 100 pupse each, 25, 17, and 16 pupae, respectively, yielded adults July 8, or just
before being placed in cold storage.
A total of 1 ,700 pupae of various ages removed to normal temperature
after refrigeration for 21, 27, 29, and 31 days were found to be dead.
Temperatures of from 38° to 40° F. — A total of 52,604 pupae were
used in experiments to determine the effect upon pupae of refrigeration
at 38° to 40° F. An average of i ,860 pupae of all ages were removed after
refrigeration for 3, 4, 6, 7, 8, 10, 12, and 14 days. The number of pupae for
each age varied from 109 to 414 and averaged 234. The number of
pupae surviving refrigeration for from 3 to 14 days is recorded in Table
IV.
256
Journal of Agricultural Research
Vol. VI, l^To. 7
Table IV. — Effect upon Mediterranean fruit-fly pupce of refrigeration for from j to 14
days at 38° to 40° F.
Age of pupae on entering storage.
Number of pupse yielding adults on removal to normal tem-
perature after refrigeration for —
3 days. 4 days
6 days. 7 days
8 days.
days.
days.
days.
1 day.
2 days
3 days
4 days
5 days
6 days
7 days
8 days
no
150
62
35
91
132
235
207
13
170
58
41
86
82
117
234
S
75
81
16
52
59
114
136
I
119
23
13
63
52
121
161
5
121
4
19
24
23
96
180
I
145
6
4
9
6
60
61
124
4
I
12
7
32
63
I
42
4
I
6
o
22
19
After refrigeration for 17 days only 3 out of 306 eight-day-old pupae,
3 out of 384 seven-day-old pupse, 3 out of 206 six-day-old pupae, i out of
162 four-day-old pupae, and 11 out of 374 two-day-old pupae yielded
adults, or only 21 out of 2,352 pupae of all ages survived.
After refrigeration for 18 days only 9 out of 701 eight-day-old pupae,
5 out of 250 seven-day-old pupae, i out of 295 five-day-old pupae, i out of
430 three-day-old pupae, and 13 out of 400 two-day-old pupae yielded
adults; or only 29 out of 2,632 pupae of all ages survived.
Nineteen days of refrigeration proved fatal to 1,911 pupae of all ages
except 2 out of 375 one-day-old pupae. No living pupae were found
among 2,031 pupae of all ages after refrigeration for 21 days, nor among
28,700 pupae of all ages after refrigeration for 35 days.
Temperatures op from 40° to 45° F. — In this experiment to deter-
mine the effect upon pupae of temperatures ranging between 40° and
45° F., 8,800 pupae from i to 10 days old were used. Each unit of
pupse contained 100 forms; hence, the numbers of pupae yielding adults
after refrigeration from i to 27 days as recorded in Table V represent
the percentages of survival.
Table V. — Effect upon Mediterranean fruit-fly pupce of refrigeration at 40° to 45° F .for
from I to 2/ days
Age of pupae on
entering storage.
1 day. .
2 days.
3 days.
5 days.
6 days.
7 days.
8 days.
9 days.
10 days.
Number of pupse yielding adults after removal to normal temperature after —
I day. 3 days. 6 days. 8 days. 10 days. 12 days. 16 days. 18 days. 24 days. 27 days
81
69
47
66
91
55
78
77
91
63
82
87
60
« 17 « 23
15
59
I
38
72
75
44
^21
13
58
54
24
49
60
27
13
13
31
57
16
«!
18, 28, 30, and 10 pupae, respectively, yielded adults just before pupae were placed in
" Besides these,
cold storage.
May 1 5. 1916 Cold Storage and Pupcs of Mediterranean Fruit Fly 257
It will be noted that only 9 out of 300 pupae survived refrigeration for
16 days, while only 4 out of 500 and i out of 500 refrigerated for 18 and
24 days, respectively, survived. Three hundred pupae refrigerated for
31 days and 200 refrigerated for 34 days were found dead on removal.
Temperatures of from 49° to 5 1 ° F. — Temperatures ranging between
49° and 5 1 ° F. and averaging about 50° have proved most interesting of
all, as these appear to be very close to the point below which the insect's
activities cease. This temperature was secured by use of an ordinary
refrigerator 42 by 34 by 18 inches. During the period from May to July,
1 914, 31,700 pupae were used in an experiment to determine the efifect
of this temperature upon pupal development. Pupae in 15 lots, of ages
ranging from i to 8 days, and averaging 3,523 pupae for each of the 8 days
represented, were held in storage for two months before removal. Fre-
quent observations were made but no pupae completed their development
and yielded adults in storage. On removal to normal temperature all of
the 31,700 pupae were found dead.
The second lot of 7,800 pupae placed in storage when 5 days old yielded
a few adults. Thus, 9 out of 7,800 yielded i, 2, 2, 3, and i adult in
storage after refrigeration for 20, 23, 44, 46, and 47 days. In other
words, it took these 9 pupae from 20 to 47 days to accompHsh the develop-
ment in refrigeration which at an outdoor temperature at that season,
July, 1 91 4, would have taken only from 4 to 5 days.
Temperatures of from 52° to 56° F. — Ten larvae pupating in a
refrigerator held at 52° to 56° F. yielded 2 and i adult in storage after
refrigeration for 38 and 52 days, respectively. The remaining 7 pupae
died.
Temperatures of from 54° to 57° F. — Temperatures of from 54° to
57° F. were obtained by using an ordinary refrigerator 46 by 27 by 18
inches. A total of 22,700 pupae were used varying in age from ^ to 9
days. Not less than 1,400 pupae, or more than 3,500 pupae of any age,
were used. In Table VI are recorded the reactions of 3,100 one-day-old
pupae to these temperatures.
From the data in Table VI it will be noted that 54° to 57° F. is not
in all cases fatal to pupal development, although a high mortality occurs.
Each outward date represents 100 pupae. As the heavy line extending
diagonally across the table indicates the dates on which pupae were
removed from refrigeration, and as the normal pupal development is
completed at this season of the year at Honolulu in from 9 to 1 2 days, the
data prove that development continues at this temperature as evidenced,
first, by the rate of emergence of adults after the pupae are removed from
refrigeration up to the thirtieth day of refrigeration, and, secondly, by
the emergence actually occurring within storage on the thirty-first day
and up to the thirty-seventh day of refrigeration. Thus development
was wholly completed and emergence had taken place at this temper-
ature among pupae removed from refrigeration after 37, 38, and 39 days.
258
Journal of Agricultural Research
Vol. VI, No. 7
Table VI. — Effect upon i-day-old Mediterranean fruit-fly pupce of refrigeration at 54° to
57° F. Pupce placed in refrigeration August 22, ipij
Date and emergence of adults.
Outward
date.
August.
September.
28
1
29
0
I
30
0
0
31
10
2
I
32
4
2
3
25
3
21
17
4
19
29
5
I
36
I
6
0
7
0
8
15
I
9
14
24
I
10
I
17
19
14
II
12
6
0
12
16
23
1
13
4
22
13
2
14
I
18
22
I
0
15
18
5
0
!°
0
16
17
II
I
0
0
17
8
23
13
4
18
4
28
24
8
I
J,
I
1
6
19
10
30
3
2
0
0
6
20
7
II
21
3
0
3
21
13
19
24
3
4
I
7
I
2
I
8
3
3
22
6
9
10
6
"
7
5
3
6
I
6
16
23
■9
7
IS
3
3
9
2
7
6
9
24
6
9
3
-2
9
8
2
0
S
3
25
5
6
6
7
I
I
4
2
26
3
3
I
I
2
27
2
J
4
28
2
2
29
Aug. 24
26
27
28
Sept. 3
^"
4
5
6
7
~
8
-
1
0
9
10
-\
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
ii
Of 1,700 pupae one-half day old when placed in refrigeration none
emerged within storage until the twenty-fourth day of refrigeration, after
which no data were secured. One hundred pupae removed after 24 days
of refrigeration produced 2, 27, and 5 adults 4, 5, and 6 days after removal,
proving that even among these very young pupae development was not
completely arrested. Out of two lots of 100 pupae each, 46 and 44 pupae,
respectively, refrigerated for 22 and 23 days, yielded adults 3, 4, and 5
days after removail from storage.
Data on file bring out an interesting fact that might be expected
from the data in Tables VI and VII. The older the pupae when placed in
refrigeration, the quicker they develop and produce adults while in
refrigeration. Thus in Table VII are recorded data on the development
of six-day-old pupae.
It will be noted that while a few one-day-old pupae require a minimum
of 31 days of refrigeration for development, 2 six-day-old pupae com-
May 15, 1916 Cold Storage and Pupce of Mediterranean Fruit Fly 259
pleted their development and produced adults in storage on the fifth
day of refrigeration, and that thereafter emergence of adults continued
until all living pupae yielded adults in storage by the end of the sixteenth
day of refrigeration except 2, which yielded adults on the fourth day
after removal after 16 days of storage. Data on 1,900 eight-day-old
pupae show that from i to 5 pupae among each of 14 different lots of 100
completed their development and yielded adults on the second day of
refrigeration, that an average of 43.5 per cent of 14 lots yielded adults
on the third day, and that emergence of adults was completed by the
seventh day except in one instance where 2 pupae yielded adults in storage
on the ninth and tenth days of refrigeration.
Table VII. — The effect upon six-day-old Mediterranean fruit-fly pupce of refrigeration at
54° to 57° F. Pupce placed in refrigeration August 22, IQIJ
Date and emergence of adults.
Outward date.
August-
September—
us
26
27
28
29
30
31
I
2
3
4
5
6
7
8
9
10
57
I
6
45
2
26
29
.17
4
I
I
I
0
0
0
22
67
10
13
13
7
II
13
II
13
II
0
28
"^■^
Sept. 3
"~~
8
13
13
12
10
10
5
21
6
19
13
12
16
13
13
9
13
13
12
15
9
13
15
9
10
0
II
12
17
19
II
10
18
13
3
0
0
I
0
I
2
I
0
0
I
I
0
2
0
0
0
0
0
0
0
0
0
2
10
7
3
2
6
I
I
0
h
0
0
a
8
2
3
I
—
3
2
3
13
0
Liii
Data on file covering observations on 1^600, 1,400, 1,700, 2,200, 3,300,
4,000, 3,100, 3,500, and 1,900 pupae, i, 2, 3, 4, 5, 6, 7, 8, and 9 days old,
respectively, show a steady increase in the pupae completing their devel-
opment and yielding adults in refrigeration, and their tabulation shows
a transition from the condition in Table VI through that of Table VII
to the condition set forth for eight-day-old pupae. One lot of two-day-old
pupae left in storage for 37 days yielded adults almost daily between the
twenty-sixth and thirty-fifth days of refrigeration. Eight lots of three-
day-old pupae left in storage from 32 to 39 days yielded adults between the
twenty-third and thirty-third days of refrigeration and none thereafter.
Fourteen lots of four-day-old pupae in storage from 25 to 39 days yielded
adults between the sixteenth and twenty-seventh days of refrigeration
and none thereafter. Eight lots of seven-day-old pupae in storage from
13 to 21 days yielded adults between the second and eleventh days of
refrigeration and none thereafter.
26o Journal of Agricultural Research voi. vi, No. 7
CONCLUSION
From the data secured during experimental work reported on the fore-
going pages, including observations on 173,318 pupae of the Mediterra-
nean fruit fly (Ceratitis capitata Wied.), it appears that no pupae survive
refrigeration for longer periods than is necessary to cause the death of
eggs and larvae in host fruits held at corresponding temperatures.
About 50° F. is the critical point below which development can not
take place and below which death will follow if refrigeration is continued
sufiEiciently long. At 49° to 51° only 9 out of 39,500 pupae yielded
adults in refrigeration 20 to 47 days after the inward date, while 3 out
of 6 held at 52° to 56° yielded adults in refrigeration 38 to 52 days after
the inward date. Many pupae can complete their entire development in
refrigeration at 54° to 57°, while higher temperatures, not considered
here, merely retard development without causing noticeable mortality.
Pupae can not withstand temperatures below 50° F. for prolonged
periods of time. Only 3 and i pupa survived refrigeration for 8 and 9
days, respectively, at 32°, while none of 4,500 pupae survived 10 days at
this temperature. Refrigeration at a temperature averaging 34°, but
ranging between 33° and 36°, proved fatal after the seventeenth day;
6,017 pupae refrigerated at this temperature for 18 and 25 days yielded
no adults, while the number to yield adults after refrigeration for 14 and
17 days was very small. No pupae survived refrigeration at 28° to 40°
but averaging 36°, for more than 10 days. A temperature of 38° to 40°
proved fatal after the nineteenth day; 30,731 pupae refrigerated for from
21 to 35 days failed to yield adults on removal to normal temperatures.
After refrigeration at 40° to 45° pupae from each of two lots removed
after refrigeration for 24 and 27 days, respectively, yielded adults; 500
pupae removed after refrigeration for from 31 to 34 days proved to be
dead.
It does not seem safe to conclude that the age of the pupa has a direct
bearing upon its ability to withstand the more ordinary ranges of cold-
storage temperatures.
EFFECT OF CLIMATIC FACTORS ON THE HYDRO-
CYANIC-ACID CONTENT OF SORGHUM
By J. J. WiLLAMAN and R. M. West, Assistant Chemists, Agricultural Experiment
Station of the University of Minnesota
[In collaboration with F. S. Harris, Agronomist, Utah Agricultural ExperiTnent Station; L. E. Call, Agron-
omist, Kansas Agricultural Experiment Station; and Beyer Aune, Superintendent, Belle Fourche Experi-
ment Farm., Newell, S. Dak.]
INTRODUCTION
The present experiments are a continuation of those carried out in
1914 (10)^ on sorghum (Sorghum vulgare). In the latter a correlation
was sought between the soil conditions, especially the supply of nitrogen, •
and the amount of the cyanogenetic glucosid (dhurrin) in the sorghum.
It was found that on fertile soils nitrogenous fertilizer has no appreciable
effect, but on poor soil added nitrogen may increase the amount of
hydrocyanic acid, though only to a small extent. Since the evidence
indicated that climate and variety may be more important factors than
soil nitrogen in determining the amount of the glucosid in this plant,
experiments were carried out during 19 15 to study the effect of climatic
conditions. It was thought that conditions of high or low temperature,
much or little available water, slow or rapid growth, might affect the
metabolism of sorghum sufficiently, not only to show the causes of the
varying amount of dhurrin, but also to throw some light on the physio-
logical function of this glucosid.
EXPERIMENTAL WORK
Seeds of two varieties of sorghum were obtained. One was Early
Amber, grown in Minnesota, and is designated in these experiments
Variety N. The other was Southern Cane, a variety similar to the
first, but grown in Missouri. It is designated Variety S. In order to
secure as widely varying climatic conditions as possible, one-twentieth-
acre plots of each variety were grown at four different State experiment
stations. A brief description of each plot follows:
1. University Farm, St. Paul, Minn. Very fertile, black loam, fair drainage.
Planted on June 3; sprouted on June 12; cultivated twice. Season very cold and
wet; sorghum three or four weeks behind the normal in development; did not reach
maturity, but was killed by frost in the soft dough stage.
2. Agricultiural Experiment Station, Logan, Utah. Irrigation farming. Plots on
McNiel farm, North Logan; the two varieties alternated with beans; soil a clay loam,
rich in maniu"e. Planted May 15; appeared aboveground on June i; irrigated on
July 9 and August 11; cultivated on June 10, June 17, July i, July 13, and August 17.
Rainfall up to June 10 was abnormally high, which kept the soil cold and retarded
growth of crops. During the rest of the season optimum moistiu-e content of soils
'Reference is made by niunber to "Literature cited," p. 272.
Journal of Agricultural Research, Vol. VI, No. 7
Dept. of Agriculture, Washington, D. C. May is, 1916
dr Minn.— 9
(261)
262
Journal of Agricultural Research
Vol. VI. No. 7
obtained. Sorghum made slow growth; leaves yellowish for first seven or eight
weeks.
3. Agricultural Experiment Station, Manhattan, Kans. Plots grown on "creek
bottom land," broken from native sod in 1913; drainage poor. Planted on June 15;
appeared aboveground on June 22; cultivated twice. Sorghum 30 days slower in
maturing than usual, owing to excessive rains.
4. Belle Fourche Experiment Farm, Newell, S. Dak. Dry farming. Planted on
June 10; appeared aboveground on June 26; cultivated on July 22 and August 7;
harvested on October 12. Season cold and wet; rainfall far above normal.
5. Belle Fourche Experiment Farm, Newell, S. Dak. Irrigation farming. Planted
on May 31; appeared aboveground on June 26; cultivated on July 23 and August 10;
irrigated on August 17; harvested on September 16, when plants were just headed out.
From the time when the plants were from 20 to 30 cm. in height, sam-
ples were taken every 10 days. They were usually cut between 9 and
12 a. m., although it has been found that the time of day makes no
difference in the amount of hydrocyanic acid present. Plants were
selected which represented the average of the plot on that date. For the
first sample the whole plant was cut into i-inch lengths and packed into
a 600 c. c. friction-top tin can with 20 c. c. of 3 per cent alcoholic sodium
hydrate and 2 c. c. chloroform for preservatives and sent to the Minne-
sota laboratory for analysis. For the other samples the leaves were cut
off where they join the sheath, and the leaves and stalks were packed
and analyzed separately. The weight of leaves and of stalks in the
total sample cut was recorded. From the fourth sample on, cans of
1,600 c. c. capacity were used. An alkaline preservative was used so as
to prevent any possible loss of hydrocyanic acid set free by enzymic
activity. Alcohol instead of water was used as a solvent for the alkali,
because it penetrates the plant tissues more readily. The chloroform
prevented any fermentative changes. In the case of the South Dakota,
Kansas, and Utah samples, from two to five days elapsed from the time
the samples were cut till they were analyzed. In the case of the Minne-
sota samples, the fresh material was analyzed. In order to test the
efficiency of the preservative, several samples from the Minnesota plots,
representing the various stages of maturity of the samples outside of
Minnesota, were analyzed for hydrocyanic acid before and after storage
in cans, with the results given in Table I.
Table I. — Efficiency of an alkaline preservative in preventing loss of hydrocyanic acid in
sorghum
Preservative treatment.
Percentage of hydrocyanic
acid in dry matter.
Fresh.
Preserved.
Preserved for four days with alcoholic sodium hydroxid and
chloroform
0. 019
. 026
. 009
. 016
0. 020
Do
. 029
. C09
. 019
Preserved for eight days with alcoholic sodium hydroxid and
chloroform
Do
May 15, 1916
Hydrocyanic-acid Content of Sorghum
263
The differences noted are within the limits of accuracy of sampling and
analyzing; hence, this method of preservation can safely be used on
sorghum plants at least through the stages of maturity represented in
these experiments.
About 50 gm. of the sample, after thorough mixing and fining with a
knife, were used to determine the percentage of dry matter. For the
determination of the hydrocyanic-acid content, from 50 to 70 gm. were
ground in a food chopper, placed in an 800 c. c. Kjeldahl flask, together
with 250 c. c. of 5 per cent tartaric acid, and distilled slowly into 10 c. c.
of 2 per cent sodium hydroxid until the distillate was nearly 100 c. c.
This completely hydrolizes the dhurrin and carries the hydrocyanic acid
over into the' alkaline distillate. The latter was made to 100 c. c. and
aliquots used for the determination of hydrocyanic acid according to the
method of Viehoefer and Johns (9) . This method was found to be easier
and more accurate than the thiocyanate method used in 1914.
The complete analytical results appear in Table II. The figures for
the amount of hydrocyanic acid in the whole plant were computed from
the relative proportion of leaves and stalks in each sample.
Table II. — Hydrocyanic-acid content of sorghum from the various experimental plots
[The percentage of hydrocyanic acid is reported on a dry-matter basis]
Date of
sampling.
Age of
plants
since
sprout-
ing.
Plot N.
Plot S.
Plot and sample
No.
Height
of
plants.
Percentage of hydrocy-
anic acid.
Height
of
plants.
Percentage of hydrocy-
anic acid.
Stalks.
Leaves.
Whole
plant.
Stalks.
Leaves.
Whole
plant.
Minnesota:
July IS
July 24
Aug. 3
Aug. 13
Aug. 24
Sept. 3
Sept. 13
Sept. 23
July 19
July 29
Aug. 7
Aug. 18
Aug. 28
Sept. 7
Sept. 17
July 16
July 27
Aug. s
Aug. 16
Aug. 25
July 26
Aug. s
Aug. 14
Aug. 24
Sept. 4
Sept. 14
July 26
Aug. 5
Aug. 14
Aug. 24
Sept. 4
Days.
33
42
52
62
73
83
93
103
49
S9
68
79
89
99
109
24
35
44
55
64
30
40
49
59
70
80
30
40
49
59
70
Cm.
25
44
68
90
135
160
188
20s
36
56
78
120
161
174
190
18
75
137
200
260
ii
61
91
142
189
282
35
SI
96
134
193
0. 114
.028
. 019
.009
.002
. 001
Trace.
Cm.
25
48
65
88
133
160
192
206
36
59
84
116
160
175
192
26
95
150
210
255
33
61
86
137
187
287
37
SI
91
134
190
0.079
• 032
0. 026
.018
.012
.000
Trace.
0.031
.021
.008
.004
.004
.001
0. 028
• ois
.023
Trace.
.000
O.OJS
.031
.011
.007
.004
. 002
.003
.003
6
8
Do.
Utah:
•034
. 019
.021
.009
.008
. 040
.031
. 029
.028
.019
Trace.
.000
.013
.023
.022
.026
•039
.026
.007
Trace.
.025
.032
•034
.041
6
7
Kansas:
. 016
.008
.003
.014
.014
.008
.000
. 000
.017
.007
. 000
.000
.030
.020
Dakota (dry farm-
ing):
. 020
.011
. 000
.002
• 030
.004
.000
.000
.013
.000
. 004
. 009
. 000
.000
.021
.008
.006
.004
6
Dakota (irriga-
tion):
.009
.006
Trace.
...do...
. 004
.000
.000
.008
Trace.
...do..
Trace.
.000
.000
Trace.
.001
Trace.
Do.
264
Journal of Agricultural Research
Vol. VI, No. 7
In figure i the percentages of hydrocyanic acid in the whole plant are
plotted against the age in days. No noteworthy differences were no-
ticed when the height of the plants was used instead of the age in days.
Figure 2 represents the growth curve of the various plots, where the
height in centimeters is plotted against the age in days since sprouting.
In order to study the relation between climatological factors and the
content of hydrocyanic acid, figures 3 and 4 were constructed. In
Fig. I. — Curves showing the hydrocyanic-acid content of sorghum on the various plots. (Percentage of
hydrocyanic-acid computed to drj'-matter basis.)
figure 3 are plotted first the precipitation (in inches) during 15-day
intervals; second, the temperature (degrees Fahrenheit), using averages
for lo-day periods, and, third, the mean relative humidity (percentage)
at 6 a. m., all for the five months May to September, inclusive. In figure
4 the history of each plot for the season is shown and includes the rain-
fall, temperature, and hydrocyanic-acid curves on the same graph.
The dates for planting, sprouting, appearance of seed panicles, and
irrigations are also shown.
May 15, 1916
Hydrocyanic-acid Content of Sorghum
265
DISCUSSION OF RESULTS
The season of 191 5 furnished some excellent extremes in weather
conditions for this experiment. Figure 2 shows that, as regards tem-
perature, the two more southern States, Kansas and Utah, form one
pair, and South Dakota and Minnesota another, with approximately
10 degrees difference between them during the growing season. Of the
two warmer stations, Utah had a low rainfall, and irrigation was resorted
-20 46 "TO 80 100"
AGE OF PLANTS IK DAYS
Fig. 2.— Curves showing the rate of growth of the sorghum on the various plots.
to; while Kansas had a very abundant rainfall, resulting even in flood
conditions during May, June, and July. The two more northern sta-
tions had about the same rainfall for the first three months of the experi-
ment, but during the period when the samples were taken the rainfall
of South Dakota dropped below that of Minnesota. This was par-
ticularly the case during August. The 191 5 rainfall of South Dakota
was above normal, and as a result the plot on irrigation ground was
irrigated only once.
266
Journal of Agricultural Research
Vol. VI, No. 7
The five plots diflfered rather widely in their rate of growth, as is shown
in figure 4. The Utah plants were 48 days old before attaining the
height required for first sample. During this time they looked yellow
and unthrifty, owing to excess moisture and cool soil. Subsequently
.0 "*>
3n
JU,
?=
t
JULY
9
9
AUOU|T
SEPTEIIBeR,,
A
\
MINNESOTA
. » 0 UTAH
■■ ■ — ^ KANSAS
DAKOTA
\
/-
-\
/
\
X
u
z
/
\
0
>-^
y
A
/I
^,
\
K
^
/''*"
s
/
\>N
\
I
^.
/
2 ^
y/
^0'
\
—
\
\]>
A
?
"^"1^
n
\
i^
\
V
./[--"
->!
K 8"
^
^
0
^,
Nnr-^
^\
^
1 1
1'^
^
:^"
'*•..,
::i:^
.-'"^
^
V
>
S
v>^>-
>-/
w
y
<o
^..
i--^
_^ . ^'
>
so
-***" ' ■ ■ '
•^c
-V—
-*^
:«— «=i
^
w
0
^,
^^
w
a.
^""^
* Ol
■-■^
..'^
3 S(I
Z
kl
>
.^^
,^
^
.--^
Ill
""~*^
>
Fig. 3. — Curves showing the precipitation, temperature, and humidity relations at the various experi-
ment stations during the growing season of 1915.
the plots grew nearly as fast as those at the other stations and gave a
higher yield of dry cane at the end of the season than did the South
Dakota plots, although the latter grew much taller. The Kansas plots
grew the most rapidly.
May 15, 1916
Hydrocyanic-acid Content of Sorghum
267
Se PTEMBER
Fig. 4. — Curves showing the contemporary climatic conditions at the various plots, together with
crop data and hydrocyanic-acid content.
36288°— 16 2
268 Journal of Agricultural Research voi. vi, No. 7
Accompanying these various conditions were also widely differing
amounts of hydrocyanic-acid glucosid. How the correlations between
these two may be explained will depend upon the function assigned to
the dhurrin in sorghum. Various uses have been attributed to glucosids
in plants, as (a) a protection against bacteria and other enemies by
means of the poison set free when some glucosids are hydrolyzed; (6)
a reserve food material in the plant; (c) the inactive form of a stimu-
lating hormone (2) set free when necessary by a glucosidase; (d) a harm-
less compound absorbing injurious products of metabolism; (e) an in-
active storage of "respiratory pigments," and other uses.^ Hydrocyanic
acid itself is thought by some investigators to be a necessary intermediate
product in protein formation (6, 8). As such, it is probably rather
transitory in the plant, and seldom occurs free in any appreciable
amount.
A discussion of each factor which might have any bearing on the
cause of the variations in cyanid content, or throw any light on the
function of dhurrin in sorghum, follows.
1. Humidity. — It is hard to perceive how the relative humidity might
have any direct bearing on the quantity of dhurrin produced. The
humidity affects primarily the rate of transpiration, and this in turn
might influence the rate of growth. The latter factor is considered in
the next paragraph. The interesting thing to note in the humidity
curves in figure 2 is the fact that the Utah curve shows a decrease dur-
ing a period of decreasing precipitation, which is natural, but the Kansas
and Minnesota curves show an increase during periods of decreasing
precipitation. It is possible that this very low humidity in Utah caused
a rate of transpiration too high for the best development of the plants,
and their growth was retarded accordingly. When the humidity was
lowest, in July and August, the plots received their two irrigations.
Following these the growth was more rapid. If the humidity affects
the amount of glucosid at all, it is by means of its effect on the nutri-
tion and growth of the plant.
2. Moisture supply. — As mentioned above, there are among the
four stations one having very high, one with very low, and two with
medium rainfall. Two plots at the last-named stations were under
irrigation and one under dry-farming methods of cultivation. In the
data as a whole there is no evident correlation between the amount of
the glucosid and the moisture supply for the five months. Arranging
the stations in the order of their moisture supply, they are Kansas, Min-
nesota, South Dakota irrigation. South Dakota dry farming, and Utah;
while arranged in the order of their cyanid content they are Minnesota,
Utah, Kansas, and South Dakota dry farming the same, and South
Dakota irrigation. However, by a closer examination of the curves for
iFor a complete discussion of the function of glucosids in plants see Armstrong, E. P., The Simple
Carbohydrates and the Glucosids. Ed. 2, p. 123-133. London, New York, 1912.
May IS, 1916 Hydrocyanic-acid Content of Sorghum 269
each plot, the following examples tending to show that high water sup-
ply is often accompanied by a low cyanid content are discernible: (i)
The normal hydrocyanic-acid curve for sorghum during the first two-
thirds of its growth is a smooth curve, with a steady decrease in the acid.
The Utah curve is an exception to this. In this plot there was for several
weeks very rapid transpiration of water, owing to low humidity; hence,
the plants and soil were reduced nearly to the minimum water require-
ment. Shortly after the first irrigation the hydrocyanic acid is seen to be
on a normal decline. Twenty days after this irrigation, however, the plots
had become comparatively dry again, and the hydrocyanic acid shows
a less decrease in variety S and an actual increase in variety N. The
second irrigation was followed by another decline in hydrocyanic acid.
By the latter part of August the need of water was once more felt, and
the cyanid in variety N, at least, had ceased to decrease. (2) The curve
for the South Dakota dry-farming plot also shows an abnormality in
that the last part of it has an upward turn in the case of variety N. It
is possible that this may be due to the smaller supply of moisture avail-
able at this time. (3) In the two South Dakota plots, both received the
same amount of rain; one was irrigated once and the other, being culti-
vated by dry-farming methods, had a larger reserve supply of water.
This would apparently give them about the same amount of water
supply, except for the fact that the irrigation was a heavy one, and the
heavy rains during May, June, and July disturbed the usual dry-farming
condition of the soil. Assuming that the irrigated plots did have more
water available, it will be seen that they also contained a less amount of
hydrocyanic acid. (4) On analyzing some sorghum plants grown in
pots in the greenhouse, they were found to contain no hydrocyanic acid.
A few weeks later some larger plants from this same group, growing in
drier soil, owing to lack of care in watering and to a larger demand made
by the plants, were found to contain some of the acid. There appears,
therefore, to be a relation between the supply of water and the amount
of dhurrin present. This may be explained on the hormone theory
With a liberal supply of water, other things being equal, the plant's
means for growth are adequate and it needs less glucosid. With a
decreasing water supply, hovv^ever, the plant may need the hormone
stimulus for growth, and more glucosid is produced. Although, as shown
by Briggs and Shantz (5), sorghum has a lower water requirement than
most cultivated plants, it is no doubt affected by changes in the supply
of moisture.
3. Temperature. — No correlation has been found between the con-
tent of dhurrin in sorghum and variations in temperature, at least for
the range of temperatures which obtained during this experiment. The
increase in hydrocyanic acid which sometimes occurs when plants are
frosted may be due to disturbed enzym balance.
270 Journal of Agricultural Research voi. vi. no. 7
4. Rate of growth. — The Kansas and Utah plots present extremes
in rate of growth and thriftiness of the sorghum plants; and they also
present cases of relatively low and high hydrocyanic-acid content,
respectively. The cane on the Minnesota plots grew more slowly than
that from South Dakota, and it also contains a very much higher hydro-
cyanic-acid content. During the first four or five weeks on the Minne-
sota plots the plants grew very poorly, the weather being cold and damp.
The plants were yellow and uneven in height, similar to those obtained
from Utah. The samples from these two stations were by far the highest
in hydrocyanic acid. In fact, the percentage in the first Minnesota
sample, variety N (0.114 per cent), is the highest ever observed in the
authors' experience with sorghum. In the Minnesota samples of 1914
those grown on the poorer sandy soil were the higher in cyanid. These
examples, together with one furnished by Avery (3), show that some
significant relation may exist between poor conditions of growth and
high dhurrin content. In opposition to this, however, is the finding of
Alway and Trumbull (i) that the yellower plants in a field contained a
smaller amount of the acid. Balfour (4) found more in plants infested
with Aphis sorghi than in others not so affected. If these facts are now
applied to the various theories mentioned above, as to the function of
glucosids, some of the possibilities are as follows: (i) If this particular
glucosid is a food storage, it is difficult to see how it could exist in largest
quantities in the unhealthy, poorly nourished, slow-growing plants. (2)
If the constituents of the glucosid act as stimulating hormones when set
free by an enzym, it is possible that when conditions of growth are poor
more of the glucosid is produced. (3) If the glucosid is an absorber of
harmful products of metabolism under disturbed metabolic conditions,
an excess of hydrocyanic acid might be produced. Of these three the
authors believe the second to be the most tenable for dhurrin, according
to the available evidence on this question.
5. Variety. — The most striking phenomenon in this experiment is the
fact that Variety vS has consistently a greater amount of hydrocyanic acid
than Variety N. That varietal difTerence is very important was brought
out also in the 1914 experiments. In fact, the authors are confident that
the most marked and constant differences in the hydrocyanic-acid con-
tent of various sorghum plants will be found to be due to variety rather
than to external conditions. A comparative study of the glucosid con-
tent of all varieties of sorghum would be interesting and valuable.
6. Distribution in the plant. — The foregoing discussion has been
based on the dhurrin content of the whole plant. As is seen from Table II,
the distribution of the glucosid between stalk and leaves in the different
plots is variable. There is in every instance a more rapid decrease in the
stalks than in the leaves, but the comparative rate of decrease varies.
The Minnesota and Utah plots had the highest amount in the stalks and
also had the slowest growth and the thinnest stalks. The Kansas and
May IS, 1916
Hydrocyanic-acid Content of Sorghum
271
the South Dakota plots, on the other hand, had little or no cyanid in the
stalks and had the most rapid growth. The Kansas stalks were very
heavy and succulent; they had developed very rapidly; and they con-
tained no cyanid whatever. The significance of this is not clear.
7. Daily variation. — In order to compare the glucosid content in
sorghum with Treub's findings (7, 8) that in Pangium edule there is a
daily variation in glucosid content with a maximum about midday, some
analyses were made at sunset and sunrise of succeeding days, with the
results given in Table III.
Table III. — Variation in the glucosid content of sorghum at different parts of the day
Variety.
Part of plant.
Percentage of hydrocyanic
acid.
Evening.
Morning.
Variety N
Do
Leaves
Stalks
0. 01 5
.013
. 020
.023
0. 012
. 014
.023
.031
Dakota Amber
Leaves
Do
Stalks
There seems to be no constant variation in sorghum between night and
day. This lends support to the view that dhurrin is not a food storage.
Although other factors have important bearing on the growth and
health of plants, those discussed above are the most readily measured
and, hence, best used as bases for comparison between widely separated
stations. It is realized that determinations of soil moisture at various
times throughout the growing season would give a much more accurate
idea of the available moisture than precipitation measurements. As
regards soil, each plot was grown on soil which has produced good crops
in the past and was cultivated according to the customary methods for
sorghum at those stations. Since the 191 4 experiments showed that soil
is a minor factor in affecting the hydrocyanic-acid content of sorghum,
the ignoring of this factor in the above comparisons is justified.
SUMMARY
Two varieties of sorghum. Southern Cane and Early Amber, ^'\•ere
grown on plots in Minnesota, Utah, Kansas, and South Dakota under
widely different climatic and cultural conditions. The amount of the
glucosid dhurrin in each plot varied considerably. The following corre-
lations relative to the amount of glucosid were found to exist.
(i) Unhealthy plants usually contain more hydrocyanic acid than
healthy ones. The unhealthy condition may be due to malnutrition, to
improper transpiration, to insect attack, or to other causes. It is possible
that under such conditions the plant produces more glucosid for the sake
of the stimulating honnones in it.
272 Journal of Agricultural Research voi. vi. No. 7
(2) The apparent effect of humidity and temperature on the amount of
cyanid in sorghum is probably due to the indirect effect on the rate of
growth.
(3) Adequate water supply is usually accompanied by low, and inade-
quate by high, hydrocyanic-acid content. This is probably due to the
need of glucosid stimulation when the water supply becomes low.
(4) The character of the growth of the plant affects the distribution
of dhurrin between leaves and stalks, there being a proportionately
smaller amount in the thick, heavy stalks than in the slender ones.
(5) There is no consistent daily variation in the amount of dhurrin,
which argues against the functioning of this glucosid as a food storage,
(6) Of the two varieties used in this experiment, the Southern Cane in
every plot but one had a higher content of hydrocyanic acid than the
Early Amber. Varietal difference is probably of more weight in deter-
mining the amount of hydrocyanic acid in sorghum than are the condi-
tions of growth.
LITERATURE CITED
(i) Alway, F. J., and Trumbull, R. S.
1910. On the occurrence of prussic acid in sorghum and maize. In Nebr.
Agr. Exp. Sta. 23d Ann. Rpt., 1909, p. 35-36.
(2) Armstrong, H. E., and Armstrong, E. F.
191 1. The function of hormones in regulating metabolism. In Ann. Bot.,
V. 25, no. 98, p. 507-519.
(3) Avery, Samuel.
1902. Laboratory notes on poison in sorghum. In Jour. Comp. Med. and Vet.
Arch., V. 23, no. 11, p. 704-706.
(4) Balfour, Andrew.
1904. Cyanogenesis in sorghum vulgare. /w ist Rpt. Wellcome Research Lab.
Gordon Mem. Col. Elhartoum, [1903/04], p. 46-48^
(5) Briggs, L. J., and Shantz, H. L.
1914. Relative water requirement of plants. In Jour. Agr. Research, v. 3,
no. I, p. 1-63, pi. 1-7.
(6) Ravenna, Ciro, and Zamorani, Mario.
1909. Nuove ricerche suUa funzione fisiologica dell' acido cianidrico nel
Sorghum vulgare. In Atti R. Accad. Lincei, Rend. CI. Sci. Fis.,
Mat. e Nat., v. 18, sem. 2, fasc. 8, p. 283-287.
(7) Treub, Melchior.
1907. Notice sur "I'effet protecteur" assigne a I'acide cyanhydrique des
plantes. In Ann. Jard. Bot. Buitenzorg, v. 21 (s. 2, v. 6), pt. i, p. 107-
114, pi. 3-4.
(8)
1909. Nouvelles recherches sur le r61e de 1 'acide cyanhydrique dans les plantes
vertes. III. In Ann. Jard. Bot. Buitenzorg, v. 23 (s. 2, v. 8), pt. i,
p. 85-118, pi. 18-23 (col.).
(9) ViEHOEVER, Amo, and Johns, C. O.
191 5. On the determination of small quantities of hydrocyanic acid. In Jour.
Amer. Chem. Soc, v. 37, no. 3, p. 601-607.
(10) Willaman, J. J., and West, R. M.
1915. Notes on the hydrocyanic-acid content of sorghum. In Jour. Agr,
Research, v. 4, no. 2, p. 179-185, 2 fig. Literature cited, p. 185.
EGG AND MANNER OF OVIPOSITION OF LYCTUS
PLANICOIvLIS '
By Thomas E. Snyder,
Assistant in Forest Entomology, Bureau of Entomology
HISTORICAL SUMMARY
The so-called "powder-post" injury to seasoned wood products is
widely distributed over the world. Of the various beetles causing this
type of injury, species of the genus Lyctus Fab. are by far the most
important. While these beetles and their damage have an extensive
literature, the place and manner of oviposition have remained obscure.
Heeger (3),^ in 1853, described and figured the egg, larva, and pupa of
a beetle attributed to a European species, Lyctus pubescens Panzer.
Duges (i),^ in 1883, described and figured the larva, pupa, and adult of
L. planicollis Le Conte (?), proving that Heeger was in error in
ascribing the larva he figured to the genus Lyctus. Xambeu (7), in
1898, described the egg and manner of oviposition of L. linearis Goeze
(canaliculatus Fab.). Recently the eggs of the native species, L. plani-
collis of the southern United States, have been found by the writer.
This egg is very unlike that described and figured by Heeger as
the egg of L. pubescens, and it differs from the egg of L. canaliculatus
as described by Xambeu, being of a most unusual type for Coleoptera.
The following brief notes on the mating and oviposition of the southern
species (L. planicollis Le Conte) were made on material being reared
either at Washington, D. C, or at Falls Church, Va., in buildings kept
dry and at a temperature above freezing.
LIFE CYCLE
MATING
The beetle passes the winter in the larval stage, but in cold weather
the larvae are more or less dormant and infested stock may consequently
pass unnoticed. Mating takes place and the eggs are deposited soon after
the adult beetles emerge from the wood in the spring. At Washington,
D. C, and Falls Church, Va., the first adults emerged from infested wood
in rearing cages during the last part of February and first part of March,
1 The specimens on which this paper is based were identified by Mr. W. S. Fisher, Specialist on Forest
Coleoptera, of the Branch of Forest Insect Investigations, Bureau of Entomology.
2 Reference is made by number to "Literature cited," p. 276.
^ According to Duges, the material on which his paper was based had been determined by two different
authorities as Lyctus planicollis Le Conte (of southern U. S.) and carbonarius Waltl. (of Mexico and Florida).
Dugfes refers to the species as planicollis in the title and carbonarius on the plate. Hopkins (s, p. 134) states
that L. carbonarius is evidently distinct from L. planicollis, and therefore Dugds's specimens are L. car-
bonarius.
Journal of Agricultural Research, Vol. VI, No. 7
Dept. of Agriculture, Washington, D. C. May 15, 1916
dq K— 32
(273)
2 74 Journal of Agricultural Research voi. vi. No. ?
in 1914 and 1915. At Baltimore, Md., adults of this species emerged
from an infested oak table in a heated building as early as January 12,
1 91 6. General emergence at Falls Church, Va., however, did not begin
until about the middle of April, 1914 and 1915. The period of maximum
activity is from the last of April to the first part of June. The last adults
emerged during the first part of July. Mating occurred commonly dur-
ing May, in 191 5.
OVIPOSITION
Oviposition began a few days after mating and was observed to take
place principally during the middle of May, in 191 5. On May 24, 191 5,
many beetles were observed on radial sections of wood with their ovi-
positors deeply inserted into the open ends of pores or large longitudinal
vessels in the wood, but the first eggs were not found till June i, 191 5.
The beetles seem to prefer to oviposit on those sections of seasoned
sapwood where the open ends of pores are most numerous. These pores
are especially prominent in "ring-porous" woods such as hickory, ash,
and oak, which are also the species most subject to attack by Lyctus
beetles. No eggs were observed on the surface of the wood, but all that
were found were in these pores.
The females remain for several minutes with .the ovipositor in the pore,
and the process is repeated at several places. The female usually as-
sumes a position in which the body is parallel to the pore and the ovi-
positor is either curved down and bent forward into the pore underneath
the body or projected directly into the open end of the pore. However,
the ovipositor, which is long and flexible and reaches from the end of the
body to the thorax when extended forward, can be projected in any
direction. At the extremity of the ovipositor are two laterally placed
palpi. In the process of inserting the ovipositor into the pores, there is a
considerable preliminary period of thorough examination with these palpi
of all parts of the pore before an egg is laid. Two or more eggs are
usually laid near together in each pore utilized. Each female deposits
eggs in several pores.
the; egg
The egg (PI. XXVIII, fig. i) is cylindrical, rounded at the ends, and has
a slender strand or process attached to the cephalic pole. It is whitish
in color, somewhat shiny, i mm. in length with the strand attachment,
0.75 mm. in length without this process, and 0.175 mm. in wdth. This
process or strand is somewhat similar to that of the eggs of certain para-
sitic Hymenoptera — that is, parasites of the cotton boll weevil in the
families Eurytomidae and Encyrtidae (6, p. 49-51, pi. 2), but this is the
only instance known to the writer of such a process on the eggs of Coleop-
tera. The egg has a granular appearance (Pi. XXVIII, fig. 2), and at
the end which terminates in the process there is an area marked with
parallel, longitudinal striae (PI. XXVIII, fig. 4). The egg of L. linearis
Mayi5. i9i6 Lyctus planicolUs 275
{canaliculatus) , as described by Xambeu, is very different from the Q:gg
of L. planicollis, since no mention is made by Xambeu of either the
strand attachment or the area of longitudinal striae, which are unusual
characters in the egg of a beetle.
The end with the process (the cephalic pole) leaves the ovipositor last/
and this strand may possibly be attached by the ovipositor to the pore
contents. The larva does not occupy much more than half the length
of the e^gg (Pi. XXVIII, fig. 3). In hatching, the larva backs out of the
egg. The eggs are easily broken, and it is probably due to this fragility
and the fact that they are inserted far into the pores that the eggs of
Lyctus beetles have apparently not been previously observed with abso-
lute certainty of their identity.
SEASONAL HISTORY
Egg laying takes place principally during the middle of May. Recently
hatched larvse were first observed on June i, 191 5. The period of incu-
bation is probably, at most, 10 days. The winter is passed in the larval
stage. General pupation occurs about the first of April; the pupal cell
(Pi. XXX) is excavated near the surface of the wood, and to this cell
the larvae retreat after cutting a transverse burrow nearly to the surface
for the exit of the adults. General emergence of the adults takes place
during May. Under normal conditions of the natural habitat of this
species (in the Gulf and South Atlantic States) activity probably occurs
earlier in the season.
There is apparently only one generation annually. But the combined
work of the many larvae of successive broods and generations burrowing
through the wood results in the complete destruction of the interior and
the conversion of the wood into fine powder — that is, "powder-posted"
wood (PI. XXIX, XXX, and XXXI).
CONCLUSIONS
Injury by "powder-post" beetles to unfinished seasoned wood products
can be prevented by simply adapting a system of inspection, classifica-
tion, and methods of disposal of stock to facts in the seasonal history of
the insects, as has been recommended for many years by Hopkins
(4, p. 6), Forest Entomologist. Such methods have been adopted by
several large manufacturing companies with marked success.
In the case of finished wood products it may often be practicable to
treat the wood with substances to prevent attack. Creosotes are effective
preventives, but they stain the wood; hence, where they can not be
used, in the light of the discovery of the place and manner of the laying
of the eggs, any substance that will close the pores will prevent oviposi-
tion in wood not previously infested. In wood from which beetles have
' This is according to the law of orientation of Hallez (2).
276 Journal of Agricultural Research voi. vi. No. 7
emerged, however, eggs might be laid within the exit holes. Paraffin
wax, varnish, or linseed oil effectively closes the pores of wood. Wood
that has been seasoned less than 8 to 10 months will not be attacked by
Lyctus beetles. In applying chemical preventives, only sapwood that
has been seasoned for 8 to 10 months and longer should be treated.
Judging from facts in the seasonal history of this species, preventives
should be applied before March i.
LITERATURE CITED
(i) DuGi&s, E.
1883. Metamorphoses du Lyctus planicollis Le Conte. In Ann. Soc. Ent. Belg.,
t. 27, p. 54-58, I pi-
(2) HaIvLEz, Paul.
1886. Loi de Torientation de I'embryon chez les insectes. In Compt. Rend.
Acad. Sci. [Paris], t. 103, p. 606-608.
(3) Hebger, Ernst.
1853. Beitrage zur Naturgeschichte der Insecten. In Sitzber. K. Akad. Wiss.
[Vienna], Math. Naturw, KL, Bd. 11, Heft 5, p. 927-942, 6 pi. Also
reprinted.
(4) Hopkins, A. D.
1910. Insects injurious to forest products. U. S. Dept. Agr. Bur. Ent. Circ.
128, 9 p.
(5) Kraus, E. J., and Hopkins, A. D.
1911 . A revision of the powder-post beetles of the family Lyctidae of the United
States and Europe. Pt. III. U. S. Dept. Agr. Btir. Ent. Tech. Ser. 20,
p. 111-138.
(6) Pierce, W. D., Cushman, R. A., and Hood, C. E.
1912. The insect enemies of the cotton boll weevil. U. S. Dept. Agr. Bur. Ent.
Bui. 100, 99 p., 3 pi., 26 fig.
(7) Xambeu, V.
1898. Mceurs et metamorphoses du Lyctus canaliculatus Fabricius. In Bui.
Soc. Sci. Nat. Quest France, t. 8, p. 69-72.
PLATE XXVIII
Lyctus planicollis:
Fig. I. — Outline of the egg, showing strand attachment.
Fig. 2. — Greatly enlarged view of end of egg, showing granular appearance.
Fig. 3. — Larva within egg, ready to hatch. Drawn by Miss M. Carmody.
Fig. 4. — Sketch of egg in pore of wood on radial section of green-ash {Fraxinus
lanceolata) ladder-rung stock, showing longitudinal striae; pore opened to show egg.
Drawn by C. T. Greene.
Lyctus planicollis
Plate XXVIII
^=^*==
Journal of Agricultural Research
Vol. VI, No. 7
Lyctus planicollis
Plate XXIX
Journal of Agricultural Research
Vol. VI, No. 7
PLATE XXIX
Lyctus planicollis:
Larval burrows in an ash shovel handle. Handle planed to show the work of the
larvae. Photographed by H. B. Kirk.
PLATE XXX
Lycius planicollis:
Pupal cells in "powder-posted" white-ash shovel handle. Photographed by
H. B. Kirk.
Lyctus planicollis
i,\' |,
Journal of Agricultural Research
Plate XXX
Vol. VI, No. 7
Lyctus planicollis
Plate XXXI
Journal of Agricultural Research
Vol. VI, No. 7
PLATE XXXI
Lyctus planicollis:
Exit holes of adults in ash shovel handles. Photographed by H. B. Kirk.
ADDITIONAL COPIES
OF THIS PUBUCATION MAY BE PROCURED FROM
THE SUPERINTENBENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
"WASHINGTON, D. C.
AT
10 CENTS PER COPY
Subscription Price, S3.00 Per Year
/
Vol. VI M^AY Z2y 1916 No. 8
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS
Page
Hypoderma defonnans, an Undescribed Needle Fungus of
Western Yellow Pine - - - - - - - 277
JAMES R. WEIR
Omix geminatella, the Unspotted Tentiform Leaf Miner of
Apple ------- - - - 289
L. HASEMAN
DEPARTMENT OF AGRICULTURE
WASHINGTON, B.C.
WASHINGTON : QOVERNMENT PRINTINO OFFICE : 1«1«
PUBI.ISHED BY AUTHORITY OF THB SECRETARY
OF AGRIGUIvtURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUE-
TURAE COLLEGES AND EXPERIMENT vSTATlONS
EDITORIAL COMMITTEE
FOR TEDB DEPARTMENT
FOR THE ASSOCIATION
KARI. F. KELLERMAN, Chairman RAYMOND PEARL
Physiolosist and Assista-nt Chief, Sureatt
of Plant Industry
EDWIN W. ALLEN
Chief,. Office of F.xpi>^-itnJ>yit Sl.af !■:>'■ :.
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Alaine Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition., The
Pennsylvania State Colleg*'
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
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 Experiment Stations should be
addressed to P^aymond Pearl, Journal of Agricultural Research, Orono, Maine,
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, May 22, 1916 ' No. 8
HYPODERMA DEFORMANS, AN UNDESCRIBED NEEDLE
FUNGUS OF THE WESTERN YELLOW PINE
By James R. Weir,
Forest Pathologist, Office of Investigations in Forest Pathology,
Bureau of Plant Industry
INTRODUCTION
In the summer of 191 3 the writer's attention was drawn to what
appeared to be a very serious needle disease of the western yellow pine
{Pdnus ponderosa Laws.) in parts of Idaho, Washington, and Montana.
That the disease has become more prevalent is shown by the receipt at
the Laboratory of Forest Pathology at Missoula, Mont., of many collec-
tions of the fungus from localities where it was not before known to
exist. These collections represent material from trees of all ages and
show the youngest needles as badly diseased as the oldest ones. The
first suspicion that the fungus might be of some economic importance
arose through the discovery of a serious infection of young reproduction
over a large area in the Whitman National Forest, Oregon. From the
fact that the fungus causes a conspicuous hypertrophy by the extension
of its mycelium into the tissues of the twigs and also through the destruc-
tion of the youngest needles, consequently causing in some localities
much damage in the forest, it seems desirable to make known its char-
TECHNICAL DESCRIPTION OF THE FUNGUS
Since the fungus does not agree with any known member of its genus,
it is described as new.
Hypoderma deformans, n. sp.
Apothecia black, shiny, averaging lo mm. in length and i mm. in breadth; may-
extend as a black line the entire length of the sheath side of the needle or be broken
up into a series of shorter apothecia, usually arranged along the middle line of the
needle, but may appear at either side and be very rarely confluent with the more
medially arranged apothecia; opening with a longitudinal medial split. Asci fusiform
(26) 26.1 to 43. sm by 159.5 ^o 207. 2/i (27.3 to 29.o,u by 171. 5 to 186. 4,u). Spores parallel
or obliquely arranged in the ascus, very generally slightly curved, uniform breadth,
rod -shaped, ends blunt, i- septate when mature, septum very conspicuous, cells
often apparently separated, pale olive, almost hyalin, eight to an ascus (40) 6.2 to
Journal of Agricultural Research, Vol. VI, No. 8
Dept. of Agriculture, Washington, D. C. May 22, 1916
dn G — 79
(277)
278 Journal of Agricultural Research voi. vi, no. 8
9.7/i by 90.67 to i3i.37,u (7.4 to 8.7^1 by 108.9 to ii-j.6,a); paraphyses numerous, fila-
mentous, swollen at the ends or recurved. Spermogonia intermixed averaging 5 mm.
in length; spermatia elongated, straight, sometimes slightly curved, hyalin, contin-
uous, averaging i by 8,u.
Type locality: Sumpter, Oreg., Whitman National Forest.
Habitat: Living needles of Pinus ponderosa.
Type material deposited in the Office of Investigations in Forest
Pathology, Bureau of Plant Industry, Washington, D. C, and in the col-
lections for study in the Laboratory of Forest Pathology in the same
office, at Missoula, Mont.
GENERAL BIOLOGY OF THE FUNGUS
The apothecia of the fungus are the most conspicuous of any of the
group on pines in the West (fig. i ) . From new infections of the previous
year fully mature apothecia with well-developed spores (fig. 2) may be
collected in early spring. From this time on the longitudinal split on the
medial line of the apothecium is plainly visible, and may remain open or
closed, depending on the humidity of the atmosphere.
iflG. I. — A side view of two apothecia oi Hypoderma deformans on needles of Pinus ponderosa, showing
the longitudinal medial split.
The splitting of the epidermis on the needle directly on the medial line of
the apothecium is a characteristic shown by nearly all of the Hysteriaceae
and in a few cases seems to be governed by a particular structure of the
overlying layers of the apothecium. Thus, Von Tubeuf ^ points out that
the pseudoparenchymous covering of the apothecium of Lophodermium
pinastri (Schrad.) instead of being one continuous homogeneous tissue
is made up of two parts which come together on the middle line of the
fruiting body. The edges of the two parts interlock by a series of short
papillae. It is on the line of these papillae, when the pressure within the
apothecium becomes suflicient, that the epidermis of the needle ruptures.
In Hypoderma deformans the rupture of the apothecium is apparently
made easier by the coalescence of filamentous elements springing from the
floor of the apothecium and meeting with the darker tissues of the
apothecial covering above. Owing to a differentiation of the covering
of the apothecium at the point of union a line of rupture is formed.
1 Tubeuf, Carl von. Studien iiber die Schuttekrankheit der Kiefer. In Arb. Biol. Abt. Land- u.
Porstw. K. Gsndhtsamt., Bd. 2, Heft i, p. 22, 1901.
May 22, 1916
Hypoderma Deformans
279
Pressure within the apothecium on approaching maturity, together with
the elongation of the central elements, causes the rupture to occur on this
line. After initiating the line of rupture, the filaments disappear and no
sign of their presence exists when the spores are mature. In all material
so far examined this mechanism is a constant characteristic. Where
two apothecia are formed side by side, the filamentous structures are in
marked contrast to the division line between the two apothecia as
formed by the union of the darker colored elements of the apothecia!
covering. Von Tubeuf found in Hypoderma strohicola Tub. (Lopho-
mm
Fig. 2. — Asci, spores, and paraphyses of Hypoderma deformans.
dermium brachysporum Rostr.) the same structure which he describes for
Lophodermium pinasiri (Schrad.), but no such structures were found in
H. deformans.
Apothecia with mature spores (fig. 3) may be found at any season of
the year. This is due to the fact that the spores do not ripen or are
not all freed simultaneously when the split first appears in the apothecium.
The process of spore liberation is observ^ed to extend over a long period
of time. A year may elapse before the apothecia have entirely liberated
their spores. During periods of drought the medial slit in the apothe-
cial covering remains closed, only opening on the return of abundant
28o
Journal of Agricultural Research
Vol. VI. No. 8
moisture. The hygroscopic movements of the lips of the apothecium
furnish the method by which the spores are forced or ejected from the
asci. As Von Tubeuf ^ has pointed out in the case of Lophodermium
pinastri, the spores are shot out from the mature asci under proper con-
ditions of moisture. This fact is easily demonstrated by inclosing short
pieces of previously moistened needles bearing mature apothecia in the
cavities of plate-glass culture slides. A microscopic study of such
preparations shows that the spores are shot out from the asci a distance
of from I to 2 mm., showing as a plainly visible deposit on the floor and
cover of the cavity. The depth of the cavity in the slides used was 2 mm.
mm
Fig. 3. — Cross section of an apothecimn of Hypodervia deformans on a needle of Pinus pmtderosa, showing
mature asci with spores, the point of first rupture, and the tissues of the leaf most seriously affected by
the mycelium of the f vmgus.
Occasionally an entire ascus was ejected and lay among the spores. In
most cases, the asci remained attached and the spores were expelled
through their terminal pores (fig. 4). Only the fully developed spores
were cast out of the apothecia. After the material had remained in the
slides a day and a half, during which time the spores were being ejected,
the cover glass of a slide was removed and the material allowed to dry
by exposure to the air of the laboratory for 30 days. The material was
washed and replaced in the cavity in the slide. Within three hours
spores from the same apothecia were expelled in considerable numbers
but not so profusely as before. The process was repeated with shorter
' Tubeuf, Carl von. Op. cit., p. 24-25-
May 22, 1916
Hypoderma Deformans
281
periods of drying till on the fourth trial no spores were liberated. An
examination of the apothecia showed the asci to be entirely empty.
This experiment not only demonstrates that the fungus has the ability
to resist protracted periods of dryness but that the period of spore libera-
tion may be much protracted, depending upon the atmospheric humidity.
During wet weather apothecia expel their spores in visible quantity
when a sharp blow is given the branch bearing infected needles.
Considering the long periods of drought in most yellow-pine regions,
it is safe to assume that an apothecium ripening in early spring may
first become emptied of its spores during the ensuing winter or even
later. This is important for the propagation of the fungus, since new
infections are possible from the time the first needles of the season appear
till the close of the growing season.
In order to determine the viability of the spores expelled from apothecia
after long dryness, a 2 per cent sugar solution was introduced into the
cavity of one of the slides containing apothecia which
had lain dry in the laboratory for two months and
the slide placed in the thermostat at 35° C. On the
fourth day spores germinated readily. The germ
tubes appeared more frequently from the ends of the
spores. A slight addition of an extract of pine needles
to the sugar solution promoted germination.
It was noticed that in collections of the fungus
made shortly after warm summer rains the asci are
frequently empty as compared with asci of mature
apothecia collected in the colder spring months.
This, it seems, may not be entirely due to a longer
period of spore liberation but also to the higher tem-
perature of the summer months. Von Tubeuf found
that increasing temperature promoted spore liberation
in Lophodermium pinastri and it is found to be true in experiments with
the yellow-pine fungus. During the winter, moistened apothecia from dry
material were mounted in two culture slides; one was placed outside the
laboratory during a period when the thermometer registered about 40°
F. and the other w^as kept in the laboratory air of about 80° F. At
the end of four hours a microscopical examination showed that a large
number of spores had been ejected from the apothecia in the slide kept
in the laboratory but none from the other slide. When the slide from
the outside was allowed to stand for a while in the warm air of the
laboratory, spores were liberated in quantity.
Although spores from various needle fungi are undoubtedly more
readily liberated during warm rains of the summer months, the frequent
drying of the foliage of the trees is probably not favorable for infection.
It is frequently observed, and as often reported, that needle fungi become
more active during the cool, protracted rainy periods of early spring and
late fall. No extensive data are at hand regarding the resistance of
Fig. 4. — The upper por-
tion of a young ascus of
Hypoderma deformans,
showing the formation
of the pore at the tip
through which the
spores are expelled.
282 Journal of Agricultural Research voi. vi, no.s
expelled spores to drought and direct light; still, the fact that dry her-
barium material a year old was found to furnish viable spores shows that
spores may exhibit considerable resistance to dry air when free from the
apothecium.
PARASITISM OF HYPODERMA DEFORMANS
An attempt to grow the fungus on culture media failed. The spores
in every case germinated and in some cases produced an abundant white
mycelium, but in the course of six months, after frequent transfers, the
mycelium turned a light yellow and died. A somewhat better result was
obtained by adding to the culture medium a strong extract made from
yellow-pine needles in water, but at the end of eight months the mycelium
died.
A quantity of needles bearing apotbecia with mature spores were col-
lected in the spring of 191 4 near Missoula, Mont., and taken to the field
station in the Priest River Valley, Idaho, for experiments on parasitism.
The fungus has not been found in this region. The needles were thor-
oughly washed in distilled water and the apothecia allowed to expel
thedr spores in small sterilized flasks. Needles and spores were shaken
up in water to which a i per cent sugar solution was added. The mix-
ture was allowed to stand one day and then thoroughly sprayed over
four 3-year-old yellow-pine seedlings having young tender shoots with
needles. The inoculated seedlings were immediately inclosed in tough,
transparent oiled paper bags and protected from injury. A second ex-
periment was initiated by binding infected needles on healthy 3-year-old
seedlings. In the part of the Priest River Valle)^ where these experi-
ments were performed the yellow pine is not common, being only spar-
ingly represented in a mixture of white pine, grand fir, spruce, hemlock,
and Douglas fir. The experiments were made on May 20. In Septem-
ber the last-formed needles of the inoculated seedlings were turning
reddish brown in spots, mostly at the tips. In the following spring. May
to June, the needles which showed infection in the fall and which had
become wholly brown developed the characteristic long, shiny black
apothecia with mature spores (PI. XXXII, fig. i). Only the needles
formed during the previous year were infected. Four control plants,
also covered with bags, were entirely free from the disease. The needles
of the seedlings on which infected needles were bound showed a much
more general infection of the last-formed needles than those by the former
described method. In these experiments every needle produced in 1914
was infected. Those of previous years remained healthy. This indi-
cates that old needles are not attacked and that the young needles may
remain attacked indefinitely after infection. All the infected needles did
not produce mature apothecia. Those merely turning brown were filled
with the mycelium of the fungus. The experiment at this point was
discontinued. In all probability, given time enough, the brown-infected
needles would have produced apothecia.
May22. i9i6 Hypoderma Deformaus 283
It has been noticed repeatedly in nature that there is great irregularity
in the time between the first browning of the needles at their tips or at
other points along the needles and the appearance of the mature apothecia.
In a few cases the cycle of development from the first appearance of the
brown color at the tips of the needles to mature apothecia has been
observed to take place within the same calendar year, or from April and
May to November. More often infected needles first showed mature
spores in the spring of the following year. It was observed in a few cases
that the needles may lie on the ground through the following winter before
the apothecia rupture. Brown needles collected in August from infected
trees and placed in damp moss in the field in a number of cases developed
apothecia before January, maturing in May and June. The apothecia, as
previously indicated, may contain asci in various stages of development,
so that mature spores are being produced throughout the year. Investi-
gation has shown, however, the greatest number of spores are expelled
during the spring rainy season, May and early June, coinciding with the
greatest vegetative period of the host. In no instance, either in the field
or in artificial inoculation, were the infected needles of young trees or seed-
lings not previously attacked by the fungus killed before they had attained
their normal size. In September or October, such needles will have
assumed a more or less uniform reddish brown color. Mostly remaining
upon the tree, they may first produce the signs of the apothecia during
the late fall and mature the spores in the following spring. At the
time the foregoing experiments were in progress small bundles of infected
needles bearing fertile apothecia were bound with similar quantities of
needles which had died from a normal cause. These were placed in moss
during May, 191 4. On examination in May of the following year the
needles which had died from a normal cause showed no signs of the fungus;
nor have they done so since that date. This apparently demonstrates
the inability of the fungus to act as a saprophyte.
The foregoing observations and experiments apparently prove the
parasitism of the fungus. This is further substantiated by the observed
evidences that young seedlings in the field succumb to the ravages of the
fungus. Furthermore, it is indicated that the period of greatest infection
is during the growing season and only the needles of the season are to any
extent susceptible to attack.
The fungus has not yet appeared in the forest nursery, but it may be
regarded as a possible nursery disease.
PATHOLOGICAL EFFECTS OF THE FUNGUS ON THE BRANCHES
OF THE HOST
A very peculiar and at the same time interesting phenomenon
caused by the growth of the mycelium of the fungus in the shoot is the
formation of spherical-shaped witches'-brooms on trees mostly past
the seedling stage. These (PI. XXXII, fig. 2) brooms in old trees often
assume large proportions. A single witches' broom may weigh as high
284 Journal of Agricultural Research voi. vi, no.s
as 100 pounds and measure 5 or 6 feet in diameter. The branch sup-
porting it will hang vertically, the broom swaying in the wind like a
great bag (PI. XXXII, fig. 3). The average size of the brooms is about
2 feet in diameter. Although a few isolated cases had been noted on the
seeming association of this needle fungus with these compact brooms,
it was not until the field season of 191 3 that this association was found
to be of common occurrence. This was all the more interesting from
the fact that the cause of these formations has been a standing question
with all who have seen them. In some cases they have been attributed
to the yellow-pine mistletoe, Razoumofskya campylopoda (Kngelm.)
Piper, an error, however, not likely to be made by anyone familiar with
the type of broom caused by this mistletoe.
The distribution of the brooms is quite general through the range of
the yellow pine in the Northwest. They are particularly abundant in
the vicinity of the great lakes of Idaho and in the dry valleys of southern
and western Montana. Climatic variation does not seem to influence
their distribution.
In order to detei-mine the cause and nature of the formation of these
brooms and the relation, if any, between them and the fungus common
on their needles, the subject has been under investigation in the field
and laboratory. A number of interesting observations have been
recorded.
The disease caused by H. deformans primarily affects the needles.
In young pines the disease occurs quite generally at first, unaccom-
panied by any kind of hypertrophy of the shoots. Later the repeated
destruction of the last-formed and older needles initiates a swell-
ing of that portion of the branch. Sometimes the entire shoot suc-
cumbs to the attack in seedlings of tender years, especially the weaker
individuals, caused, no doubt, by the rapid drying out of the shoot. In
growths of 7 to 10 years the fungus confined itself to the needles of the
season, with the result that on the infection of these a second crop some-
times appears afDOut the terminal bud, which may or may not become
infected but may remain in a stunted, deformed condition. They help,
however, to maintain the shoot in a living condition. In a far greater
measure than in any other member of the order the mycelium of H.
deformans penetrates the leaf sheath and eventually perennates in the
tissues of the shoot, causing a marked enlargement of the parts infected.
The fungus, however, fruits only on the needles.
An additional result of the infection of the terminal shoots and the
continued production of food materials by the older, uninfected needles
is the stimulation of all lateral and adventitious buds either between
the primary terminal buds or at the last two or three nodes. Eventually,
the food materials are more and more diverted from the main shoot,
resulting in a gnarled and curved bunch of short branches. Young trees
May22, i9i6 Hvpoderma Deformafis 285
4 to 8 years old when uniformly infected are frequently observed
with the terminal portions of every principal branch in the process of
"brooming." The fact that the fungus sometimes occurs without the
least sign of a hypertrophy of the branch does not indicate that it is not
capable of producing such physiological and morphological changes.
The fact remains that on all young growth almost always the twigs
bearing the infected needles are abnormally swollen or branched. The
fungus has not been found by the writer on mistletoe brooms or on any
form of broom caused by insect or other animal injury. On large and
mature trees H. deformans very rarely occurs on any part of the tree
except the needles of these brooms. These abnormalities are scattered
promiscuously over the tree, but principally on the lov.er branches.
This indicates the nature of an infection. The more recent infections
on old trees are usually distributed or isolated on particular branches.
Serious injury seldom results from the growth of the brooms on more
mature growth. Very rarely may the brooms become so heavy as to
split-off the supporting branch.
As the result of an examination of the witches' brooms on yellow pine
in the Bitter Root and Missoula River valleys, Montana, and the Coeur
d'Alene region of Idaho, with respect to the presence of H. defor-
mans on the brooms and the number, position, and distribution of the
brooms on the tree, the following data were obtained :
On 107 trees examined, the average number of witches' brooms per
tree was 3.2. These brooms generally appeared on the lower part of the
crown on the side facing the prevailing winds. The average number
of brooms per tree bearing needles showing apothecia of H. defor-
m.ans was also 3.2.
These figures support the view that the peculiar brooms so common on
yellow pine are the result of fungus infection and that the fungus respon-
sible is H. deformans.
In the parts of northern Washington, Montana, and Idaho so far
visited, H. deformans has not been found to attack the yellow-pine
reproduction in as great a degree as in regions farther to the west
and south. This is probably due to a greater mixture of species. The
fungus is not able to spread with the same rapidity as in the more typ-
ical yellow-pine stands. The infected young growth usually continues
alive indefinitely, and deformed branches appear, eventually resulting
in an entire retardation of growth, and finally die. This process may
require several seasons, but the infected pines never attain a very large
size. Such deformed trees usually are attacked by bark beetles, such
attacks hastening their decline.
In parts of Oregon in the yellow-pine belt the fungus was found to be
very destructive. During an investigation of the larch mistletoe
(Razoumofskya laricis Piper) in the vicinity of Sumpter, Oreg., the
286
Journal of Agricultural Research
Vol. VI, No. 8
yellow-pine reproduction, especially on south slopes under mature cover,
was observed to be turning brown and in many cases dying. On exam-
ination the needles of these seedlings showed that they were infected
with H. deformans. This is a grazing region, and the forest has been
continuously grazed by large bands of sheep for many years. The
stems of the young pines in numerous cases bore near their bases one or
more wounds of a shape and nature indicating that they were produced
by the treading of grazing animals. Since little information is at hand
on the effects on forest production of wounding by grazing animals, it
seemed worth while to make a detailed study of the case so far as time
would allow, with the double object of determining which injury — viz,
the needle fungus or the wounding — was responsible for the sickly con-
dition of the young pines. It must be remembered, however, that the
seedlings were growing under the canopy of a mature yellow-pine stand;
consequently they were not growing rapidly in height. Four one-tenth-
acre plots were laid off on representative south slope sites and every
seedling on the plots carefully pulled up and bound in bundles. These
bundles were sent to the laboratory and afterwards carefully diagnosed.
The normal condition of root system and crown and general vigor of
seedlings were judged from a knowledge of normal young pines of the
same age, free from disease and wounds, growing in the same regions and
under the same slope conditions. The results of this study were em-
bodied in a preliminary table from which Table I was condensed as being
more readily undestandable.
Table I. — Number of seedlings on 4 one-tenth-acre plots, average age and height, con-
dition of infection with Hypoderma deformans, present condition of wounding and root
system, south-slope type
Condition of seedlings.
Number
on^tenth- ^S^"
acre plots.
Average
height.
vigor and general appearance of seedlings.
Seedlings neither
v,'oiinded nor infected.
Seedlings infected with
Hypoderma deform-
ans but not wounded.
Seedlings both woitnded
and infected.
Seedlings wounded but
not infected.
40
67
127
49
Years.
10.7
13-0
12. 1
12.8
Feet.
2.8
1-3
1.4
2.14
Healthy green color, vigorous, normal, well de-
veloped root system.
Few green needles, most all badly infected and
either dead or dying; twigs twisted or broomed;
poorly developed root system; general picture
of a starving condition.
Do.
Wounds mostly healed. Time required to heal,
I to 4 years. Seedlings normal. Wounding
apparently not affecting growth. Root system
normal.
DISTRIBUTION OF HYPODERMA DEFORMANS
The disease of yellow-pine needles caused by H. deformans is widely
distributed throughout the northwestern part of the United States
and western Canada. Its distribution in other parts of the West
is not known, although the fungus has undoubtedly been collected
May22. i9i6 Hypodevma Deformafis 287
by other observers.^ The writer has observed and collected H. de-
formans in the National Forests of the Northwest as follows: Sioux,
Helena, Deerlodge, Jefferson, Missoula, Coeur d'Alene, St. Joe, Clear-
water, Selway, Bitterroot, Pend Oreille, Kaniksu, Nez Perce, Lolo,
Cabinet, Flathead, Kootenai, and Whitman. The late J. F. Pernot,
forest examiner, supplied specimens from the Deschutes, Wallowa,
Malheur, Crater, Colville, and Wenatchee National Forests. Along
the Thompson River in British Columbia the fungus was occasionally
found by the writer in the summer of 191 3.
CONCLUSIONS AND RECOMMENDATIONS
A very conspicuous disease on yellow-pine needles in many parts
of the Northwest, the cause of which has for several seasons remained
unknown, is found to be caused by a fungus which is described as a new
species under the name " Hypoderma deformans."
H. deformans is a true parasite and attacks the foliage of all age
classes; and in some of the more exposed sites of the typical yellow-
pine belt of Montana, Oregon, Washington, and Idaho, young seed,
lings at first suffer great suppression and are finally killed.
The first sign of infection of the needles is usually a slight browning
of the tips; or in the regions of heavy infection the entire needle may
gradually assume a straw-yellow color, deepening to a brown on the
first appearance of the apothecia.
Because of the destruction of the youngest needles and the penetra-
tion of the mycehum of the fungus in the tissues of the stems of the
host, the terminal shoots do not attain their proper development, but
become stunted and deformed, eventually producing a witches' broom.
These witches' brooms on young yellow-pine saplings or older trees are
often very conspicuous and often occur in such numbers as to make
either an individual tree or an entire stand look very ragged and un-
sightly.
Up to the present time the disease has not been found in the forest
nursery, but it may be regarded as a possible nursery disease. Since the
vegetative mycelium of the fungus may hibernate in the shoots of seed-
lings after the infected needles have fallen, the fungus may make its
appearance in the forest nursery and may be unknowingly transferred
to the planting areas.
The presence of the fungus on mature forest trees is very readily recog-
nized by the foHage browning up in patches or by the formation of
brooms. Since the fungus does not affect the merchantabiUty of the
tree, except by influencing the increment in cases of very severe infec-
tion, all trees of the regulation diameter classes should be marked for
1 Meinecke describes a very destructive needle fungus, under the name "Hypoderma," on yellow and
Jeffrey pines, which apparently is the same fungus as the one described in these pages. (Meinecke, E. P.M.
Forest Tree Diseases Common in California and Xevada, p. 34. Washington, 1914. Pub. by U. S. Dept4
Agr. Forest Serv.)
288 Journal of Agricultural Research voi. vi. xo. s
cutting. The brooms never produce cones and the normal parts of the
supporting branch are usually sterile. The branches bearing patches of
infected needles or brooms should be piled and burned' as soon as possible.
This may be done in the course of the regular brush-piling operations.
If young trees below the regulation cutting diameter are so badly
"broomed" that in the opinion of the forest officer the increment of
the tree will be seriously impaired, and whenever the cost is not prohibi-
tive, such trees should be lopped and immediately burned. The chief
reason for such procedure is to protect the reproduction from infection,
thus insuring a healthier forest in the future.
PLATE XXXII
Fig. I. — Needles of Pinus ponderosa infected with Hypoderma deformans, showing
the apothecia. Natural size.
Fig. 2. — Branches of Pinus ponderosa deformed and broomed by Hypoderma defor-
mans.
Fig. 3. — A branch of Pinus ponderosa, showing how it will hang vertically when
supporting a large broom caused by Hypoderma deformxins.
Hypoderma deformans
Plate XXXI
Journal of Agricultural Research
Vol. VI, No. 8
ORNIX GEMINATKLLA, THE UNSPOTTED TENTIFORM
LEAF MINER OF APPLE
By L. Haseman,^
Entomologist, Missouri Agricultural Experiment Station
INTRODUCTION
The small, unspotted tentiform leaf miner (Ornix geminatella Pack.) has
been extremely abundant in Missouri in recent years and has attracted
the attention of fruit growers throughout the State. It has confined
itself largely to bearing apple {Malus sylvesiris) orchards, though con-
siderable injury has been done to apple foliage in nurseries. Fortu-
nately, it is most abundant in the late summer and early fall, so that its
work is of less importance to the trees. As with many insect pests, it
seems to run in cycles. It was most abundant during the summers of
191 1 and 1 91 2, reaching a climax in 191 2. Since 191 2 it has attracted
little attention.
It confines its work to the leaves and spends most of its larval life
inside the leaf as a true miner. The caterpillar therefore is small, though
the characteristic elevated, or tentiform, dead patches which it produces
on the leaves are quite noticeable. In some cases as many as 15 mines
have been found on a single large apple leaf (PI. XXXIII, fig. 14, 15).
The pest was so abundant and so widely distributed throughout the
State that a careful study of its life history, habits, and control v/as under-
taken.
HISTORY OF THE PEST
The moth was first described and figured by Packard (7, p. 353)^ in
1 869 as Liihocolletes geminatella. The description and figures are incom-
plete and not entirely accurate, owdng perhaps to incomplete observa-
tions. Since its first discovery it has been collected by various workers
and was redescribed by Chambers (2) as L. prunivorella. Other closely
related micros have been mistaken for it, and some careful observers have
given very inaccurate descriptions of its v/ork and habits.
DISTRIBUTION OF THE LEAF MINER
Packard reported it as being abundant in New England on pear and
apple; Lowe (6) reported it as being very abundant on apple in New
' The writer wishes to acknowledge his indebtedness to the late Miss Mary E. Murtfeldt, of Kirkwood,
Mo., to Miss Annette F. Braun, of Cincinnati, Ohio, and to Mr. August Busch, of the Smithsonian Insti-
tution, Washington, D. C.,ior assisting with the naming of the leaf miner; and to Dr. L. O. Howard. Chief
of the Bureau of Entomology, and to Mr. A. A. Girault, of the same Bureau, for the determination of the
parasites. He is also especially indebted to Prof. C. R. Crosby, of Cornell University, for helpful sugges-
tions and for assistance in naming the leaf miner and the parasites.
' Reference is made by number to "Literature cited. " p. 295.
Journal of Agricultural Research, Vol VI, No. 8
Dept. of Agriculture, Washington, D. C. May 22, 1916
dt Mo.— I
(289)
290 J our7ial of Agricultural Research voi. vi, no. s
York, and Brunn (i) reported it from Ithaca, N. Y. Forbes (4, p. 57)
reported it from Illinois, New York, Colorado, Kentucky, Michigan, and
Massachusetts; and Jar\ds (5, p. 49) reported it as being common in
Connecticut. Dietz (3) reported it from the Middle and Northern States
of the Atlantic slope, though he confused species. In a recent attempt
to determine its present distribution the writer has been able to get defi-
nite records from but one additional State, Ohio. It is probable that it
is found from the Atlantic States to Colorado, but being so small and
inconspicuous, except when abundant, fruit growers and entomologists
have overlooked the insect and its work.
LIFE HISTORY OF THE MINER
The writer has not been able to find any report of the complete life
history of the pest. Such records as are available deal with the insect
and its development and work in the summer or more often for a short
period in the late fall. In some cases very careful data have been re-
corded, but many of the records and descriptions are decidedly at fault.
The following records for the insect in Missouri have been collected since
the summer of 191 1 and include new data on the life history, develop-
ment, and habits of the pest.
EGG
The ^gg is extremely small, slightly oblong, varying from 0.254 to 0.4
mm. in length and from 0.18 to 0.29 mm. in breadth, only slightly ele-
vated and firmly cemented invariably to the lower surface of the leaf.
(PI. XXXIII, fig. 3.) It is so small that it can scarcely be detected with
a hand lens, and the writer has failed to find the unhatched eggs on foli-
age, though many have been collected and studied soon after hatching,
when the young caterpillar had just begun to start its mine. The adults
have refused to lay eggs in captivity in small vials; therefore, these rec-
ords are for the freshly hatched eggs.
THE LARVA
On hatching, the larva is footless and resembles a microscopic flat-
headed borer. It always seems to break through the part of the shell
which is cemented to the leaf and enters the tissue of the leaf at once.
The freshly hatched caterpillar is less than a millimeter in length. It
grows rapidly and when mature is about 6 mm. in length. In its de-
velopment it passes through four distinct larval stages. There is con-
siderable variation in size, but the following measurements are the
average of many specimens.
In the first stage the caterpillar is pale, with a slight yellowish tinge
to the head. The head and thorax are enlarged and it is footless. It
molts when it is yet less than 2 mm. in length, and the head capsule is
about 0.18 mm. in breadth. (PI. XXXIII, fig. 4.)
May22, i9i6 Omix geminatella 291
In the second stage the body is pale, the head becomes brownish, a
black blotch begins to appear on the first thoracic segment, legs are still
absent, the head capsule is about 0.27 mm. broad, and the caterpillar
is about 2.2 mm. long. (Pi. XXXIII, fig. 5, 6.)
In the third stage the body is at first pale, but darkens with age;
the thoracic and abdominal legs appear; the thoracic blotch breaks up
into four irregular spots; the head becomes darker and is about 0.35 mm.
in breadth, while the caterpillar is about 4.5 mm. in length. (PI. XXXIII,
fig. 7-)
In the fourth stage the caterpillar is about 6 mm. long and the head
capsule is 0.49 mm. broad; the body takes on an olive-gray color, sharply
contrasting with the conspicuous white tubercles; the head becomes
darker, and along its hind margin appears a row of four small black
spots which parallel the similar row of larger spots on the first thoracic
segment. (PI. XXXIII, fig. 8, 9.)
THE MINE
While the caterpillar is changing from a pale, flat, footless, micro-
scopic caterpillar to a conspicuously marked, cylindrical, active one,
its mine also undergoes distinct changes. At first the mine is serpentine
in form; but after it is from 4 to 8 mm. in length and is usually curved
upon itself, the caterpillar begins to transform it into a blotch mine.
(PI. XXXIII, fig. 13.) The blotch mine begins by the third day, and
about that time the caterpillar changes to the second stage. At fifst
the blotch appears only on the lower side of the leaf. The lower layer
of the leaf is separated from the upper by the flat caterpillar, and soon
the severed lower layer dies and turns brown. The mine remains in
the blotch stage about four or five days, and during that time the cater-
pillar changes to the third stage. When complete, the blotch is from
I to 2 cm. long and usually occupies all the space between two of the
main lateral veins of a leaf. On preparing to produce the tentiform
mine, the caterpillar spins silk threads on the floor of the mine, which
causes the lower dead layer of the leaf to become folded lengthwise of
the mine. These threads, with others spun later under the roof of the
mine, cause the upward projection of the mine. Just about this time
the caterpillar changes to the fourth stage and begins to feed on the
chlorophyll cells, and this in time gives the unspotted effect when a
clear net work of veins appears. During the latter part of June it was
found that in from a week to 10 days after the young caterpillar begins
to feed, the mine is changed from the serpentine through the blotch to
the tentiform type. The majority of the feeding and growth occurs in
the third and fourth stages, and after the tentiform mine is made it
requires from four to seven days to eat out all the chlorophyll cells and
give it the completed, unspotted, tentiform appearance. The larval
life in the mine is therefore about two weeks. The caterpillar leaves
292 Journal of Agricultural Research voi. vi, no.s
the mine through a small hole in the floor of the mine and after crawling
about for a varying length of time prepares to make a cocoon in which
to pupate.
COCOON
The cocoon is almost invariably made on the upper surface along the
edge of the leaf or at its very tip. On preparing to make the cocoon the
caterpillar first rasps off and eats a patch of the surface layer of cells
along the edge of the leaf, about 4 mm. wide and twice that in length.
This causes a withering of the tissue and a slight folding over of the edge
of the leaf. Then begins the work of spinning silk. First a loose layer
of silk threads is spun from a line about 2 mm. from the edge of the leaf
to the inner edge of the patch rasped off. Then follows a second layer
of threads which are drawn very tight as they are placed. The anterior
two-thirds of the body of the caterpillar enters into this work with great
energy and force. The caterpillar's silk press must be a strong one.
This layer only slightly draws up the edge of the leaf, but after trans-
versed bands are used to tie the tight threads in bundles the edge of
the leaf is perceptibly folded. At this time a second layer of foundation
threads are spun underneath and then the work of drawing tight
threads is continued along one end of the future cocoon. In half an
hour the leaf edge is half drawn over and the hardest part of the work
is completed. After the edge is tied down tightly the inclosed space is
thoroughly lined with snow-white silk, so that a very dense semicircular
cocoon 8 mm. long is formed. (PI. XXXIII, fig. 12.)
PUPA
The mature caterpillar pupates soon after the cocoon is completed.
The pupa is about 4 mm. long, exclusive of the antennal sheaths which
project fully a millimeter beyond the tip of the body (PL XXXIII, fig,
10, 11). The pupa darkens with age, becoming dark brown on the
dorsum and yellowish brown on the venter. The leg, wing, and antennal
sheaths are all distinct. The pupal period varies from a few days to a
week in midsummer.
MOTH
The newly-emerged adult on assuming its full splendor is truly a
beautiful creature when viewed through a microscope. When left undis-
turbed it will stand perfectly still for hours, with the head elevated and
the tip of the wings and abdomen lightly touching the surface on which
it rests (PI. XXXIII, fig. 2). This is its characteristic pose, and it holds
it so perfectly that prolonged exposures for enlargements can safely be
made. While in this pose the light flashes from every properly arranged
scale as from polished metal, and one who is only familiar with the appear-
ance of museum specimens can hardly appreciate the peacock-like
splendor of this seemingl}' proud little creature.
May 22, 1916 Ornix geminatella 293
Brunn's (i) description of the adult is very good. To the unaided
eye the moth is slate-gray with a slight tinge of brown, being lighter in
rufifled specimens. The ventral surface of the body is lighter in color.
The markings on the front two pairs of legs are similar. The tarsal
segments are white, tipped with black; the tibia and femur vary from
dark brown to black with lighter patches; the coxae are mottled with
white and dark scales. The tarsal segments of the hind legs are brownish
with white basal bands, while the tibia, femur, and coxa are much lighter,
being nearly the same color as the lower surface of the abdomen. The
palpi are prominent and banded with white and dark scales. The
brownish proboscis is unusually long, reaching to beyond the base of the
abdomen which, though it has not been observed to do so, would lead
one to conclude that the moth feeds. The antennae are brownish in
color and distinctly annulate with whitish. In life they are closely
pressed along the sides of the body and reach to be5^ond the tip of the
abdomen and wings.
The surface of the forewings is beautifully mottled with light and
dark scales. The light scales are arranged in eight or nine more or less
distinct transverse bands. In museum specimens it is difficult to dis-
tinguish these bands. Near the tip of the forewings in fresh specimens,
is a distinct black patch of scales bordered without by three alternating,
narrow, white and black curving bands, giving to the tip of the wings a
distinct peacock spot. On the hinder margin of the front wings the
black and white scales forming the terminal peacock spot give way to
long, light-colored hair. This border of delicate hair ceases near the mid-
dle of the hinder margin of the wing. The hind wings are slender and
armed on the hinder margin with a broad band of delicate light-colored
hair. This band becomes narrower toward the tip of the wing. The
costal band is scarcely as broad as the wing (PI. XXXIII, fig. i).
The moth has a wing expanse of from 7 to 9 mm. and is approximately
5 mm. long when at rest with the wungs folded.
NUMBER OF BROODS
This species winters in the pupa stage in a carefully prepared cocoon
protected by the folded-over edge of a leaf. In the spring the adults are
abundant by the first week in May. By the middle of May the typical
tentiform mines begin to appear, and the adults of the first spring brood
begin to emerge by the last of May. The life cycle is completed in from
four to five weeks. The broods overlap, but beginning with May a
fairly well-defined brood can be made out for each month until November.
The larvae of the October brood pupate and live through the \vinter on
fallen leaves. After the moths emerge a considerable period of time elapses
before the mines begin to appear. This is undoubtedly due to the fact
that the moth, with its well-developed proboscis, feeds for a time before
ovipositing.
36289°— 16 2
294 Jotnmal of Agricultural Research voi. vi, No. s
FOOD PLANTS OF THE LEAF MINER
This leaf miner is primarily a pest of the foliage of the apple. There
is where it abounds. However, the small caterpillars have been found
developing in considerable numbers in the leaves of the crab-apple (Malus
sp.), and to a less extent in the leaves of the haw (Crataegus spp.), plum
{Primus spp.), cherry (Prtmus spp.), and pear {Pyrus spp.). In the case
of the last four trees only an occasional mine has been observed. Cham-
bers (2) and others have also reared it from mines in the leaves of the
wild cherry {Prunus spp.).
CONTROL OF THE LEAF MINER
While this miner may develop in such numbers that from 90 to 95 per
cent of all leaves on apple trees may contain from i to 10 or 15 mines, it
must be said that it is not an especially alarming pest of the orchard
(PI. XXXIII, fig. 14, 15). The pest increases in abundance as the
summer and fall advance, so that by September or October much of the
foliage may be consumed, but by that time the tree has about completed
its growth and matured its crop. However, when conditions are favor-
able and the pest is abundant, steps should be taken to prevent it from
reappearing in injurious numbers the next season.
Since the caterpillar enters the leaf immediately on hatching and
remains in the mine until mature and ready to spin its cocoon for pupat-
ing, arsenical and contact sprays are of no value. Applications of
sprays have given the writer absolutely no results. From the general
nature of the pest and its habits, there seems to be no feasible means of
controlHng it during the growing season. Since it passes the winter
as the pupa in cocoons on fallen leaves, it can be effectively controlled
by destroying the leaves early in the spring. The most practical method
of destroying the pupae on the leaves is to use a disk for shallow cultiva-
tion before the first of March so as to work under the leaves before the
moths begin to emerge. Summer cultivation will not help, since the
pest is not found on the ground at that time. In a small home orchard
the leaves can be raked together and burned or piled and used for leaf
mold. If they are not burned, they should be covered with enough
soil or stable manure to hasten the decay of the leaves and prevent the
moths from emerging in the spring.
PARASITES OF THE TENTIFORM LEAF MINER
It w^ould seem that a caterpillar of this type, which lives protected
inside the leaf from the time it hatches from the egg until it is ready to
pupate, would be as well protected from natural enemies as from arti-
ficial treatment given by man. This does not prove to be the case,
however, for the pest is heavily parasitized. It resembles other insect
pests which are subject to the attacks of parasites in that under favorable
May 22, 1916 Ornix geminatella 295
conditions it increases rapidly and then when the parasites get the
upper hand it suddenly disappears. In the summer of 191 2 it reached
a climax as regards abundance. During the fall the parasites increased
in such numbers that but few of the caterpillars escaped to pupate and
pass the winter. The check, owing to the beneficial work of the para-
sites, was complete, for the miner has not attracted attention since 191 2.
As the investigation of the miner progressed, it was observ^ed that
many of the mines went no farther than the blotch stage, while others
arrived at the tentiform stage; but from them no caterpillars emerged.
In such mines would be found the dried skin of the caterpillar and the
lar^^a or pupa of a parasite. Only casual obser\^ations were made on
the habits and life cycles of the different species of parasites. One of
the common species was found to attack the more mature caterpillars
and pupate in a small, oval, white cocoon suspended in the tentiform
mine. Others destroyed the younger miners and pupated without
producing cocoons in the blotch mines. The grub of one of the parasites
was observed to attack the miner just behind the third pair of thoracic
legs, paralyzing and eventually destroying it.
The collection of parasites was first submitted to Prof. Crosby, who,
from a portion of the collection, identified two species: Synipiesis
nigrifemora Ash. and 5. tischerae Ash. Later Mr. Girault examined
the collection and identified two new species, 5. nieieGri Girault and
Eulophus lineaticoxa Girault, and one previously recognized species, 5.
dolichogaster Ash. Besides these five species, there were a number of
males which were not determined. Brunn (i) reared two species of
Sympiesis from the mines of this insect. They v/ere recorded under
the manuscript names of 5. mmuhis Howard and 5. liihocolletidis Howard ;
but the descriptions by Howard were apparently never published, and
Ashmead later redescribed the latter species as S. nigrifemora Ash.
LITERATURE CITED
(i) Erunn, a. E.
1883. Tineidae infesting apple trees at Ithaca. Cornell Univ. Agr. Exp. Sta.
2d Ann. Rpt. 1882/S3, p. 148-162, pi. 5-6.
(2) Chambers, V. T.
1873. Micro-Lepidoptera. In Canad. Ent., v. 5, no. 3, p. 44-50.
(3) DiETZ, W. G.
1907. The North American species of the genus Omix Tr. In Trans. Amer.
Ent. Soc, V. 2,^, no. 2/3, p. 287-297, pi. 4.
(4) Forbes, S. A.
1889. The apple Ornix. In 15th Ann. Rpt. State Ent. 111. [1884/86], p. 51-57.
(5) Jarvis, C. D.
1906. The apple leaf-miner. A new pest of the apple. Conn. Storrs Agr.
Exp. Sta. Bui. 45, p. 37-55.
(6) Lowe, V. H.
1900. Two apple leaf miners. In N. Y. Agr. Exp. Sta. Bui. 180, p. 131-135,
pi. 6-7.
(7) Packard, A. S.
1869. Guide to the study of insects. 702 p., illus., 11 pi. Saiem.
PLATE XXXIII
Ornix geniinatella Pack.:
Fig. I. — Moth expanded. X lo.
Fig. 2. — Moth at rest on leaf. X 2>^.
Fig. 3. — Egg on lower surface of leaf; also tunnel made by miner on leaving the
egg. X 80.
Fig. 4. — Dorsal view of first larval stage; below, side view of head and thorax.
X 18.
Fig. 5. — Dorsal view of second larval stage. X 18.
Fig. 6. — Side view of second larval stage. X 18.
Fig. 7. — Dorsal view of third larval stage, showing edge of thoracic legs. X 18.
Fig. 8. — Dorsal view of fourth larval stage. X 18.
Fig. 9. — Side view of fourth larval stage. X 18.
Fig. 10. — Ventral view of pupa. X 18.
Fig. II. — Dorsal view of same. X 18.
Fig. 12. — Lower surface of leaf with numerous partly developed mines; also two
cocoons, one exposed. The cocoon is usually on the upper surface of the leaf. X i.
Fig. 13. — Portion of leaf showing a mine in process of development. The serpen-
tine mine was completed on June 24, the small darkly shaded area of the blotch mine
June 25, the second area on June 27, the third area on June 29, and on June 30 the
blotch was completed and then transformed to the tentiform mine. X 2. Egg more
enlarged.
Fig. 14. — A small twig showing leaves badly curled and injured by numerous mines.
XK.
Fig. 15. — Leaf much distorted with 10 mines almost completed; also one cocoon
appears at the tip of tlie leaf. Nattual size.
(296)
Ornix geminatella
Plate XXXllI
4
^dfc^^
^<
"I
Journal of Agricultural Research
Vol. VI, No.
ADDITIONAL COPIES
OF THIS PUBUCATION MAY BE PROCtJKED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
10 CENTS PER COPY
Subscription Price, $3.00 Per Year
V
Vol. VI MAY 29, 1916 No. 9
JOURNAL OF
CONTENTS
A Western Fieldrot of the Irish Potato Caused by Fusarium
radicicola * - - - - - - - - - 297
O. A. PRATT
Comparative Study of the Root Systems and Leaf Areas of
Com and the Sorghums - - - - « -311
E. C. MILLER
Production of Clear and Sterilized Anti-Hog-Cholera Serum 333
M. DORSET and R. R. HENLEY
DEPARTMENT OF AGRICULTOEE
WASHINGTON, D.C
WA6HIN0TON : OOVEnNMENT PSINTINQ OFFICE : 1818
PUBWSHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Indttstry
EDWIN W. AtLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Experitmnt
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition , The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
tlu University of Minnesota
All correspondence regarding articles from the .Department of Agriculture
should be addressed to Karl F. Kellerman, Journal of Agrictrltural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orono, Maine.
JOMAL OF AGEICOLTIALISEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, May 29, 191 6 No. 9
A WESTERN FIELDROT OF THE IRISH POTATO TUBER
CAUSED BY FUSARIUM RADICICOLA
By O. A. Pratt, ^^
Assistant Pathologist, Office of Cotton and Truck Disease Investigations,
Bureau 0/ Plant Industry »,«rAMCAl.
INTRODUCTION
Tuber- rots of the Irish potato (Solanum tuberosum) which are common to
the arid^ West may be grouped into two classes: Storage- rots and field-
rots. This paper is concerned only with certain rots attacking the potato
tuber while growing in the field. From the tuber-rots under discussion, the
fungus Fusarium r adicicola WoWenw. was isolated. Carpenter ^ in 191 5
demonstrated that F. radicicola could, under laboratory conditions,
cause decays in potato tubers similar in every way to these rots. His
experiments, however, v/ere conducted wholly in the laboratories of the
Department of Agriculture, in Washington, D. C. It was therefore
thought practicable to present this paper, which gives the results of
experiments performed under field conditions in the irrigated West.
These experiments substantiate the results obtained by Carpenter and
further establish the relationship of F. radicicola to the field tuber-rots
under consideration.
THE DISEASE
Under the head of fieldrot are considered several types of decay occur-
ring in potato tubers while yet in the field — a stem-end rot, a lenticel
rot, and a rot proceeding from eye infections. Eye infections in the
field are not as common as stem-end and lenticel infections. These
types of rot are known as " stem-end rot," " field dryrot," or " blackrot."
The name "blackrot" best describes them, for the decayed tissues are
nearly black in color when the tubers are taken from the field. The rot may
be further described as a comparatively dry rot, dark to nearly black in
color, proceeding from the stem end, lenticels, and occasionally from
the eyes of the tuber. The decay is first recognized by the blackened,
sunken appearance of the stem end, or, in the case of lenticel and eye
' The observations and experiments set forth in this paper were confined principally to southern Idaho.
' Carpenter, C. W. Some potato tuber-rots caused by species of Fusarium. In Jour. Agr. Research,
V. s, no. 5, p. 183-210, pi. A-B (col), 14-19. 1915.
Journal of Agricultural Research, Vol. VI, No. 9
Dept. of Agriculture, Washington, D. C. May 29, 1916
dni G — 80
(297)
298 Journal of Agricultural Research voi. vi, no. 9
infections, by the blackened, more or less sunken spots on the surface
of the tuber. Tubers collected in a commercial potato field and infected
in this manner are shown in Plate XXXIV, figures i to 6. This black color
is lost in part as the infection becomes older, the infected tissues taking
on various shades from nearly black to sepia brown. In connection with
the stem-end rot, the fungus often proceeds down the vascular tissue,
killing and blackening the network of bundles. Figures 5 and 6 in
Plate XXXIV show sections of a tuber infected in this manner. Often it is
possible to break away the cortical tissues and lay bare the blackened
network. Lenticel infection proceeds outward in all directions from the
point of infection and may or may not extend down to the main vas-
cular system. Very frequently in the case of eye infections the vas-
cular strand connecting the eye with the main vascular system is black-
ened, but it is seldom that such infection extends far into the main
vascular ring. Blackrot is confined principally to potatoes of the Idaho
Rural, Pearl, and other round types.
Closely related to the blackrot of potatoes of the round types is a
jelly-end rot attacking principally varieties of the Burbank group.
Jelly-end-infected tubers of the Netted Gem variety are shown in Plate
XXXV, figures i to 3. The jelly-end rot of the Burbank group differs
from the blackrot of round types of potatoes in that it is a softrot, light
to dark brown in color, while the blackrot is a comparatively dry rot,
black or nearly black in color. Jelly-end rot may be described as a soft,
wet rot of the tubers proceeding from the stem end downward through
the tuber attacking all tissues but apparently advancing somewhat more
rapidly through the vascular bundles. Examination of tubers infected
with jelly-end rot, however, often reveals no perceptible discoloration of
the vascular tissue below the line of the rot in the other tissues. As the
decay becomes older, the stem end becomes somewhat shriveled and
dried, often closely resembling the type of decay caused in storage by
F. trichothecioides Wollenw.^ Lenticel and eye infections are seldom
found in connection with the jelly-end rot of the Burbank group.
Occasionally a softrot of the seed end is also found. A Netted Gem
tuber infected at both the seed end and the stem end is shown in Plate
XXXV, figure i. F. radicicola was isolated from both ends of this
tuber. There was apparently no infection in the vascular tissues con-
necting the tvv^o regions of decay.
At first it was thought that the jelly-end rot of the Burbank group and
the blackrot of round types of potatoes were two distinct diseases, but
inoculations made in 191 4 into the stem ends of Netted Gem and Idaho
Rural tubers with F. radicicola led to the belief that they might be
caused by the same organism. Material collected in the field, whether
jelly-end rot or blackrot, when placed in a moist chamber for a few days
1 Jamieson, Clara O. , and WoUenweber, H. W. An txtemal dry rot of potato tubers caused by Fusarium
trichothecioides, Wollenw. In Jour. Washington Acad. Sci., v. 2, no. 6, p. 146-152, illus. 1912.
May 29, 1916 Fieldrot of Potato Tubers 299
usually showed tufts of F. radicicola. Infected tubers of Idaho Rural
potatoes kept 10 days in a moist chamber at room temperature are
shown in Plate XXXV, figures 4 and 5. Tufts of F. radicicola have
appeared. Inoculations in 191 5 left no doubt in the writer's mind that
F. radicicola was capable of causing both types of rot.
DISTRIBUTION AND ECONOMIC IMPORTANCE
F. radicicola is apparently widely distributed. Wollenweber ^ states
that its habitat is "on partly decayed tubers and roots of plants,
such as Solanum tuberosum in Europe and America (collected by Wollen-
weber) and Iponwea batatas in the United States of America (collected
by Harter and Field)." Carpenter^ makes the following statement as
to its habitat: "On partly decayed tubers and roots of plants. Cause
of potato dryrot and jelly-end rot. Identified from the following:
Ipomoea batatas (collected by Mr. L. L. Harter) ; Musa sapientum (col-
lected by Mr. S. F. Ashby, Jamaica, Porto Rico) ; Cucumis sativus (col-
lected by Mr. F. V. Rand, West Haven, Conn.); soil (collected by Mr.
F. C. Werkenthin, Austin, Tex.)."
The writer has isolated F. radicicola from the roots of poplar trees
(Populus deltoides) at Jerome, Idaho, where he found it associated with
crownrot. The fact that the fungus appears on potato tubers when
disease-free seed potatoes are planted on raw desert lands suggests that
it may be well distributed throughout the desert soils. Orton ^ in
1909 reported jelly-end rot of potatoes from the San Joaquin Valley, in
California.
F. radicicola has been reported on potatoes from Idaho, Oregon,
and California by Wollenweber * and from Idaho, Oregon, California,
Nevada, Mississippi, New York, Virginia, and the District of Columbia
by Carpenter.^ The writer has isolated this fungus from decayed potato
tubers from the following localities in Idaho: Idaho Falls, Blackfoot,
Aberdeen, Rupert, Murtaugh, Twin Falls, Filer, Kimberly, Jerome,
Wendell, Gooding, King Hill, and Caldwell, and has observed the rot in
potato fields in many other localities in the State. The disease appar-
ently appears at its worst under dry-land-farming conditions and in raw
desert land planted to potatoes before having been in other crops. On
comparing rotted tubers collected by himself in Idaho with specimens
sent to the Department of Agriculture from California and Oregon he
was convinced that the rots were of one and the same nature. He has
also observed rots identical in outward appearance with those found in
Idaho, in Portland, Oreg., Seattle, Wash., and British Columbia.
' Wollenweber, H. W. Identification of species of Fusarium occurring on the sweet potato, Ipomoea
batatas. In Jour. Agr. Research, v. 2, no. 4, p. 257. 1914.
2 Carpenter, C. W. Op cit., p. 206.
•■' Orton, W. .v. Potato diseases in San Joaquin County, Cal. U. S. Dept. Agr., Bur. Plant Indus. Circ.
23, 14 p. 1909. .
* Wollenweber, H. W. Op. cit.
' Carpenter, C. W. Op. cit.
^oo Journal of Agricultural Research voi. vi, N0.9
In the irrigated portions of Idaho the economic importance of the dis-
ease has varied greatly from year to year. In 191 3 the writer was usually
able to find only an occasional rotted tuber in any one commercial field.
In a few fields which had been planted on raw desert land and poorly
cared for he found as high as 80 per cent of the tubers infected with stem-
end blackrot and lenticel rot. The year 1914 might be called an epidemic
year. In one 50-acre field of Netted Gems near Jerome, Idaho, he found
as high as 40 per cent of the crop infected with jelly-end rot. Similar
conditions were observed in many other fields in the irrigated portions of
southern Idaho. Stem-end blackrot and lenticel rot were also found
very abundant in the fields of Idaho Rurals. It is significant that in 191 4
a freeze occurred in June which killed the vines to the ground, the plants
coming up anew and producing a crop. Often the origin of infection
could be traced from the frozen tip of the vine down through the stem
to the infected tubers. Although infected tubers were found in most of
the commercial fields visited in 191 5, the disease this year was of slight
importance.
EXPERIMENTAL WORK
PREUMINARY EXPERIMENT IN I914
In the fall of 191 4 ten Idaho Rural tubers and ten Netted Gem tubers
were disinfected by dipping in formaldehyde and were punctured at the
stem end with a needle carrying spores from a culture of F. radicicola which
had been isolated from a potato tuber infected with blackrot. After
inoculation the tubers were placed in moist chambers, where they re-
mained for something over a month. An examination of the tubers
showed that infection had been produced in every tuber inoculated. The
infection in the Idaho Rurals was similar in all respects to the blackrot
occurring in the field. The infection in the Netted Gems was not quite
so dark in color as that produced in the Idaho Rurals and resembled
certain stages of jelly-end rot collected in the field. No checks were
prepared.
LABORATORY EXPERIMENTS IN I915
On August 6, young and apparently healthy potato tubers of the
Netted Gem and Idaho Rural varieties were selected, carefully washed,
and disinfected in a solution of formaldehyde (i : 240). After disinfec-
tion the tubers were dried and inoculated with F. radicicola. The
methods of inoculation were as follows: (i) By spraying with a spore
suspension; (2) by wounding the tubers with a needle bearing spores;
and (3) by dipping the broken stolon ends in a spore suspension. In
method 3 the tubers were taken from the field with their stolons at-
tached. After disinfection each stolon was broken off afresh at from i
to 2 inches from its junction with the tuber and inoculated as stated in
the foregoing.
May 29, 1916
Fieldrot of Potato Tubers
301
Fifty tubers each of Idaho Rural and Netted Gem, respectively, were
inoculated by methods i and 2, and twenty-five tubers each of Idaho Rural
and Netted Gem were inoculated by method 3. Checks on each experiment
were prepared in the same manner, except that in method i the tubers
were sprayed with sterile water, in method 2 the tubers were wounded
with a sterile needle, and in method 3 the broken stolon ends were
dipped in sterile water. Inoculated tubers and checks were placed in
moist chambers and put in the culture room of the Experiment Station
laboratory. During the course of these experiments the culture-room
temperature varied from a minimum of 20° to a maximum of 29° C.
Temperatures were taken daily at 8.30 a. m. and 5.30 p. m. After a
month the tubers were examined. Table I gives a summary of the
experiments and the number of tubers found infected.
Table I. — Summary and results of laboratory inoculations of Solanum tuberosum
Method
No.
Method of inoculation and parts inoculated.
Variety.
Number
of tubers
inocu-
lated.
Number
of tubers
infected.
[Tubers sprayed with suspension of spores
fNetted Gem
\Idaho Rural
fNetted Gem
\Idaho Rural
fNetted Gem
\ldaho Rural
fNetted Gem
\ldaho Rural
fNetted Gem
\Idaho Rural
fNetted Gem
\ldaho Rural
50
50
50
50
50
so
5°
50
25
25
25
25
48
so
I
[check. Tubers sprayed with sterile water
2
3
Tubers punctured with inoculated needle at stem end . .
1 Check. Tubers; stem end punctured with sterile
\ needle.
[Tubers; broken stolon ends dipped in spore suspen-
1 sion.
jChcck. Tubers; broken stolon ends dipped in sterile
t water.
0
50
50
0
0
25
19
0
0
Of the 50 Netted Gem tubers sprayed with the spore suspension, 48
showed infection. Stem-end infection was present in each of the inocu-
lated tubers. Lenticel infections were present on most of the tubers,
and eye infections were also found. Every Idaho Rural tuber sprayed
with the spore suspension showed infection at the stem end. The ma-
jority showed lenticel infections and several showed eye infections. Len-
ticel infections, induced by spraying with the spore suspension, are shown
in Plate XXXVI, figure 3. In figure 4 of Plate XXXVI is shown the
same tuber after remaining several days longer in the moist chamber.
Tufts of F. radicicola have appeared over the surface of the decayed areas.
A stem-end infection of an Idaho Rural tuber sprayed with the spore
suspension is shown in Plate XXXVI, figure 5. Evejy tuber, whether
Netted Gem or Idaho Rural, developed infection when punctured at the
stem end with a needle carrying the spores of the fungus. Decays in-
duced in this manner are shown on Plate XXXVI, figures i and 2.
Twenty-five stem-end tuber infections resulted from the inoculation of
the broken stolon ends in the Netted Gems, and 19 in the Idaho Rurals.
The decay resulting from this method of inoculation was similar in every
302 Journal of Agricultural Research voi. vi, no. 9
way to that produced by the other methods. A stem-end infection
resulting from the inoculation of the broken stolon end under labora-
tory conditions is shown in Plate XXXVI, figure 6. In Plate XXXVI,
figure 7, is shown an Idaho Rural tuber cut to expose the blackening of
the vascular tissue which resulted from the inoculation of the tuber
stolon. None of the checks were infected. The fungus was recovered
from the decayed tissues each time the attempt was made.
EXPERIMENTS IN THE FIELD IN 1915
On August II, in a plot in which disease-free Idaho Rural and Netted
Gem seed potatoes had been planted, apparently healthy potato plants
were selected. The soil was removed from around the plants in such
a manner as to expose the tubers without disturbing their position.
Three growing tubers under each plant were then inoculated with F.
radicicola, after which the soil was replaced, care being exercised to place
moist soil next to the tubers. The methods of inoculation were, respec-
tively, as follows: (i) By spraying the tubers with a spore suspension;
(2) by wounding each tuber stolon with a needle bearing spores at from i to
2 inches from its junction with the tuber; (3) by wounding the upper sur-
face of each tuber with a needle bearing spores, and (4) by puncturing each
tuber at the stem end with a spore-bearing needle. Ten plants each of
Idaho Rural and Netted Gem potatoes were used in each experiment.
As a check on each experiment, a similar number of apparently healthy
Idaho Rural and Netted Gem plants were selected and a similar number
of growing tubers treated in the same manner, except that in the case of
experiment i the tubers were sprayed with sterile water, and in numbers
2, 3, and 4 a sterile needle was used in place of a spore-bearing needle.
A fifth experiment was set up in which 10 apparently healthy Idaho
Rural and 10 apparently healthy Netted Gem plants, growing in the
same plot with those employed in the four experiments just described,
were used. In this experiment, the stem of each plant was punctured
at the crown with a needle carrying spores of F. radicicola. Checks
were prepared in the same manner, except that the stem of each plant
was punctured with a sterile needle.
The soil of the plot in which these experiments were made was very
dry and no irrigation water could be applied after the inoculations were
made. During the course of the experiments (August 1 1 to September 6)
the minimum soil temperature recorded was 66° and the maximum 84° F.
The soil temperature was taken at a depth at which the potato tubers
were found lying by burying the bulb of a soil thermograph under a
potato plant. A little less than a month after making the inoculations
an examination of all the plants was made. Table II gives a summary
of the experiments and the results obtained from inoculating growing
potato plants and tubers with F. radicicola.
May 29, 1916
Fieldrot of Potato Tubers
303
Table II. — Summary and results of inoculating growing potato plants and tubers with
Fusarium, radicicola
Ex-
Number
Number
ment
No.
Method of inoculation.
Variety.
of inocu-
of tubers
lations.
iiJected.
f Tubers soraved with susoension of snores
/Idaho Rural
\Netted Gem
30
17
/Idaho Rural
INetted Gem
30
30
0
f Tuber stolons punctured with inoculated needle
/Idaho Rural
\Netted Gem
30
27
30
23
1 Check. Tuber stolons punctured with sterile needle. ,
/Idaho Rural
INetted Gem
30
0
I
30
0
f Tubers ounctured with inoculated needle
/Idaho Rural
INetted Gem
30
30
3
/Idaho Rural
INetted Gem
30
0
fTubers punctured at stem end with inoculated needle.
/Idaho Rural
INetted Gem
30
30
J
30
30
4
ICheck. Tubers punctured at stem end with sterile /Idaho Rural
I needle INetted Gem
30
0
30
0
IStem of plant punctured at crown with inoculated 1 (Idaho Rural
1 needle \Netted Gem
10
0
10
0
S
[Check. Plant stem punctured at crown with sterile | /Idaho Rural
I needle \Netted Gem
10
0
10
0
Of the 30 Idaho Rural tubers sprayed, 15 showed infection with stem-
end and lenticel rot. Of the 30 Netted Gem tubers sprayed, 17 showed
stem-end rot. Lenticel rot did not occur on all of the Netted Gem
tubers and where it did occur the infections were very slight. The
thicker skin of the Netted Gem probably renders it more resistant to
fungus attacks than the Idaho Rural. The failure of a part of the
sprayed tubers to develop infection can probably be attributed to the
extremely dry condition of the soil. Infections resulting from spraying
the growing tubers with a suspension of the spores of F. radicicola
are shown in Plate XXXVI, figures i to 4. In figure 4, Plate XXXVI, is
shown an eye infection which has extended down into the vascular sys-
tem. F. radicicola was recovered from the discolored vascular tissue
of this tuber. None of the checks showed any infection. Twenty-
seven Idaho Rural tubers infected with stem-end rot resulted from the
puncturing of the 30 tuber stolons. The three which failed to develop
infection were under the same plant. Twenty-three of the Netted Gem
tubers whose stolons were inoculated showed stem-end infection. Seven
showed no evidence of infection in the tubers, though the stolons were
black and dead up to within about one-eighth of an inch of their juncture
with the tubers. Where infection in the tuber was found the line of
infection could easily be traced down the stolon from the point of inocu-
lation into the tuber.
Tuber infections resulting from the inoculation of the stolons in the field
are shown in Plate XXXVII, figures 5 to 8. Both stem-end rot and vas-
cular infection are shown. Figure 8, Plate XXXVII, represents a Netted
Gem tuber with stem-end infection resulting from the inoculation of the
304 Journal of Agricultural Research voi. vi, no. 9
stolon. The rot in this case was nearly black in color, soft, and resembled
the earlier stages of the jelly-end rot often found in commercial fields.
Vascular infection also developed in this tuber. The fungus was recov-
ered from all infected tissues whenever the attempt was made. None
of the checks were infected. Infection resulted in all cases where tubers
were punctured with a needle carrying the spores of the fungus. None
of the checks were infected. In the case of the checks the punctures
could be seen easily but were healed over in each case. The inocula-
tions made into the stems of potato plants failed to give very decisive
results. In each case a blackening of the tissue adjacent to the puncture
was observ^ed. This blackening extended up and down from the point
of puncture for from one-eighth to one-half an inch and in most cases
also extended into the pith.
BIvACKROT
The infections, whether at the stem end, at the lenticels, or at the
eyes, produced by the artificial inoculation of Idaho Rural tubers with
F. radicicola, could not be distinguished in any way from the infections
on decayed tubers collected in the commercial fields. The infections
resulting from the inoculation of growing tubers in the station plots
when final examination was made were not as deep or as far advanced
as many infections occurring naturally in the field, but this can easily
be explained by the late date at which the inoculations were made. In
fact, at the time the inoculations were made, tubers with well-advanced
decay were being found in commercial fields. On the other hand, tubers
with decay no farther advanced than that resulting from the inocula-
tions have often been found in the field late in the season. In every
case where an attempt was made, the fungus was recovered.
Tubers infected by inoculation in the field, by spraying with the spore
suspension, by the puncture of the tuber with an inoculating needle,
and by puncture of the tuber stolons, were placed in moist chambers,
and in each case, after a few days, tufts of F. radicicola appeared.
Blackrot-infected tubers in commercial fields, after being kept in a moist
chamber from 3 to 10 days at temperatures ranging from 65° to 75° F.,
invariably threw out tufts of this fungus (PI. XXXV, fig. 4, 5). Isolations
made from the cortical and medullary tissues of blackrot-infected tubers
have never yielded any fungus other than F. radicicola, which could be
considered as the cause of the disease. Isolations made from stem-end
blackrot-infected Idaho Rurals, Pearls, and other round types of potatoes
have occasionally yielded F. oxysporum, especially when the culture was
made from or near the vascular tissue. The failure to obtain F. oxyspo-
rum from lenticel and eye infections of tubers collected in commercial
fields leads the writer to conclude that when F. oxysporum is found in
stem-end infections it probably entered as a vascular parasite, independ-
May 29, 1916 Fieldrot of Potato Tubers 305
ent of F. radicicola. F. oxysporum has never been found in connection
with the stem-end blackrot of western potatoes to the exclusion of F.
radicicola.
Fully 50 per cent of all cultures made from the decayed cortical and
medullary tissues of tubers infected with stem-end and lenticel rot have
remained sterile. This may have been due to improper cultural condi-
tions, but it is believed that the discoloration of the tuber tissue often
extends some distance beyond the point actually reached by the invading
fungus. Stem-end blackrot-infected tubers often show a black net
necrosis. Isolations made from the black network of bundles, if made
some distance below the stem end, often fail to reveal any fungus. On
the other hand, many such cultures have revealed F. radicicola, and
occasionally both F. radicicola and F. oxysporum. That F. radicicola is
capable of causing the blackened net, as well as the stem-end blackrot,
is fully demonstrated by the results of artificial inoculations PI. (XXXVI,
fig. 7, and PI. XXXVII, fig. 6, 8), though the fungus may not always be
present throughout the entire length of the blackened bundle area.
JELLY-END ROT
Whenever the inoculation of Netted Gem tubers took effect at the stem
end, an infection typical of certain types of jelly-end rot found in the
commercial fields was produced. In the moist chamber under laboratory
conditions infections at the stem end induced by puncturing the tubers,
by spraying with a spore suspension, or by puncture of the stolons with
an inoculating needle were fairly typical of the advanced stages of jelly-
end rot, being soft and watery. Under field conditions, infections at the
stem end induced by spraying the tubers with the spore suspension, by
puncturing with an inoculating needle, or by the inoculation of the stolons
were in no case as pronounced as the infections found occurring naturally
in the field. Those induced by a puncture at the stem end were deeper
than those produced by the other methods.
The failure of the inoculations in the field to develop as severe cases of
infection as those occurring in nature may be attributed to the late date
on which the inoculations were made and to the very dry condition of the
soil. Aside from the depth of the infection at the stem end, the stem-end
decays induced by artificial inoculation were very similar in appearance
to infections found occurring naturally in commercial fields of Netted
Gem potatoes. Wherever the attempt was made, F. radicicola was
recovered from the stem-end infections induced by the inoculations. It
is evident, therefore, that F. radicicola is capable of producing a jelly-
end rot of the potato tuber. However, isolations made from such rotted
tubers taken from the field have not always revealed F. radicicola to the
exclusion of other fungi. F. oxysporum is frequently obtained.
3o6 Journal of Agricultural Research voi. vi, no. 9
Wollenweber * reports the isolation of F. orthoceras from jelly-end tubers
and thought it the probable cause of the disease. The writer has twice
isolated F. trichothecioides from such tubers fresh from the field.
Artificial infection of the growing tuber with F. trichothecioides under
western conditions has never been accomplished. Under conditions of
high humidity Jamieson and Wollenweber ^ were able to produce an
infection in the growing tuber wuth this fungus, but their results are
not believed to be indicative of what actually takes place in nature in
the irrigated West. Tubers infected with jelly-end rot, when kept in a
moist chamber for a few days, invariably threw out tufts of F. radici-
cola through the lenticels, although from these same tubers with well-
advanced stem-end rot other fungi, notably F. oxysporum, have been
isolated from the interior of the tuber. Carpenter ^ has shown that F.
oxysporum is capable of producing a similar rot of the potato tuber,
and from its frequent occurrence in connection with jelly-end-rot-
infected tubers it must be considered as one of the factors involved in
producing this type of rot. Other Fusarium species, either indepen-
dently or in conjunction with F. radicicola, may be in part responsible
for the disease.
STORAGE EXPERIMENTS
In the fall of 1914 two ordinary 2 -bushel sacks filled with Netted
Gems infected with jelly-end rot were secured. With a soft blue pencil,
a line was drawn around each tuber in such a manner that the blue line
separated the decayed from the healthy tissue. The tubers were then
sacked and put in storage in the potato cellar of the Jerome Experiment
Station, at Jerome, Idaho. Fifty tubers each of Pearls and Idaho
Rurals infected with stem-end and lenticel blackrot were secured. On
each tuber a blue line was drawn around the stem end at the margin of
the infected and healthy tissues. Lenticel infections were marked in
the same manner. The marked Pearl and Idaho Rural tubers were
then sacked and placed in storage near the similarly treated Netted
Gems infected with jelly-end rot.
The storage period was from November 15, 191 4, to April 12, 191 5.
The temperature of the cellar during this period ranged from 32° to
48° F. During the last six weeks of the storage period the minimum
temperature was 36°, and for the greater part of this time the tem-
perature approached the maximum of 48°. On April 12 the tubers
were removed from the sacks and examined one by one to determine
whether the rot had continued to develop. In no case could any per-
ceptible advance in the decay be found. It is apparent that neither jelly-
' Wollenweber, H. W. Studies on the Fusarium problem. In Phytopathology, v. 3, no. i. p. 24-50.
I fig., pi. s. 1913-
2 Jamieson, Clara O., and Wollenweber, H. W. An external dry rot of potato tubers caused by Fusa-
rium trichothecioides, WoUenw. In Jour. Washington Acad. Sci., v. 2, no. 6, p. 146-152, illus. 1912.
« Carpenter, C. W. Op. cit.
May 29, 1916
Fieldrot of Potato Tubers
307
end rot nor blackrot makes any progress in storage at a temperature of
48° or under.
This conclusion is further substantiated by results obtained in storing
several sacks of blackrot-infected Idaho Rural and Pearl tubers for
experimental use in the fall of 191 3. Although the infected stock re-
mained in the cellar until the middle of May, 1914, when the cellar tem-
peratures had risen to something over 50° F., the tubers were apparently
as sound as at the time they were put in storage. Carpenter ^ has found
that when tubers were inoculated with F. radicicola and kept at a tem-
perature of 12° C. (approximately 53° F.) no rot developed.
EFFECT OF PLANTING INFECTED SEED
In the spring of 191 5 three plots were planted with infected seed
potatoes. Plot i was planted with Idaho Rural potatoes every
seed piece of which showed infection with F. radicicola, stem-end
blackrot, or lenticel rot. The presence of the fungus was verified by
artificial cultures. Plot 2 was planted with Pearl potatoes every seed
piece of which was infected with F. radicicola, stem-end blackrot, or
lenticel rot, the presence of the fungus being verified by artificial
cultures. Plot 3 was planted with Netted Gem potatoes infected
with jelly-end rot. The seed pieces were cut from the stem end, care
being exercised to see that at least one healthy eye was present on
each seed piece. Cultures from this seed gave a variety of fungi,
including F. radicicola and F. oxysporum. Check plots were planted
with the same varieties. The seed selected for the check plots was
entirely free from disease and was disinfected for i % hours in a solu-
tion of mercuric chlorid (4 ounces of mercuric chlorid to 30 gallons of
water). All of the plots were planted on alfalfa land which had never
before been planted to potatoes. The soil was a heavy clay loam of
lava-ash formation. Irrigation was given on July 4 and 5, July 16,
July 31, and August i. Throughout the season the plots were kept in
a good state of tilth, but they suff"ered somewhat from lack of moisture
during the latter part of August. Table III shows the percentage of
disease in the harvested product.
Table III. — Percentage of disease in harvested potatoes
Plot
Variety.
Condition of seed.
Percentage of dis-
ease in tubers.
No.
Vascular
infection.
Tuber-
rots.
I
Idaho Rural
Pearl
Infected with blackrot
96
44
i6
40
14
10
82
do
40
Netted Gem
Idaho Rural
Pearl
Infected with jelly-end rot
5
do
6
Netted Gem
do
> Carpenter, C. W. Op. cit.
3o8 Journal of Agricultural Research voi. vi, no. 9
The vascular infection present in plots 1 and 2 was all of the heavy
black type demonstrated to be caused by F. radicicola. Numerous
cultures from the vascular systems of tubers from these plots gave
the fungus. The percentages of rot include all phases of blackrot, in-
cluding stem-end, lenticel, and eye infections. Strangely enough, no
tuber-rots developed in plot 3. Of the tubers from plot 3, 16 per cent
showed vascular infection, of which 14 per cent were of the type
usually ascribed to F. oxysporum and 2 per cent were of the black
type caused by F. radicicola. Cultures made from the vascular sys-
tems of infected tubers in this plot give F. oxysporum in all cases
of light-brown discoloration and F. radicicola in all cases of black
vascular discoloration. In the check plots, i per cent of blackrot ap-
peared in plot 5. The others were free from all tuber- rots. The vas-
cular infection present in the check plots was for the most part of
the type ascribed to F. oxysporum. A few tubers showing blackened
vascular bundles were found, and F. radicicola was isolated from such
tissues whenever the attempt was made.
The results clearly show that seed infected with blackrot will produce
infection in the resulting product. From the fact that no jelly-end rot
resulted from planting jelly-end-infected seed, the conclusion should not
be drawn that such seed can not cause infection in the resulting product,
but rather that it requires conditions for its development different from
those required for the development of blackrot.
CONTROL OF BLACKROT
Absolute control of blackrot will be difficult. When potatoes are
planted on alfalfa or grain lands blackrot is rarely found if the crop has
had sufficient water to make good growth conditions possible. Plantings
of disease-free seed potatoes on raw desert lands in 191 5 gave as high as
II per cent of tubers infected with blackrot in the harvested product,
whereas plantings of disease-free tubers on alfalfa or grain lands were
usually free from the disease, although as high as 5 per cent of infected
potatoes were found in the harvested product of one plot on alfalfa
land. Judging from the results of three years' observations in commer-
cial fields, it is apparent that losses from blackrot can be reduced to a
minimum by planting only on land which has been in cultivation for a
number of years and by giving the growing crop the proper amount of
water, care, and attention. The crop should be kept in a good growing
condition until maturity or frost. Jelly-end rot, on the other hand, has
been found in fields where all the conditions of growth were apparently
ideal. Some adverse condition, however, is probably responsible for its
development. Further research upon jelly-end rot and its cause and
occurrence is highly desirable.
Both jelly-end rot and blackrot-infected tubers may be stored with
safety, provided the storage cellar is fairly well ventilated and the tem-
perature kept below 50° F,
May29. iQio Fieldrot of Potato Tubers 309
SUMMARY
(i) Fusarium radicicola Wollenw. is the cause of a field blackrot of
potato tubers in southern Idaho. The disease is confined principally to
potatoes of the round type, such as Idaho Rural and Pearl.
(2) F. radicicola is capable of causing a jelly-end rot of potatoes simi-
lar to the jelly-end rot of the Burbank group found in southern Idaho,
but under actual field conditions other factors are apparently in part
responsible.
(3) Neither blackrot nor jelly-end rot makes any progress in storage
at or below a temperature of 50° F.
(4) Seed pieces infected with blackrot will bring about infection in
the following crop.
(5) F. radicicola is apparently well distributed throughout the desert
soils.
(6) Blackrot may be controlled fairly well by planting potatoes only
on lands which have been in other crops for a number of years and by
providing good conditions for growth.
PLATE XXXIV
Fig. 1,2,3, 4- — Types of stem-end blackrot, lenticel rot, and eye rot in Idaho Rural
potato tubers. Field material.
Fig. 5, 6. — Longitudinal and cross sections of an Idaho Rural tuber infected with
blackrot. Note the blackened vascular system. Field material.
(310)
Fieldroi of Potato
Plate XXXIV
Journal of Agricultural Research
Vol. VI, No. 9
Fieldrot of Potato
Plate XXXV
Journal of Agricultural Research
Vol. VI, No. 9
PLATE XXXV
Fig. I. — Netted Gem potato tuber infected with jelly-end rot. A soft bud-end
infection may also be seen. Field material.
Fig. 2. — Stem-end view of a Netted Gem tuber infected with jelly-end rot. Field
material.
Fig. 3. — Longitudinal section of a Netted Gem tuber infected with jelly-end rot.
Field material.
Fig. 4. — Idaho Rural tuber infected with stem-end and lenticel blackrot, after
having been kept 10 days in a moist chamber. Tufts of Fusariwm radicicola have
appeared. Field material.
Fig. 5. — Idaho Rural tuber infected with lenticel blackrot after having been kept
in a moist chamber for 10 days. A single tuft of F. radicicola has appeared. Field
material.
PLATE XXXVI
Fig. I, 2. — Stem-end blackrot produced by stem-end punctures with a needle
carrying Fusariumradicico la. Netted Gem and Idaho Rural potato tubers. Laboratory-
inoculations.
Fig. 3. — Lenticel blackrot produced by spraying the tuber with a spore suspension
of F. radicicola. Netted Gem tuber. Laboratory inoculation.
Fig. 4. — Same tuber as shown in figure 3; after having been kept a few days longer
in the moist chamber. Note the tufts of F. radicicola that have appeared.
Fig. 5. — Stem-end blackrot produced by spraying an Idaho Rural tuber with a
spore suspension of F. radicicola. Laboratory inoculation.
Fig. 6. — Stem-end blackrot produced by the inoculation of the tuber stolon. Idaho
Rural tuber. Laboratory inoculation.
Fig. 7. — Blackened vascular system produced by the inoculation of the tuber
stolon. Idaho Rural tuber. Laboratory inoculation.
Fieldror of Potato
Plate XXXVI
n^
/
V
>^
Journal of Agricultural Research
Vol. VI, No. 9
Fieldrot of Potato
Plate XXXVII
X
8
Journal of Agricultural Research
^
Vol. VI, No. 9
PLATE XXXVII
Fig. I, 2, 3. — Stem-end and lenticel blackrot produced by spraying the growing
tubers with a spore suspension of Fusarhtm radicicola. Idaho Rural potato tubers.
Fiel^ inoculations.
Fig. 4. — Eye infection produced by spraying the growing tuber with a spore sus-
pension of F. radicicola. Netted Gem tuber. Field inoculation.
Fig. 5, 6, 7. — Stem-end blackrot produced by the inoculation of the stolons of
growing Idaho Rural tuber. Field inoculation.
Fig. 8. — Stem-end rot of Netted Gem tuber produced by inoculating the stolon of
the growing tuber.
36290°— IG 2
COMPARATIVE STUDY OF THE ROOT SYSTEMS AND
LEAF AREAS OF CORN AND THE SORGHUMS
By Edwin C. Miller,'
Assistant Plant Physiologist, Department of Botany, Kansas Agricultural Experiment
Station
INTRODUCTION
During the summers of 191 4 and 191 5 a series of investigations was
conducted to determine the fundamental characteristics possessed by
the sorghum plants (Andropogon sorghum) which enable them to with-
stand severe climatic conditions better than the com plant (Zea mays).
The results of these investigations will be reported in a series of articles
as rapidly as the data are assembled. This paper deals with the com-
parative study of the root systems and leaf areas of corn, Blackhull kafir,
and Dwarf milo. These experiments were carried on at the State Branch
Experiment Station at Garden City, Kans. This Station is located in
the southwestern part of the State, in latitude 37° 58' north and longitude
100° 55' west (Greenwich), and has an elevation of 2,940 feet.
EXPERIMENTAL METHODS
CLIMATIC DATA
The instruments for obtaining the weather data consisted of a thermo-
graph, a hydrograph, a soil thermograph, maximum and minimum
thermometers, a psychrometer, a rain gauge, an evaporation tank, and
two anemometers. The maximum and minimum thermometers, thermo-
graph, and hydrograph were kept in a standard shelter 4 feet above the
ground. One of the anemometers measured the wind velocity at a height
of 2 feet and the other at a height of 8 feet. The 2-foot anemometer was
connected with a clock attachment so that the wind velocity for each hour
was recorded. The bulb of the soil thermograph was buried to a depth
of I foot.
A portion of the weather records for the growing seasons of 191 4 and
1 91 5, grouped in 5-day periods, is given in Table I. This table shows
that the climatic conditions of 191 4 and 191 5 were in marked contrast.
The total rainfall for the year 1914 amounted to only 9.7 inches, while
that for 1 91 5 totaled 26.77 inches.
' Acknowledgments are due Messrs. J. G. Lill and C. B. BrowTi, of the United States Department of
Agriculture, for their aid in obtaining the weather and soil data, and to Mr. M. C. Sewell, formerly super-
intendent of the Experiment Station at Garden City, Kans., for general assistance in this work.
Journal of Agricultural Research, Vol. VI, No. 9
Dept. of Agriculture, Washington, D. C. May 29, 1916
dx Kans. —4
(3")
312
Journal of Agricultural Research
Vol. VI, No. 9
Table I. — Summary of the climatic conditions at Garden City, Kans., for the growing
vionths of igi4 and igij
Year and jnonth.
1914.
May
Do
Do
Do
Do
Do
June
Do
Do
Do
Do
Do
July
Do
Do
Do
Do
Do
August
Do
Do
Do
Do
Do
September. .
Do
Do
Do
Do
Do
1915-
May
Do
Do
Do
Do
Do
June
Do
Do
Do
Do
Do
July
Do
Do
Do
Do
Do
August
Do
Do
Days (iu
elusive).
I- 5
6-10
11-15
16-20
21-25
25-31
I- 5
6-10
11-15
16-20
21-25
26-30
I- 5
6-10
11-15
16-20
21-25
26-31
I- 5
6-10
11-15
16-20
21-25
25-31
I- 5
6-10
11-15
16-20
21-25
26-30
I- 5
6-10
11-15
16-20
20-25
25-31
I- 5
6-10
11-15
16-20
21-25
26-30
I- 5
6-10
11-15
16-20
21-25
25-31
I- 5
6-10
11-15
Air temperature (° F.).
Average of —
58
65
53
62
72
69
76
77
76
76
82
77
74
77
86
76
81
83
77
77
77
82
77
73
77
79
75
77
63
67
53
56
71
46
67
55
65
64
66
71
69
72
66
76
81
72
74
75
69
Maxi-
mum.
68
78
61
79
87
89
88
89
94
94
85
91
99
87
94
98
93
91
91
99
91
87
94
96
89
90
80
86
65
69
87
55
78
65
75
85
79
84
77
90
97
84
85
74
83
80
83
Mini-
mum.
47
51
44
55
59
57
65
64
63
62
69
59
62
60
69
62
65
66
65
62
62
64
61
60
60
64
58
60
44
51
44
55
39
57
47
58
52
53
61
58
59
55
60
67
62
61
64
56
60
61
Maxi-
mum.
78
92
72
79
90
89
92
91
96
99
98
103
94
93
103
lOI
98
102
95
95
95
102
99
94
103
102
96
97
85
90
76
81
94
68
90
72
81
86
87
95
91
88
S3
96
lOI
96
91
90
90
94
86
Mini-
mum.
44
41
38
50
57
49
62
51
59
58
64
51
53
53
64
58
64
64
61
56
58
62
50
54
55
59
48
56
37
47
31
32
46
32
44
39
55
36
50
56
56
57
49
54
64
56
^6
62
51
56
59
Precip-
itation
Inches.
I. 40
• 19
. 20
.72
. 12
I. 00
• 19
. 21
.61
•39
.04
•15
. 10
T.
■^'^
T.
• 19
.06
. 01
T.
01
03
79
.07
I- 15
.64
.94
.62
.69
•57
• 51
.06
• 15
• 13
• 24
.90
5- II
. 10
Evapora-
tion.
Inches.
0-9S3
1.484
I- 135
•596
1.584
1.294
1-432
1.728
1.520
1.409
1-991
I. 862
I. 200
1.440
1.822
1. 416
I- 451
2. 074
1-477
1.792
1-474
1-959
1-745
1-563
1-739
I. 501
1-653
1-390
1-343
I. 740
I. 187
.985
1-857
1-324
I. 069
I. 169
•738
1.386
I. 490
1.485
I. 181
1-419
1-451
1-732
1-743
1.407
1-397
1.528
I. 012
.860
.927
Wind
velocity
per
hour.
Miles.
9.0
II. 8
10. 9
13.6
10. 2
6.9
13-0
15-2
9
6
9
7-7
5-7
5-7
6. I
8.0
7.0
7-5
7-4
7-5
8.6
II. 4
7.6
6-4
II. I
7-7
10.8
12. 2
8.6
8. I
8-7
8.6
8.0
8.8
8-5
5-5
6.8
5-8
4.9
2.7
May 29, 1916 Root Systems and Leaf Areas of Corn and Sorghum 313
Table; I. — Summary of the climatic conditions at Garden City, Kans.,for the growijig
months of 1914 and igis — Continued
Year and month.
Days (in-
clusive).
Air temperature (° F.).
Precip-
Evapora-
Average of
Maxi-
Mini-
itation.
tion.
Mean.
Maxi-
mum.
Mini-
mum.
mum.
mum.
Inches.
Inches.
61
80
61
84
57
0. 03
0.790
70
Si
60
H
57
.46
I. 018
t..S
77
50
«5
40
1-313
68
^?,
55
«7
51
.82
1.424
60
81
50
91
54
I. 029
71
84
60
97
53
T.
.983
6q
82
55
«7
39
. 20
I. 072
66
76
5«
84
50
I. 00
.864
56
67
48
7«
44
•25
.665
Wind
velocity
per
hour.
1915-
August
Do
Do
September. .
Do
Do
Do
Do
Do
16—20
21-25
25-31
I- 5
6-10
11-15
16-20
21-25
25-30
Miles.
3-2
4.4
4-7
7-4
6.3
7.2
5-2
18. 2
4.4
During the growing months of ]\Iay, June, July, August, and Septem-
ber in 1914 the rainfall amounted to only 6.42 inches, while during the
same months in 191 5 it amounted to 17. 88 inches. Table II gives the
number of inches of rainfall for each month during 191 4 and 191 5.
Table II. — Rainfall {in inches) at Garden City, Kans., in IQ14 and igis
Month.
January .
February
March . . .
April . . . .
May
June ....
Year.
None.
Trace.
Trace.
1.74
1.44
None.
2- 53
. 18
2. 67
4-39
2. 96
Month.
July
August. .. .
September
October. . .
November.
December.
0.56
.64
•15
1.48
Trace.
.06
1.66
6.60
2. 27
I. 79
. 12
1.6
The summer of 191 4 was much warmer than that of 191 5, and the
evaporation for each of the five growing months, with but one excep-
tion, was appreciably lower in the latter year than in the former. The
evaporation from a free water surface for each month of the growing
season is given in Table III.
Table III. — Evaporation {in inches) at Garden City, Kans., for the growing months of
IQ14 and igi5
May .
June
July
7.046
9.942
9-403
7-593
7.699
9.258
Month.
Year.
August
September.
10. 010
9.366
5. 920
6.037
314
Journal of Agricultural Researcn
Vol. VI, No. 9
The evaporation during 5-day periods for the two growing seasons
is shown graphically in figure i .
GENERAIv OUTLINE OF THE WORK
The experiments herein reported were conducted with Pride of Saline
corn, Blackhull kafir, and Dwarf milo. The plants were grown both in
the field and in large galvanized-iron cans. The investigations with
the plants in the field included (i) the isolation of the root systems of
-£'.<'
2.3
2.3
Z./
2.0
/ .3
\'--
\/.s
\ /.^
.6
.7
.e
.s
.•^
• 3
.2
./
O
^
-n
\
\
\
ml
\
i
\
1
\
\
s
/
\
,
^1
\
J
\
\
/
A
1
\
\
/
\
/
/
^/
1
j
\
/
\
\
/
\\
/ /
J
V
\
/
/
j
\l
\
/
\
f
y
' 1
V
L-'/
\
<^
/
^
i
1
/
\
\^
1
1
\
/
\
/
--
I
f
\
S
k
1
\
X
/
\
/
f
\
/
\
I
l\
V
A
/
1
1
\
,'
\
\
}
\
f \
/
\
1
/
y
\
\
\
^-
/
^
'
\
\
1
f
s
\
}
r
\
L_
L_
' •' -••'--'•'- '■''■'' Jfe -^ i ^ !^ Jj ^ ; i ^ ^ l^j {§< i ^ 5§ i %
ar^^ V '^ ' V ''^ V *
Jj) N ^5
•<: ^ ^
AiAy^
>JO/V£
fjaLi<
S£P7£/*fe£^
Fig. I. — Evaporation from a free water surface (tank) at Garden City, Kans., during the growing seasons
of 1914 and 1913.
corn, Blackhull kafir, and Dwarf milo at three stages of their growth;
(2) a study of these root systems in relation to their general extent, as
well as the number of their primary and secondary roots; (3) a com-
parative study of the leaf and sheath areas of these three plants at four
periods of their growth; (4) a study of the soil-moisture content and
the depth of root penetration.
The plants grown in the large iron containers furnished the material
for a study of the relative dry weights of the roots and aerial portions
of corn, Blackhull kafir, and Dwarf milo.
May 29, 1916 Root Systems and Leaf Areas of Corn and Sorghum 315
CULTURAIv METHODS
The soil in which the plants were grown is known as a sandy loam of
the Richfield series and shows very little difference in its texture in the
upper 10 feet. Tables VIII and IX give the moisture equivalent and the
wilting coefficient (i, p. 56-73) ^ for the soil at each foot to a depth of
10 feet on the plots which were used in 191 4 and 191 5, respectively.
The plants were grown on plots which had been in Dwarf milo the
previous season. The land was plovv-ed in the fall to a depth of 6 inches
and then irrigated with approximately 8 to 10 inches of water or until
the soil was saturated to a depth of from 3 to 4 feet. It received no
further attention until spring, when it received several shallow cultiva-
tions, was then harrowed, and before planting was leveled with a float.
In order that the plants might be under the same conditions, the corn,
kafir, and milo were planted in alternate rows on the same plots. On
May 23, 1 9 14, and on May 26, 191 5, the crops were surface-planted in
rows 44 inches apart. After the plants were a few inches in height the
corn in the rows was thinned to a distance of i^ to 2 feet between the
plants, BlackhuU kafir from i to i^^ feet, and the Dwarf milo from 8
inches to i foot. The plants were kept free from suckers at all times
during the growing season. The plots were scraped with a hoe as often
as was necessary to keep them free from weeds, but no other cultivation
was given. After the fall irrigation the plots received no water other
than that from the rainfall.
The relative weights of the root systems and aerial portions of the
corn, BlackhuU kafir, and Dwarf milo were obtained from plants grown
in large sealed metal containers. These cans were made of 22-gauge
galvanized iron and were 24 inches in height with a diameter of about 15
inches; and in this experiment each can contained from 100 to no kgm.
of soil. The surface foot of the field soil was worked through a X-i^ich
mesh screen and then thoroughly tamped in the cans. This soil was
in good tilth, and for both seasons had a m.oisture content of 20 to 21
per cent (dry basis). This moisture content v/as kept approximately
constant during the growing season by weighing the cans every 48 hours
and then replacing the water that had been lost by the method used by
Briggs and Shantz (2) in their work on the water requirement of plants.
Different numbers of plants were grown in each can, as will be shown in
the tables that record the data for this part of the work.
ISOLATION OF ROOT SYSTEMS IN THE FIELD
The root systems of plants growing under field conditions were iso-
lated by a modification of the method devised by King (5).^ This
method, stated briefly, consists of the isolation of a prism of soil con-
1 Referenceismadeby number to "Literature cited," p. 331.
' The work of other investigators concerning the development of root systems will be mentioned in this
article only in so far as it is necessary to give a clear discussion of the experiments reported. The studies
that have been made by other investigators on the development of the root systems of agricultural plants
have been reviewed in detail elsewhere by the writer.
3i6 Journal of Agricultural Research voi. vi, no. 9
taining the plants whose root S3^stems are desired and then placing over
this block of earth a wire cage of such a shape and size as to fit closely to
the vertical sides of the block. Numerous small wires are then run
through the prism of earth and fastened to each side of the cage. The
plants are fastened to the cage at the surface of the soil and the roots
washed free from dirt by means of a stream of water. When the earth
is washed away, the main roots remain suspended on the cross wires in
the same position that they occupied in the soil.
This method is open to criticism, first, because in order to use it with
any degree of satisfaction the prism of soil must be limited to about 18
inches in thickness, and on this account one obtains only a section of the
root system. Furthermore, the main roots of the plant may not be in the
prism of soil which has been isolated; therefore, when the soil is washed
away, only a poor representation of the root system is obtained. Finall}^
although the primary roots of the plant remain on the wires in the same
position that they occupied in the soil, it is impossible to retain all the
finer roots in their normal position. No method has been devised, so
far as is knovv^n to the writer, whereby the root systems of mature plants
growing in the field under natural conditions can be isolated intact.
The method of Rotmistrov (6) for obtaining complete root systems is
open to criticism because root systems certainly would not develop
normall)^ in so small a volume of soil. For a comparative study of the
general nature of the root systems of plants, growing under field condi-
tions, the modified method of King as used in these experiments seems
to be the least objectionable.
In the work reported in this paper, sections of the root systems were
obtained crosswise of the rows. The prisms of soil varied from 15 to 18
inches in thicl^ness and were isolated by digging a trench 2% feet wide
around them. After the isolation of a prism of soil, a wooden frame-
work of light material was fitted snugly over it. Ordinary wire fencing
with a 4- to 6-inch rectangular mesh was placed on tvv-o sides of the frame-
work (PI. XXXVIII, fig. 1,2). This was found to be much more sat-
isfactory than the poultry netting used by King and Ten Eyck, since
the small mesh of such netting made it impossible to photograph the root
systems with any degree of satisfaction after they had been isolated.
The plant stubs and root crowns were held in place by wiring them to
narrovv^ strips of inch boards which were placed crosswise of the soil
block at the surface of the soil and nailed to both sides of the framework
of the cage. This method is much more convenient and simple than the
one used by King (5) and Ten Eyck (9, 10, 11). In order to hold the
plants in place, these investigators removed the upper portion of the
soil surrounding the crown of the plant, and replaced it by a plaster of
Paris cast.
For cross wires, ordinary broom wire was found to be the most satis-
factory. Owing to the compactness of the soil, a X"inch iron rod pointed
May J9. 1916 Root Svstems and Leaf Areas of Corn and Sorghum 317
at one end and provided with a wooden handle at the other was em-
ployed to make a passage through the soil block for the cross wires (PI,
XXXVIII, fig. 2). In the upper 2 feet of soil the cross wires were
pushed through the block of soil at inter\'als of 3 to 4 inches on both
the vertical and horizontal wires of the cage, while below that depth
they were placed at the intersections only of the vertical and horizontal
wires. In the isolation of the root systems of two mature plants, be-
tween 200 and 250 cross wires were pushed through the soil prism.
vSeveral methods of washing the soil away from the roots were tried,
but the following was found the most desirable: The trench around the
block of soil was partially filled with water from an irrigation ditch near
by; and then by means of a pitcher pump connected with a ^-inch pipe
of convenient length the water was pumped into a piece of galvanized-
iron eaves trough and allowed to flow gently on the prism of soil (PI.
XXXVIII, fig. 3). In this manner the same water could be used over
and over again. As soon as any of the larger .roots were exposed they
were carefully tied to the cross wires so that they would not be moved
from their original position by the further washing. When the dirt that
had been washed from the soil prism had filled the trenches to the sur-
face of the water, the washing was discontinued and the water allowed
to soak away. The soil that had been washed into the trenches was
then removed, the trench again partially filled with water, and the wash-
ing continued. This routine, especially in working with mature plants,
had to be repeated several times. After the soil had been washed from
all the roots, the cages containing them were taken up, the unused cross
wires removed and the root systems studied and photographed.
ISOLATION OF THE ROOT SYSTEMS FROM LARGE VESSELS
The following method was used in the isolation of the root systems of
the plants that were grown in large galvanized-iron cans :
As soon as the aerial portions of the plants were harv^ested, the soil
contained in the can was emptied upon a cleared space; and all the larger
roots were removed from the soil by carefully working it over, a handful
at a time. In order to separate the soil from the root particles still re-
maining in it, as much of the soil as possible was shaken through a sieve
with a -j^-inch mesh. In this manner all the finer root portions, together
with the larger soil particles, remained upon the screen. The root
remnants and the soil particles on the sieve were then placed in a vessel
and covered with a large excess of water, which was stirred vigorously
until all the lumps of soil had disintegrated. All the root remnants
floated to the surface of the water, and as soon as the soil in the vessel
had settled, they were removed by pouring the water upon the fine
sieve. All the roots which were obtained from each can were placed
upon the fine screen and washed carefully a number of times until, so
3i8
Journal of Agricultural Research
Vol. VI. No. 9
far as could be seen, they were free from sand particles. The roots were
then dried in a hot-air oven at 105° C. and their dry weight obtained.
DETERMINATION OF THE LEAF AREA
For obtaining the leaf and sheath areas five representative plants of
the com, kafir, and milo, respectively, were selected at the desired stage
of growth. Their leaves and sheaths were cut into convenient pieces,
and the outlines of these portions were carefully traced with a hard lead
pencil on ordinary unruled paper. The outlines thus obtained were
traced with a polar planimeter and the inclosed area calculated. In
dealing with that portion of the leaf which was not yet fully unfolded,
care was taken that the measurements included only that surface of the
unfolding leaf that was exposed to the air.
GENERAL DISCUSSION OF EXPERIMENTAL DATA
EXTENT OF THE ROOT SYSTEMS
The root systems of corn, kafir, and milo growing in the field were
isolated at four stages of growth in 191 4 and at three stages in 191 5.
A summary of the general extent of the root systems of these plants is
given in Table IV.
Table IV. — General summary of the root systetns isolated during the sutnmers of IQI4
and jgi5 at Garden City, Kans.
1914.
June 24
July 17
Aug. I
Aug. 25
1915-
July 10
Sept. 3
Crop.
Com, Pride of Saline
Kafir. Blackhull ....
Milo, Dwarf
Com, Pride of Saline
Kafir, Blackhull ....
Milo, Dwarf
Corn, Pride of Saline
Kafir, Blackhull ....
Milo, Dwarf
Corn, Pride of Saline
Kafir, Blackhull
Milo, Dwarf
Com, Pride of Saline
Kafir, Blackhull ....
Milo, Dwarf
Corn, Pride of Saline
Kafir, Blackhull . . . .
Milo, Dwarf
Corn, Pride of Saline
Kafir, Blackhull . . . .
Milo, Dwarf
Height
of
plants.
Ft. in.
I 6
3 6
2 6
3 6
Greatest
depth of
root pen-
etration.
I 3
1 6
2 o
4 6
4 6
4 6
6 o
6 o
6 o
Greatest
lateral
extent of
roots.
Ft. in.
3 o
3 o
3 6
3 10
3 6
3 8
3 8
Greatest
length of
a single
root.
Ft. tn.
3 3
3 8
4 6
General remarks.
4 fully unfolded and 4 par-
tially unfolded leaves.
Do.
Do.
8 fully and 6 partially un-
folded leaves.
6 fully and 4 partially un-
folded leaves.
"Rooting."
"Shooting."
Heading.
Seed in milk stage.
End of vegetative growth,
grains glazed.
Seed in milk.
Seed fully ripe.
4 fully and 4 partially un-
folded leaves.
Do.
Do.
Tassel peeping.
7 fully unfolded and 5 par-
tially unfolded leaves.
Blooming.
Early milk stage.
Blooming.
Seed in milk stage.
May 29, 1916 Root Systcms and Leaf A reas of Corn and Sorghum 319
Stage I. — At this period of growth, the plants of Dwarf milo and Black-
hull kafir had reached a height of i foot and had four fully and four par-
tially unfolded leaves, while the corn plants with the same number of
leaves had a height of i foot 6 inches. In 191 4 the plants reached this
stage on June 24, four weeks from the time of planting the seed; but in
1915, omng to cool weather, they did not reach this stage until July 10,
six weeks after seeding (PL XLIII, fig. i).
In 1 91 4 the greatest depth reached by the root system of the corn plant
at this stage was i foot 4 inches, while the greatest depth of the kafir and
milo roots was i foot 6 inches. At this time the roots of the corn ex-
tended horizontally to a distance of 2 feet 9 inches, while in the same
direction the roots of both kafir and milo extended 3 feet (PL XXXIX,
fig- 3)- The depth of root penetration for corn and kafir at this stage was
practically the same in 1915 as in 1914, but D^varf milo exceeded the
depth reached the previous year by 6 inches. The maximum lateral ex-
tent of the corn roots was the same as in 1914, but it was i foot less for
the kafir and milo (PL XXXIX, fig. 2, 4).
At this time the differences exhibited by these three plants in their
method of rooting were very slight. The number of primary roots which
penetrated to a depth of a foot was between 12 and 15 for each plant, but
more of the first primary roots of the com took a horizontal direction
than did those of the kafir and milo. On this account more of the primary
roots of the latter penetrated to the maximum depth than did those of the
com plant. The secondary roots of all the plants grew both upv/ard and
downward from the primary roots, so that at this stage the upper foot of
soil was filled with a network of roots to within X inch of the surface.
Stage II. — The root system.s at this period of growth were isolated only
in 1 91 4. At this time the corn plants had reached a height of 3^ feet and
had 8 fully and 6 partially unfolded leaves, while Blackhull kafir, with
approximately the same numxber of leaves, had a height of 2}4 feet. The
Dwarf milo plants had from 9 to 10 fully unfolded leaves, including the
"boot" leaf, and stood 2^ feet high. The plants reached this stage on
July 17, seven weeks from the time of planting (PL XLIV, fig. i).
The greatest depth reached by the corn roots at this time v\-as 3 feet,
while the maximum depth for Blackhull kafir and Dwarf milo w^as 2 feet
6 inches and 2 feet 9 inches, respectively. The greatest lateral extent
reached by the roots of com and Dwarf milo at this period was 3 feet,
while the roots of standard kafir extended horizontally for a distance of 4
feet. The tendency of the first primary roots of the com to take a more
horizontal direction than those of the sorghums is well shown at this
stage (PL XXXIX, fig. i).
It was found that the later roots of the com take the same general
direction as do those of Blackhull kafir and Dwarf milo, and that the
maximum depth of root penertation is practically the same for all three
plants.
320 Journal of Agricultural Research voi. vi, no. 9
Stage III. — In 1914 the roots of the three plants were isolated about
the first of August, 10 weeks from the time of planting. The corn at this
stage was shooting and had a height of ^}4 feet, while Blackhull kafir
was heading and stood 4 feet high. The seed of the Dwarf milo was in
the milk stage, and the plant had reached a height of 3 feet.
The greatest depth of root penetration at this stage was 4 feet for all
the plants. The maximum lateral extent of the roots of corn was 2j/^
feet, while the roots of both Blackhull kafir and* Dwarf milo showed a
maximum horizontal extent of 3^^ feet (PI. XL, fig. 2).
The roots at this stage were isolated on July 17, 191 5, when the plants
had reached the same age at which they were examined the previous
year. The corn at this date stood 5 feet high, and the tassel was just
beginning to show. Blackhull kafir stood 3^^ feet high and had seven
fully and five partially unfolded leaves. The Dwarf milo was blooming
and had a height of 3 feet.
The maximum depth and lateral extent of the roots at this stage
was found to be approximately the same for all three plants. The
greatest depth of the roots was 43^ feet, while the greatest extent in a
horizontal direction was approxim^ately 37^ feet.
Stage IV. — The root systems at this stage were isolated on August
25, 1 914, when the plants were 13 weeks old. The com had reached a
height of 6 feet and the grain was in a glazed condition. The seed
of Blackhull kafir was in the milk stage and the plants Avhich stood
5 feet high had reached their maximum vegetative growth. The seed
of the Dwarf milo was fully ripe, and the plants had made little if any
growth since the previous stage (PI. XLIV, fig. 2).
The roots of all three plants were found to reach a maximum depth
of 6 feet, while the greatest lateral extent for all three was between
3 and 4 feet (PI. XL, fig. i).
In 1 91 5 the plants had not reached their full vegetative growth until
September 3, and even at that date they were not nearly as mature
as those examined at the same age in 1914. The corn was 7 feet high,
and the grain was in the early milk stage. Blackhull kafir was in
bloom and had a height of 6 feet, while the grain of the Dwarf milo
was in the milk stage and the plants stood 2>% feet high.
The maximum depth of the root systems was 6 feet for each plant, while
while the maximum extent horizontally for each was 373 feet (Pi. XLI,
fig. 1,2).
Both the primary and secondary roots of Dwarf milo and Blackhull
kafir at all stages of growth were more fibrous than those of the corn.
The length of the secondary roots was found to be approximately the
same for the three plants at any given stage of growth. The secondary
roots of kafir and Dwarf milo broke so easily in the washing process that
it was almost impossible to obtain them intact from the soil which was
used in this experiment (Pi. XLII, fig. 1,2).
May 29, 1916 Root Systems and Leaf Areas of Corn and Sorghum 321
NUMBER OF SECONDARY ROOTS
It has been shown in the foregoing discussion of the isolation of the
root systems of com, BlackhuU kafir, and Dwarf milo at the various
periods of grov/th, that no marked differences were to be observed
between these plants in regard to the number and general extent of their
primary roots. It was thought advisable on this account to make a
study of the number of secondary roots possessed by the three plants
at different stages of growth.
After the isolated root systems had been studied and photographed
the primary roots of each plant were cut into inch lengths and the number
of the secondary roots originating from each unit of length was deter-
mined under a dissecting microscope. The results of this investiga-
tion for all the stages of root growth examined in 191 4 and 191 5 are
shown in Table V. It was found from 321 observations of the roots of
the corn, 311 of Dwarf milo and 210 of BlackhuU kafir that the number
of secondary roots per unit of length of primary root was approximately
twice as great for the two sorghums as for the com. This fact stands
out strikingly not only for each year but for all the different stages
of the development of the root systems (PI. XLII, fig. 1,2).
Table V. — Number of secondary roots per unit of length of primary roots of corn, kafir,
and milo in IQ14 and IQIS at Garden City, Kans.
Year and crop.
Stage of
growth
(height of
plants in
feet).
Number of
observa-
tions.
Average
number of
roots per
inch.
Average
number of
roots per
centimeter.
1914.
iK
ZZ
15
6
Com,
Pride of Saline
6
6
37
57
32
17
12
II
7
5
4
Milo,
Dwarf
1
I
I
3
21
54
72
25
29
26
10
12
10
Kafir
BlackhuU
1
1.
5
40
60
31
26
12
10
Corn,
1915-
Pride of Saline
5
7
50
65
47
16
12
12
6
5
5
Milo,
Dwarf
I
3
3K
24
70
70
23
25
9
10
Kafir
BlackhuU
{
I
6
40
70
20
20
8
8
WEIGHT OF THE ROOTS AND AERIAL PORTIONS OF THE PLANTS
A comparative study was made of the dry weight of the aerial parts
and roots of corn, BlackhuU kafir, and Dwarf milo in 191 4, and for these
three plants and Dwarf BlackhuU kafir in 1915. The root systems that
322
Journal of Agricultural Research
Vol. VI, No. 9
were isolated for this study were obtained from mature plants which were
grown primarily for transpiration studies in the large metal cans pre-
viously described. The plants made a vigorous growth and compared
very favorably in every way with the plants that were grown under
field conditions.
Three corn plants were grown in each can during both seasons. In
1 91 4 the corn reached a height of 5 feet, and in 191 5 it stood 6 feet high,
but no grain was produced in either season. In 1914 six Dwarf milo
plants were grown in each can, but in 191 5 the number of plants was
reduced to three to each can. Six Blackhull kafir plants were grovm to
each can in 191 4 and three plants to each can in 191 5.
The Dwarf milo reached a height of 3 feet in 1914, while in 1915 it
stood 4>^ feet high. The Blackhull kafir plants attained a height of 5
feet in 1914, but in 1915 they reached a height of 6 feet. Dwarf Black-
hull kafir was planted during the season of 191 5 only, and three plants
. were grov/n in each can. These plants reached a height of 4>^ feet. The
results for the two seasons are shown in Table VI.
Table VI.' — Relative -weight of the roots and aerial portions of corn, kafir, and milo in
IQI4 and igij at Garden City, Kan';.
1914
Crop and can No.
Number
of plants.
Weight of
stem, leaves,
and grain.
Weight of
stem and
leaves.
Weight of
roots.
Ratio of the
weight of
stem, leaves,
and grain to
weight of
the roots.
Ratio of the
weight of
the stem
and leaves
to the weight
of the roots.
Milo, Dwarf:
6
6
6
6
6
6
Gm.
187-3
161. 5
173-9
184.4
161. 7
159-7
Gm.
"5-5
121. I
128.7
105. I
102. 9
91. 2
Gm.
11. 7
10.7
12. 9
12. 0
12. 0
9-5
16
lU
15-3
13-4
16.8
9.8
"•3
9.9
8-7
8-5
9.6
•3
A
C
6
Averatre ratio .
15. 0 9. 6
4
5
5
6
4
6
217.9
234.1
212. 6
219. 5
175-6
257-3
163.4
167.4
157- I
159.0
123.6
180. 0
16.5
12. 9
14.2
13.8
10. 9
20.8
Kafir, Blackhull:
7
13.2
18. I
14.9
15-9
16. I
12.3
9-9
12. 9
8
0
II. 0
10
II- 5
II-3
8.8
II
12
Average ratio.
15.0
ID. 9
3
3
3
3
150. 6
153-9
131- 4
163-7
13-7
15-9
15.6
16. 4
Corn, Pride of Sa-
line:
rz
10.7
9.6
8.4
ij.
TC
16
9-9
Average ratio .
9.6
May 29, 1916 Root Systems and Leaf Areas of Corn and Sorghum 323
Table VI. — Relative weight of the roots and aerial portions of corn, kafir, and milo in
1914 and 1915 at Garden City, Kans. — Continued
1915
Crop and can No.
Milo, Dwarf:
Average ratio .
Kafir, Dwarf Black-
hull:
Number
of plants.
13-
14.
15-
16.
17-
Average ratio
Kafir, Blackhull:
18
21
53
54
55
56
57
58
Average ratio
QDm, Pride of Sa-
line:
24
25-
26.
27.
28.
29.
42.
43-
Average-ratio .
Weight of
stem, leaves,
and grain.
Gm.
214. 6
226. 4
231.4
223.3
233-3
217. 6
230-5
Weight of
stem and
leaves.
Gm.
III. 5
III. 8
125.8
121. 3
123-7
no. o
115-8
"7-5
249.7
221.8
2 57-8
168.8
230.2
341-7
219.3
299.7
287
3^°- 3
342.8
333-^
354-2
142.7
133-4
137-9
97-1
135- I
215. o
147.2
207.3
206. 3
213. I
253-2
219.5
244.6
205.6
252-5
234-4
202. 4
211. 2
228. 3
239-7
249-3
Weight of
roots.
13-5
12. 7
14. o
22. 4
15.0
14. o
16.8
15.0
16. o
13.6
15-4
10. 4
16. 9
19. o
14. 6
25.0
23-5
14.7
21. o
20. I
14.7
Ratio of the
weight of
stem, leaves,
and grain to
weight of
the roots.
15.8
17.8
16.5
a (9. 9)
15-5
15-5
13-7
15.0
15.6
16.3
16. 7
16. 2
13.6
15-7
17.9
15.0
11.9
12. 2
(21. I)
16. 3
16.6
(24- o)
30-5
33- I
26. o
28.2
33-^
31-7
24- 7
28. I
14.9
Ratio of the
weight of
the stem
and leaves
to the weight
of the roots.
8
7
6
8.0
8.9
9.8
8.9
9-3
8.0
8.9
II- 3
10. o
8.2
o (14. 4)
12. o
10. 9
a (16. 6)
6-7
7.6
9.0
7-1
6-3
7-2
9-7
" Not included in the average.
The root systems of Dwarf milo and Blackhull kafir were isolated from
six cans in 191 4 and from eight cans in 191 5. The average ratio of the
dry weight of the grain and of the stem and leaves of Dwarf milo to the
dry weight of the roots was as 15 to i in 191 4 and as 15.7 to i in 191 5.
324 Journal of Agricultural Research voi. vi, No. 9
The dry weight of the stem and leaves was 9.6 times the weight of the
roots in 1914, and 8 times their weight in 1915. In 1914 the dry weight
of the grain, stem, and leaves of Blackhull kafir was 15 times that of the
roots, while the ratio of the dry weight of the stem and leaves to the dry
weight of the roots was as 10.9 to i. The average ratio of the weight of
all the aerial parts to the root weight in 191 5 was as 14.9 to i, while the
weight of the stem and leaves v\^as lo.i times that of the roots. In 1914
root systems of corn were obtained from 4 cans and from 10 cans in 191 5.
The average ratio of the weight of the stem and leaves to the weight of
the roots was 9.6 in 1914 and 7.8 in 1915. The roots of Dwarf Blackhull
kafir were isolated from five cans in 191 5. The weight of all the aerial
parts was 15.7 times that of the roots, while the ratio of the weight of
the stem and leaves to the weight of the roots was 8.9 to i.
For the purpose of comparison the results obtained by various investi-
gators for the relative weights of the tops and roots of plants are given
here. It must be borne in mind, however, that the relative weights of
the roots and aerial portions of plants vary according to the conditions
under which they are grown. It has been shown (4, 8, 12) that, among
other factors, the soil-moisture content and the amount of available plant
nutrients are important in determining the ratio of the weight of the
tops of plants to their root weight. Hellriegel (3) found the ratio of the
aerial portions of mature barley and oat plants to the w^eight of their
roots to be 1 1.6 to I, and 6.6 to i, respectively. Schulze (7) reports the
ratio of the weight of the aerial portions to the weight of the roots to be
10.8, 13.5, and 1 1. 1, respectively, for mature wheat, barley, and oat
plants. King (5) found the weight of the aerial part of mature com to
be 7 times that of the root weight, while Kiesselbach (4) found the ratio
of the weight of the tops to the root weight to be 8.5 for com plants
grown in a soil with a water content of 98 per cent and 5.2 for plants
growing in a soil with a water content of 20 per cent.
SOII^-MOISTUR^ CONTENT AND THB DEPTH OF ROOT PENETRATION
In order to be able more exactly to define the conditions under which
the plants used for root examinations were grown, soil samples for
moisture determinations were taken at intervals of from 10 to 14 days
from the plots upon which the corn, standard kafir, and Dwarf milo
grew. Since the moisture content of the soil was determined a few days
before or a few days after the isolation of the various root systems, it
was possible to compare the depth of the penetration of the roots with the
depth of the moisture depletion of the soil.
The results of these observations are given in Table VII. The moisture
content of the soil for each foot to a depth of 10 feet is shown for several
periods of the two growing seasons. The depth of the root penetration
was determined from the root systems isolated at the various stages
May29. i9i6 Root Systems aiid Leaf Aveas of Covu and Sovghum 325
which have already been described. The moisture equivalent, together
with the wilting coefficient obtained from it by the formula of Briggs and
Shantz (i, p. 56-73) for each foot of soil, is also included therein.
Table VII. — Soil-moisture content and depth of root penetration of corn, kafir, and
niilo in 1914 and igiS at Garden City, Kans.
Date.
1914.
June s
July 2
10
21
29
Aug. 9
22
Sept. 6
Wilting coefficient
of Briggs and
Shantz
Moisture equiva-
lent
1915.
June 18
29
July 12
24
Aug. 6
16
25
Wilting coefEcient
of Briggs and
Shantz
Moisture equiva-
lent
Percentage of moisture at a depth of-
foot.
22.9
14- 6
II. 8
10. 6
8.7
9.4
8.4
7- 7
20. 3
16. 2
13-3
24-4
feet.
22. S
20. 2
17- I
13-3
13- I
13- S
13-4
12. 2
feet.
21. 1
21. 2
14-5
26. 7
14. I
25-9
14- S
26. 7
21. 7
20.8
20. 2
17-8
15-4
17-7
15-9
14.9
27-5
feet.
22.8
23.6
19.4
16.8
14- S
14.4
12. 9
16.3
30.0
18. 5
17-9
17.8
17- 2
i5. o
17. 2
13-6
25- I
feet.
17- 1
31-5
IS- 5
16.8
19.0
17-4
16. I
16. 4
IS. 6
13-4
24. 6
6
feet.
16. 1
29-6
16.0
16. 7
IS- 5
16.5
15-4
IS-S
16. 2
II. 9
21. 9
feet.
feet.
15-7
29.0
16.6
17. 1
19. 8
18.9
15-7
2S. 9
19-1
19-3
18.6
19.0
19. 9
12. I
22. 3
feet.
16. 4
16. 7
18.4
20. 6
19- S
15.0
27. 6
19. 2
19-7
20. 2
20. 4
20. 9
13.0
23-9
Greatest depth of
rcotj.
Feet.
1%
'434
6
Feel.
2Va.
6
.Mi!o.
Feet.
1%
3
4^4
6
The season of 1914 was especially favorable for such an observation,
since the rainfall for the last half of June amounted to only 0.44 inch, and
for July and August 0.56 and 0.64 inch, respectively. This amount of
rainfall, a little over 1% inches for the 2^ months, came at 12 different
periods, so that with the exception of the first foot of soil no influence was
exerted by the rainfall upon the original soil-moisture content. The sea-
son of 1 91 5 was not so favorable for an observation of this kind, but the
results, while not so striking as those of 1914, show the same facts. It
should be borne in mind in studying Table VII that in 191 4 the soil sam-
ples which were taken on July 2 and 21 were procured from five to six
days after the isolation of the root systems whose depths are recorded for
that date. Furthermore, in 191 5 the samples for July 12 and August 6
were taken two and six days, respectively, after the recorded depths of
the root systems.
The results of these experiments for both seasons seem to show that
there was little if any depiction of the soil moisture below the depth to
which the roots penetrated.
36290°— 16 3
326
Journal of Agricultural Research
Vol. Yl, No. 9
LEAF AND SHEATH AREAS ^
The leaf and sheath areas of com, Blackhull kafir, and Dwarf mile were
determined at four stages of growth in 1914. The results of these meas-
urements are shown in Table VIII. Figures 2 and 3 represent these
areas graphically. •
Table VIII. — Dry weight, leaf avA sheath areas of corn, kafir, and viilo at different
stages of growth in IQI4 at Garden City, Kans.
Plant and period of growth.
Corn. Pride of Saline. June 24, 1914. one
month from time of planting.
Average.
Kafir, Blackhull, June 24, 1914, one month
from time of planting.
Average.
Mile, Dwarf, June 24, 1914, one month from
time of planting.
Average.
Com, Pride of Saline. July 7, 1914, six weeks
from time of planting.
Average.
Kafir, Blackhull, July 7, 1914, six weeks
from time of planting
Average.
Milo, Dwarf, July 7, 1914, six weeks from
time of planting
Average .
Com, Pride of Saline, July 21. 1914, eight
weeks after planting
Dry
Plant weight
^'^"^ ' of leaves
and
stems.
No.
Average.
Kafir, Blackhull, July 21, 1914, eight weeks
from time of planting
Average .
Cm,
IS- 2
9.6
9. 2
7-5
7-'5
S-4
o. I
6.6
7.0
6-S
50.0
54-7
51.0
27.S
31-4
31-0
29.9
27- S
29-5
2S.3
23.4
24.5
26.8
114. 6
137-2
149.2
140. I
102.8
128. 7
70. I
75- 9
60.7
63.5
67. 8
Leaf area.
Sq. in.
272.2
230.6
285.7
205.8
210. 2
241. o
141-3
291.9
145. o
138.0
II6.8
146.7
140.0
138-3
149-0
15s- 6
141. o
902.7
842-3
828.6
877-3
754-3
372-3
43S.0
484.0
420.7
338.2
40S.6
iq. cm
1,756
1,487
1,842
1,327
I-3S5
I.SS3
961
1 003
787
909
5,822
S,433
5, 344'
5,6,8
4,86s
5, 424'
2,401
2,82s
3, 122
2,635
402.4
456.4
363-9
343-2
397-3
392-
1,231.8
1,423-7
1,378-2
I, 266. 6
1,366.6
1,333-4
987.
96s-
829.
893-
689.
63.5
873-0
2,595
2,943
2,347
2, 214
2,562
2.532
7,945
9,182
8,889
8,169.
8, Si
8, 600,
6,367
6,228
5,349
5.766
4.445
5'63i-3
Sheath area.
Sq. in.
24. I
15-0
15.9
IS- 7
9.8
8.4
Sq. cm.
155-4
96.7
154-4
S2.6
93-9
9-8
71. 2
67.8
81.2
54- o
67-3
68. 3
26.3
35-8
42- 7
35-0
37-4
37-0
32.2
31-9
34-
127.7
127. 2
13S-8
124.8
92.7
122.3
69. I
57-8
50-4
82. 8
42.9
116.6
102.8
100. 9
65-5
63-2
54-2
77-3
62. 3
67.8
62.6
66.8
58.8
63- 6
459-0
43 7-3
523-5
348- 3
434-2
440.4
169.6
231-2
275-7
264.4
189-3
226.0
241-5
238.6
207.4
205.4
211. 2
220. 8
822.2
820.7
895-5
805.2
598.4
7S8. 8
445-8
334- S
325-0
534-4
276.9
383-3
1 In this paper tlie term "leaf area " means the surface inclosed by the margins of the leaves,
leaf surface exposed to the air therefore would be twice the leaf area.
The total
iiay29. isi6 Root Systems and Leaf Areas of Corn and Sorghum 327
Table VIII. — Dry weight, leaf and sheath areas of corn, kafir, and milo at different
stages of growth in igidat Garden City, Kans. — Continued
Plant and period of growth.
Plant
No.
Day
weight
of leaves
and
stems.
Leaf
area.
Sheath area.
STAGE ni— continued.
Milo, Dwarf, July 21, 1914, eight weeks from
time of planting. Leaf growth completed.
f I
2
1 ^
4
I S
Gm.
48.6
54-1
49-8
57- 0
47-7
Sq. in.
606. 4
664.6
S8S-S
593-3
572- S
Sq. cm.
3. 911. 4
4, 286. 8
3, 796- 2
3,827-0
3,693-0
Sq. in.
75-0
45-8
44-6
53-7
40-5
Sq. cm..
471.2
293-8
287-8
346. 49
260. 9
SI- 4
605. I
3,902.9
SI- 4
332-4
1 ;
STAGE IV.
Com, Pride of Saline, Aug. 4, 1914, ten weeks
from time of planting. Leaf growth com-
pleted
167.4
197. I
171. 0
1,273-6
1,630. 7
1,324-6
8,215.0
10,517-7
8, 543- 7
192. 1
2S9-4
210- 9
1,239-3
1,737-9
1,360.6
178-3
1,409. 6
9,092. 1
224- I
1,445.9
1 i
Kafir, Blackhull, Aug. 4, 1914, ten weeks
from time of planting. Leaf growth com-
84-5
123. 0
113-9
83.0
734-4
992.5
917.0
S71.2
5. 059- 3
6,401.9
5,914.9
5,6:9-2
83-7
94.0
103- S
94-8
540.2
606.3
667.9
611. 5
A •
loi. 1
891-3
5. 748- 8
93-2
606.5
erag
M
70.6
54-6
70.2
69-5
85.9
67.6
102. I
89-4
Milo, Dwarf, Aug. 4, 1914, ten weeks from
time of planting. Leaf growth completed
6^3.5
576.6
Average
66.2
605. I
3.902.9
EC- 2
536- 2
SUMMARY
Plant and period of growth.
Height of
plants.
Num-
ber of
leaves. 1
Dry
weight
of stem
and
leaves.
Leaf
irea.'
Sheath area.
Square
centi-
meter of
leaf area
per gram
of dry
weight.
Stage I, June 24, 1914:
Corn
Feet.
1-5
I.O
1. 0
2.S
i-S
2. 0
4
2-5
2-S
6
4
3
Cm.
45
30
30
75
45
60
120
73
I So
120
90
4F4P
4F4P
4F4P
6F6P
6F4P
6F3P
9FsP
7F3P
9
14-13
12-14
9-10
Gm.
II- S
8.1
6-5
51
29-5
25- 1
12S-7
68.5
SI- 4
178- 5
151. I
66.2
Sq. in.
241
146
141
841
40S
392
1,333
873
60s
1,409
891
6c 3
Cm.
1-533
945
909
S.244
2,63s
2,532
8, 600
5,631
3,902
9,092
5,748
3,902
Sq. in.
18
12
9
68
35
34
122
59
SI
224
93
86
Sq. cm.
116
77
63
440
226
220
7SS
3S3
332
1,445
606
556
133-0
Kafir
116- 6
Milo
139.8
Stage II, July 7, 1914:
Corn
102-8
Kafir
89-3
Milo
100. 0
Stage III, July 21, 1914:
Corn
66.8
86.2
Milo
73-9
Stage IV, August 4, 1914:
Corn
50.0
Kafir
55-8
Milo
58.9
1 F=- Leaves fully imfolded ; P= Leaves partially imfolded. ' Leaf surface equals twice these figures.
Stage I. — The plants reached this stage one month from the time of
planting. Each plant showed four fully and four partially unfolded
leaves. The Dwarf milo and Blackhull kafir plants had reached a height
of I foot, while the com plants stood i>< feet high (PI. XLIII, fig. i).
The leaf areas at this stage measured 1,553, 945, and 909 sq. cm. for corn,
Blackhull kafir, and Dwarf milo, respectively, while the sheath areas of
328
Journal of Agricultural Research
Voi. VI, No. 9
A^/lO
these plants taken in the same order amounted to 1 16, "jj, and 63 sq. cm.
It is seen at this stage that the leaf area of com was 1.7 times that of
Dwarf milo and i .64 times that of the Blackhull kafir.
Stage II. — The com plants at this time had a height of 2^ feet and
possessed six fully and six partially unfolded leaves. The Blackhull
kafir measured i^feet in height and showed six fully and four partially
unfolded leaves, while the Dwarf milo stood 2 feet high and had six
fully and three par-
tially unfolded leaves.
The plants reached
this condition six
weeks from the date
of planting (PI. XUII,
fig. 2). At this time
the leaf area of the
com had increased to
5,424 sq. cm., while
that of the Blackhull
kafir and Dwarf milo
measured 2,635 and
2,532 sq. cm., respec-
CO/PA/
^77^ GET /.
<9^/Si(p./M
B63S C/772\
M/LO
PS32 C/?!^
STy^G^ 2.
COf?/V
t573 SQ./M
S63/ Cm^
A^/LO
33 OS cm^
sST^^GjETsS.
/^OS SQ./AC
•3092 cm^
SS/ SQ. /M
S74i9 cm^
60SSQ./M
3902cm^
tively, the leaf area
of the corn having in-
creased to twice that
of the Dwarf milo or
Blackhull kafir. The
leaf area of the two
sorghums increased
at the same rate up
to this stage. The
sheaths of all three
plants showed an area
approximately three
times larger than they
did when examined in
the first stage.
Stage III.— The
plants at this period were 8 weeks old. The com stood 4 feet high
and had nine fully and five partially unfolded leaves. Blackhull kafir
and Dwarf milo had each reached a height of 2^ feet. The former
had seven fully and three partially unfolded leaves, while the latter
was in the "booting stage" and possessed nine fully grown leaves (Pi.
XLIV, fig. i). The Dwarf milo at this stage had reached its full leaf
development and showed a leaf area of 3,902 sq. cm. The leaf area of
the com plant was 2.2 times this, or 8,600 sq. cm. The leaf area of
Fig. 2.— Comparison of the leaf areas of Pride of Saline corn, Black-
hull kafir, and Dwarf milo at four stages of the growth of these
plants during the season of 1914.
May 29, 1916 Root Systetus and Leaf Areas of Corn and Sorghum 329
Blackhull kafir had increased to 5,631 sq. cm. and was 1.44 times the leaf
extent of the Dwarf milo. The sheath area of the corn, Blackhull kafir,
and Dwarf milo measured 788, 383, and 332 sq. cm., respectively.
Stage IV. — The plants at this stage had reached an age of 10 weeks
and had completed their leaf development. The corn plants had from
14 to 15 leaves and the standard kafir from 12 to 14 leaves. The com
plants were 6 feet high, the standard kafir 4 feet high, while the Dwarf
milo had reached a height of 3 feet (PI. XLIV, fig 2). The leaf area of
the corn plant at maturity was 9,092 sq. cm., an area 2.3 times that of
the mature Dwarf milo, and 1.53 times that of the Blackhull kafir. The
sheath area of these
three plants was 1,445, CO/rW /f/^F//? A^/LO
605, and 556 sq. cm.,
respectively, for com.
and
S7y=iG^ /.
^^O
c/77^
2SO
^TaHG£^2.
333^
Crr?2
332
cm 2
ST^G£'s3.
SS6
C/77S
Blackhull kafir,
Dwarf milo.
SUMMARY
The root systems of
Pride of Saline com,
Blackhull kafir, and
Dwarf milo plants
which were grown in
alternate rows were
isolated in the field at
four stages of growth
in 1914 and at three
stages of growth in
1915. All told, the
root systems of 33
plants were isolated
and studied. It v/as
found that for a given
stage of growth each
plant possessed the same number of primary roots and that the general
extent of these roots in both a horizontal and vertical direction was the
same for all three plants. The maximum depth of root penetration for
mature Dwarf milo, Blackhull kafir, and corn was found to be 6 feet for
both the years 1914 and 1915. It was found that Blackhull kafir and
Dwarf milo possessed approximately twice as many secondary roots per
unit of primary root as did the com plant. This is true not only for
both years but also for all stages of the root systems examined. Both
primary and secondary roots of the sorghums were found to be more
fibrous than those of the com plant.
ST^GE-'^,
Fig. 3— a graphic illustration of the sheath areas of Pride of Saline
corn. Blackhull kafir, and Dwarf milo at four stages of the growth of
these plants during the season of 1914.
330 Journal of Agricultural Research voi. vi, ko. 9
The relation of the weight of the dry matter of the aerial portions of
mature plants to the weight of the roots was determined in 1914 for
36 Dwarf milo plants, 30 Blackhull kafir plants, and 12 corn plants.
In 1 91 5 the same determinations were made for 24 Dwarf milo plants,
14 Dwarf Blackhull kafir plants, 23 Blackhull kafir plants, and 24 corn
plants.
The average ratio of the dry weight of the grain, stem, and leaves of
standard kafir to the dry weight of the roots was found to be 15 and 14.9
for the years 1914 and 191 5, respectively, while the dry weight of the
stem and leaves of the same plant was on the average 10.9 times that of
the root weight in 1914 and lo.i times the root weight in 1915. The
ratio of the dry weight of the stem, leaves, and grain of Dwarf milo to
the weight of the roots was found to be as 15.7 to i in 1914, and as 15 to
I in 1 91 5, and the weight of the stem and leaves of the same plants was
9.6 and 8 times, respectively, the weight of the roots in 1914 and 1915.
The weight of the stem and leaves of Pride of vSaline com was 9.6 times
the root weight in 1914, while in 191 5 the weight of the stem and leaves
of the com was 7.8 times the weight of the root system. The aerial parts
of Dwarf Blackhull kafir examined in 191 5 showed a weight 15.7 times
that of the roots, while the weight of the stem and leaves amounted to
8.9 times the weight of the underground portion.
The results of the experiments for the two years in regard to the soil-
moisture content and depth of root penetration seem to show that under
the conditions of this experiment very little, if any, depletion of soil
moisture took place below the depth of root penetration.
The average leaf areas of five representative plants of corn, Blackhull
kafir, and Dwarf milo were obtained at stages when the plants were 4,
6, 8, and 10 weeks of age. The last stage examined showed that the
plants had completed their full-leaf development. In all the stages of
growth the com plant was found to have the greatest leaf area. Taking
the stages of growth in order, one finds that the leaf area of the com
plant w^as 1.7, 2.0, 2.2, and 2.3 times the leaf area of Dwarf milo and 1.6,
1.9, 1.5, and 1.5 times that of Blackhull kafir.
In comparing the plants of Dwarf milo, Blackhull kafir, and Pride
of Saline com, it will be seen that in all stages of their growth these
two sorghum plants have a primary root system that is just as
extensive as that of the corn plant. In addition, the Dwarf milo and
Blackhull kafir possess twice as many secondary roots as the com at any
stage of its growth. The leaf area of the corn plant at all stages of its
growth is approximately twice as great as that of the Dwarf milo and
never less than 1.5 times that of Blackhull kafir.
It is apparent, therefore, that the Dwarf milo and Blackhull kafir plants
would have the advantage over the corn plant under any climatic con-
dition that would tend to bring about a loss of water from these plants.
May =9.1916 Root Systeiiis and Leaf Aveas of Corti and Sorghuiu 331
The two sorghums have, in the first place, as compared to the com
plant, only one-half the leaf surface exposed for the evaporation of water;
and in the second place they have a root system which, judging from the
number of secondary roots, would be twice as efficient in the absorption
of water from the soil.
LITERATURE CITED
(i) Briggs, L. J., and Shantz, H. L.
1912. The wilting coefficient for different plants and its indirect determination.
U. S. Dept. Aer. Eur. Plant Indus. Eul. 230, 77 p., 9 fig., 2 pi.
(2)
1913. The water requirement of plants. I. — Investigations in the Great Plains
in 1910 and 1911. U. S. Dept. Agr. Bur. Plant Indus. Bui. 284, 49 p.,
II pi.
(3) Hellriegel, Hermann.
1883. Beitrage zu den naturvvissenschaftlichen Grundlagen des Ackerbaus . . .
Abschnitt 2. — Wurzel und Bodenvolumen. p. 118-280, 7 fig.
(4) KlESSELBACH, T. A.
19 10. Transpiration experiments with the com plant. In Nebr. Agr. Exp.
Sta. 23d Ann. Rpt. [1909], p. 125-139, 2 fig.
(5) King, F. H.
1893. Natural distribution of roots in field soils. In Wis. Agr. Exp. Sta. 9th
Ann. Rpt. [1891]/ 92, p. 112-120, fig. 9-17.
(6) ROTMISTROV, V. G.
1909. Root- Systems of Cultivated Plants of One Year's Gro^vth. 57 p., illus.
Odessa.
(7) SCHULZE, B.
1914? Studien iiber die Bewurzelung unserer Kultiupfianzen. In Festschrift,
50. JubilaumAgr. Chem. Versuchs.u.Kontroll-Stat., Breslau, p. 67-95,
10 pi.
(8) Seelhorst, C. von, and Freckmann, W.
1903. Der Einfluss der Wassergehaltes des Bodens auf die Ernten und die
Ausbildung verschiedener Getreide-Varietaten. In Joiy. Landw.,
Bd. 51, Heft 3, p. 253-269.
(9) Ten Eyck, A. M.
1899. A study of the root systems of wheat, oats, flax, corn, potatoes and sugar
beets and the soil in which they grew. N. Dak. Agr. Exp. Sta. Bui.
36, p. 333-362.
(10)
1900. A study of the root systems of cultivated plants grown as farm crops.
N. Dak. Agr. Exp. Sta. Bui. 43, p. 535-550, 15 fig.
(11)
1904. The roots of plants. Kan. Agr. Exp. Sta. Bui. 127, p. 199-252, i fig.,
26 pi.
(12) Tucker, M., and Seelhorst, C. von.
1898. Der Einfluss, welchen der Wassergehalt und der Reichtum des Bodens
auf die Ausbildung der Wurzein und der oberirdischen Organe der
Haferpflanze ausiiben. In Jour. Landw., Bd. 46, Heft i, p. 52-63.
PLATE XXXVIII
Fig. I. — Method used in isolating root systems in the field. View of two soil prisms
ready for washing. The trenches here shown are 3 feet wide, 12 feet long, and 6 feet
deep.
Fig. 2. — Method used in isolating root systems. This figure shows the method of
placing the cross wires through the soil block.
Fig. 3. — Method of washing used in the isolation of the root systems. The trench
was partially filled with water, which was continuously pumped upon the prism of
soil by means of a pitcher pump.
(332)
Root Systems and Leaf Areas of Corn and Sorghum
Plate XXXVIIl
Journal of Agricultural Research
Vol. VI, No. 9
Root Systems and Leaf Areas of Corn and Sorghum
Plate XXXIX
m
Kl!^
IMMI1I
Journal of Agricultural Research
Vol. VI, No. 9
PLATE XXXIX
Fig. I. — Root system of a com plant that had reached a height of 3 feet 6 inches.
Seed planted May 23, 1914. Root system isolated on July 17, 1914. Greatest depth
of root penetration, 3 feet. Greatest lateral extent of the roots, 3 feet 6 inches.
Fig. 2 . — Root systems of two corn plants with a height of i foot 6 inches. Seed
planted on May 26, 1915. Root systems obtained on July 10, 1915. Greatest depth
of roots, I foot 3 inches. Greatest lateral extent of roots, 2 feet 10 inches.
Fig. 3. — Root system of a Dwarf milo plant at the age of 4 weeks. Seed planted
on May 23, 1914. Root system obtained on June 24, 1914. Plant stood i foot high.
Greatest depth of root penetration, i foot 6 inches. Greatest lateral extent of roots,
3 feet.
Fig. 4. — Root systems of two Blackhull kafir plants i foot in height. Seed planted
on May 26, 1915. Root systems isolated on July 10, 1915. Greatest depth of root
penetration, i foot 6 inches. Greatest lateral extent of roots, 2 feet.
PLATE XL
Fig. I. — Root systems of two mature corn plants. These plants stood 6 feet high,
and the grain was in the glazed condition. Seed planted on May 23, 1914. Root
systems obtained on August 25, 1914. Greatest lateral extent of the roots, 3 feet. ,
Greatest depth of root penetration, 6 feet. The lower portion of the root cage is not
shovvTi here, but the roots which penetrated the sixth foot are shown in a horizontal
position at the bottom of the cage.
Fig. 2. — Root system of a com plant at the tim.e of "shooting." Height of plant,
5 feet 6 inches. Seed planted on May 23, 1914. Root system obtained on August i,
1915. Greatest depth of root penetration, 4 feet. Greatest lateral extent of the roots,
2 feet 6 inches.
Root Systems and Leaf Areas of Corn and Sorghum
Plate XL
Feet
Journal of Agricultural Research
Vol. VI, No. 9
Root Systems and Leaf Areas of Corn and Sorghum
Plate XLI
SIll^^ipsMfiiBlMSI^ifBigiEii
■ilBiilniiBi&iiHffiil^'^SiSiMiii-
■liMliWPlMlBnBIBliaiiiaBMWIBlllBllllW
'lliaiHHIMIiBIT ~
aiieiiiiiiiifaniWieilMHWi
ill 6
Journal of Agricultural Researcli
Vol. VI, No. 9
PLATE XLI
Fig. I. — Root systems of two Blackhull kafir plants at the time they had reached
a height of 6 feet and were blooming. Seed planted on May 26, 19 15. Root sys-
tems isolated on September 3, 1915. Greatest depth of root penetration, 6 feet.
Greatest lateral extent of the roots, 3 feet 8 inches.
Fig. 2. — Root system of two Dwarf milo plants at the time the seed was in the milk
Stage. The plants stood 3 feet 6 inches high. Seed planted on May 26, 1915. Root
systems isolated on September 3, 1915. Greatest vertical penetration of the roots,
6 feet. Greatest lateral extent of the roots, 3 feet 8 inches.
PLATE XLII
Fig. I. — Portion of a primary root of Pride of Saline com, showing the number and
relative size of the secondary roots. Both the primary and secondary roots of the com
are larger than those of the Dwarf milo or standard kafir.
Fig. 2. — Portions of the primary roots of BlackhuU kafir, showing the number and
relative size of the secondary roots. Both the primary and secondary roots of Dwarf
milo and BlackhuU kafir are smaller and more fibrous than those of the com. The
number of secondary roots per unit of length of primary root is twice as great for
BlackhuU kafir and Dwarf milo as for the com.
Root Systems and Leaf Areas of Corn and Sorghum
Plate XLII
Journal of Agricultural Researcli
Vol. VI, No. 9
Root Systems and Leaf Areas of Corn and Sorghum
Plate XLIII
Journal of Agricultural Research
Vol. VI, No. 9
PLATE XLIII
Fig. I. — Pride of Saline com, Dwarf milo, and BlackhuU kafir plants, showing their
relative leaf and sheath areas at 4 weeks of age. Seed planted on May 23, 1914.
Leaf areas determined on June 24, 1914.
Fig. 2. — Pride of Saline com, Dwarf milo, and BlackhuU kafir plants, showing their
relative leaf and sheath areas at 6 weeks of age. Seed planted on May 23, 1914.
Leaf areas determined on July 7, 1914.
PLATE XLIV
Fig. I. — Pride of Saline corn, Dwarf milo, and BlackhuU kafir plants, showing
their relative leaf and sheath areas at 8 weeks of age. vSeed planted on May 23, 19 14,
Leaf areas determined on July 21, 19 14.
Fig. 2. — Pride of Saline com, Dwarf milo, and Elackhull kafir plants, showing
their relative leaf and sheath areas at 10 weeks of age. At this time the plants have
completed their leaf development. Seed planted on ]^Iay 23, 1914. Leaf areas
determined on August 4, 1914.
Root Systems and Leaf Areas of Corn and Sorghum
Plate XLIV
Journal of Agricultural Research
Vol. VI, No.9
PRODUCTION OF CLEAR AND STERILIZED
ANTI-HOG-CHOLERA SERUM
[preliminary paper]
By M. Dorset, Chief, and R. R. Henley, Chemist, Biochemic Division, Bureau of
A n im al I ndust ry
INTRODUCTION
In the United States the anti-hog-cholera serum of commerce for the
most part consists of the defibrinated blood of hyperimmunized hogs.
The red corpuscles contained in such commercial serum are not only
devoid of protective qualities but are objectionable for a number of rea-
sons. The practice of using the defibrinated hog's blood was adopted
because of the difficulty experienced in separating completely the clear
serum from the fibrin and the blood corpuscles.
Hog blood, when allowed to undergo spontaneous coagulation, ordi-
narily yields but a small proportion of clear serum. In practice not
more than 30 or 35 per cent can be secured, the remainder of the serum
being held firmly within the large clot. If, instead of allowing the blood
to clot spontaneously, immediate defibrination be practiced, a yield of
defibrinated blood varying from 90 to 95 per cent may usually be obtained.
This defibrinated blood contains all of the antibodies present in the
blood when drawn, whereas, if the blood is allovv-ed to coagulate and the
separated clear serum alone is used, there must be a large loss of anti-
bodies, because part of the serum is held back in the clot.
The occurrence of the foot-and-mouth disease in the United States
and the accidental infection of certain lots of hog-cholera serum and
virus with this disease have demonstrated the urgent need for some
method of treating these products which will serve to remove the possi-
bility of either of them being a medium for its dissemination. In order
to insure the freedom of hog-cholera serum from the virus of the
foot-and-mouth disease, it is not sufficient merely to filter the product
through bacteria-proof filters, because the virus of this disease itself is
known to pass through bacteria-proof filters. It is likewise known
that the virus of the foot-and-mouth disease is more or less resistant to
the preservatives which are commonly used and which are suitable
for the preserv^ation of serum. There seems to be, therefore, only one
means by which the serum may be sterilized in so far as the virus of
the foot-and-mouth disease is concerned, and thqt is by the application
of heat. The best European authorities state that this virus is killed
when heated at a temperature of 50° C. for 12 hours. It also seems
Journal of Agricultural Research, Vol. VI, No. 9
Dept. of Agriculture, Washington, D. C. May 29, 1916
em A— 22
(333)
334 Journal of Agricultural Research voi. vi, no. 9
well established that the virus is killed by 5 minutes' exposure to a
temperature of 60°.
Experimental work has shown that defibrinated hog-cholera-immune
blood may be heated to 50° C. for 1 2 hours without destroying the anti-
bodies and without materially altering the physical character of the
defibrinated blood. Heating to higher temperatures — 60°, for example —
results in more or less complete coagulation of the defibrinated blood,
and therefore in the destruction of the serum in so far as its commercial
worth is concerned. While heating at 50° for 12 hours might appear
to be satisfactory, in practice it would be difficult and expensive to carry
out such a process.
Experiments with clear serum, separated from the red cells, have
shown that, unlike the defibrinated blood, which coagulates at 60°, the
serum, separated from the red blood cells, withstands heating at 60°
for 30 minutes without alteration of its physical characters and without
noticeable impairment of its antitoxic power.
With the above facts in mind, renewed efforts have been made to
devise a cheap and simple process for preparing hog-cholera antitoxin
in the form of a clear serum free from the red blood corpuscles and from
corpuscular debris.
PREPARATION OF THE SERU]\I
If ordinary defibrinated hog's blood be subjected to centrifugalization,
there may be secured ordinarily about 50 per cent of serum. The time
required will naturally depend to a large extent upon the precipitating
force developed by the centrifuge. We have found that a force equiva-
lent to approximately 1,700 times gravity serves to attain this result in
from 20 to 30 minutes. The serum which separates is usually cloudy,
and, owing to the fact that the red blood corpuscles are not firmly packed,
it is impossible to remove all of the serum without at the same time
carrying over some of the red cells. Therefore, simple centrifugalization
has not seemed practicable for the following reasons: (i) Antibodies are
lost because of inability to separate all of the serum from the corpuscles,
(2) the serum secured is generally not clear, and (3) the removal of the
serum from the cells is a difficult and tedious procedure.
In endeavoring to overcome the difficulties enumerated above, we have
used extracts of the seed of different varieties of the common garden
bean (Phascolus multifiorus and P. vulgaris). Extracts of these beans
are known to possess the property of agglutinating the red corpuscles of
hog's blood, and they are said to be nontoxic.^ Our own experience has
shown that, although the extracts ^ exert no general systemic effect
upon rabbits, guinea pigs, or hogs, certain varieties of these beans do
yield extracts which act as intense local irritants, resulting, in guinea pigs
1 Mendel, L. B. Observations on vegetable hxmagglutinins. /« Arch. Fisiol., v. 7, p. 168-177. 1909.
2 Extracts made with water or normal salt solution.
May 29. 1916 Clear and Sterilized Anti-Hog-Cholera Serum 335
at least, in swelling, followed by necrosis of tissue and the formation of
suppurating abscesses at the sites of injection. The extracts of the
scarlet runner bean (P. multiflorus) and of the pink kidney bean (P. vul-
garis) are both intensely irritating, v.'hile extracts of the common white
navy bean {P. vulgaris) are entirely lacking in this irritating property.
While both the scarlet runner and the kidney bean are very powerful
agglutinants, they have been rejected, at least temporarily, and extracts
of the common white navy bean have been used exclusively in our later
work.
Very minute amounts of the extracts of the navy bean serve to agglu-
tinate large quantities of defibrinated hog's blood ; and when such agglu-
tinated blood is centrifugalized, the red cells pack together and form a
rather stiff jelly-like mass in the tube. With a precipitating force of
about 1 ,700 times gravity about 50 per cent of serum may be separated
in 15 minutes. The serum is clear and may be readily poured from the
tube.
In order to secure a greater yield of serum and a more firmly packed
clot of red corpuscles, we find that the addition of a small quantity of
sodium chlorid is very effective. The addition of i per cent of sodium
chlorid to defibrinated hog's blood after agglutination from the addition
of bean extract has begun will increase the yield of serum from 50 per
cent without the salt to 70 per cent when the salt is added.
Considerable experimental work has led to the adoption of certain
conditions of work as being most favorable to the production of the maxi-
mum amount of clear serum from defibrinated hog's blood. While
experience may later show that some changes in procedure are desirable,
it seems best to describe here the exact method, which is now being
applied in these laboratories, of producing a clear sterile serum, heated
to avoid the possibility of foot-and-mouth disease infection.
Preparation of bean extract. — One hundred gm. of coarsely ground
white navy beans are allowed to soak for one hour in 500 c. c. of distilled
water, with occasional stirring. The pulp is strained through cheese-
cloth or cotton and mixed with powdered kieselguhr and filtered until
clear. A filter of paper pulp mixed with some kieselguhr has been found
to be efficient. The clear filtered extract is passed through a bacteria-
proof filter of infusorial earth.
Preparation of defibrinated blood for centrifugalizing. — To
each 100 c. c. of the cool defibrinated blood add i c. c. of the sterile
bean extract and stir to secure a uniform mixture. Allow the mixture
to stand until agglutination is clearly evident. This can be determined
by examining a small amount in a glass or tube. Agglutination is usually
apparent within five minutes after adding the bean extract. There should
then be added i gm. of finely powdered sodium chlorid. The salt is
stirred in until dissolved, and the mixture of defibrinated blood, bean
extract, and salt is allowed to stand for about 15 minutes.
36290°— 16 4
336
Journal of Agricultural Research
Vol. VI, No. 9
CentripugaIvIzing. — The defibrinated blood mixture is placed in suit-
able containers, preferably some what elongated, and rotated in a centrifuge
for 15 minutes at a speed sufficient to produce in the cups a precipitating
force equal to approximately i ,700 times gravity. At the end of this
period the serum may be poured from the cups into suitable containers.
Heating the serum. — The clear serum obtained by centrifugalizing is
placed in a container v^^hich is surrounded by a jacket of v^^ater. The
temperature of the water in the outer jacket at the beginning of the
heating should not exceed 63° C. The serum in the inner container
is slowly stirred during the heating process, the temperature of the
outer jacket being maintained between 6i° and 62°. A thermometer
should be kept constantly in the serum and care should be taken to
see that the temperature of the serum, once it has reached 60° C, does
not fall below that point and that it does not rise materially above it.^
Continuous heating for 30 minutes at 60° C. is required. Upon the com-
pletion of the heating, the serum should be rapidly cooled. After cooling,
I part of a 5 per cent solution of phenol should be added to 9 parts of
the serum.
Filtering the serum. — After the phenol has been added a slight
precipitate may at times form in the serum; therefore it is desirable to
allow several days to elapse between the addition of the phenol and the
final filtration through infusorial earth.
EXPERIMENTAL RESULTS
To illustrate the yield of clear serum obtained by the application of the
described method to the preparation of anti-hog-cholera serum, there
is given in Table I a statement of the yield of clear serum obtained from
three different lots of defibrinated immune blood and one lot of defibri-
nated hog-cholera virus.
Table I. — Yield of clear seruvi from defibrinated anti-hog-cholera serum and virus
under a precipitating force of 1,700 tim,es gravity applied for 12 minutes
Blood.
Bean
Sodium
extract
chlorid
added.
added.
Per cent.
Per cent.
None.
None.
None.
Serum
yield.
Hog-cholera serum from defibrinated immune blood 3895 .
Do
Do
Do
Serum from defibrinated immvme blood 3866 and 2165
Serum from defibrinated immune blood 2166
Serum from defibrinated hog-cholera virus 377 and 379. . .
Per cent.
47J4
49
70
70
74
70
Table II gives the results of potency tests of one lot of serum prepared
by use of the bean and sodium chlorid mixture. As will be seen, a test
was made of the whole defibrinated blood, of the clear serum separated
1 Thermometers used should be standardized, and the temperature of the serum should not be allowed
to exceed 60.5° C.
May 29. 1916 Clear and Sterilized Anti-Hog-Cholera Serum
337
from such defibrinated blood by the use of bean extract and sodium
chlorid, and of the cell residues from which the clear serum was removed.
In preparing the cells for injection they were taken up in distilled water
and made to a volume' corresponding to the volume of defibrinated
blood from which they were derived. Thus hog 2149 received all of the
cell residue from 200 c. c. of defibrinated blood and hog 2150 received
all of the cell material from 100 c. c. of defibrinated blood. The serum
which was obtained from the defibrinated blood was used to inoculate
hogs 2155 to 2158, inclusive.
Table II. — Test of serurn separated by use of bean extract and sodiujn chlorid in igido-
Hog
No.
Weight.
Date in-
oculated.
Protective material
injected.
Quanti-
ty of pro-
tective
material
injected.
Quanti-
ty of
virus.
Results.
Date
died.
2143
Pounds.
70
65
70
6s
60
65
50
S5
45
50
Mar. 24
...do
...do. .. .
...do.. ..
Phenolized defi-
brinated blood
3895-
Phenolized defi-
brinated blood
.^895 (unwashed).
Cells from defibri-
nated blood 3895.
do
C.c.
20
10
200
100
16
16
8
8
C.c.
2
2
2
2
2
2
2
2
2
2
Remained normal
throughout test.
do
2149
Injured in fighting
Mar. 27; off feed
Mar. 28 to Apr. 4.
Very slight hemor-
rhagic lesions.
Went off feed Mar.
27; very sick Mar.
3 to Apr. II. Ex-
tensive hemorrhagic
lesions.
Remained normal
throughout test.
do
Apr. 4
Apr. II
21SS
...do.. ..
...do.. ..
Clear serum from
defibrinated
blood 3895,
heated.
do
...do.. ..
.... do
.. . do
2158
...do.. ..
do
do..
...do....
...do.. ..
Control
Well-marked lesions
of hog cholera on
post-mortem exam-
ination.
Extensive lesions of
hog cholera on post-
mortem examina-
tion.
Apr. 12
Apr. 11
do
" No inflammation or swelling at point of injection on any pigs in this test. Thriftiness of pigs remain-
ing normal not impaired.
From the fact that both of the pigs injected with the cell material
contracted hog cholera and died it seems clear that, in this experiment
at least, the amount of antibodies left behind with the cells was negligible.
The bean-extract-sodium-chlorid method of separating the corpuscles
from defibrinated hogs' blood has been applied repeatedly in these labora-
tories and always with success. There seems to be no reason why the
process should not be entirely satisfactory for use in the practical pro-
duction of anti-hog-cholera serum. There appears to be little or no loss
in antibodies; the serum secured is generally clear; and it may be re-
moved from the agglutinated cells easily by pouring from the cups.
The method also would seem to tend toward a certain concentration of
338 Journal of Agricultural Research voi. \^, N0.9
the antibodies of the blood, and it is also to be recommended on account
of the fact that it results in a large yield of serum.
The fact that this serum may be heated for half an hour at 60° C.
without noticeable impairment of its potency is of much practical
importance because there is thus afforded a ready means for safeguard-
ing it against infection with the virus of the foot-and-mouth disease.
Anyone contemplating the practical application of the process is
urged, at the beginning at least, to follow the method described herein,
and to use only the common white navy bean for preparing the bean
extract. It is hoped that the method will soon be adopted on a large
scale by commercial producers of serum.
ADDITIONAL COPIES
OF THIS PUBLICATION MAY BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
AVASHINGTON, D. C.
AT
15 CENTS PER COPY
SUBSCRIPTION Price, £3.00 Pee Year
V
Vol. VI JUNK 5, 1916 No. lO
JOURNAL OF
AGRIGUIvTURAL
RESEARCH
CONTENTS
Page
Silver-Scurf of the Irish Potato Caused by Spondylocladium
atrovirens --------- 339
EUGENE S. SCHDLTZ
Woolly Pear Aphis - - - - - - - '- 351
A. C. BAKER and W. M. DAVIDSON
Pathological Histology of Strawberries Affected by Species
of Botrytis and Rhizopus - -- - - - 361
NEH E. STEVENS
Life Histories and Methods of Rearing H^essian-Fly Para-
sites - - - - -- -.J,. 367
C. M. PACKARD
DEPARTMENT OF AGRIGUITURE
AX^SHTNGTON, B.C.
WASHINQTON : GOVERNMrNT PRINTING OFFICE : 181*
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL
i/:?/«6> Agriailtural Experiment
P/i^sfologist avd Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Cf'i^f, Office of flxperiment Stations
CHARLES L. MARLATT
Assistant Chief Bureau of Entomology
Biolng-s.
S:citior:
H. P. ARMSBY
D:
ute of Animal Nutrilion, Tin
State Callegr^
E. M. FREEMAN
Botanist, Pijni Pathologist, and Assistajit
Dean, .■^gyiculiural Expi ri>ne>it Station of
All correspondence regarding articles from tlie Department of Agriculture
should be addressed to Karl F. Kellerman, " ' Agricultural Research.
Wa;hh:v'.\:^. D. C. ' ' ; ■ ,
All lence regarding articles from Experiment Stations should be
addressea :o Kiiymond Pearl, Journal of Agricultural Research, Orono, Maine.
JOMALOFAGEICDITIALISEARCB
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, June 5, 1916 No. 10
SILVER-SCURF OF THE IRISH POTATO CAUSED BY
SPONDYLOCLADIUM ATROVIRENS
By Eugene S. Schultz,^
Expert in Potato Investigations, Cotton and Truck Disease Investigations,
Bureau of Plant Industry
INTRODUCTION
Silver-scurf of the Irish potato (Solanum tuberosum) , caused by Spon-
dylocladium atrovirens, has been known in Europe since 1871, when it
was discovered by Harz (6) on new potatoes in Vienna; but there is no
record of its appearance in this country until mentioned by Clinton (4) in
1908. Notwithstanding its comparatively recent discovery, its general
distribution in the eastern United States was shown by Melhus (7), 1913,
who also raised the question as to its importance as a new potato disease
in America, while its appearance in the Northwest was first reported in
1914 by Bailey (2) and later, in 1915, by O'Gara (8).
Reports of studies made by former investigators contain contradictory
assertions, especially on the effect of this organism upon the host. It is
evident, therefore, that further study of the symptoms, manner of infec-
tion, and physiology of the organism is desirable in order to understand
more fully the significance of this disease, which has already become
widely distributed in this country.
STUDIES OF THE FUNGUS
MORPHOLOGY
Spondylocladium atrovirens, one of the black molds, is classified accord-
ing to Saccardo (9, p. 483) in the Fungi Imperfecti under the Dematieae.
The genus Spondylocladium is characterized by its dark multiseptate
conidiophores, which bear the many-celled conidia pleurogenously in the
form of whorls.
Conidiophore and conidia formation can be studied either in hanging-
drop or agar cultures. When the organism was cultured on agar plates
1 The sincere thanks of the author are due to Dr. I.E. Melhus, Bureau of Plant Industry, for many help-
ful suggestions during the progress of this study and the preparation of the manuscript, and also to Prof.
L. R. Jones, of the University of Wisconsin, in whose laboratory a part of the work was conducted.
Journal of Agricultural Research, Vol. VI, No. lo
Dept. of Agriculture, Washington, D. C. June 5, 1916
dy G— 81
(339)
340 Journal of Agricultural Research voi. vi. no. io
held at room temperature, conidiophores and conidia appeared in lo to
12 days, which indicates that 5. atrovirens is one of the slow-growing
fungi.
The conidia are formed first either at the apex or the distal end of the
intermediate cells. Under certain apparently abnormal conditions,
however, they appear at the ends of what seem to be ordinary branches
of the mycelium, but in that case the character of the normal conidio-
phore is absent. The lowest whorls of conidia are borne about halfway
between the base and the apex of the conidiophores, and the conidia are
attached at the broad end (PI. XLVI, fig. 2).
Germination of the conidia takes place by means of germ tubes. These
are produced from either pole, generally from the distal or pointed end,
as well as from any cell of the conidium, as observed by previous investi-
gators. Germination in water occurs within 24 to 40 hours; and in a
few days the somewhat hyalin, knoblike protrusion, which is character-
istic of the early stages of germ formation, develops a multiseptate,
branched mycelium which is of a much lighter color than the conidio-
phores, conidia, or portions of the old mycelium. This is very hyalin
and continues so up to the time of conidiophore formation, at which
time dark-brown, thickened cells are formed in different parts, and from
these specialized cells are produced the many-septate, dark-brown coni-
diophores, which attain a length of 5 mm. and are perceptibly wider
than the surrounding mycelium (PI. XLVI, fig. i).
Because of the wide variation found in the size of the spores, Appel and
Clinton (4, p. 359) suggested the possibility of there being two species
of the fungus — that is, a large-spore and a small-spore species. Several
series of 18 measurements were made by the writer on conidia taken from
tubers imported from Germany and tubers from various parts of the
United States. A wide variation in dimensions occurred in the conidia
from all the various tubers used in the experiment. The conidia taken
direct from the surface of the tuber from Germany varied from 22 to 42/i
(mostly 30 to 40) in length, 6 to 12/i (mostly 6 to 8) in width at greatest
diameter, and were 4 to 8 (mostly 5 to 6) septate; conidia taken from the
progeny of tubers from Maine grown in Washington, D. C, varied from
30.4 to 56.2^1 (mostly 30 to 40) in length, 7.6 to 9.5JU (mostly 7.6 to 8.5)
in width, and were from 4 to 7 septate; while the conidia taken from tubers
from Rhode Island, West Virginia, Washington, D. C, Oregon, Wash-
ington, and Wisconsin averaged 32.6 to 40// in length, 7.5 to 8.5jU in
width, and were 5- to 7-septate.
In order to study more fully the variation of spore dimensions, several
series of measurements were made on conidia produced from a single
spore strain. ♦ The difference in dimensions obtained in this case ranged
from 18 to 64/i (mostly 30,4 to 40) in length, and 7 to S.i/x in width,
and 5 to 6 septa.
Junes, 1916 Siher-Scurj of Irish Potato 341
From this it is apparent that, even though considerable variation in
spore dimensions occurred on infected tubers from different locaHties,
nevertheless an even greater variation resulted in the case of spores from
a single spore strain. This shows that normally a wide variation exists,
and consequently it does not appear necessary to form small-spore and
large-spore species.
REACTION OF THE FUNGUS TO LIGHT
In order to secure a better knowledge of the relation of 5. atrovirens to
its environment so that its life history might be better understood,
experiments on some of the physiological characteristics of this organism
were conducted.
The reaction to Hght is of special interest in connection with the effect
of storage conditions upon the development of the fungus on potatoes.
In this study the writer used the plate-dilution method, the conidia
being sufficiently diluted on Lima-bean agar plates to be observed
individually. Immediately after the plates were poured, each was
wrapped in carbon paper, the entire dish being covered except an aperture
from I to 2 cm. in diameter at the side, and the plates were then arranged
with the apertures facing the light from the window.
The plates were examined at the end of three days and it was found
that the mycelial branches developed on the side of the hyphas farthest
away from the window and that the majority of these grew in the opposite
direction from the source of the light. The position of germ-tube forma-
tion does not appear to be influenced by the light, germination sometimes
taking place from the side closest to the source of light; but as soon as the
germ tube receives the heliotropic stimulus — that is, when it is a few
millimeters long — it invariably turns away from the light, and subsequent
mycelial development is formed on the side of the conidium farthest
from the source of light. Instead of appearing at the center of the colony,
therefore, the conidia are found at the margin exposed to the light, and at
the end of 5 to 10 days the entire colony appears as if a gentle breeze had
blown the hypha in one general direction away from the light (PI. XLVI,
fig. 3). These results also confirm Eichinger's (5) observations.
The reaction of this fungus to light in culture media demonstrated
that it is negatively heliotropic. In view of the fact that infection
of the tubers in the field takes place in the dark, negative heliotropism
here does not obtain. In order to determine whether this heliotropic
property favored tuber infection, artificial inoculations were made on
tubers in the light. In this case no perceptible difference occurred,
since infection appeared on all parts of the tubers alike.
REACTION OF THE FUNGUS TO MOISTURE
Like most fungi, S. atrovirens requires considerable moisture for
development; but, owing to the absence of accurate instruments for
342 Journal of Agricultural Research voi. \a, no. w
measuring the degree of moisture, only approximate data regarding
moisture reaction can be given. It was noted in field studies that a
higher percentage of infection occurred in the lower and more moist
sections of the field than in the higher areas, and that in laboratory
infection experiments the fungus develops best when the surface of the
tuber is kept moist but not supersaturated. By placing tubers sufficiently
near water so that a heavy film of moisture was constantly present,
it was found that sporulation was inhibited to a greater degree on the
side of the tuber near the water than on the opposite side, which indi-
cates that excess moisture may check the growth of the fungus.
Although the fungus prefers moisture for growth, it can withstand
drying without the entire loss of its virility. This was shown by the
fact that transfers from agar cultures i6 months old continued to grow,
although only a small percentage of the conidia germinated. Notwith-
standing the fact that these cultures had been kept at room tempera-
ture and were dried to such an extent that simply a dry, brittle mass of
media and fungus remained, both viable conidia and mycelium were found.
REACTION OF THE FUNGUS TO TEMPERATURE
Conidia in corn meal and oat agar and in water and naturally infected
and artificially inoculated potato tubers were used in studies to determine
the effect of temperature on 5. atrovirens. In the case of media spore-
dilution plates were prepared, the spores being sufficiently far apart so
that individual colonies were retained. The same dilution was used on
each plate and all were inoculated at temperatures ranging from 2° to
31° C. The water cultures were used in making hanging-drop prepara-
tions on Van Tiegham cells and in small Petri dishes, the spore suspensions
in this case also being made in such manner that some of the spores re-
mained on the surface, although germination occurred to a slight extent
also beneath the surface. The naturally infected and artificially inocu-
lated tubers were placed in pint bottles containing some pebbles and a
few cubic centimeters of water, with a piece of cheesecloth extending from
the contents of the bottle to its mouth, thus forming a moist chamber.
These bottles were incubated in the same way as the media cultures.
In the eight series of Petri-dish cultures microscopic germination was
noted at 3°, 4°, and 5° C, but no macroscopic colonies developed; at
temperatures ranging from 6° to 28° macroscopic colonies were obtained,
21° to 27° being the optimum for abundance of growth; while at 30° or
31° no macroscopic growth was apparent (Pi. XLVII). These tem-
perature limits for growth were confirmed by the water cultures, which
were used as checks on the media cultures subjected to the highest and
the lowest temperatures. In the case of three series of these water cul-
tures which were subjected to a temperature of from —5° to —10° C.
for four days and then brought to room temperature, 80 per cent of the
June 5, 1916 Silver-Scurf of Irish Potato 343
conidia germinated within 48 hours, and pieces of the mycelium in the
cultures also showed growth. Agar culture and cultures on sweet-clover
stems subjected to the same temperature also remained viable, as indi-
cated by subsequent transfers, hanging-drop cultures showing that both
conidia and mycelium retained their vitality.
In the test with naturally infected and artificially inoculated potatoes
sporulation occurred on the former at temperatures ranging from 6° to
27° and on the latter at a range of from 1 2° to 27° C. In cultures on agar
media and sweet-clover stems subjected to 35° and 50° further growth
was inhibited at the former temperature, but the fungus remained alive
after two weeks' exposure, while it was killed when subjected to 50° for
three days.
REACTION OF THE FUNGUS TO MEDIA
Since 5. atromrcns is a relatively slow-growing organism, tests were
made with media of different grades of acidity with a view of facilitating
growth in culture. The media used for this purpose were synthetic,
Lima-bean, string-bean, oat, potato, corn-meal, and beef agar, all of
which varied in reaction from -f 15 to — 15 Fuller's scale.
Two plates each of these media equally diluted with conidia from the
same culture were poured, and all were incubated at room temperature.
Examinations of the colony development, including nature and extent of
growth and sporulation, were made at 4-, 6-, and 12-day intervals and
showed that 5. atrovirens developed slightly faster on potato and Lima-
bean agar than on string-bean, corn-meal, or oat agar; that growth was
much retarded on beef agar; that mycelial development was very de
cidedly inhibited on S3^nthetic agar; that sporulation occurred slightly
sooner on oat agar than on the other agars; and that the hyphae on
fruiting remained lighter in color on Lima-bean and beef agars than on
other agars.
The optimum reaction appeared to depend largely on the kind of
medium. On potato agar no perceptible difference in growth appeared
between -f 10 and — 10, but mycelial development was much retarded at
-1- 15. On corn-meal agar only -j- 1, o, — i, —3, —5, and — 15 reactions
v/ere run, because of the fact that hydrolysis took place when there was a
higher degree of acidity. In this series -f i reaction was the optimum
for growth, and in this case the mycelium became dark earlier than was
the case in the minus reactions, owing possibly to the hydrolytic action
of the acid on the media. On Lima-bean agar with -f 5 to — 3 reactions
the apparent growth of the fungus was not much changed, but with 5 to
10 and —3 to —10 reactions mycelial growth was perceptibly retarded.
On beef agar optimum reactions ranged from o to +1, very little differ-
ence appeared in the colonies at -f 3 to — 3, growth was gradually retarded
at 5 to 15, and no colonies were macroscopically visible at the end of 10
days on reactions ranging from — 5 to — 15.
344 Journal of Agricultural Research voi. vi, no. io
Besides this test of different reactions of the medium, a series of
nutrition tests was conducted, a full nutrient agar, including carbon,
nitrogen, oxygen, hydrogen, potassium, phosphorus, magnesium, sulphur,
and iron, being used. With one exception each set of the media con-
tained one element less than the full nutrient culture; in other words, the
experiment was arranged as follows: (i) Check containing water agar,
(2) full nutrient, (3) full nutrient minus nitrogen, (4) full nutrient minus
potassium, (5) full nutrient minus phosphorus, (6) full nutrient minus
magnesium, (7) full nutrient minus sulphur, (8) full nutrient minus iron,
(9) full nutrient minus carbon, (10) full nutrient minus all minerals..
Two plates of each kind of agar were inoculated with conidia and two
with mycelium from the same culture of S. atrovirens, and all were incu-
bated in the laboratory at room temperature.
Examinations at the end of 15 and 20 days indicated that sporulation
occurred only on the plates from which sugar was omitted — that is, Nos.
I and 9 — the colonies on these plates being of a light color and spreading
character and from 1.5 to 2.5 cm. in diameter and that no sporulation
occurred on the plates from which sugar had been omitted, the mycelium
in these being dark and densely compacted and only 0.75 to 1.25 cm. in
diameter.
This preliminary study of the reactions of media on 5. atrovirens indi-
cates that neutral or slightly acid reactions are more favorable for the
growth of this fungus; that the kind of medium determines the effect of
higher reactions on this organism as shown by the alkaline reactions of
beef agar compared with the same reactions of potato or the other agars;
that compounds in one kind of medium may be formed which are seem-
ingly toxic, whereas in a different kind of medium the same adjustment
produces no such inhibitory effects; and that the presence of 5 per cent
of cane sugar in a nutrient agar inhibited sporulation, but induced dark,
heavy, compact mycelial growth, while the absence of sugar caused
sporulation and a more spreading mycelial development.
HISTOLOGY
Studies were made to determine the relation of 5. atrovirens to the
potato. Both normal and affected material from the eye end of Irish
Cobbler, Green Mountain, and Minnesota Triumph tubers badly infected
normally and artificially was taken from the center and from the margin,
that from the latter with and without lenticels or eyes. This material
was embedded, sectioned, and stained according to ordinary cytological
methods. From these studies it was evident that the mycelium may
enter the tuber through the lenticels or between the lenticels through
the epidermis.
After the fungus gains entrance the hyphae invariably form within the
cells, where they appear as a single branch of the mycelium; or they
June s. 1916 Siher-Scurf of Irish Potato 345
may shorten and thicken to form a short and many-celled mass of
hyphae, from which the conidiophores subsequently arise. In severe
cases of infection the cells appear to be disintegrated by the invasion to
such an extent that only two or three instead of six or more cork layers
remain above the living parenchyma. In experiments with potato roots
grown under sterile conditions and inoculated with conidia and mycelium
of the fungus, the mycelium grew on the surface, but did not penetrate
the parenchyma, which indicates that the roots are less subject to infec-
tion than the tubers.
So far as the author has been able to determine, the fungus hyphse
confine their activity to the corky layers. In no case has it been found
in the living parenchyma. This superficial infection causes a loosening
of the corky and epidermal cell layers, so that these subsequently slough
off. In this manner transpiration may proceed with greater facility and
thus affect the parenchyma layers.
That 5. atrovirens prefers this relatively heavy corky layer is further
apparent from the fact that it grows very sparingly on the cut surface
of the tubers where the loosened surface cells are invaded. Further-
more, its very limited presence on roots, stems, and stolons also indicates
that it prefers the heavier, corky layers of the potato tuber.
EFFECTS OF THE FUNGUS ON THE HOST
The progress of the disease after tuber infection may be divided into
two stages, the early and the late. In the former the infected areas are
light-brown and have a glazed appearance, the latter characteristic
becoming especially pronounced when the infected surface is moistened.
Sometimes the margins of these areas are slightly fimbricated. The
discoloration, which is found on newly infected tubers at harv^est time,
is often so inconspicuous as to pass unnoticed, even on close examina-
tion, unless the tubers are washed. When infected tubers are placed in
moist chambers, the brownish areas become olive-colored, owing to the
formation of conidiophores and conidia. The late stage is characterized
by the shrinking and shriveling of the diseased areas and sloughing off of
the epidermis and may be subdivided into two stages : The spot or patch
infection (PI. XLV, fig. 2) and general infection (PI. XLV, fig. i). In
the former slightly sunken isolated areas on the surface show the shrivel-
ing, and late in the storage season these areas become shriveled and
sunken.
In the case of general infection the entire surface is covered with
infected areas and the epidermal and corky layers may shrink to such
an extent that distinct folds or ridges appear. In the red-skinned
varieties the color is completely destroyed. This again largely only
mars the appearance and not their food value, but still they must be
sold at a sacrifice. Potatoes stored under moisture and temperature
346 Journal of Agricultural Research voi. vi, no. 10
conditions favorable to sporulation often become so badly infected that
they become a dull-black, the tubers having the appearance of having
been dusted with soot. Several such bins were observed in Maine in
May and June, 1914.
In case of slight infection in the field the infected areas are often
found in isolated spots close to the stem end of the tuber. This was the
case in practically every infected tuber harvested from the silver-scurf
experimental plot at Caribou, Me., in the fall of 1914 and coincides with
the observations of Appel and Laubert (i). While no reasons for this
phenomenon are given by these investigators, from experiments and
obser\^ations so far made it appears that infection is brought about
through contact of the stem end of the young tuber with the infected
mother tuber (PI. XLVIII). This is indicated by the fact that in many
cases where there was but slight contact only small areas about the
point of the stolon attachment showed infection, while in the case of
extensive contact infection was more widespread. It is further indi-
cated by the fact that only one or two tubers closest to the mother tuber
showed infection in counts made when the crop was about three-fourths
grown, while in counts made later, after the conidia had become gener-
ally distributed, a large percentage of the tubers were infected.
Although infection appears to take place through the stem end, both
stem ends and eye ends are subject to infection, general infection of
both resulting from artificial inoculations.
In view of the fact that investigators like Bohutinsky (3) have attrib-
uted to 5. airovircns foliage symptoms such as leaf roll, mosaic, etc.,
inoculations upon stems, stolons, and roots of the potato plant were
made, both under field and greenhouse conditions. Two distinct pro-
cedures were followed: In one set of experiments viable spores were
sprayed upon the stems, stolons, and roots; in the other virile mycelium
was inserted into the inoculated portions. Checks were also run.
Experiments in this order were run during 1914 and 191 5, and in every
case the inoculated plants behaved like the checks — viz, no perceptible
infection occurred — showing again the inability of this organism to
invade the vine tissues of the host.
METHODS OF DISSEMINATION
The fungus lives over by means of the mycelium, conidia, and sclerotia
within the infected areas, so that under favorable conditions of moisture
and temperature sporulation occurs and infection may spread even in
storage. Not only do the infected tubers carry the disease to new sec-
tions, but they may carry it over from one season to another in the soil
and in this way infect the new crop. This was the case in the author's
field studies in Maine, viable conidia being found on the surface of
June 5, 1916 Silver-Scurf of Irish Potato 347
mother tubers taken on August 2, 1914, the date of the last examination,
from an oat field at Houlton, in which they undoubtedly over-wintered
in the soil. Many of these volunteer plants occurred in fields in which
rotation had not been practiced, the deep snows which covered the
ground the previous winter having protected the tubers.
Whether the fungus may live over in the soil from which the tuber
host has been removed is not yet known, but that it may do so is not
improbable, in view of what occurs in the case of fungi having a similar
life history. Investigations to determine this point are now in progress.
Several series of experiments were undertaken to ascertain how readily
5. atrovirens spreads from infected to healthy tubers and whether infec-
tion in this way might occur during the entire storage season. Inverted
bell jars were used in these experiments to secure moist chambers which
would hold a sufficient number of tubers for a satisfactory test and at the
same time retain uniform moisture conditions. A wire rack of X-inch
mesh was placed in each jar to support the potatoes and to prevent con-
tact w^ith the water in the jars, the inside of each jar was lined with blot-
ting paper to conserve the moisture and prevent the entrance of excessive
light, and the mouth was covered with window glass. Four varieties of
potatoes were used: Rural New Yorker, Green Mountain, Irish Cobbler,
and Bliss Triumph. A spore suspension of conidia which had been
grown in pure culture on sweet-clover stems for four weeks was sprayed
on the tubers with an atomizer, and for several days thereafter water
was sprayed into the jars with the atomizer to keep the air saturated.
A similar lot of healthy tubers was arranged as a check.
The first series was begun at Houlton, Me., on March 26, 1914; and
within three weeks the entire surface of the inoculated tubers was covered
with dark-brown conidiophores and conidia, while the checks were free
from infection. Additional tests were made at Caribou, Me., on July 20,
1 914; Washington, D. C, in December, 19 14; Madison, Wis., on March 25,
1915; and Presque Isle, Me., on August 2, 1915. In each case infection
occurred wdthin three weeks after inoculation.
Similar infection experiments were conducted upon young tubers just
harvested, as well as upon tubers still attached to the vines. In case of
the tubers attached to the vines the soil was removed and a spore sus-
pension was applied with an atomizer, whereupon the tubers were again
covered with earth. Checks also were made. In each of these tests
infection appeared upon tubers varying in diameter from i and 2 cm.
to full-grown tubers (PI. XLVII). Checks showed no infection.
From these results it is apparent that infection from S. atrovirens may
take place at any stage in the development of the tubers and at any time
throughout the storage season.
348
Jourrial of Agricultural Research
\'ol. VI, No. 10
METHODS OF CONTROL
Melhus (7) found in laboratory experiments that neither double
strength of mercuric chlorid (1:500) nor formalin applied for longer
than the ordinary periods would completely inhibit the development of
5. atrovirens on the potato and that both injured the tubers to such an
extent that germination was decidedly inhibited. He also found that
in many cases sporulation was inhibited on the surface of infected tubers
treated with solutions of mercuric chlorid heated by a method devised
by him for heating the solution for brief periods at temperatures near
the thermal death point of protoplasm.
In view of these results, field tests were conducted during 191 4 and
1 91 5, both in Maine and at Norfolk, Va. Infected tubers were treated
in double strength and heated solutions of mercuric chlorid. In Maine
the treated tubers were planted on virgin soil.
As noted in Table I, the temperature fluctuated slightly, owing to the
lower temperature of the tubers than that of the solution in which they
were immersed. This table indicates that there was a decrease in the
percentage of infected progeny in the treated rows as compared with the
check. However, in no case was there a complete control of the infection.
Similar tests in 191 5 also indicated that even though silver-scurf may be
inhibited to some extent; nevertheless, no treatment served as a complete
control.
Table I.
-Effect of warm sohition of mercuric chlorid on silver-scurf of the Irish
potato
Row
No.
Strength
of solu-
tion.
Tem-
perature
of solu-
tion.
Time
of im-
mer-
sion.
Num-
ber of
hills.
Num-
ber of
hills
in-
fected.
Per
centage
of hills
in-
fected.
Aver-
age per-
centage
infected
hills in
rows
14. I5>
and 16.
Weight
of
healthy
tubers.
Weight
of in-
fected
tubers.
Percent-
age of
infected
tubers.
Aver-
age per-
centage
of in-
fected
tubers
in rows
14. IS.
and 16.
I^ . . .
Per ct.
°C.
Con-
trol.
47-52
47-48
45-49
Min.
5
5
10
78
63
93
75
40
29
6
8
51. 28
46.03
6.45
10. 66
21.04
Pounds.
85
84
102
75
Pounds.
II. 25
10
2-5
II. 67
10. 13
.98
3.21
14. ..
15...
16...
0. 2
. 2
. 2
4. 91
In October, 1914, four pecks of tubers infected with S. atrovirens were
subjected to a i to 1,000 solution of mercuric chlorid ranging from 45°
to 53° C. for four minutes with a view of ascertaining the effect of treat-
ing infected tubers with that solution before storing. After treat-
ment the tubers were placed in new muslin peck sacks and part of the
lot stored at Caribou and part at Washington. At the same time sepa-
rate lots of untreated infected and clean tubers were stored. On exami-
Junes, i9i6 Silver-Scurf of Irish Potato 349
nation of these lots in June, 191 5, the fungus was found fruiting on both
treated and untreated infected tubers, but no infection was found on
the untreated clean tubers.
As the treatments described do not absolutely control silver-scurf and
as clean, tubers only escaped infection, it is evident that disease-free seed
should be selected in the fall and should be kept from contact with
infected tubers in storage. Moreover, in view of the inhibitory effect of
very low temperatures on the development of the fungus, the tubers
should be stored at the lowest temperature permissible.
SUMMARY
A study of silver-scurf of the Irish potato, caused by Spondylocladium
atrovirens Harz, shows that, notwithstanding the wide range in spore
dimensions, which led certain investigators to believe there might be a
large-spore and a small-spore species in this country, there is but one
species, as proved by the fact that conidia ranging from 18 to 64/^ were
produced by a single spore culture.
5. atrovirens is negatively heliotropic. This, however, does not mate-
rially influence tuber infection in nature.
Severe drying of the conidia and mycelium in agar culture at room
temperature does not kill the fungus.
S. atrovirens withstands a wide range of temperature. Its growth is
inhibited at 2° to 3° C, but it is not killed at — 10°. Its optimum temper-
ature is 21° to 27°, maximum 30° C.
Optimum reaction to media varies with the kind used, neutral to
slightly acid reactions being most favorable to the development of the
fungus. Five per cent of cane sugar in nutrient agar inhibited sporula-
tion.
The fungus enters the tuber through the lenticels or the epidermal
layers between the lenticels. The mycelium invades and disorganizes
the epidermal and corky layers, leaving in bad cases only one or two
instead of six or more layers, thus apparently accelerating transpiration.
The disease may be carried from place to place by infected tubers, in
which it lives over from one season to another, or to the succeeding crop
by the infected tubers which remain in the field over the winter.
Under favorable moisture and temperature conditions potatoes may
become infected throughout the entire storage season. Both old and
young tubers are subject to infection.
Inoculations on living stems, stolons, and roots in the field and labora-
tory experiments produced no infection.
Warm solutions of mercuric chlorid have a more toxic effect on 5.
atrovirens than cold solutions.
350 Journal of Agricultural Research voi. vi, no. lo
LITERATURE CITED
(i) Appel, Otto, and Laubert, R.
1907. Die Konidienform und die pathologische Bedeutung des KartofFelpilzes
Phellomyces sclerotiophorus Frank. In Arb. K. Biol. Anst. Land- u.
Forstw., Bd. 5, Heft 7, p. 435-441, pL 11.
(2) Bailey, F. D.
1914. Notes on potato diseases from the northwest. In Ph3rtopathology, v. 4,
no. 4, p. 321-322, pi. 10.
(3) BOHUTINSKY, G.
1910. Beitrage zur Erforschung der Blattrollkrankheit. In Ztschr. Landw.
Versuchsw. Oesterr., Jahrg. 13, Heft 7, p. 607-633, 3 fig.
(4) Clinton, G. P.
1908. Scurf, Spondylocladium atrovirens Harz. In Conn. Agr. Exp. Sta.
Ann. Rpt. 31/32, [i907]/o8, p. 357-359- Quotes (p. 359) statement by
Appel.
(5) ElCHINGER, A.
1909. Zur Kenntnis einiger Schalenpilze der Kartoffel. In Ann. Mycol.,
V. 7, no. 4, p. 356-364, 3 fig.
(6) Harz, C. O.
187 1. Einige neue Hyphomyceten Berlin's und Wien's nebst Beitragen zur
Systematik derselben. In Bui. Soc. Imp. Nat. Moscou, t. 44, pt. i,
p. 129.
(7) Melhus, I. E.
1913. Silver scurf, a disease of the potato. In U. S. Dept. Agr, Bur. Plant.
Indus. Circ. 127, p. 15-24, 4fig-
(8) O'Gara, p. J.
1915. Occurrence of silver scurf of potatoes in the Salt Lake Valley, Utah.
In Science, n. s. v. 41, no. 1047, p. 131-132.
(9) Saccardo, p. a.
1886. Sylloge Fungorum ... v. 4. Patavii.
PLATE XLV
Fig. I. — Potato tubers showing shriveling and a silvery appearance caused by
Spondylocladium atrovirens.
Fig. 2. — Tuber nattu"ally infected by 5. atrovirens, showing the segregated area
type of infection, a condition developing in some cases later in the storage season.
Fig. 3. — Immature potato tuber artificially inoculated with conidia of 5. atrovirens,
July, 1913, at Houlton, Me. Infected area covered with dark-brown tufts of
conidiophores and conidia. Infection was effected in a moist chamber at room
temperature.
Silver-Scurf of Irish Potato
Plate XLV
Journal of Agricultural Research
Vol. VI, No. 10
Silver-Scurf of Irish Potato
Plate XLVI
-. " ^ -f- .
•*> »•*..•,■ *'- , ■■*/':• -^■%- ^-'.z
• *-^'-€ ^'
Journal of Agricultural Research
Vol. VI, No. 10
PLATE XLVI
Fig. I. — Photomicrograph of Spofidylocladimti atrovirens on coTn-mesil agar, showing
method of development of conidiophores and conidia in the early stages.
Fig. 2. — Photomicrograph of 5. atrovirens in hanging-drop culture, showing develop-
ment of conidiophore and conidia in mature stages.
Fig. 3. — Negative heliotropism of S. atrovirens on com-meal agar exposed on one
side to daylight from April 8 to April 24, 191 5, in laboratory at room temperature.
PLATE XLVII
Effect of temperature upon mycelial development of Spondylocladium atrovirens in
pure culture on corn-meal agar at end of four weeks.
Pet
71 Dish No.
Temperature (°C.)
Petri Disli No.
Temperature (°C.)
1
5
6
10
15
l6
^
-7,1
4 27
8
l8
Silver-Scurf of Irish Potato
Plate XLVII
(U
CD
..^^^:^''^"-.-
>■• ■■» -.>•' .-■ J ■"■■ • ■■ ■-. . •*
:i
,*. -i-^ .7
V . '■ .' '■: A ' '
lA
Journal of Agricultural Research
Vol. VI, No. 10
Silver-Scurf of Irish Potato
Plate XLVIII
Journal of Agricultural Research
Vol. VI, No. 10
PLATE XLVIII
Contact infection. A part of the new tubers becoming infected with Spondylocla-
dium atrovirens by means of contact with the infected mother tuber. In this case it
is a distinctly stem-end infection. Harvested on September 19, 1915, at Presque
Isle, Me.
37767°— iG 2
WOOLLY PEAR APHIS ^
By A. C. Baker, Eniomological Assistant, and W. M. Davidson, Scientific Assistant,
Deciduous Fruit Insect Investigations, Bureau of Entomology
INTRODUCTION
For some years a species of Eriosoma has been known to attack pear
roots in California. It has, however, been considered to be the woolly
apple aphis, Eriosoma lanigerum Hausmann, since both in habit and in
structure the two species somewhat resemble each other. To the species
on the pear, which, after careful study, proves to be un described, the
name '^Eriosoma pyricola" is herein given, and a brief account of the
species is attempted.
HISTORY OF THE INSECT
Mr. Frank T. Swett is authority for the statement that the woolly pear
aphis has been in California for more than 20 years. Ten years ago he
says the species ruined about 2,000 French seedlings in one block, while
occasional apple seedlings, planted along with them, made normal
growth. Attention has frequently been called to the immunity of apple
seedlings planted close to infested pear seedlings in nurseries and orchards.
During September and October, 1897, Mr. Theodore Pergande received
specimens of a species of Eriosoma on pear roots from Prof. F. M. Web-
ster, of Wooster, Ohio. Through the kindness of Mr. Pergande we have
been able to examine these specimens, and they prove to be identical
with our California material. It is quite possible, therefore, that the
species may be present in other parts of the country, notably in Oregon.
It is noteworthy that the Ohio specimens were taken from roots of pear
stock received from France the preceding spring.
The species occurs over practically all the pear sections of northern and
central California, and in some regions is a very destructive pest. To
entomologists the extent of its presence has been known only for the last
three or four years, but reports from orchardists and field observers indi-
cate that it has been parasitic upon pear roots for a much longer period.
HABITS OF THE INSECT
The insect works entirely underground. The species that has been
found feeding on the aerial portions of Nelis, Easter Beurre, and other
pears is the woolly apple aphis, E. lanigerum. The woolly pear aphis
' What is probably the same species has been treated as a pear pest in California under the name Eriosoma
lanigera by Geo. P. Weldon. (The woolly aphis as a pear pest. In Mo. Bui. State Com. Hort. [Cal.], v. 4,
no. 9, p. 441-444, fig. 94-95. 1915 )
Journal of Agricultural Research, Vol. VI, No. 10
Dept. of Agriculture, Washington, D. C. June 5, 1916
ea . K— 33
(351)
352 Journal of Agricultural Research voi. vi, no. io
appears to attack the roots of all types of pears, and it is especially
injurious to the French wild stock so largely used in California as a
stock for the Bartlett. Quince roots are fed upon, but much less freely,
and the quince may be credited with a considerable degree of immunity.
The Kieffer stock is attacked, but it is possible that Japanese stock may
show immunity to a satisfactory degree. Observations to date indicate
that both these stocks are more resistant than that from France. It
should be said that the individual plants of the wild stock from France
vary greatly, and there appears to be among the plants some variation in
intrinsic vigor or in power to resist the woolly aphis. However, the
majority of the imported seedlings show no satisfactory evidence of a
power of resistance, and a different stock is very desirable.
The insect works especially upon the smaller fibrous rootlets and may
be encountered on any such rootlets within the topmost 3 feet of soil and
perhaps deeper. Infestations are usually heavier on the rootlets near
the trunk, but frequently the aphides are as abundant 10 or 12 feet from
the stem. In a badly infested orchard the soil on being overturned
may in places be found to be white with the wool and skins of the insects.
The aphides attack less frequently larger roots up to X inch in diameter
and sometimes settle on still larger roots or on the main stem where
abrasions have set up a callus growth. They often colonize the under-
ground portions of sucker growth, feeding on the succulent stalks. After
the insects have forsaken a rootlet, fungi sometimes appear and com-
plete its destruction.
This method of feeding upon the fibrous rootlets is somewhat analo-
gous to the habits of the grape phylloxera {Phylloxera vitijoliac Fitch)
on the resistant types of grapevines in that chiefly the smaller rootlets
are attacked. It is directly opposed to the habits of the woolly apple
aphis and of the grape phylloxera on nonresistant types of vines, for
both these insects feed upon the larger roots and cause the formation of
tuberlike lesions. The woolly pear aphis rarely forms any perceptible
lesions, but it destroys great numbers of young rootlets, especially in
late summer and autumn. In old trees this sometimes results in a
dwarfing of growth and in a generally unthrifty appearance and condi-
tion. The majority of old infested trees do not show evident injury
ascribable to the aphis, although it is presumable that they are suf-
fering to some extent. They remain thrifty on account of their intrinsic
vigor. In many instances where old trees were showing injury, extra
cultivation of the soil and better irrigation practice resulted in the
establishment of thrifty conditions, even though this method did not
appear to reduce the numbers of the aphis. The effect on the crop is
hard to estimate and can not be satisfactorily specified, but in general
it is such as may result from the diversion of the flow of sap in the tree.
Junes, i9i6 Woolly Pear Aphis 353
With trees under 4 years of age, conditions of injury are different.
Heavy infestation of a tree of weak vigor or resistance may result in the
death of the tree. Badly stunted growth and the early falling of foliage
are characteristic of aphis injury on young trees. Injury and death
are due to heavy summer and autumn infestations on the fibrous root-
lets and to the inability of the tree to replace the destroyed roots quickly
enough to afford plant food for the vegetative portion. Frequently the
trees are saved and relief comes from the production in the fall months
of a high percentage of migrants which leave behind them for the winter
only a small infestation of wingless individuals; and since the aphides
increase but slowly in spring, the tree is enabled to send forth new root-
lets without danger of having them rapidly destroyed. Sometimes
young trees in no wise stunted have been observed to cast their leaves
prematurely, and upon examination have been found to be heavily
infested with the aphis. It would appear from the absence of stunted
growth that these trees did not have, or were not adversely influenced by,
an infestation until their summer growth was about completed, and
that the simultaneous destruction of feeding rootlets cut off the flow of
sap suddenly. The fact that trees were stunted was an indication that
the injurious effects of feeding by the aphides were felt earlier in the
season.
In addition to trees noticeably stunted and others prematurely defo-
liated are found still others which show no external evidence of infesta-
tion and yet upon examination prove to be heavily infested. This
phenomenon is frequently noticeable among young trees or in nursery
rows, and hints at a power of resistance.
In orchards and districts where conditions favor large productions of
winged forms, or migrants, spring and early summer infestations are small,
denoting that few insects passed the winter on the roots. After the month
of June, however, such infestations multiply rapidly and become very large
by September, the month in which the fall migrants are produced in
greatest abundance. After September there remain small wingless colo-
nies which increase but little until the summer following. The winged
forms are produced in abundance on heavy dry clay soils which crack in
summer and autumn. Irrigated orchards produce them in smaller num-
bers than those that receive no moisture from ]\Iay to October. On loam,
silt, and light-clay soils the winged forms are much less abundantly pro-
duced. On such soils the infestation remains largely or wholly wingless
the year around, and the conditions are generally unfavorable to such
heavy infestations as occur on the heavy clays. The aphides appear to
lack freedom of movement, and frequently their colonies are unable to
increase perceptibly through summer. Occasionally the wingless infesta-
tions are severe the year round; where this is so, in the early part of the
year there is caused a considerable stunting of growth and more or less
354 Journal of Agricultural Research voi. vi. no. io
weakening, unless the trees can put out plenty of new rootlets to replace
those injured and destroyed. This condition has been noted especially
on light-clay soils where poor cultivation was employed.
SPREAD OF THE INSECT
In nurseries under favorable conditions the spread of the insect may be
rapid. A half-acre pear nursery examined on June 9, 191 5, failed to show
infestation, though the aphis was probably present. When visited four
months later, on October 16, it was found that more than half the trees
examined were infested, some quite heavily. In large orchards where the
soil is permeated throughout with rootlets the aphis doubtless is very easily
diffused through the soil. In young orchards conditions indicate that
not much spread takes place from tree to tree. Infested young orchards
generally point to the nursery as the source of infestation, but the possi-
bility of infestation through the winged forms, or migrants, must be con-
sidered. A knowledge of the full life cycle of the insect alone can clear
up this point.
BIOLOGY AND DESCRIPTION OF THE INSECT
The wingless individuals live chiefly on the small rootlets and less fre-
quently on roots and the underground portions of the sucker growth.
They are always somewhat elongate and are for the most part pale
yellowish red, but they may vary from a pale pink or yellow to deep red.
They are rather sparsely clothed with long, curling, woolly, or cottony
filaments, of which there are four or six on each segment. Toward the
end of each instar these filaments are longer than the body — often three
times as long. There is a sparse whitish powder on the body, more abun-
dant at the caudal end. The cornicles appear as dusky-rimmed pores.
The young are pale yellowish red and elongate.
The pupae develop on the same portions of the tree as the wingless
forms. They are very elongate in form and are clothed as are the wing-
less. The wing pads are inconspicuous and are white or light gray. As
a rule pupae on a rootlet develop almost simultaneously. The winged
forms issue together, after which the narrow, elongate, cast pupal skins
are conspicuous in little heaps, and are easily distinguishable from those
of the wingless forms.
In the Walnut Creek district pupae and winged migrants were collected
in appreciable numbers from August 25 to November 17, and as late as
December 22 a nymph was found. These forms were most abundant in
September, and this obser\^ation apparently holds true for other localities
in California. Wingless colonies collected at San Jose, Cal., on June 10
and thereafter, kept in Petri dishes with moist sand in a cellar, produced
pupae on July 20 and migrants from July 24 to August 7. This appeared
to be abnormally early in the year for the production of winged forms, and
Junes. i9i6 Woolly Pear Aphis 355
it may be that the environment and conditions hastened it. Under favor-
able conditions of soil the migrants were produced in great abundance on
both young and old pear trees. In many cases, especially on young trees,
it appeared that fully 90 per cent of the aphides observ'ed at one time were
pupse, and in other instances observations in October and later after the
winged forms had departed indicated that almost the entire infestation
had developed into migrants. On old trees there remained on the average
a larger residue of wingless forms. On unfavorable types of soil the
winged forms are produced in far less abundance. It appears to be a
rule that the heavier and drier the soil the larger the percentage of pupse
developing. It sometimes happens that the migrants are unable to rise
to the surface of the ground and become imprisoned in pockets in the soil.
In one instance two living sexual females were found in such a pocket
beside dead migrants.
The winged forms have been noticed on pear foliage and on the trunk,
but with one exception ^ no deposition of sexes has been obser\^ed on the
pear. On cork and American elms (Ulmus spp.) migrants were ob-
ser\'ed to deposit the sexes in cracks in the bark and on the lower surface
of leaves. In one instance the migration from a nursery of pear trees to
a group of young elms 200 yards distant could be traced. The migrants
fly readily and strongly and are stimulated by the sun's rays, being more
active on warm than on cool days. On the elms they were more abun-
dant on trees with rough bark than on the smooth-barked plants.
The migrants vary considerably in size. They are rather elongate,
shining black or dark green, with a tuft of white v/ool on the caudal seg-
ment; otherwise, there is no flocculence. The lower surface is dark
green, sparsely pov\'dered at the sutures. The antennse, eyes, and a por-
tion of the legs are black. The base of the femora and the middle por-
tion of the tibice are yellowish brown or amber. The wings have narrow
black veins and a greenish blue stigma. The wing insertions are some-
times brown, but are more often yellowish. In recently molted indi-
viduals there is sometimes a smoky-brown patch at the base of the fore
wings.
To obtain the sexes, migrants vrere confined in stender dishes and in
small rubber cells mounted on microscope slides with cover glasses as
lids. Some were kept in a lighted room in which the temperature varied
very considerably, at times rising to 75° and at other times falling to
55° F. Others were kept in a dark cellar' where the temperature varied
but little and averaged about 61° F. Under cellar conditions the
migrants deposited more sexual forms than under the conditions ob-
taining in the room. Some of the dishes were kept dry and others
moistened to different degrees. In the moistened dishes the sex pro-
' In August, 191 1, at San Jose, Cal., a migrant was noticed depositing sexes on the upper surface of a pear
leaf.
356 Journal of Agricultural Research voi. vi, no. io
duction was better than in the dry ones, although too much moisture
prevented the sexual forms from freeing themselves from the pellicles.
Whether the migrants had flown or not did not seem to bear any influ-
ence on the deposition of the sexual forms. In most of the dishes more
than half of the sexed forms were not extruded, but died unborn. In the
rubber cells five-eighths of an inch in diameter and three-sixteenths of
an inch in height the migrants did best singly, while the larger stender
dishes provided space for a number. In all the dishes pieces of pear or
elm bark were provided, but the migrants rarely deposited the sexes on
these, nearly always extruding them on the filter paper also provided.
It frequently happened that the sexes after having been extruded be-
came entangled with the wings or legs of the parents or with each other.
The sexes were deposited in rapid succession. The migrants rarely lived
beyond three days after they were placed in the dishes, whether they
deposited sexual forms or not. None lived longer than six days.
They died immediately after the sexes had been extruded and very few
deposited their full complement.
All the sexes deposited were not noted; but about four-fifths of them
totaled 109 individuals, of which a little over half (58) were females.
Only a few matured, and the majority died unmolted. Undoubtedly
the cause of this was the abnormal condition of the environment. How-
ever, it appears to be proved that the sexes are produced in about equal
numbers, and observations in the field corroborate this. Four fall mi-
grants dissected on October 27 and 28 had contained, respectively, 5, 7, 8,
and 9 young. In the dishes not more than seven sexes were ever
dropped by an individual. The number of males and females depos-
ited by individual migrants was found to range from seven females and
no males to five males and one female. Probably a larger series would
have furnished a migrant producing only males. As a rule the produc-
tion of sexes was about evenly divided between male and female.
The sexes have no woolly covering such as that occurring on the sexes
of Eriosoma lanigerum, but are bare and shining. The female, however,
at the time of depositing the winter egg, has a patch of short white wool
on either side of her body and with this she contrives to clothe partly
the winter or impregnated egg. The sexes are active, the male especi-
ally so, both immediately after extrusion and following the casting of
their fourth and final skin. Between casting their first and fourth
skins they remain inactive unless disturbed. Normally they seek crev-
ices in the bark, but in the dishes they frequently molted on filter paper
or on the sides and floor.
The sexes mature in from 7 to 1 1 days and molt four times — that is,
about every other day. Being beakless, they take no food.
The males are smaller than the females, the latter being enlarged by
reason of the egg within the body. The male at first is light green, with
June 5, 1916
Woolly Pear Aphis
357
hyalin antennae and legs and black eyes of three facets. The insect
becomes darker with age and the mature individual is dark olive-green,
sometimes tinted with lilac or purple, the central part of the abdomen
being darkest. The male is always narrow in shape. The female varies
in color from a light orange to a dark red. The eyes and appendages
are as in the male. The majority are orange or a light crimson-lake.
They are mxuch stouter than the males and are longer and stand much
higher. A mature female measured alive was 0.67 mm. long by 0.33
mm. in maximum width. A mature male was 0.43 mm. long by 0.21
mm. in maximum width.
Copulation occurs as soon as the sexes are mature. It appears that
unless the female is fertilized directly after she has cast her last skin she
will fail to deposit the winter egg. The male may live at least a week
after he is mature, but apparently he can exercise the sexual function
only immediately after he has cast the last skin. The females deposit
the impregnated egg imm^ediately after copulation, and after its deposi-
tion they may live for a day or two at the most. The winter or impreg-
nated egg is laid normally in crevices or scars of the bark of the elm. In
the dishes it was laid sometimes on the outside of the bark, and both
elm and pear bark were used. It was never laid elsewhere than in the
bark. The egg measures about 0.444 ™ni. by 0.225 mm., is short oval,
reddish yellow, and shining. The end first extruded is reddish and bare,
while the other extremity is yellowish and usually covered with short
white wool provided by the female. Winter eggs were deposited in
dishes between October 15 and November 12. Undoubtedly they occur
in nature as early as September 5, and may be laid as late as the middle
of November. Toward the end of October some were collected under the
bark of elms under observation. Table I is a comparison of the biology
of Eriosoma pyricola with that of E. lanigerum.
Table I. — Comparison of biology of Eriosoma pyricola ziiih that of Eriosoma lanigerum
in California '
Eriosoma lanigerum on apple and varieties of pear.
Eriosoma pyricola on pear.
Aerial and radical.
Attacks trunks, branches, and
causes knotty swellings on roots.
twigs;
Fall migrants rarely abundant; apparently
not influenced by conditions.
Radical only.
Attacks chiefly fibrous rootlets; rarely
causes lesions; occasionally settles on
larger roots.
Fall migrants very abundant under fav-
orable conditions.
• The full cycle of these species has not been 'worked out in California, but there appear to be no records
of spring generations of E. lanigerum observed on elm.
The fall migrants of E. pyricola may be distinguished from those
of E. lanigerum and E. americanum as shown in Table II.
358
Journal of Agricultural Research
Vol. Yl, No. 10
TablS II. — Comparison of the fall migrants of Eriosoma pyricola, E. lanigerum, and
E. americanum
E. pyricola.
E. lanigerum..
E. americanum.
Stigma short, greenish blue.
Veins narrow without brown
Stigma somewhat elon-
gate, yellowish or gray.
Veins narrow and without
Stigma elongate, gray.
Veins broad, with brown-
margins.
Body naked except for cau-
dal segment.
Distal sensoria of antennal
segments V and VI with
fringe.
brown margins.
Body with some woolly
clothing.
Distal sensoria of antennal
segments V and VI
without fringe.
ish margins.
Body with slight woolly
covering.
Distal sensoria of antennal
segments V and VI with-
out fringe.
The new species is easily distinguished from E. ulmi Linnseus from
the fact that segment V bears prominent transverse sensoria. The
wingless forms can be distinguished from those of E. lanigerum by the
structure of the compound wax pores, and the winged forms by the
antennae. The winged forms of E. pyricola are remarkably like those of
E. lanuginosa Hartig. The proportions are almost exactly the same.
The only difference seems to be the fringing of the sensorium on segment
V. The wingless forms and the pupae have the prominent wax pores
figured. No such pores occur in our specimens of E. lanuginosa, but
very similar ones do occur in E. ulmi. At first it was thought that
two species were present in the collected material, but careful rearing
experiments by the junior writer have shown the connection between
all the forms. It does not seem probable that such prominent wax-
secreting structures would be present in one form of the species and not
in all forms.
Eriosoma pyricola, n. sp.
Wingless viviparous female. — General form elongate. Antennal segments in
length as follows: I, 0.048 mm.; II, 0.048 mm.; Ill, o.i mm.; IV, 0.04 mm.; V,
0.048 mm.; VI, 0.064 mm. (unguis, 0.032 mm.); segments armed with hairs (fig.
I, E), which are considerably longer than those met with in lanigerum (fig. i, D),
and with a large distal fringed sensorium on segments V and VI, as well as some smaller
ones on VI. Compound wax pores very prominent and circular (fig. i,/), those on
the abdomen containing about 20 cells. Abdomen sparsely covered with hairs about
0.16 mm. long; cornicles circular, their rims more heavily chitinized on their inner
margins than elsewhere. Wax reservoir apparently present as in E. lanigerum
(visible as a clear yellow area in mounted specimens). Hind tibiae about 0.44 mm.
long; hind tarsus, 0.112 mm.; rostrum extending beyond the second pair of coxae.
Length, 1.92 mm.; width, 0.96 mm. The hairs on the antennas of the young are
especially prominent (fig. i, /).
Young forms yellowish pink, older ones pink to red. Antennte, legs, and labium
dusky; eyes dark red, very minute.
Intermediates. — In the collection, Q. 6399, are a number of specimens which
would be taken at first glance for wingless viviparous females. A careful study, how-
ever, proves them to be intermediates. No trace of wing pads can be found, but the
eyes clearly show the intermediate nature of the specimens. In the normal wingless
June 5, 1916
Woolly Pear Aphis
359
/r _ ^
Fig. 1. — Comparative structure of antennje and wax pores of Eriosoma spp.: A , distal segments of antenna
of winged viviparous female of E. pyricola; B, distal segments of antenna of winged viviparous female of
E. ulmi; C, distal segments of antenna of wingless viviparous female of E. amcriccnum; D, distal segments
of antenna of wingless viviparous female of E. lanigerum; E, distal segments of antenna of wingless vivi-
parous female of E. pyricola; F, distal segments of antenna of winged viviparous female of E. americanum:
G. distal segments of antenna of winged viviparous female of E. lanigerjirn; H, compound wax pore of E.
lanigerum; I, compound wax pore of E. pyricola; J, distal segments of antenna of first instar wingless
viviparous female of E. pyricola.
360 Journal of Agricultural Research voi. vi, no. 10
forms the eyes are composed of three facets and are very minute, whereas in these
specimens the eyes are large and composed of numerous facets, thus approaching the
compound eyes of the winged form. All other characters met with are those of the
wingless viviparous female.
Pupa. — Antennal segments in length as follows: 1,0.048 mm.; 11,0.064 mm.; Ill,
0.192 mm.; IV, 0.064 mm.; V, 0.08 mm.; VI, 0.08 mm.; segments armed with hairs
and sensoria as in the wingless female. Wing pads about 0.64 mm. long. Compound
wax pores similar to those of the wingless females. Hind tibia, 0.432 mm.; hind
tarsus, 0.128 mm. Body with long hairs as in the wingless form. Length, 2.32 mm.;
width, 0.96 mm.
Pinkish, with a brick-red diffusion.; wing pads whitish yellow; wool sparse, erect.
Winged viviparous female (fall migrant). — Antennal segments in length as
follows: 1,0.048 mm.; 11,0.064 mm.; Ill, 0.432 mm.; IV, o. 112 mm.; V, 0.112 mm.;
VI, 0.08 mm. (unguis, 0.032 mm.); segments I and II armed with a few hairs; segment
III armed with about 20 transverse sensoria, which extend a little over half way around
the segment as in E. lanigerum, the dorsal side of the segment armed with numerous
prominent hairs; segment IV similar to segment III and armed with four or five
transverse sensoria; segment V (fig. 1, A) armed with three or four transverse sen-
soria and a distal fringed sensorium, a few hairs, and many rows of setse; segment VI
similar to segment V, but without transverse sensoria. The fringed sensorium at the
base of the unguis varies in shape. Fore wing somewhat similar to that of E. ameri-
canum; stigma short and rounded at the distal extremity. Hind tibia, 0.88 mm.;
hind tarsus, 0.128 mm. Form elongate ; length, 1.76 mm.; width, 0.72 mm.; fore wing,
2.4 by 0.88 mm. Without wool.
Dark brown or very dark green. Base of fem.ora and tibiae yellowish gray. Stigma
bluish gray. Abdomen shining.
Described from wingless females, intermediates, pupae, and winged
viviparous females in balsam mounts.
Type: Cat. No. 20083, U. S. National Museum.
PATHOLOGICAL HISTOLOGY OF STRAWBERRIES AF-
FECTED BY SPECIES OF BOTRYTIS AND RHIZOPUS
By Neil E. Stevens,
Pathologist, Fruit Disease ItivestigatiotJ-s, Bureau of Plant Industry
INTRODUCTION
The fungi causing rots of strawberries (Fraqaria spp.) in transit from
the Southern States have been under investigation by Dr. C. L. Shear,
Mr. R. B. Wilcox, and the writer for the past two years. From the first
it has been apparent that two species were chiefly responsible for their
decay during shipment and on the market. These were Botrytis (cinerea?)
and Rhizopus {nigricans}).'^ The effect of these two fungi on ripe straw-
berries is strikingly different. Berries injured by Botrytis sp. show a
characteristic dryrot — that is, they retain their shape, shrivel somewhat,
and no leaking of juice is evident; whereas berries rotted by Rhizopus
sp. quickly flatten out, with the loss of a large amount of juice. Such
berries are characterized as "leaks" by growers and dealers.
F. L. Stevens - has already recognized a species of Rhizopus as the
probable cause of leak. He, however, considers (p. 950) that Botrytis sp.
"is the primary cause of the molding, that the Botrytis initiates the decay,
opening the way to such other saprophytes as may be present; of such
saprophytes, Rhizopus is by far the most prominent and most abundant."
In order to determine if possible the relations of these fungi in rotting
strawberries and in particular what differences exist in their method of
attack on the fruit, a study of strawberries affected by these fungi was
undertaken.
EXPERIMENTAL METHODS
The strawberries examined were chiefly of the Klondike variety grown
in Louisiana during the season of 191 5. Berries of other varieties grown
in South Carolina and at Arlington Experimental Farm, Va., in 191 5, as
well as the Missionary and Klondike varieties from Florida in 191 6, were
used for comparison. Naturally infected berries as well as sound berries
inoculated with spores and mycelium from pure cultures were used in
both cases.
The material was fixed in a solution of equal parts of absolute alcohol
and glacial acetic acid. This fluid penetrates very rapidly, so that
whole strawberries are satisfactorily fixed. In the case of large berries,
' In the present uncertainty regarding the taxonomy of these genera it seems unwise to attempt a definite
determination of the species. Permanent mounts of the material described are preserved, however, and
cultures of the species considered are retained for further study.
^ Stevens, F. L- A destructive strawberry disease. In Science, n. s., v. 39, no. 1017, p. 949-950. 1914.
Journal of Agricultural Research, Vol. VI, No. 10
Dept. of Agriculture, Washington, D. C. June 5, 1916
dz G— 82
(361)
362 Journal of Agricuhural Research voi. vi, no. 10
however, the ends were cut off to hasten penetration. Strawberry cells
are so large that rather thick sections, from 10 to 20M, were found most
desirable. The walls of the strawberry cells and of the fungus hyphge
are so similar that differential staining was rather difficult. The best
differentiation was secured by a combination of methylene blue and
clove-oil eosin, using a water solution of tannin as a mordant. This
method was suggested to the writer by Mr. Charles S. Ridgeway, of the
Bureau of Plant Industry. The hyphse, however, are so large as to be
easily distinguished when the sections are properly stained with the
more permanent stains, as safranin, Delafield's hematoxylon, or even
Bismarck brown.
RESULTS OF INFECTION OF STRAWBERRIES BY BOTRYTIS SP.
Botrytis sp. has long been a favorite subject for the investigation of the
relations of host and parasite. The somewhat conflicting views held by
different investigators as to the nature of its attack on the host are well
summarized by Brown ^ in a recent paper. In general, all writers agree
on the presence of a cell-wall dissolving enzym, but differ widely as to
the cause of the toxic action of the fungus.
As already stated, strawberries rotted by Botrytis sp. retain their
shape, shrivel slightly, and even in a moist chamber there is no evident
leaking. The moisture is apparently lost so slowly that it evaporates
from the surface of the berry. A microscopic examination shows that
the fungus has penetrated all parts of the berry; indeed, the cells are
in many places embedded in the mass of mycelium and are apparently
held together by it. The fungus is evidently capable of readily dis-
solving the middle lamella and of penetrating the cell walls themselves.
Often hyphae grow between the cells of the host for some distance and
then penetrate the cells (PI. XLIX, A). Not infrequently cells containing
numerous hyphae have the shrunken and distorted protoplasmic contents
still present (PL XLIX, B, C, D). Sometimes hyphae occur in adjacent
cells whose separating wall remains intact and apparently unchanged
(PI. XLIX, B) ; or they may pass from one cell into the next, either where
the cells are in contact or across an intercellular space (PI. XLIX).
It is interesting to observ^e that hyphae usually enter a cell at the
angle where it joins two other cells; Plate XLIX, D, F, and G, shows
examples. The hypha passes between two cells, apparently by dis-
solving the middle lamella, and then penetrates the wall of the cell
with which it comes in direct contact. Occasionally a hypha seems to
push back a portion of the cell wall before penetrating (PI. XLIX, G).
The fungus may, however, penetrate the wall at a considerable dis-
tance from the intersection of the cells (PI. XLIX, A, E)\ or it may
1 Brown, William. Studies in the physiology of parasitism. I. The action of Botrytis cinerea. In
Ann. Bot., v. 29, no. 115, p. 313-34S- 1915-
June 5. 1916 Histology of Strawberries 363
pass the point of intersection and penetrate a short distance beyond
(PI. XUX, H).
Brown/ working with thin disks of tissue cut from various plants,
particularly tubers of the potato and roots of the turnip, immersed in a
strong extract from the germ tubes of Botrytis cinerea, noted that the
separation of the cells followed the line of the cell walls, the cells on
either side being left intact. His idea of the destruction of the cells is
that the middle lamella is first dissolved, in consequence of which the
tissue readily falls apart along the line of the m^iddle lamella. Very soon
the remainder of the cell wall disintegrates and the whole structure
becomes very fragile.^ In no case was complete solution of the cell w^all
observed. Death of the cells ^ takes place at a late phase in the process
of disorganization of the cell walls. He observed also that in all cases
if a cell w^all was disintegrated death of the cell ensued; on the other
hand, if the cell wall was not affected neither were the living contents of
the cell.*
Brown's conclusions satisfactorily explain the condition found by the
writer in strawberry cells attacked by Botrytis sp. Certainly the fungus
is able to penetrate the cells of the host while they are still fairly normal
in appearance and while the cytoplasm is still distinguishable (PI. XLIX,
B, D, G). The writer did not find, however, in any of the strawberries
examined cells which were unaffected by the action of the fungus.
RESULTS OF INFECTION OF STRAWBERRIES BY RHIZOPUS SP.
In contrast to the condition of strawberries rotted by Botrytis sp.,
berries rotted by Rhizopus sp. show the following characteristics. The
berries soon become flattened, with considerable loss of juice. Micro-
scopic examination shows that the hyphae are characteristically close
to the surface of the berry, the majority being found in the outer six or
eight cell layers. Hyphae rarely or never penetrate the cells of the berry
under field conditions or when kept in moist chamber. The nuclei of
the cells persist in apparently normal condition until the cytoplasm of
the cell has almost entirely collapsed.
The crowding of the fungus in the outer portion of the berry is very
noticeable. Indeed hyphae frequently grow for some distance imme-
diately beneath the epidermis. Plate XLIX, /, shows a portion of such
a hypha in a section cut nearly tangential to the surface of the berry.
The sm^all, thick-v^-alled cells (heavy, lines) on the right are epidermal
cells; the larger, thin-walled cells (light lines) on the left are storage cells.
The hypha, which could be traced across several sections, grows between
these two layers of cells for a considerable distance without penetrating
either. A similar condition is shown in vertical section in Plate XLIX,
' Brown, William. Op. cit., p. 333. 'Ibid., p. 347.
2 Ibid., p. 333. < Ibid., p. 345.
364 Journal of Agricultural Research voi. vi. no. 10
K, L. In the latter case the fungus has penetrated the epidermis and the
external hyphse are sporangiophores.
It is evident from a study of the sections that Rhizopus sp. does not
readily penetrate the unbroken epidermis from the outside. Hyphae
are found which extend for some distance along the surface of the berry
without penetrating. Plate XLIX, /, shows a portion of such a hypha;
even the germ tubes seem unable to penetrate readily and often grow
for some distance (PI. XLIX, M) over the surface without penetrating.
Under field conditions or in moist chamber in the laboratory Rhizopus
sp. apparently very rarely enters the host cells. Although several
hundred slides were examined no single instance was found in which a
hypha had penetrated a cell wall. Plate XLIX, I-L, shows that the
hyphae typically grow between the cells along the middle lamella. The
effect of the fungus on the host cells is readily seen by the contraction
of the protoplasm. Plate L shows strawberry cells in various stages of
degeneration close to hyphae of Rhizopus sp.
Plate L, A, shows the normal appearance of one of the smaller storage
cells of the strawberry. In this case the cytoplasm contains numerous
small vacuoles. Frequently, especially in larger cells, there is a single
large vacuole. Plate L, B, shows a similar cell in which the protoplasm
has begun to contract away from the wall. This cell was separated from
the nearest hyphae by three layers of cells. In Plate L, C, hyphae of
Rhizopus sp. are shown in contact with two host cells (a branch hypha
overlies one cell). The protoplasm of these cells is much shrunken, but
the cell walls retain their normal position, and the nuclei are unchanged.
Plate L, D, E, F, and G, show progressively later stages in the breaking
down of cells adjoining hyphae. In some (PI. L, D, F) the wall has begun
to collapse. In all except Plate XLIX, G, in which there was very
little cytoplasm remaining, the nucleus shows no signs of degeneration.
This persistence of the nucleus in apparently normal condition after
the contraction of the protoplasm has progressed considerably is one of
the most striking characteristics in berries attacked by Rhizopus sp. and
is in sharp contrast to the condition found in berries rotted by Botrytis sp.
Often in a cell in which the cytoplasm has largely disappeared and the
wall is partly collapsed the nucleus appears large and typical, as in an
intact cell (PI. L, /). Frequently the cell wall collapses so rapidly that
no space is left between it and the contracted protoplasm (PI. L, H, I).
EFFECT OF RHIZOPUS SP. ON BERRIES IN EXTREMELY DRY AIR
In connection with experiments on the humidity relations of the fungus,
berries inoculated with Rhizopus sp. were placed in a desiccator with
concentrated sulphuric acid. Under these extremely dry conditions the
berry "leaked" in the characteristic manner, but the habit of growth of
the fungus was changed in two important particulars. •
Junes, i9i6 Histology of Strawberries 365
Fungus hyphae were found in all parts of the berry, being abundant
even in the center, within the circle of vascular bundles. Apparently
the extreme dryness of the surrounding air made the intercellular spaces
within the berry more favorable for its growth than the outer ones.
Under these severe conditions the cells of the berry collapsed so generally
that the relations of the fungus hyphae to the walls could usually be
studied only in cells near vascular bundles. It was evident that while,
in general, the hyphae grew between the cells of the host (PI. L, L) they
were frequently found inside the cells as well (PI. L, K, M). It is worthy
of note that in these berries several instances were found where hyphae
had punctured the cells and the nucleus of the cell was unchanged in
appearance (PI. XLJX, K).
COMPARISON OF THE FUNGI
The difference in the histological relations of the two fungi with the
strawberry may be briefly summarized as follows :
Botrytis sp. penetrates all parts of the berry, growing within the cells
as well as between them and ramifies through the tissues of the strawberry,
surrounding and filling them with a network of mycelium. The cells of
the berry seem to be quickly killed by the fungus; at least the proto-
plasm shrinks away from the cell wall and becomes disorganized so that
no nucleus can be distinguished.
The mycelium of Rhizopus sp., on the other hand, is found chiefly in
the outer portion of the berry. The hyphae grow between the cells,
separating them and apparently extracting the cell sap. The nuclei of
the cells persist unchanged until a late stage in the breaking down of the
cytoplasm. When the fungus is grown on berries in a dry atmosphere, its
action is somewhat different. The mycelium penetrates to the center
of the berry, and hyphae are frequently found inside cells.
It is difficult to trace an exact causal relation between the histological
differences in the attack of these fungi on the strawberry and the fact
that they cause quite different types of rot. The fact that Rhizopus sp.
separates the cells of the berries so completely may readily account for
the berries affected with this fungus becoming so soft and easily flattened.
On the other hand, the mycelium of Botrytis sp., by penetrating all parts
of the strawberry, helps to hold it in shape and converts it into a mummy.
It is possible that the juice of the berries affected by Rhizopus sp. is
pressed out by the collapse of the berries, owing to the mere separation
of the cells. This is, however, hardly an adequate explanation of the
phenomenon.
While it is not proposed at the present time to review the rather
voluminous literature on either of the fungi under consideration, a
closely parallel case described by Behrens * should be mentioned in this
' Behrens, Johannes. Beitriige zur Kenntnis dcr Obstfauhiis. In Ccntbl. Bakt. [etc.], Abt. 2, Bd. 4,
No. 12, p. 515-516. 1S98.
377G7°— 16 3
366 Journal of Agricultural Research voi. vi, no. 10
connection. He observed in 1S95 ripe tomatoes affected by Mucor sto-
lonijcr which reduced the pulp of the tomato to an almost fluid mass. A
species of Fusisporium found at the same time on the tomatoes produced
a dry-rot quite in contrast to the wet condition produced by the species
of Mucor. Behrens found on microscopic examination that the mycelium
of Fusisporium sp. penetrated the cells of the host, while the mycelium
of Khicor stolonijer grew entirely in the intercellular spaces.
The relation of these fungi to each other in their attack on the berry
is much clearer. In comparatively few cases have both fungi been
found on the same berry and in no instance has the writer found a berry
in which Rhizopus sp. had followed in a place originally infected by
Botrytis sp.
Numerous cases have, of course, been found in which there were two
fungi in the same berry; for instance, Botrytis sp. and Fusarium sp.,
Botrytis sp. and Alternaria sp., Rhizopus sp. and Fusarium sp. These
fungi do not, however, seem to have entered in the same place, but rather
from different portions of the berry. The mycelia of the two fungi
sometimes mingle in the tissues of the berry — for example, Botrytis sp.
and F^lsarium sp., Rhizopus sp. and Fusarium sp. — or they may occupy
different portions of the berry with a marked line of division between
them, each apparently being unable to invade tissue occupied by the
other fungus — for example, Botrytis sp. and Alternaria sp.
These observations do not preclude the possibility of Rhizopus sp.
following in an area originally infected by Botrytis sp. or some other
fungus, and this may occur in the field or in badly affected berries which
are thrown out as culls in packing. They do, however, plainly indicate
that Rhizopus sp. is not dependent on the presence of any other fungus
in its attack on strawberries during shipment and on the market.
PLATE XLIX
A-H, Strawbeny cells attacked by Botrytis sp. (X 210) : A, H^'pha g^o^^^ng partly
between and partly within strawberry cells; B, hyphae inside strawberry cells in which
remnants of the protoplasm may still be distinguished ; C, hypha passing from one cell
into another across a short intercellular space; D, E, F, G, H, hyphae entering cells in
various ways (in G the hypha has pushed back a portion of the cell wall before breaking
through). I-M, Strawberry cells attacked hy Rhizopussp. {X 210): 7, Hypha growing
between the epidermis and the adjacent layer of storage cells; /, hypha gro^ving over
the surface of the strawberry; K, hyphae growing underneath the epidermal layer and
between the storage cells; L, Rhizopus sp. growing between epidermal cells (basal
portions of sporangiophores above and rhizoids below epidermis); M, germinating
spore in cavity formed by a seed.
Histology of Strawberries
Plate XLIX
Journal of Afjricultural Research
Vol. VI, No. 10
Histology of Strawberries
Plate L
Journal of Agricultural Research
Vol. VI, No. 10
PLATE L
Strawberry cells attacked by Rhizopus sp. A, Normal storage cell of strawberry;
B, storage cell (near hyplise) showing a slight contraction of the protoplasm; C, D, E,
F, G, progressive contraction of protoplasm of host cells near hyphse (the cell walls
have contracted very little) ; H, I, J, strawberry cells near hyphae in which the cell
wall has crumpled with the contraction of the protoplasm; K, M, hyphae inside cells;
L, hyphae growing between cells of the strawberry; K, L, M are drawn from berries
which had been rotted in the desiccator. (X 210.)
LIFE HISTORIES AND METHODS OF REARING
HESSIAN-FLY PARASITES
By C. M. Packard,^
Scientific Assistant, Cereal and Forage Insect Investigations, Bureau of Entomology
INTRODUCTION
The most effective factors in the control of the Hessian fly {Mayetiola
destriictor Say) in the past have been its parasites. There are seasons,
however, when the parasites become scarce and the Hessian fly exceed-
ingly abundant. Again, in the same season the Hessian fly seems prac-
tically free from parasites in some localities while in others its parasites
are numerous. A thorough knowledge of the life histories, field habits,
relative efficiency, and effective methods of artificial propagation and dis-
semination of the different parasites, therefore, might make it practicable
to introduce the most efficient species from localities where they are
abundant into other localities where the host is working destruction
unchecked by its enemies. It might also be possible to propagate arti-
ficially and to disseminate the parasites during periods when they have
become scarce in the fields, and thereby shorten the period of destructive
abundance of the Hessian fly. Up to the present time very little accurate
and detailed information seems to have been recorded regarding the life
stages, habits, and efficiency of Hessian-fly parasites. It has been uncer
tain whether or not some of the species involved were true parasites.
Some results in this direction have been accomplished by the author
during the last two seasons, and the purpose of this paper is to make
public these results arid the methods used in attaining them.
The life histories and methods of rearing three hymenopterous parasites
are treated in this paper: Eupelmus allynii French, Merisus destructor Say,
and {Merisus) Micromelus subapterus Riley. The seasonal history and field
habits of these parasites wall require another season's observation before
they can be effectively treated. The scope of this paper is therefore
limited to the life histories and relationships of these species to one another
and to their common host as determined under laboratory conditions.
' The writer wishes to acknowledge his indebtedness to ^lessrs. E. O. G. Kelly, W. R. Walton, A. B.
Gahan, W. R. McConncll, and J. A. Hyslop, all of the Bureau of Entomology, for helpful advice; to Mr.
Kelly for making the work possible, and to Mr. Gahan for determining all specimens.
Journal of Agricultural Research, Vol. VI, No. lo
Dcpt. of Agriculture, Washington, D. C. June 5, 1916
dv K— 34
(367)
368 Journal of Agricultural Research voi. vi. no. 10
METHODS OF BREEDING AND REARING
The adult parasites used in all experiments were kept in modified
forms of the Doten cage.^ One form, used when it was desired to con-
fine a number of parasites together, consisted of two large, straight-sided
vials of the same diameter, the mouths of which snugly fitted into a
paper tube i inch long. This paper tube was held in shape by a layer
of adhesive plaster around the outside. The cage was prevented from
rolling by sticking a square of heavy cardboard to one side of the cqu-
necting tube. A label was pasted to the upper side of the tube for iden-
tification. One vial was kept dry and clean, while water and honey were
supplied in the other.
The other form of Doten cage, used chiefly for isolating pairs and indi-
viduals, was simply a small, straight-sided vial into the mouth of which
was fitted the open end of a slightly smaller, straight-sided vial. A
small label was pasted on the side of the larger vial for identification.
Cages of this kind were prevented from rolling by keeping them in
shallow boxes with corrugated pasteboard-lined bottoms. Food and
water were placed in the smaller vial.
In both forms of cages the water and the honey used for food were
placed separately in small droplets on the upper surface inside the food
vial. The honey used was the extracted form diluted with an equal
amount of water. It was necessary to exercise considerable care not to
place too large a drop of honey in a cage, because of its tendency to run
down on the inside of the vial and to entangle the insects. Fresh water
and honey were placed in the cages daily, and at least once a week the
food vials of the cages were carefully cleaned to remove dried or soured
honey. Replenishing the food and water in the cages once a day seemed
sufficient to supply the needs of the parasites. It was often found neces-
sary to make up a fresh supply of the honey because of souring or mold-
ing, especially in hot weather. Sterilizing the fresh supply by placing
the dropper bottle containing it in boiling water for a few minutes caused
it to remain sweet and usable much longer.
BREEDING THE PARASITES
To determine all the life stages from egg to adult involved the processes
of exposing Hessian-fly puparia to parasites, dissecting the parasite eggs
from the host puparia, and rearing, in little glass-cell cages devised
for the purpose, the resulting parasite larvae on Hessian-fly larvae which
were also dissected from puparia. Hessian-fly puparia contained in
sections of wheat stems were first exposed to the adult parasites by
placing the stems in the vial cages containing the adults. The stems
remained in the cage for a day, or until a parasite was seen to oviposit in
a flaxseed, when they were removed and the puparia dissected. The
1 Doten, S. B. Concerning the relation of food to reproductive activity and longevity in certain hy-
menopterous parasites. Nev. Agr. Expt. Sta. Tech. Bui. 78, 30 p., 10 pi. 1911.
June s, 1916 Hessian-Fly Parasites 369
eggs of the three species studied were always found between the inner
surface of the puparium and the larva itself. They were transferred
separately, each to an unparasitized Hessian-fly larva which had been
previously dissected from its puparium and placed in a little glass-cell
cage of the following description :
Flat glass plates i inch by 1% inches square were used, in which hollows
about the size and shape of a Hessian-fly puparium were ground in one
surface, one hollow per plate, this work being done with a small car-
borundum grinder. After a host larva and a parasite &g<y had been
placed in a hollow, the cell was closed by covering it with an ordinary
glass cover slip. The cover glass was held in place by two little dabs of
honey on its underside. The cell was not sealed by a complete ring of
the adhesive because of the desirability of diffusion of atmospheric
moisture under the cover slip. Honey seemed to be the ideal adhesive
for this purpose, since it had no odor harmful to the inmates of the cell;
it held the cover-glass tight against the slide; it did not dry so hard as to
prevent the cover-glass from being easily removed; and a supply of it
was always convenient. A label was pasted on the glass plate near one
end for identification. The complete development of the parasite from
egg to adult on its host could then be observed under the binocular in
this little cell without disturbing the parasite or the host in the least.
With each of the three species the period from oviposition to emergence
of adult, when individuals were reared in glass cells, approximated very
closely the period from egg to adult when individuals were reared under
the same meteorological conditions in Hessian-fly flaxseeds. Hence, the
length of each stage of development as determined from individuals
reared in glass cells may be considered normal.
It was discovered that the lar\^ge of all three species molted while making
their growth within the little cells. The length of the instars was not ob-
serv^ed, but the number of molts was determined by transferring to a balsam
mount on a microscope slide all the material left behind in the little glass
cell where a single individual had made its growth. In cases where the
larva had pupated, the last molted skin was added to the mount. In
cases where the full-grown larva had not pupated, the mandibles borne
by the larva were dissected from it and added to the mount. To deter-
mine the number of molts of a single individual, the mount of the material
it left behind was examined under the microscope and the number of
pairs of mandibles in the material ascertained. In all cases the cell in
which the larva made its growth was known to be absolutely clean when
the host and the e:gg from which the parasite larva hatched were placed
in it ; hence, it was known that all pairs of mandibles found in a mount
belonged to the same larva. Cells which contained simultaneously the
remains of more than one parasite larva were not used in determining the
number of molts. No attempt was made to determine the number of
molts of individuals which had made their growth inside flaxseeds.
370 Journal of Agricultural Research voi. vi, no. io
EUPELMUS ALLYNII
THE EGG
The egg of E. allynii French (PI. LI, fig. i) is elliptical in shape, with a
thin stalk of varying length on one or both ends. In some cases the stalk
seems to be entirely absent from one end. The egg is grayish white in
color. The long axis of the body of the egg averages 0.35 mm., the short
axis o. 1 4 mm. in length. As a parasite of the Hessian fly, the observations
at Wellington, Kans., indicate that the egg is normally deposited in the
puparium of the host. Females were repeatedly observed by Mr. E. O. G.
Kelly and the author to be ver}^ numerous in fields, ovipositing in Hessian-
fly flaxseeds where these constituted the only stage of the fly to be found.
In one instance, however, a wheat stem containing nearly grown Hessian-
fly larvae, but no flaxseeds, was placed in a vial cage containing females
of E. allynii. Upon dissecting this stem two eggs of this parasite were
found inside the leaf sheath close beside the Hessian-fly larvae. Whether
or not the parasite is able to complete its development on Hessian-fly
larvae before they have formed puparia is still unsettled.
Hundreds of flaxseeds in which E. allynii had oviposited have been
dissected and the eggs of the parasite have always been found inside the
puparium but external to the inclosed Hessian-fly larv^a or pupa. Some-
times they were unattached, but more often the egg was fastened to the
inner surface of the puparium by a little netlike structure made apparently
of fine, white threads tangled together (PI. LI, fig. 2). The threads form-
ing the net appeared to be identical in diameter, color, and material with
the egg stalks. The edges of this little net or mat were fastened down all
around the egg, holding it securely in place. Sometimes the net was
partly fastened to the host larva in addition to the puparium. In all
experiments E. allynii oviposited seemingly indiscriminately in flaxseeds
already containing parasite larvae as v^-ell as in those containing Hessian-
fly larvae. The incubation periods of 109 eggs varied from 1% days to 5
days. The egg stage was shorter in summer temperatures, observations
being made during a period from July to November.
THE LARVA
Upon becoming fully formed inside the egg the larva (PI. LH, fig. 2)
breaks through one end of the chorion and after crawling around a little
attaches itself to the external surface of the host larva. The parasite
lai'va bears strong mandibles and feeds externally on the Hessian fly by
puncturing the epidermis of the host and sucking out the body liquids.
Larvae reared in glass cells became full grown in from 7 to 10 days.
After becoming full grown many of the lar\'a£ were inactive for months;
others pupated at once. In the warm summer temperatures most of the
larvae reared pupated at once upon completing their growth, while larvae
reared in the fall pupated only in occasional instances.
June 5, 1916 Hessian-Fly Parasites 371
The larvae reared in glass cells normally pass through five instars.
Nearly all mounts made of the material left behind by larvae which had
finished feeding showed a total of five pairs of larval mandibles, while in
the remaining mounts from two to four pairs were found which always
correspond in size and shape to some one pair in the complete series.
Five was the maximum number found in any one instance, and in cases
where less than five were present it appeared that some of the molts had
been lost in manipulation. Where five pairs of mandibles were found in a
single mount, the sizes increased fairly uniformly from the second molt
to the last. The mandibles and head shields of newly hatched larvae
appeared to be m.ore heavily chitinized than those of later instars, except
the last, and somewhat larger than those of the second instar. The man-
dibles of all instars are similar in shape. They articulate laterally with
the head and fold together across the mouth, the ends overlapping. They
are decidedly curved, taper to points, and are brown and chitinous. The
sharp distal portions of the mandibles enlarge suddenly into a compara-
tively broad base bearing a chitinous lobe on the ventral side. The fol-
lowing average measurements will show the relative sizes of molted man-
dibles. These measurements represent the distance in a straight line,
from the tip of the mandible to the shoulder, where the mandible suddenly
enlarges into the broad basal portion.
Moit No. Length of mandible.
I o. 016 mm.
2 016 mm.
3 024 mm.
4. 032 mm.
5 048 mm.
The full-grown larva is grayish white, averaging about 3 mm. long and
0.9 mm. in diameter, with 13 body segments besides the head. There
are no tubercles on the head, but there is a row of four hairs evenly spaced
across the top. The front of the head bears a pair of hairs, one on each
side, just outside of each of which is a very short, white, conical projection,
apparently antennae. There is a short bristle near the base of each man-
dible. The mouth is chitinized along its upper edge, this brown, chiti-
nous rim extending around the bases of the mandibles and bearing six
toothlike lobes pointing downward along the portion of the edge betv/een
the mandibles (PI. LII, fig. i). A subdorsal and sublateral row of fine,
white hairs runs the full length of the body on each side, one hair per seg-
ment in each row. The first three body segments bear several additional
rows. What appears to be the anal segment is divided into a dorsal and
a ventral lobe by a transverse invagination across the end. The dorsal lobe
bears two pairs of short, fine hairs, one pair close together near each lateral
end of the lobe. The ventral lobe bears a short hair at each lateral end.
The body hairs are evidently tactile organs, since when any of them are
touched, the larva wriggles and bites viciously at the point of contact.
372 Jouryial of Agricultural Research voi. vi, no. io
Larvae of this species seem to be better equipped, more vigorous, and more
capable of defending themselves than the larvae of Micromelus subapterus
and Merisus destructor. E. allynii was reared from &gg to adult on larvae
of both the other species just mentioned as well as on the Hessian fly.
In one case, however, a newly hatched larva of E. allynii placed on a
full-grown lar\^a of M. subapterus in a glass cell was killed by the latter
almost immediately. A few instances were observed where larvae of E.
allynii killed ether individuals of the same species present in the same
Hessian-fly puparium.
THE PUPA
The lar\^a forms a naked pupa (PI. LI, fig. 3, 4) inside the puparium
of the host. The first step in the process is the excretion of all waste
matter from the body, leaving the larv^a pure white. The pupa is then
formed and the last larval skin cast off. The newly formed pupa is
nearly white, but turns dark within a few hours. The pupal stage
of 30 specimens reared in glass cells varied from 9 to 24 days. The
pupal period of those pupating in the summer averaged 13 days, while
the pupal periods of those reared late in the fall became as long as 24
days in some cases. The arrival of cold weather retards pupal develop-
ment, but whether or not the pupae are able to survive severe winter
temperatures has still to be determined. When the adult has completely
developed, the pupal skin is cast off inside the host puparium, and the
adult gnaws a round hole through the flaxseed near one end, penetrating
the leaf sheath covering the flaxseed, through which it emerges.
THE ADULT
After remaining quiet until dry, the adult becomes very active. Adults
do not seem to fly more than a few feet at a time, using their wings merely
to go from stem to stem. They do this so quickly and often that it is dif-
ficult to observe a single individual in the field very long. The females
run quickly up and down the wheat stems, vibrating their antennas
rapidly against the side of the stem until they come to a place where a
Hessian-fly puparium is located. Here they feel back and forth above
the flaxseed until they locate the exact spot which suits them for ovipo-
sition. Then, facing upward, the tip of the abdomen is bent down until
it touches the stem and raised away again, leaving the ovipositor pressed
vertically against the stem supported from its articulation with the mid-
dle ventral portion of the abdomen. The leaf sheath and puparium are
pierced by what under the microscope appears to be a sort of drilling
motion of the ovipositor, which seems to be rotated part way around and
back again. Oviposition takes several minutes.
Males placed in the same cage with females usually attempt to mate
with them at once. Mr. W. R. McConnell has ascertained that this
species can reproduce parthenogenetically. The question of the sex
jiiaes, 1916 Hessian-Fly Parasites 373
of parthenogenetic progeny has not yet been definitely settled. Mated
females produced both male and female progeny. Two mated adults
kept separately in vial cages from the time they emerged from pupae
until they died each laid a total of 58 eggs. This number actually
was found in each case by dissection of flaxseeds which had been exposed
to the adult. A few eggs may have been lost in dissection. These
adults remained alive for periods of 48 and 56 days and were oviposit-
ing during periods of 29 and 46 days, respectively. Another adult,
caught in the field while ovipositing in a flaxseed, remained alive in
a vial cage and oviposited in flaxseeds during a period of 57 days. An
unmated female was kept alive in a vial cage for 83 days. How long
adults normally live in the field is not known.
In one experiment Mr. W. H. Larrimer, of the Eureau of Entomology,
exposed stems of Elymus canadensis containing galls of Isosoma sp. to
two Eupelmus allynii females which previously had been ovipositing in
Hessian-fly puparia. They at once oviposited in the galls. The galls
were dissected and the inclosed larvae of Isosoma sp., together with the
eggs of E. allynii found in the galls, were transferred to glass-cell cages,
one larva of Isosoma sp. and one parasite e^gg to each cell. The para-
sites proceeded to complete their development to adults on the larvae
of Isosoma sp. Progeny were also bred on the Hessian fly from the same
parents used by Mr. Larrimer. These parents and their progeny were all
determined by Mr. Gahan as E. allynii.
MERISUS DESTRUCTOR
THE EGG
The Q.gg of Merisus destructor Say (PI. LI, fig. 5) is elongate, kidney-
shaped, circular in cross section, with one end smaller than the other.
It is white, wuth the surface apparently smooth. The average length of
eggs measured was 0.4 mm., the average diameter at thickest point, o.i
mm. Hundreds of the eggs were dissected from flaxseeds, in which they
had been deposited, and in all cases they were found external to the host
larva or pupa inside the puparium. Some eggs apparently bore a short
pedicel on one end, which seemed to be fastened to the inside of the host
puparium. Ordinarily, however, the eggs were found free.
M. destructor, like E. allynii, normally oviposits in the Hessian-fly
flaxseed, according to the observations of Mr. Kelly and the author at
Wellington, Kans. It was very abundant in the fields at times when
no other stage of the Hessian fly was present. The females were
repeatedly observed ovipositing in puparia in the field. In cages they
also oviposited readily in flaxseeds contained in sections of wheat stems
as well as in naked flaxseeds removed from stems. They did not ovi-
posit readily in sections of stems containing only partially grown Hessian-
fly larvae, although they seemed interested in them. In one instance,
374 Journal of Agricultural Research voi.vi.no.io
however, a female M. destructor oviposited in a stem containing nothing
but partially grown larvae. Upon dissection the egg was found sticking
to the stem underneath the leaf sheath, close to one of the larvae. It is
not yet known whether or not M. destructor can develop to maturity on
partially grown Hessian-fly lar^^ae. The egg stages of 96 specimens
placed on Hessian-fly lan-ae in glass slides varied from i }4 days in hot
July weather to 4 days in cool September weather. The larva emerges
from the egg by breaking through one end. After crawling around a
little the larvae reared in glass cells fastened themselves with their mandi-
bles to the outside of the host lar^'ae in order to feed.
THE LARVA
The full-grown larv^a of M. destructor (PI. LI I, fig. 4) is white with the
dingy brown contents of the alimentary tract visible through the integu-
ments. There are two pairs of slightly raised circular tubercles on the
front of the head near the top. The lower pair are slightly farther apart
than the upper pair and each bears a small conical projection, evidently
an antenna, varying from white to pale brown in color and about 0.02
mm. long. The median ventral surface of the head bears the round
suctorial mouth opening. The only mouth appendages distinguishable
are a pair of brown chitinous mandibles borne laterally and closing
together across the mouth with their tips overlapping (PI. LII, fig. 3).
The distal portion of the mandible is conical, tapering gradually to a
sharp point. The proximal end is suddenly enlarged, evidently to pro-
vide for muscle fastenings. One subdorsal and one sublateral row of
very short and inconspicuous setae on each side of the body are clearly
distinguishable in some specimens, extending the full length of the body,
one seta per segment in each row. On some specimens there appear to
be two ventral and two dorsal rows of scarcely discernible setae on the
first three body segments only. There are thirteen body segments be-
sides the head, the anal segment being divided into a dorsal and a ventral
lobe by a horizontal fold across the end. The dorsal lobe bears four
very short, fine setae in a row across the end, the setae composing the
row being usually in two lateial pairs. The ventral anal lobe bears only
two setae, one near each lateral end of the lobe. The length of the
full-grown larvae averages 2.5 mm., the largest diameter, 0.7 mm.
Balsam mounts of all the material left behind in the little glass cells
by pupating lar\'se nearly always contained five pairs of mandibles.
Mounts of all the material left in the cell by full-grown larvae which had
ceased to feed, together with the mandibles dissected from such larvae,
also nearly always contained five pairs of mandibles. In every mount
the pairs varied uniformly in size from those resembling the ones borne
by. newly hatched larvae to those borne by full grown larvae. Mandi-
bles of newly hatched larv'ae were somewhat hooked. All the remain-
ing pairs were similar in shape, and corresponding pairs in all the mounts
Junes, i9i6 Hessian-Fly Parasites 375
were almost identical in size. As was the case with E. allynii, the
mandibles of the newly hatched lan^a appeared to be heavier, more
powerful, and somewhat larger than the mandibles borne by the second-
instar larva. Also, the head shield appeared to be more heavily chiti-
nized in the first instar than in the later ones. Beginning with the
second instar, the successive pairs of mandibles apparently increase
fairly uniformly in size with each molt. In the mounts where five
pairs of mandibles could not be found, those which were found corre-
spond in size and shape to some one of the pairs in the complete series
and it was evident that certain pairs had been lost in making the mount.
No more than five pairs were found in any one case. All the findings
lead to the conclusion that larvae of M. destructor normally pass through
five instars in making their growth.
The relative sizes of the molted mandibles are shown below. The
measurements represent the distance in a straight line from the tip of
the mandible to the shoulder where it suddenly enlarges into the broad
base.
Molt No. Length of mandible.
I o. 014 mm.
2 014 mm.
3 020 mm.
4 024 mm.
5 032 mm.
The larvae develop readily on Hessian-fiy lar\'ae and pupae, both in
flaxseeds and in glass cells, unless the host pupa has nearly completed
its development. Several newly hatched larv^ae in flaxseeds and glass
cells containing Hessian-fly pupae which were nearly developed killed
the pupae, but died from lack of sufficient food to complete their growth.
The larvae are evidently cannibalistic upon occasion. In one flaxseed
which had been exposed to ovipositing females, a young larv^a of M.
destructor was found which had been feeding, as also the shrunken remains
of another young lar\^a. Evidently the healthy larv^a had found and
killed the other and was feeding on the Hessian-fly lar\'a when the flax-
seed was dissected. Full-grown larv'se in glass cells punctured and killed
eggs and larvae of M. destructor which were placed in the cells with them.
Larvae of M. destructor were able to become full grown by feeding on
larvae of M. suhapterus also.
The periods required by 36 larvae to make their growth when reared in
glass cells varied from 7 to 1 1 days. Cool weather appeared to make
growth slower. After becoming full-grown the majority of the lar\-ae of
M. dcstnictor reared in glass cells remained quiescent for months, though
still alive and able to wriggle vigorously when touched. Lar\'ae reared
in flaxseeds exhibited the same characteristic. In other words, the larvae
seem to have a tendency to estivate and hibernate until another warm
season before pupating. Larvae of A/, destructor were actually found to
3 7^ Journal of Agricultural Research voi. vi. No. lo
have hibernated in stubble of wheat cut the previous June. Eight per cent
of the flaxseeds in stubble gathered from a field in southeastern Kansas
in late March contained live, full-grown parasite larvae which afterwards
became adult and were determined by Mr. Gahan as M. dcstrtictor.
THE PUPA
The period from the formation of the pupa (PI. "LI, fig. 6) to the emer-
gence of the adult varied from 7 to 14 days in 2 1 specimens carried through
this stage in glass cells. Those pupating in April and September, when
cooler temperatures prevailed, took longer to develop than those which
pupated during the hot weather of July and August. The larvae form
naked pupae inside the puparium of the host. The process of pupation as
observed in glass cells begins with the excretion of all waste matter
from the body of the larva, which then becomes pure white. In a few
hours the pupa is formed. The eyes begin to turn reddish in about a
day and by the fourth day are a very dark red. The body of the pupa
is by the fourth day a creamy white, and by the sixth day the head
and thorax are black. Within another day the abdomen turns black
except for the base of the abdomen, which assumes the light brown as
found in males and some females. The emergence of the adult follows
vdthin a day or so after the pupa has turned dark. Cool weather retards
development. The adult casts off the pupal skin inside the host puparium
and emerges by gnawing a round hole through the side of the flaxseed
and the wheat leaf sheath covering it just large enough for the adult para-
site to crawl through.
THE ADULT
Adults soon become active after emerging from flaxseeds. In the
spring males emerged two or three days before the females in cages con-
taining stubble collected from the fields where it had stood during the
winter. Mating took place at once when the females emerged. Oviposi-
tion takes place in the following manner : The females run up and down the
wheat stalks, vibrating their antennae rapidly against the side of the stem.
When they come to a place where there is a flaxseed underneath the leaf
sheath, they stop and excitedly feel up and down over the place where
the flaxseed is located. They face upward to oviposit, with the body
parallel to the puparium. They locate the proper place for oviposition
with the tip of the abdomen and then raise it away from the stem, leaving
the ovipositor unsheathed and pointing perpendicularly against the stem
from its articulation with the middle of the abdomen. In less than a
minute the ovipositor is forced through the leaf sheath and the puparium.
In penetrating the flaxseed the ovipositor is seemingly rotated like a drill
part way round and back again. Oviposition takes 5 to 10 minutes, and
dissections of flaxseeds indicate that a single egg is laid at a time. One
female kept isolated in a vial cage laid a total of 39 eggs in puparia exposed
Junes. i9t6 Hessian-Fly Parasites ^'j'j
to her and later dissected. Some may have been lost in dissection. This
female was laying eggs during a period of six weeks. Other females were
kept alive and active in confinement for periods of over two months.
Some stems of Elymus canadensis containing galls formed by a species
of Isosoma were placed in a vial cage containing females of M. destrtcctor.
Almost immediately one of the females became interested in the galls,
feeling over them with her antennae. She then attempted to oviposit,
endeavoring persistently to penetrate the gall with her ovipositor, but
without success. Mr. W. H. Larrimer finally succeeded in getting the
females to oviposit in the Isosoma galls and found the eggs inside the
galls but external to the larvas of Isosoma sp. He actually reared a few
specimens of M. destructor from egg to adult on the Isosoma larvae in glass
cells. The parents used in this experiment and the progeny which were
reared were determined as Merisus destructor by Mr. Gahan.
MICROMELUS SUBAPTERUS
Heretofore it has been uncertain that the winged and wingless forms
of Micromelus suhaptcrus Riley were the same species. It has been
proved, however, that the two forms are specifically identical by breeding
a wingless female from a winged parent. Further evidence indicating
that the winged and wingless forms are the same species is the fact that
wingless males mated with winged females as readily as with the wing-
less form. The method by which the wingless female was bred from the
winged parent is as follows: The winged parent deposited an egg in a
Hessian-fly puparium known to have been previously unparasitized. The
t.gg was removed from the puparium and from it a wingless adult was
reared on a healthy Hessian-fly larva, which also had been dissected from
its puparium. Mr. Gahan found this wingless offspring of a winged adult
to be identical with winged specimens of unknown parentage.
THE EGG
The egg of Micromelus subapterus (PI. LI, fig. 7) resembles that of Merisus
destructor in size and shape. It is elongate, kidney-shaped, with one end
longer than the other, circular in cross section, white in color, with sur-
face of shell smooth, and about 0.38 mm. long by 0.09 mm. in diameter
at the thickest point. It has no stalk.
All the observations made at the Wellington (Kans.) station lead to
the conclusion that the egg is normally laid in the Hessian-fly puparium.
In cages the adults oviposit readily in flaxseeds, the eggs being placed
inside the puparia but external to the inclosed Hessian-fly larvae and
unattached. This was the case both when stems of fly-infested wheat
1 Mr. A. B. Gahan makes the following statement: " The real generic position of this species is in doubt.
It was originally described by Riley under the name Merisus (Ilomoporus) subapterus Riley, and later
referred to Boeotomus by Osborn and other writers. N. V. Kurdiumov has more recently placed the
species in the genus Micromelus. Doctor Ashmead reduced Boeotomus to synonymy with Micromelus."
• 37767°— 16 4
378 Journal of Agricultural Research voi. vi, no. 10
were exposed to the parasite in vial cages and when ovipositing females
were placed in large glass chimneys containing growing wheat infested
with the Hessian fly. Occasionally the females have been observ^ed appar-
ently to oviposit in stems containing only larvge ; and although careful
dissections of these stems were made, no eggs were found. Further proof
that M. suhapterus normally oviposits in flaxseeds was obtained by dis-
secting puparia collected in fields where this parasite was numerous at
the time the collection was made. Both eggs and young lars'ae of a parasite
were present in the flaxseeds and when reared to maturity in the labora-
tory were found to be M. suhapterus. The t^gg stage in 1 19 cases varied
from I >^ to 5 days. Low temperatures in fall and spring retarded embry-
onic development. The larvse reared in glass cells emerged from the
eggshells by breaking through one end, and after crawling around a short
time settled down in one place to feed.
THU L.'VRVA
The full-grown larva of M. suhapterus (PI. II, fig. 6) averages 2 mm.
long by 0.75 mm. in thickness. It is white, with the pale-brown con-
tents of the alimentary tract showing through the body. There are two
pairs of slightly raised circular tubercles on the front of the head near
the top. The lower pair are slightly farther apart than the upper pair,
the former each bearing a small conical projection, evidently the an-
tennae, varying from white to brown and about 0.015 "^"i- long. The
median ventral surface of the head bears the round, suctorial mouth
opening. The only mouth appendages distinguishable are a pair of
very small brown chitinous mandibles borne laterally and closing to-
gether across the mouth (PI. LII, fig. 5). The distal ends of the man-
dibles are sharp and needlelike. The proximal ends are suddenly
enlarged, evidently to provide for muscle fastenings. A minute pit,
which sometimes appears to have a brown center, occurs on each side
of the mouth. The body is entirely glabrous, so far as could be deter-
mined, except for the anal segment, oval in shape, with the anal end
the more pointed. There are 13 segments besides the head, the anal
segment being divided into a dorsal and ventral lobe by a horizontal
fold across the end. The dorsal lobe bears four short, very fine setae in
a transverse row, these usually being in lateral pairs. The ventral anal
lobe bears only two very short, fine setae, one near each lateral end of
the lobe.
The number of instars passed through by larvae of M. suhapterus in
making their growth appeared to be five. Five pairs of molted mandi-
bles increasing uniformly in size, from the small pair resembling those
borne by newly hatched larvae to the large pair molted off when full-
grown lar\^ae pupated, were present in almost eveiy mount made of
the material left behind in a cell where a larva had developed. In
Junes. i9i6 Hessian-Fly Parasites 379
mounts where five pairs could not be found, each of those present cor-
responded to some one of the pairs in the complete series. No more
than five pairs were found in a single mount. As in the two species
of parasites previously discussed, the head shield of the newly hatched
larva was more heavily chitinized than those of later instars. The
mandibles appeared to be more powerful for their size than those of
any later instar, and in some cases they were actually larger than the
second-instar mandibles. The approximate sizes of the respective pairs
of molted mandibles follow. The measurements represent the distance
from the tip of the mandible to the shoulder where it suddenly enlarges
into the broad base.
Molt No. Length of mandible.
I o. 012 mm.
2 012 mm.
3 016 mm.
4 020 mm.
5 028 mm.
Larvae of Micromelus subapierus do not seem as capable of moving
around and reattaching themselves to the host as are the larvae of Eupel-
mus allynii and Merisus destructor. Larvae reared in glass cells crawled
about a little immediately after hatching before they settled down to
feed, but they usually completed a large part of their growth without
leaving the original feeding point on the external surface of the host.
This species not only developed on Hessian-fly lar\^ae in puparia, but in
some instances fed on the lar\^£e of other parasites. One egg of Micromelus
subapierus was placed on a full-grown lar\-a of the same species in a glass
cell. The egg hatched and the little larva became full grown on the large
larva, almost completely devouring it. Another egg of M. subapierus was
placed on a full-grown larva of Merisus destructor and the little larva
hatching from the egg became full grown on the larva of Merisus destruc-
tor. Experiments like these, however, usually resulted in the destruction
of the egg or young larva of M. subapterus and the survival of the full-
grown larva of the same or the other species as the case happened to be.
Larvae of M. subapterus apparently could make their growth on the
Hessian-fly pupa as well as on the larva unless the former had partially
developed. Where the host pupa had already completed a large part of
its development, both the host and the parasite generally died, the latter
apparently for lack of sufficient suitable food. Larv^ae of M. subapterus
appeared to be the least able to defend themselves where the larv^ae of
more than one species occurred in the same flaxseed. They also seemed
the least capable of successfully establishing a feeding point on the host
larva, at least when reared in little glass cells. They seemed more delicate
in structure and less vigorous.
The respective periods required for 36 larvae to make their growth
varied from 7 to 10 days. A large proportion of the larvae after finishing
380 Journal of Agricultural Research voi. vi, no. 10
their growth remained in a quiescent state in the little glass cells for
months. Others pupated at once upon completing their growth.
THE PUPA
In general, the process of pupation as observed in glass cells is as follows :
The full-grown larva excretes all waste matter from the body, leaving it
perfectly white. Within a day after this operation the pupa (PI. LI,
fig. 8) is formed and is at first perfectly white, the last larval skin being
found at the anal end of the pupa. In another day or so the pupa begins
to turn a pale brown, and the eyes turn reddish. The pupa finally
becomes entirely black as development progresses, the head and thorax
changing first, and remains so until the adult emerges.
The pupa is formed naked inside the puparium of the host. The adult
emerges by casting off the pupal skin inside the host puparium and then
cutting a round hole through the side of the flaxseed near one end. The
length of the pupal period varied in 21 instances from 7 to 13 days. Cool
weather retarded the development of the pupae. A larger proportion of
the larvae reared in the cooler weather of fall pupated at once upon attain-
ing their growth than was the case with the larvae reared in the hot
weather of midsummer, indicating a tendency of the larvae of this species
to estivate.
THE ADULT
Newly emerged adults became active almost at once upon emerging
from the host puparium. Males placed in the same cage with females
began mating at once. Females that had been mated seemed to oviposit
more readily than unmated females. Both Mr. McConnell and the writer
found that this species was arrhenotokous in every instance where this
point v/as determined. Females have been kept alive in cages as long
as six months, and one female oviposited after having been kept alive
over five months. It was usual for them to live and oviposit for at least
a month in vial cages. One female actively oviposited during a period of
75 days and laid a total of 103 eggs. Another female laid a total of 45
eggs. The number of eggs laid by a single female was determined by
exposing flaxseeds to an isolated individual and dissecting them to find
the number of eggs the parasite had laid in each.
In ovipositing the female would run up and down the stems of the
plants, vibrating her antennae against the surface. When she came to a
place in the stem where a flaxseed was located, she would stop, feel up
and down over the spot with her antennae, and then lower the tip of her
abdomen. When she had found the point that suited her for oviposition,
the end of the abdomen was raised, leaving the ovipositor standing
vertically against the side of the stem from its articulation with the
middle of the abdomen. In penetrating the leaf sheath and puparium
the parasite seemed to rotate the ovipositor with a drilling motion in
Junes, i9i6 Hessimi-Fly PavasUes 381
addition to the downward pressure exerted on it. The female always
took a position heading up the stem in ovipositing. The whole process
generally took five minutes or more.
CONCLUSION
The writer's experiments and observations have all led to the inference
that only one specimen of any of the three species studied ever matures
in a single Hessian-fly puparium. In every instance where more than
one egg or larva was placed on the same host or in the same cell, one
survived and the rest were killed by that one, or starved to death. This
was true whether the two or more larvae were of the same or different
species.*
1 For correct figures of the adults of all three of the species treated in this paper, see U. S. Dept. Agr.
Fanners' Bui. 640. (Webster, F. M. The Hessian fly. 20 p., 17 fig. 1915.)
PLATE LI
Fig. I. — Egg of Eupelmus allynii.
Fig. 2. — Egg of Eupelmus allynii in situ.
Fig. 2>, 4- — Pupa of Eupelmus allynii.
Fig. 5. — Egg of Merisus destructor.
Fig. 6. — Pupa of Merisus destructor.
Fig. 7. — Egg of Micromelus subapterus.
Fig. 8. — Pupa of Microm,elus subapterus.
(382)
Hessian-Fly Parasites
Plate LI
Journal of Agricultural Research
4
Vol. VI, No. 10
Hessian-Fly Parasites
Plate LI I
3
Journal of Agricultural Research
Vol. VI, No. 10
Fig. I
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
PLATE LII
— Mandibles of full-grown larva of Eupelmus allynii.
— Larva of Eupelmus allynii.
— Mandibles of full-grown larv'a of Merisus destructor.
— Larv^a of Merisus destructor.
— Mandibles of full-grown larva of Micromelus subapterus.
— Larva of Micromelus subapterus.
ADDITIONAL COPIES
OF THIS PUBLICATION MAT BE PKOCURED FROM
THE Sin-EMINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
20 CENTS PER COPY
Subscription Price, $-3 Per Year
V
Vol. VI June: 12, iQi<. no^ ii
JOURNAL O
RESEARCH
CONTENTS
Page
Effect of Rontgen Says on the Tobacco, or Cigarette, Beetle
and the Results of Experiments with a New Form of
Rontgen Tube - - --- - _ - 383
G. A. RUIWER
Stimulating Influence of Arsenic upon the Nitrogen-Fixing
Organisms of the Soil - - - - - - - 389
J. E. GREAVES
Transmission and Control of Bacterial Wilt of Cucurbits - 417
FREDERICK V. RAND and ELLA M. A. ENLOWS
DEPARTMENT OF AGRICULTURE
WASHINGTON, D.C
\k\
WAeHINOTON : GOVERNMENT PRINTING OFFICE : l«t»
mm
PUBLISHED BY AUTHORITY OF THE SECRETARY
OE AGRICUTTURE, WITH THE COOPERATION
OE THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Physiologist and Assistant Chief. Bureau
of Plant Industry '
EDWIN W. ALLEN
Chief, Office of Experittient Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL
Biohgiil, Maine Agricultural Experiment Station
H. P. ARJklSBY
Director, Instiliife of A nimal Nutrition, The Penn-
syliiania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant Dean,
Agricultural Experiment Station of the Univer-
sity of Mmnciofa
AU correspondence regarding articles from the Department of Agriculture
should be addressed to Earl F. Kellerman, Journal of Agricultural Research,
Washington, D. C.
All correspondence regarding articles from Experiment Stations should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orotio, Maine.
JOim OF AGRIdTURAL RESEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, June 12, 1916 No. 11
EFFECT OF RONTGEN RAYS ON THE TOBACCO, OR
CIGARETTE, BEETLE AND THE RESULTS OF EX-
PERIMENTS WITH A NEW FORM OF RONTGEN TUBE
By G. A. Runner,
Entomological Assistant, Southern Field Crop Insect Investigations,
Bureau of Entomology
INTRODUCTION
The Rontgen tube used in experiments on the efifect of Rontgen rays
on the tobacco, or cigarette, beetle (Lasioderma serricorne Fabricius)
described in this paper is a new form designed by Coolidge.^ By this type
of tube a much more powerful Rontgen-ray radiation can be maintained
than was possible with the apparatus used in experiments of a similar na-
ture previously made by the writer. The intensity and the penetrating
power of the Rontgen rays produced are both under the complete control
of the operator, and many of the factors limiting the use of other types
of tubes for the special purpose desired are absent. The tube can be
operated continuously for long periods without showing an appreciable
change in either the intensity or the penetrating power of its resulting
radiation. The starting and running voltage are the same. The resulting
radiation is therefore homogeneous and of any desired penetrating power.
The ordinary forms of tubes used in previous experiments were incapa-
ble of being operated continuously without change in penetrating power.
Owing to the fluctuation in intensity and penetrating power incidental
to frequent adjustment, it was impossible to tell with any degree of
accuracy the dosage and amount of radiation.
In previous experimental work with Rontgen rays it had been found
that in sterilizing cigars or tobacco, small dosages are ineffective, from
a practical standpoint. To be effective, the radiation must be intense,
and it is evident that if the process can be successfully applied to com-
mercial work, the apparatus used must be capable of producing and
ma'ntaining such radiation during the entire period required for the
material treated to pass through the exposure chamber of the machine.
' Coolidge, W. D. A powerful Rontgen ray tube with a pure electron discharge. In Phys. Rev., s. 2.
V. 2, no. 6, p. 409-430, 6 fig. 1913.
Journal of Agricultural Research, Vol. VI, No. 11
Dcpt. of Agriculture, Washington, D. C. June 12, 1916
^" /ON ^-J5
(383)
384
Journal of Agricultural Research
Vol. VI, No. II
8 ^
^
^
it V
"'a
03
M c4
1 1 <"
t^ 0,-S s
;i-"S
;S^'
§1
.a>.
63
5«
••d
B..2
> i!
■OtJ
a a ^
g Ji
5 u " c
w >; S n
0^ u
n 01 2 a
- C 3
ja o y15 o
§ -go
d 6 6 d 6 d_,'2'2.Hfd di; W S H
« ^ w H
,-,-v. oooooooo
BT3 a "O "O T3 -O "O "O "O Tl
«
•"oo oooooo O-S
w-O-O -O "O -O T5 TS "0 "C M
9 M
o ■ • 0)
^ : : :::::::"
>o o 1-1 ;:::::::
c c >, 66666666
a o, ts -a -o t3 -o Tf "O "o "0
ooo oooooooo
wPwQ
2
(A U
ex .a
2_
■5^
3-^
•O <n33
0)3.
§3 §3 3 5^3 &
oooooooo
oooooooo
W : :
!_■ I-' O u O O O u
V mo t»00 Ov O ~
June 12, 1916
Effect of Rontgen Rays on Tobacco Beetle
385
s>^ "Sis
•O— m a- g 5j Q tj
£ (u_ 3 a & «-s «,
""jS rt CO --s M
3 « > .SJ3 4)
•" aj3 a = -5ja a
■—' V
: : : : : 0! ^
K 4) 41 4) 41 U ^
_>•>.>_> > ^ "a
o> m •» « « •< <;
a *
82
. ^
— £f .
a-d.s
o a 3
^%^
o.-^ 5; 5^ a ?B ?«
!r,o04jw«wi-»3
a^ >;
1.4 ■<
1^ 5
■a
386 Journal of Agricultural Research voi.vi.No.it
EXPERIMENTAL WORK
Eggs for the experiments were obtained by placing large numbers of
tobacco beetles in jars containing leaf tobacco which had been sterilized
by heat. The eggs were then placed between slabs of chewing tobacco
in wooden boxes. The covers of the boxes were tightly sealed with
adhesive tape. Control boxes containing approximately the same
number of eggs as the treated boxes were prepared in a similar manner-
Infested tobacco containing larvae, pupae, and adults was also exposed
in sealed wooden boxes. After exposure the insects were transferred to
wooden boxes containing granulated tobacco which had been sterilized
by heat, A corresponding number of specimens were kept as controls.
Exposure to the rays was made by placing the containers directly
under the Rontgen tube at a distance of 7.5 inches from its focal spot.
In order to guard against any efifect of heat, a fan was kept blowing on
the container while the exposure was made. The maximum temperature
registered by a thermometer placed in the chamber was 91° F.
In the series of experiments tabulated 150 milliampere minutes (current
of 15 milliamperes for 10 minutes or a current of 10 milliamperes for 15
minutes), with a voltage of 65,000, was the minimum dosage applied.
The material used in the experiments was kept under observation until
January 10, 191 6. Table I gives the details of the experiments. The
notes included show the condition at different times. During the colder
months the material was kept in an automatically regulated electric
incubator in which suitable breeding conditions were maintained. The
temperature was kept at 86° F. and the humidity at 80.
Eggs from exposed beetles were kept under daily observation. Part
were kept in cells on microscope slides and part were kept on the leaf
tobacco on which they were laid and placed between slabs of chewing
tobacco. Most of the eggs which failed to hatch became shrunken and
changed in color in about 10 days. Part remained plump and apparently
normal for a considerable time. In eggs which were over 2 days old
and in which embryonic development was well advanced when treated
the partly developed larvae could be seen within by examination with a
microscope.
As will be seen in Table I (experiments 11, 14, and 18), hatching took
place in some of the eggs which were over 3 days old. In experiment 14,
which was made with eggs nearly hatched when treated, part of the
eggs hatched, even though the dosage of 150 milliampere minutes, which
was effective with the newly laid eggs, had been increased to 600 milli-
ampere minutes.
Results of previous experiments, as well as those tabulated, indicate
that in treatment of the egg stage heavier dosages are required to sterilize
eggs which are nearing the end of the incubation period than are required
to sterilize eggs newly laid.
June 12. 1916 Effect of Rontgen Rays on Tobacco Beetle 387
In these experiments the larvae hatched from treated eggs failed to
develop. In several other series of experiments with Rontgen rays made
by the writer and also in experiments made by Morgan and the writer/
eggs given lighter dosage hatched and development seemed normal,
several generations of tobacco beetles being reared from some of the
tobacco and cigars which contained treated eggs.
In the two experiments with larvae (No. 19 and 20), no immediate
effect as the result of exposure to the rays was noted. After a time the
larvae became inactive, somewhat shrunken, and changed in color, and
no evidence of feeding could be observed. Nearly all remained in an
inactive or dormant condition for long periods before death. Two larvse
exposed on June 7 (experiment 20) remained alive until January 10, 19 16.
All check larvae used in this experiment had transformed to the adult
stage by July 11. All treated larvae died before reaching the pupal
stage. With conditions under which the material used in the experi-
ments was kept, the normal larval period of the tobacco beetle is about
40 days. All larvae used in the experiments were partly grown when the
experiment was made. No further growth could be noticed. In general,
the effect of the heavy exposure given (600 milliampere minutes, voltage
65,000, distance from focal spot of Rontgen tube 7.5 inches) seems to
have been to stop development and activity and to produce an inactive
or dormant condition, and greatly to prolong the larval period.
The results of all previous experiments with larvae given comparatively
light exposures had shown entirely negative results.
In the experiment with pupae (experiment 21) the number of pupae
used was not sufficiently large to permit the drawing of positive con-
clusions. Of the 20 specimens treated, only 4 reached the adult stage.
These seemed normal, but died without laying eggs.
In the two experiments with adults (experiments 22 and 23), the
results obtained were very similar. The exposure given apparently did
not affect the length of life or the activity. Mating was observ^ed and
large numbers of eggs were laid. None of the eggs from the exposed
beetles hatched, while eggs from the check beetles hatched normally.
Egg clusters of the tent caterpillar (Malacosoma americana Fabricius)
and the white-marked tussock moth (Noiolophus leticostigma Smith and
Abbot) were used. With both of these species the period of incubation
is very long, eggs deposited in summer or fall not hatching until the
following season. An exposure of 150 milliampere minutes was given.
Other conditions of the experiment were the same as in experiment 7
made with eggs of the tobacco beetle, details of which are given in
Table I. The experiment was made on April 16. The egg clusters
treated contained something over 1,000 eggs of each species. The same
number of clusters were kept as checks. Both experiments gave nega-
' Morgan, A. C, and Runner, G. A. Some experiments with Rontgen rays upon the cigarette beetle
l,asioderma serricome Fabr. /» Jour. Econ. Ent., v. 6, no. 2, p. 226-230. 191J.
388 Journal of Agricultural Research voi. vi, no. h
tive results, hatching being apparently normal in treated eggs of both
species.
The eggs of both the tent caterpillar and the tussock moth were nearing
the end of the incubation period when treated. In eggs of the tent
caterpillar embryonic development is practically completed in the fall,
the larvae remaining in the eggshells over the winter and emerging on
the appearance of warm weather in the spring.
SUMMARY
Under laboratory conditions tests made with a Rontgen-ray tube per-
mitting a high-energy input and giving an intense and powerful radiation
gave results which promise that the X-ray process may be successfully
used in treatment of cigars or tobacco infested with the tobacco, or ciga-
rette, beetle.
Heavy dosages must be given, as is indicated by the exposure given in
the series of experiments tabulated in this paper.
In treatment of the Q.gg stage, heavier exposures are required to sterilize
eggs which are near the hatching point than are required to sterilize eggs
newly laid.
In experiments performed by the writer a dosage equivalent to 150
milHampere minutes exposure with a spark gap of 5.5 inches gave satis-
factory results with eggs in tobacco placed 7.5 inches from the focal spot
of the tube. With this exposure the eggs in which embryonic develop-
ment was well advanced hatched, but in all cases where these larvae were
kept under observation they failed to reach the adult stage.
The minimum lethal dosage at a given distance from the focal spot of
the Rontgen tube used has not been determined.
In two separate experiments adults were given an exposure of 600
milliampere minutes (amperage X time), with a spark gap of 5.5 inches,
giving an approximate voltage of 65,000, with humidity at ^j. The
distance from the focal spot of the Rontgen tube was 7.5 inches. The
results are as follows :
(i) No effect on length of life was apparent, as the beetles died at about
the same rate as the same number of beetles kept as a check.
(2) Large numbers of eggs were deposited after exposure. These eggs
were infertile. Eggs laid by the check beetles hatched normally.
Larvae were given an exposure of 600 milliampere minutes, other con-
ditions of the experiment being the same as in the experiments with
adults given above. While no immediate effect was apparent, the treat-
ment had the effect of stopping activity and development, the larvae
remaining in a dormant condition for a prolonged period. All treated
larvae died before reaching the pupal stage.
STIMULATING INFLUENCE OF ARSENIC UPON THE
NITROGEN-FIXING ORGANISMS OF THE SOIL
By J. E. Greaves,
Bacteriologist, Utah Agricultural Experiment Station
INTRODUCTION
Arsenic, when applied to a soil, has been found to stimulate the
ammonifying (Greaves, 1913c)* and especially the nitrifying organisms
of that soil. The stimulation varied greatly with the form, quantity,
and method of applying the arsenic. Furthermore it was found that
very large quantities of arsenic had to be applied to a soil before its
toxic effect became marked. This toxic effect became pronounced
only when quantities of arsenic which far exceeded those found in any
of the cultivated soils (Greaves, 1913b) had been applied. Therefore
it was desirable to determine its influence and mode of action upon the
nitrogen-fixing powers of the soil. For, even though arsenic does not
inhibit the action of the ammonifiers or nitrifiers, if it stops or materially
retards the nitrogen-fixing organism, it can not be said that arsenic is not
injurious to the soil flora. To determine this point the following study
has been made.
EXPERIMENTAL WORK
The soil used in the first part of this work was the same as that used
by the author in the previous series. It is a typical bench soil, a
sandy loam fairly high in calcium and iron content and supplied with
an abundance of all the essential elements of plant food with the excep-
tion of nitrogen, which was low, a characteristic of arid soils.
The determination of the nitrogen-fixing powers of the soil was made
as follows: Tumblers covered with Petri dishes were sterilized, and into
these were weighed loo-gm. portions of the air-dried soil and 2 gm. of
mannite, which were then carefully mixed. Sodium arsenate was added
from a standard solution with the proper proportion of sterile distilled
water and the mixture thoroughly stirred with a sterile spatula. The
other arsenical compounds were added in the dry state and then care-
fully mixed. Sufficient sterile distilled water was added to make the
moisture content of the soil 18 per cent. The tumblers and contents
were weighed and the moisture content made up weekly to the initial
concentration.
' Bibliographic citations in parentheses refer to " Literature cited," p. 414-416.
Journal of Agricultural Research, Vol. VI, No. 11
Dept. of Agriculture, Washington, D. C. June 12, 1916
dw Utah— a
(389)
390
Journal of Agricultural Research
Vol. VI, No. II
The samples were incubated at 28° to 30° C. for 18 days and the total
nitrogen determined. The tumblers and contents at the end of this
time were placed in an electric incubator and kept at 95° C. until dry.
The soil was then ground in a mortar, after which 20-gm. portions were
weighed and placed in Kjeldahl JBasks. The nitrogen was then deter-
mined according to the Lipman and Sharp (191 2) method. The deter-
minations were all made in duplicate and compared with sterile blanks,
so that each result reported is the average of two or more closely agree-
ing determinations. The compounds used were sodium arsenate, lead
arsenate, cupric aceto-arsenite (Paris green), arsenic trisulphid, and zinc
arsenite. In each case the quantity of the compound added was such
as to give equivalent quantities of arsenic. The results reported as
milligrams of nitrogen per 100 gm. of soil are given in Table I.
Table I. — Quantity of nitrogen {milligrams) fixed in loo gin. of soil during 18 days -with
varying amoxmts and different forms of arsenic
P. p. m.
O
20
40
80
120
160
200
240
280
320
360
400
O
Sodium
Lead
Paris
Arsenic
arsenate.
arsenate.
green.
trisulphid.
18.2
16. I
15.22
9.8
22.4
16. 0
13-72
II. 2
14. 0
16. 4
13.02
14. 0
14. 0
18.9
14. 00
15-4
15-0
21. 0
8.82
16. 2
14.4
21. 0
8.32
16. 4
14. 0
21.7
7.42
14. 0
12. 6
16.8
6.72
12.8
0
16. I
6. 02
II. 2
0
16. 0
6. 00
II. 2
0
16.6
6. 02
9.8
0
16.8
5.22
9.8
18.2
16. I
15. 22
9.8
Zinc
arsenite.
9. I
II. 9
9-7
9.6
10.5
9-7
8.4
8.4
8.4
9.0
9.1
9.1
9.1
In this series the concentration of the arsenic was not carried above
400 p. p. m., for previous work had shown that the main stimulation
occurs below this concentration. Furthermore the arsenic occurring in
agricultural soils seldom exceeds 150 p. p. m., so it is likely that in agri-
cultural soils it will never be found to exceed the quantity used in this
work.
The results reported in the above table bring out some very inter-
esting facts and show that the nitrogen-fixing organisms are very similar
to the nitrifying organisms in so far as their relations to arsenic are
concerned. The addition of 20 p. p. m. of sodium arsenate stimulates
their action and 40 p. p. m. or more have a toxic influence. When the
concentration of arsenic reaches 280 p. p. m., it stops all nitrogen-fixing
activity. The toxic influence which becomes so very prominent above
this concentration must be due entirely to the arsenic and not to the
sodium ion, as Lipman and Sharp (191 2) have added many times this
junei2, i9i6 Influence of Arsenic up07i Soil Organisms 391
quantity of sodium in the form of sulphates, chlorids, and carbonates to
the soil without retarding its nitrogen-fixing power.
The lead arsenate at the lower concentrations has no influence upon
the nitrogen-fixing powers of the soil, but when the concentration
reaches 80 p. p. m. a stimulating influence becomes quite perceptible.
This continues until the concentration exceeds 200 p. p. m. Above this
concentration the nitrogen fixed, within experimental error, is the same
as that fixed in the untreated soil. It is interesting to note that the
compound does not become toxic, even when the quantity added reaches
400 parts of arsenic per million parts of soil. This series shows a very
close similarity to the nitrification series previously reported, and it is
quite likely that part of the stimulating influence is due to the lead ion.
Paris green is toxic even in the lowest concentration used, and the
toxicity increases as the quantity of Paris green added increases. This
toxicity is due mainly to the copper ion. However, as was shown in the
ammonification and nitrification work, the quantity of soluble arsenic
present would be much higher where the Paris green was added than
where the other compounds were used. The fact that no stimulation
occurs in the Paris-green series points to the conclusion that the toxicity
of the copper must increase much more rapidly than the stimulating
influence of the arsenic. Yet it is quite possible that if a lower con-
centration of the substance had been taken a stimulation would have
been noted.
Arsenic trisulphid stimulates in the lowest concentration tested and
increases in stimulating influence until a concentration of 160 p. p. m.
is reached. In concentrations above this its stimulating influence de-
creases. In concentration above 320 p. p. m. there is fixed no more
nitrogen in the presence than in the absence of arsenic. But even at
the highest concentration tested (400 p. p. m.) this compound exerts no
tonic influence on the nitrogen fixers.
Zinc arsenite probably stimulates slightly in low concentrations, but
aside from this it has little apparent influence on the nitrogen-gathering
organisms. Had fresh soil been used in this series, greater stimulation
would have been noted, as was found by later work.
The amount of nitrogen fixed in the untreated soil of the above series
shows a marked variation. This is probably due to various factors, chief
among which is the fact that the nitrogen-fixing powers of the soil with
sodium arsenate, lead arsenate, and Paris green were made in the order
named on the air-dried soil soon after it had been brought to the labora-
tory. In the case of the arsenic trisulphid and zinc arsenite the soil had
been in the laboratory in an air-dried condition for about two months
before the determinations were made, but each set of samples within each
series was handled in exactly the same manner, and the samples are
directly comparable within each set, as has been the case in the previous
392
Journal of Agricultural Research
Vol. VI, No. II
discussion. In order to make those containing different forms of arsenic
more nearly comparable with each other — that is, the lead arsenate with
the arsenic trisulphid, etc. — the nitrogen fixed in the untreated soil has
been taken as loo, and from this the ratio has been calculated with each
of the concentrations of arsenic. This gives us more nearly comparable
results, which are shown in figure i.
Comparing these results with those obtained for the ammonification
and nitrification series (Greaves, 1913c), we find a marked similarity exist-
ing between them. In all of the series there is a marked stimulation
with all of the compounds except Paris green. The arsenic trisulphid
/so
/30
^
r=
==^
\
y
^^'
\
\
^
^
V'
\
fc^.
' /
-t.
1^
N
I
Y
,^^'
e^'
'^,
\
ilk
a\
I
,#^
r
t
'^
■^J-nsv,
//O
^<§>>».>.
S.^
F
\'
r'
— '
^
\
N
I- —
S
N
\^
\
_-— '
I
N
^^
' "
'■— '
^
&0
eo
i
\
--**
,--
—
h
hir 1
'
-^
V
■^^
h
^.
7^
GO
^0
•^o
30
20
/o
0
\
1
\
1
"-^
:
'"*^
^^.
\
^
'^ii
SiL
^
\
^-^
-^
\
\
\
O A?0 ^OO 300 ^tTOO
Fig. I. — Graph showing the action of five compounds of arsenic on nitrogen fixation in dry soil. (Nitrogen
fixed in untreated soil equals loo.)
stimulates growth much more in the nitrogen-fixing series than it does
in the other series. The arsenic trisulphid has the greatest stimulating
influence, followed in the order named by lead arsenate, zinc arsenite,
and sodium arsenate. Paris green was the only compound tested which
exerted no stimulating influence. It may be seen that the maximum
stimulation was not obtained when equivalent quantities of arsenic in
the various forms are applied to the soil. Hence, it seems possible that
a relationship may exist among the various cases in the water-soluble
arsenic found. In order to answer this, determinations were made of
the water-soluble arsenic existing in the soil. The soil and arsenic,
juneij, i9i6 Influence of Arsenic upon Soil Organisms
393
together with 2 gm. of mannite, were placed in sterile tumblers, the water
content made up to 18 per cent, and then incubated at 28° C. for 18 days.
At the end of this period the soil was transferred by means of i ,000 c. c. of
carbon-dioxid-free distilled water to large acid bottles. The mixture
was left in these bottles, with occasional shaking, for 8 days, then filtered
and the arsenic determined in an aliquot part (Greaves, 1913d). In
another set the various forms of arsenic were mixed with loo-gm. portions
of soil and 2 gm. of mannite and the water-soluble arsenic determined as
above without incubation.
The results are given in Table II as milligrams of water-soluble arsenic
occurring in 100 gm. of the soil both before and after the three weeks'
incubation. Each reported result is the average of three or more closely
agreeing determinations.
Table II. — Quantity of water-soluble arsenic (in milligrams) in 100 gm. of soil before
and after three vjeeks' incubation
Treatment.
Lead arsenate.
Arsenic trisul-
phid.
Sodium arse-
nate.
Arsenic added
16. GO
I. 04
1.26
16. 00
.14
1.42
2. CO
Arsenic found before incubation
1.08
Arsenic found after incubation
T '\'\
Average
I- 15
.78
I. 26
The arsenic in each case became more soluble as bacterial activity
progressed. This is especially marked in the soil containing arsenic
trisulphid, which yielded 10 times the water-soluble arsenic after incuba-
tion that it did before. A remarkably close agreement is found to exist
among the results obtained for water-soluble arsenic at the close of the
incubation period, which shows that the maximum stimulating influ-
ence is obtained when soil contains between 10 and 15 p. p. m. of
water-soluble arsenic. This is a quantity that exceeds that found in
agricultural soil (Greaves, 1913b); hence, the influence of the arsenic
occurring in soil must be to increase and not to retard nitrogen fixation.
The maximum fixation varies with the form of arsenic applied. This is
undoubtedly due, as will be pointed out later, to the elements accom-
panying the arsenic, which may have either a retarding or an accelerat-
ing influence upon the bacterial activity.
The finding of this marked stimulating influence of arsenic upon the
nitrogen-fixing powers of soil raises a number of very interesting and
important questions. Some of these are: (i) Does this stimulating
influence exist in other soil or is there something inherent within this
particular soil which makes its bacterial flora susceptible to the influence
of arsenic? (2) Is the stimulating influence brought about by the retard-
ing of injurious species or is it a direct stimulant to the soil organisms?
394 Journal of Agricultural Research voi. vi, no. n
(3) Do the arsenic and arsenic compounds act as a source of energy to
the nitrogen-fixing organisms or do they so influence the soil flora that
it can utilize more economically the carbon compounds available? (4)
What nitrogen-fixing organisms are there in the soil which are influenced
by arsenic ?
In order to find whether arsenic influences the nitrogen-fixing powers
of other soils in a similar manner, three other soils were tested with and
without arsenic. The soils vary greatly in chemical and physical com-
position. Soil A is a black loam of very light texture and, for an arid
soil, high in nitrogen and humus. It is well supplied with phosphorus,
potassium, and calcium carbonate and grew potatoes for 23 years.
After this it was planted to oats for 2 years, and during the past 4 years
has been planted in alfalfa. It has received some manure. Soil B is a
sandy loam of much lighter color than soil A and contained much less
humus and nitrogen, but an abundance of other elements. It has been
cultivated for 28 years and during this time has been fallowed two sum-
mers. The remainder of the time it has been planted in wheat. Soil C
is a heavy clay almost devoid of humus. The nitrogen is low, but the
soil is well supplied with phosphorus, potassium, and calcium carbonate.
While wet it is exceedingly sticky, and on drying it bakes like adobe. It
has been tilled for 23 years, and during this time it has been fallowed for
3 years. The remainder of the time it has been in wheat. While it has
received no manure during this time, it is still very productive. All of
the soils are very fertile and well supplied with Azotobacter, and previous
work has shown them to have high nitrogen-fixing powers.
The soils were all air-dried in the dark for 24 hours, ground in a mortar,
sieved, weighed, and placed in sterile tumblers. Some were mixed with
mannite and arsenic, others with mannite, while still others received
only arsenic. They were all incubated in the regular manner, and the
nitrogen determined as in the previous series. The results are given in
Table III. Each reported result is the average of six closely agreeing
determinations.
A marked stimulation is found in every case where the arsenic and
mannite were applied to the soil, as compared with the results obtained
where the mannite only was applied. The action of the various arsenical
compounds follows the same order in each of these soils that it did in the
first soil tested, being greatest with the lead arsenate and least with the
sodium arsenate. The nitrogen fixed in the presence of arsenic but in
the absence of mannite is usually considerably higher than that fixed in
the presence of mannite and absence of arsenic. It would not be right
to conclude from these results that the arsenic compounds furnish a
source of energy to the nitrogen-fixing organisms, for these soils (Greaves,
1 914, p. 456) have been found to fix appreciable quantities of nitrogen
when incubated with an optimum moisture content without the addition
of any carbon compound. It is likely that the arsenic makes the nitrogen-
June 12, 1916
Influence of Arsenic upon Soil Organisyns
395
gathering organism use more economically its usual source of carbon,
which in the absence of mannite is probably the plant debris which has
been slowly added to the soil. The belief that this is the case is strength-
ened by the fact that soil rich in organic matter (soil A) acts practically
the same in the absence of mannite and presence of arsenic as it does
when both arsenic and mannite are added to the soil. The clay soil
(C), which is low in organic matter, acts about the same in the absence of
arsenic as in the absence of mannite. It is interesting to note that in
soils B and C the total fixation in the soil containing mannite plus that
fixed by the soil containing arsenic approximates the total fixation in the
series in which both arsenic and mannite are present.
Table III. — Quantity of nitrogen {in milligrams) fixed in 100 gm. of soil with and -with-
out arsenic
LEAD ARSENATE
SoU.
16 mgm. of
arsenic, 2 gm.
of mannite,
added to soil.
16 mgm. of
arsenic,
no mannite,
added to soil.
No arsenic, 3
gm. of mannite,
added to soil.
Total of
columns
2 and 3.
A
17.0
16.8
10. 5
16.8
9.8
5-3
6.7
4.0
6.3
24-5
13-8
II. 6
B
C
Average
14.7
10. 6
6.0
16. 6
ARSENIC TRISULPHID
A
16.3
12. 6
10. 6
15-6
7.0
5-6
13-8
7.6
4. 2
29. 4
14. 6
B
c
9.8
Average
13- I
9-4
8-5
17.9
SODIUM ARSENATE
A
7.8
7.0
9.2
6.3
4.9
8.4
6.3
7.0
12 60
R
8. 20
r
15.40
Average
8. 0
6.5
5- 5
12. 0
In all of the tests so far reported the incubation period has been 18
days. Longer periods of incubation may give results very different
from those so far obtained, for the stimulating influence of arsenic may
be of short duration, and we may find later a slowing up of the reaction,
or, inasmuch as we are dealing with the algebraic sum of many reac-
tions which are taking place in the soil, we may find it to be negative.
An attempt was made to determine this by the following experiment :
loo-gm. portions of the high-humus soil (A) were mixed with 0.0728
396
Journal of Agricultural Research
Vol. VI, No. II
gm. of lead arsenate and the moisture content made up to i8 per
cent and then weighed. One-half of the samples thus prepared were
sterilized in the autoclave and all of them placed in an incubator at a
temperature of from 28° to 30° C. The moisture was made up weekly to
its initial content. Beginning at the end of 20 days, six samples, three
autoclaved and three not autoclaved, were used for the making of dupli-
cate total-nitrogen determinations. The average excess of nitrogen in
the unsterilized soil over that in the sterilized is given in Table IV.
Table IV. — Quantity of nitrogen (in inilligrams) fixed in lOO gm. of soil containing
0.0728 gm. of lead arse?taie
Time incubated
The greatest quantity of nitrogen was obtained at the end of 20 days.
During the next 10 days, however, 24.72 mgm. of combined nitrogen
disappeared. During the next 14 days there was a loss of only 4 mgm.
From this time on there was a gradual increase in the amount of com-
bined nitrogen found within the soil up to the end of the experiment, but
even after 172 days' incubation there was less nitrogen in the soil than
there was at the end of 20 days.
The great loss of nitrogen can not be entirely charged up to the arsenic
added, for other workers (Ashby, 1907; Hoffmann and Hammer, 1910,
p. 164) have noted, when working with impure cultures, a loss of nitrogen
on prolonged incubation in the absence of arsenic. The loss is probably
due to the soil's becoming compact, vnth the production of anaerobic
conditions. This, assisted by the protozoa (Miller, 1914, p. 217), which
appropriate too large a share of the limited supply of oxygen in the soil,
prevents entirely the activity of the aerobic nitrogen-fixing organisms
and greatly stimulates the activity of the denitrifying organisms of the
soil. This can, however, only partly account for the phenomena; other-
wise there would be a continual decrease in the nitrogen as the soil
became more compact.
The fact that aeration plays a considerable part in the reaction is
brought out by the following experiment, which differs from the pre-
ceding only in that the soil was thoroughly stirred, thus aerating it each
time before making up the moisture content. The results of this experi-
ment are given in Table V.
June 12, 1916
Inflti^nce of Arsenic upon Soil Organisms
397
Table V. — Quantity of nitrogen {in inilligrams) fixed in 100 gm. of aerated soil with and
without the addition of arsenic after different periods of incubation
Days incubated.
20
44
Nitrogen
fixed in
soil con-
taining
0.0728 gm.
of lead
arsenate.
5. 8S
8.26
Nitrogen
fixed in
tintreated
soil.
2.58
3-Q2
3-78
Days incubated.
66.
96.
162
Nitrogen
fixed in
soil con-
taining
0.0728 gm.
of lead
arsenate.
9-38
4.90
2.80
Nitrogen
fixed in
untreated
soil.
14.00
2. 52
4. 20
These results show conclusively that it was the lack of air in the
former series which caused such great losses of nitrogen and that they
could in no way be attributed to the arsenic added. This series was
stirred but once a week and after the stirring the moisture content was
made up to the optimum so that the soil became quite compact. It is
quite likely that greater care in the aeration of the soil would have reduced
very materially the loss of nitrogen which was observed in this series.
In the first stages of the experiment the soil containing arsenic gained
the greater quantity of nitrogen, while in the later stages the soils con-
taining no arsenic were the highest. If, however, an average of the
quantity found in each soil is taken, it will be found to be considerably
higher in the soil containing arsenic than in the other.
It was thought that some of the questions referred to in the first part
of this article could be answered more readily with the solution method
than with soil. For this reason a series was incubated using a solution
of the following composition:
Dibasic potassium phosphate (K2HPO4) .... 0.2 gm.
Magnesium sulphate (MgS04) 2 gm.
Calcium chlorid (CaCla) 02 gm.
Ferric chlorid (FeaClg) i drop (10 per cent solution).
This was made up to 1,000 c. c. with tap water and distributed in 100
c. c. portions into 750 c. c. Erlenmeyer flasks. One gm. of calcium car-
bonate was added to each, and the flasks were then sterilized and inocu-
lated. One series was inoculated with Azoiohacter vinclandii. This was
done by making a suspension in sterile tap water of the organism and
adding 5 c. c. of this suspension to each flask. In the other series the
inoculating medium was 10 gm. of soil. The solutions were incubated
at 28° to 30° C. for 18 days, and then the nitrogen determined in the
manner previously outlined. The results are given in Table VI and are
reported as milligrams of nitrogen fixed in 100 c. c. of the solution. Each
reported result is the average of three closely agreeing determinations.
398
Journal of Agricultural Research
Vol. VI. No. II
Table VI. — Quantity of nitrogen {in milligrams) fixed in loo c. c. of nutritive solution
■with and without the addition of arsenic
Treatment.
Inoculated
•with Azotobac-
ter vinelandii.
Soil +0.0728
gm. of sterilized
lead arsenate.
Soil +0.0728
gm. of unsteril-
ized lead
arsenate.
Nutritive solution + i-5 gm- of mannite. . . .
Nutritive solution + i-5 gm. of mannite and
0.0728 gm. of lead arsenate
Nutritive solution + 0.0728 gm. of lead
arsenate
Nutritive solution + i-5 gm. of mannite
and 0.0272 gm. of arsenic trisulphid
Nutritive solution + 0.0272 gm. of arsenic
trisulphid
14. 12
c
o
•5
15.16
14.79
1-45
5-98
.28
15-77
13-72
•52
2.05
.08
After the first series had been completed, it was thought possible that
the heat in the autoclave had changed the solubility of the arsenical
compounds and that this was the reason there was no fixation in the
solution with arsenic. For this reason analyses were made of the soluble
arsenic in 100 c. c. of the nutritive solution containing arsenic both
before and after autoclaving. The determinations were made as pre-
viously outlined. The lead arsenate yielded 0.91 mgm. of soluble arsenic
before autoclaving and 0.85 mgm. after autoclaving. The arsenic tri-
sulphid yielded 0.40 mgm. before autoclaving and 0.42 mgm. after
autoclaving.
The results indicate conclusively that the toxicity of the compound is
not due to a difference in the solubility of the compound produced by the
heat. In order to make sure of this, a series was arranged in which the
arsenic was added just before inoculation and after the solution had been
autoclaved. These results are given in the last column of Table VI
and are slightly lower than those previously obtained with the arsenic.
The A. vinelandii fixed no nitrogen in the presence of the arsenic. Even
where the soil was used as the inoculating medium, the lead arsenate
retarded nitrogen fixations to a certain extent. The toxic influence of
the arsenic sulphid is very pronounced. These results show the care
which must be used in drawing conclusions from the Remy-solution
method as to what is to be expected in soils. They greatly strengthen
the contention of Jonsson (1896) that the fact that Nobbe (1884) found
arsenic solutions to be toxic to seedlings in water culture and concluded
that arsenic, even in small quantities, is extremely toxic to plants does
not indicate that these solutions will be toxic when in the soil. The
results herein reported show arsenic to be extremely toxic to nitrogen-
fixing organisms while in solution, but the same concentration in the
soil is not only devoid of toxicity but acts as a powerful stimulant. This
therefore establishes for the bacteria what Kanda (1904, p. 16) found
to be true for the higher plants — namely, that dilute solutions of sub-
June 12, 1916 Influence of Arsenic upon Soil Organisms
399
stances may be toxic when used in water culture, but that the same
quantities when placed in the soil may act as stimulants.
The results reported for A. mnelandii, when considered in connec-
tion with those obtained for the soil, make very problematic the part
played by Azotobacter, especially A. vinelandii, in these soils. The
exact mode of action of the arsenic also remains a question. For these
reasons the soil used in the first series was plated and the main nitrogen-
fixing organisms isolated. Three types of Azotobacter were obtained.
These have been designated Azotobacter A, Azotobacter B, and Azoto-
bacter C. Azotobacter A has a nitrogen-fixing power of 6.86 mgm. of
nitrogen per gram of mannite in Ashby solution, Azotobacter B a nitro-
gen-fixing power of 5.00 mgm., and Azotobacter C a nitrogen-fixing power
of 6.44 mgm. of nitrogen.
The preceding results have shown that little information of value can
be obtained by the solution method. Therefore another series was
planned in which loo-gm. portions of the soil used in the first series were
weighed into covered sterile tumblers and autoclaved at a temperature
of 120° C. for 30 minutes, cooled, and the moisture content made up to
18 per cent. The soil was then inoculated with the various organisms
which had been isolated. The soil portions were incubated for 18 days,
the moisture content kept constant, and then the total nitrogen deter-
mined. Sterile blanks were incubated and analyzed as checks. Each
reported result is the average of four or more closely agreeing determina-
tions, so that the analytical error has been reduced to a minimum. The
results are given in Table VII.
Table VII. — Quantity of nitrogen (in milligrams) fixed in 100 gm. of soil wiik and
-without arsenic and inoculated uiih various nitrogenfixing organisms
Milligrams of nitrogen fixed in loo gm. of soil
treated with —
Inoculating organism.
2 gm. of man-
nite. 0.0728 gm.
of lead arsenate.
2 gm, of man-
nite, no
arsenic.
0.0728 gm. of
lead arsenate,
no mannite.
Azotobacter A
15.60
24. 15
18.20
26.31
18.40
21. 70
14.70
18.20
22. 05
17.70
3.01
8.80
Azotobacter B
Azotobacter C
4.90
5-81
6.65
Azotobacter A and B
Azotobacter A, B, and C
The results reported above show for each organism a fixation much
higher in the soil than was found in the solution. The results without
arsenic, but with mannite, are as high as are reported in Table I with
both mannite and arsenic combined, a fact which would seem to indicate
that arsenic acts upon injurious species. This, however, does not account
for the entire phenomenon, for we find in this series a very small fixation
of nitrogen in the absence of mannite and presence of arsenic, while in
37768°— 16 2
400
Journal of Agricultural Research
Vol. VI, No. II
the ordinary soil with its mixed flora as great a fixation was obtained in
the presence of arsenic as in the presence of only mannite. This probably
indicates that some of the stimulation is due either to the fact that the
arsenic acts upon allied species which are gathering carbon that can be
used by the Azotobacter, or else to the fact that some species, possibly
the cellulose ferments, are stimulated so that they render available to the
Azotobacter the carbon-carrying compounds of the soil faster in the pres-
ence of arsenic than in its absence. Only one of the organisms isolated,
Azotobacter B, is directly stimulated by arsenic. The stimulation,
however, is very large in this case. It also fixes large quantities of nitro-
gen in the presence of arsenic and absence of mannite. These results
are complicated by the carbonaceous material which occurs in the soil.
For this reason a series similar to the above was incubated, using silica
sand in place of the soil. The silica used was devoid of organic matter
and had the following composition :
Per cent.
Silicon dioxid (SiOj) 97.5
Ferrous oxid (FeO) i
Aluminum oxid (AI2O3) 1.7
Calcium oxid (CaO) 2
One-hundred gm. portions of this were sterilized in covered tumblers,
and to each was added i gm. of calcium carbonate and 18 c. c. of sterile
distilled water containing 0.02 gm. of potassium phosphate, 0.02 gm. of
magnesium sulphate, and 0.002 gm. of calcium chlorid. The tumblers
were inoculated with the various nitrogen-fixing organisms incubated
with a constant moisture content at 28° C. for 18 days, and then the
nitrogen determined as in the previous series. They were all compared
with sterile blanks. The results are given in Table VIII as milligrams
of nitrogen fixed in 100 gm. of sand. Each reported result is the aver-
age of six or more closely agreeing determinations.
Table VIII. — Quantity of nitrogen {in milligrams) fixed in 100 gm,. of quartz sand with
and without arsenic
Inoculating material.
Sand and Ash-
by solution,
+0.0728 gm. of
lead arsenate.
Sand and Ash-
by solution, no
arsenic.
Sand and Ash-
by solution, no
mannite,
+0.0728 gm. of
lead arsenate.
10 c. c. of soil extract
Azotobacter A
19. 60
17. 01
13.84
15. 10
10.50
22. 61
12. 60
16.80
4.70
0
Azotobacter B
0
Azotobacter C
0
Qualitatively, the above results are the same as those obtained with
the soil. Azotobacter B was the only one of the three organisms stimu-
lated by the arsenic. Where the mixed flora were used, the stimulation
was very marked, but the fixation in the absence of arsenic where either
Azotobacter A or Azotobacter C was used is about the same as that
June 12, 1916 Influence of Arsenic upon Soil Organisms 401
obtained in the presence of arsenic where the soil extract was used. This
fact would seem to indicate that the main stimulation brought about by
arsenic is due to its action upon injurious species. The results obtained
in the presence of arsenic and absence of mannite indicate that the Azo-
tobacter can not use the arsenic as a source of energy. The small fixation
where the soil extract was used may be due to the nitrogen-fixing organ-
isms obtaining a small quantity of carbon compounds from algae which
may have grown in the complex flora.
The results given in Table VII pointed strongly to the conclusion that
the stimulating influence of the arsenic was due in part to an indirect
action upon the nitrogen-fixing organisms, possibly an action which it
exerts upon the cellulose ferment. A series was therefore arranged in
which the cellulose ferments were used in connection with the Azoto-
bacter.
In this series loo-gm. portions of the high humus soil (A) were placed
in covered tumblers and sterilized in the autoclave and then treated as
in Table IX. The Azotobacter was inoculated into 100 c. c. of Ashby
solution. After three days the solution was thoroughly shaken and 5 c. c.
of the solution were added to the sterile soil. The cellulose ferment was
added by making a suspension of the organism in sterile distilled water
and adding 5 c. c. of this to the soil. The moisture content was made up
to 1 8 per cent and incubated for 1 8 days. Six samples of each were used,
so that the results reported are the averages of six closely agreeing deter-
minations. The results are given in Table IX. The cellulose ferments
used were Bacillus rossicus, isolated by Kellerman, McBeth, and others
(1913) from Geneva (N. Y.) soils, and Pseudomonas efjusa, isolated by the
same investigators from the soils used in this work.
Table IX. — Quantity of nitrogen (in milligrams) fixed in 100 gin. of soil with and
without arsenic in the presence and absence of cellulose ferments
Treatment.
Nitrogen
gained.
14.
14.
26.
70
28
18
28.
00
13-
22.
30
68
14.
46
21.
00
■15-
20
19.
60
14.
00
21.
00
Azotobacter chroococcum
Azotobacter chroococcum, 0.0728 gm. lead arsenate
Azotobacter chroococcum, Bacillus rossicus
Azotobacter chroococcum, Bacillus rossicus, 0.0728 gm. of lead arsenate. . . .
Azotobacter chroococcum, Pseudomonas effusa
Azotobacter chroococcum, Pseudomonas effusa, 0.0728 gm. of lead arsenate .
Azotobacter B
Azotobacter B, 0.0728 gm. of lead arsenate
Azotobacter B, Bacillus rossicus
Azotobacter B, Bacillus rossicus, 0.0728 gm. of lead arsenate
Azotobacter B, Pseudomonas effusa
Azotobacter B, Pseudomonas effusa, 0.0728 gm. of lead arsenate
In this series, as in the previous series in which A . chroococcum was used,
it did not fix as much nitrogen in the presence of arsenic as it did in the
absence of it. A. chroococcum fixes nearly twice the quantity in the
402
Journal of Agricultural Research
Vol. VI, No. 11
presence of B. rossicus as in its absence, and when arsenic is added to the
two there is an even greater fixation. This is also the case with P.
effusa; measured in terms of the increased nitrogen fixed by A. chroo-
coccum, it may therefore be safely concluded that both of the cellulose
ferments are stimulated by lead arsenate.
The Azotobacter B differs from the A. chroococcum in that it is
directly stimulated by the arsenic, but is not as greatly helped by the
cellulose ferment. In this case the lead arsenate greatly stimulates the
activity of the cellulose ferments, and the stimulating influence is much
greater with P. effusa, the normal habitat of which is this soil, than
it is with B. rossicus. Hence, from this work it is safe to conclude that
the cellulose organisms, so far as arsenic is concerned, obey the same
laws as do the ammonifying, nitrifying, and nitrogen-fixing organisms
of the soil.
It has been noted throughout all of this work that the soil taken direct
from the field was stimulated to a much greater extent by the arsenical
compounds than was the air-dried soil. Furthermore, it was noted that
the soil which had stood in the laboratory for a great length of time was
stimulated only very slightly by arsenic. For these reasons a series of
experiments was planned to throw more light upon this substance or
organism which disappears on drying.
Fred (191 1) has suggested the use of filter paper for the separation of
the protozoa. Later this has been shown by Kopeloff and others (191 5)
to be quite effective. Using this suggestion, loo-gm. portions of soil
were placed in tumblers. To half of them was added 0.0728 gm. of lead
arsenate, and the mixture was autoclaved until free from bacterial life.
They were all inoculated with 10 c. c. of a solution obtained by shaking
100 gm. of soil in 1,000 c. c. of sterile water and then filtering through
three thicknesses of a fine grade of quantitative filter paper, after which
they v/ere incubated and nitrogen determined as in the previous set.
The results are given in Table X as milligrams of nitrogen per 100 gm.
of soil. All results are averages of six determinations made on that
number of incubated samples.
Table X. — Quantity of nitrogen (in milligra^ns) fixed in loo gm. of sterile soil inocti-
lated with filtered soil extract, with and without arsenic
Time incubated.
0.072S gm. of
lead arsenate.
No lead
natc
arse-
20
Days.
16. 10
3.0S
2.80
1.94
.28
.78
14. 70
2. 52
•30
• 14
.28
AA
66
06
162
. 8?
June 12, 1916 Influence of Arsenic upon Soil Organisms
403
It probably would have been better if in every case untreated soil
could have been incubated with the variously treated soil, but this so
greatly increased the number of determinations that it was not thought
advisable. Furthermore, all the work has been done on the high-humus
soil, A, without the addition of any carbohydrate, and repeated deter-
minations have shown that the arsenic more than doubles the nitrogen
fixed in the soil in 20 days, so that the absence of the stimulation can be
safely attributed to the treatment. In the above results, it is readily
seen that the soil extract on passing through filter paper loses to a very
great extent its power of being stimulated by arsenic. Hence, it is safe
to conclude that the main stimulating influence of arsenic upon nitrogen
fixation is due to its suppressing something which is found in the soil and
which is removed by the filter paper.
That this factor is to a great extent the same as is removed by heat is
shown by the results reported in Table XI. The arrangement of this
series of experiments was as follows: loo-gm. portions of the soil were
weighed into covered tumblers. To one-half of the set was added
arsenic — 0.0728 gm. to each 100 gm. of soil. The tumblers were all
carefully sterilized and half of them were placed in the incubator in the
sterile condition. To the others was added a soil extract prepared by
shaking one part of soil with two parts of sterile distilled water for three
minutes. After standing for about five minutes the liquid was decanted
and 10 c. c. portions of this were used to inoculate the soil. Before
inoculating, this extract was placed in thin-walled test tubes in 10 c. c.
portions and then held at the required temperature for exactly 15 min-
utes before adding to the soil. The moisture content was made up to 18
per cent and the whole incubated for 20 days. Each reported result is
the average of six closely agreeing determinations.
Table XI.
-Quantity of nitrogen {in milligrams) fixed in 100 gm. of soil, v:ith and
•without arsenic, inoculated with soil extract
Temperature of soil extract (°C.).
0.0728 gm.
of lead arsenate
added.
No arsenic
added.
Room
50....
55----
60....
65....
70....
75---
80....
85....
8.77
9. 24
14. 28
12. 60
13-85
12. 18
12.88
13- 44
II- 54
5- II
9. 00
14. 14
16.38
14.42
13.02
11-34
12. 66
10.36
The heating of the soil extract to a temperature of 55° C. for 15 minutes
changes the soil so that it is no longer stimulated by arsenic. The
heating of the soil extract to a higher temperature stimulates its nitrogen-
404 Journal of Agricultural Research voi. vi, no. n
fixing properties. It is not, however, increased by the addition of
arsenic. Hence, it would appear as if the substance which is suppressed
by the arsenic is very thermolabile and is easily injured by drying, for it
has been repeatedly brought to our attention that the drying of the
soil prevents the arsenic from greatly stimulating its nitrogen-fixing
properties. Harden and Young (191 1, p. 72; 1906) have shown that
the addition of arsenates to a yeast-juice sugar solution greatly accelerates
the rate of fermentation of such a mixture. The close analogy existing
between the chemical properties of phosphorus and arsenic led to the
idea that possibly the arsenic replaced the phosphorus in the reaction
characteristic of phosphorus, but they found that this is not the case,
for while the arsenic has an optimum concentration, as has the phos-
phorus, there was no direct relationship between the amount of arsenate
added and the extra amount of fermentation, the arsenic in this way
acting more like a catalyzer than does the phosphorus. Furthermore it
was shown that fermentation can not proceed in the absence of phos-
phorus, even though there be present either arsenates or arsenites. The
arsenic acts mainly as a liberator of the phosphorus from the hexosephos-
phates and does not of itself enter into the vital reactions of the cell as
does the phosphorus.
These facts make it likely that a similar action may be exerted by the
arsenic upon the bacteria. For these reasons a series of experiments
was arranged in which the phosphorus had been replaced by arsenic.
These were carried on in the nitrogen-free quartz sand. To each 100 gm.
of the sand there was added the quantity of carefully tested nutrient
without phosphorus found in 100 c. c. of Ashby's solution. To one-half
of them was added the phosphorus, while to the other half there was
added 0.0728 gm. of lead arsenate. They were each inoculated with
I c. c. of a soil extract and then incubated the regular length of time.
The nitrogen determinations were made on them and sterile blanks with
the following results: When incubated with complete Ashby's solution
and 0.0728 gm. of lead arsenate, 100 gm. of sand fixed 11.62 mgm. of
nitrogen. Similar samples without phosphorus but with arsenic fixed
0.03 mgm., while without phosphorus or arsenic there was fixed o.oi
mgm. of nitrogen. The results for the set with the complete nutritive
media show that sufficient of the soil extract was taken to get the nitrogen-
fixing organism, and the results without phosphorus show that there was
not sufficient phosphorus in the i c. c. of soil extract to furnish phosphorus
for the organisms. These results show conclusively that arsenic can not
replace phosphorus in the vital activities of the nitrogen-fixing organisms
of the soil, and establish for this set of organisms what Stoklasa (1897)
has established for the higher phanerogams, Molisch (Lafar, 191 1, p. 37)
for algae, Giinther (1897) for the molds, and Harden and Young (1906)
for the yeasts.
jimei2, 1916 Influence of Arsenic upon Soil Organisms 405
There is still the possibility that the arsenic liberates the phosphorus
from its insoluble compounds in the soil and thus makes it more available
to the micro-organisms. If this be the case, one would think that the
addition of soluble phosphates to the soil investigated would increase its
nitrogen-fixing powers. Experiments, however, did not bear out this
assumption, for just as large a quantity of nitrogen was fixed in the
absence of the soluble phosphate as in its presence. This was probably
due to the fact that the soil under investigation was well supplied in the
natural condition with soluble phosphorus. But that the arsenic did
have an influence upon the solubility of the phosphorus of the soil was
shown by the following experiment: loo-gm. portions of the soil were
placed in covered tumblers. Of these, 24 received 0.0728 gm. of lead
arsenate each, while the other 24 received none. The moisture was made
up to 18 per cent and incubated for 20 days. At the end of this time the
water-soluble phosphorus was determined in 12 of the treated and 12 of
the untreated soils by extracting with 500 c. c. of distilled water and deter-
mining the phosphorus in the extract (Greaves, 1910). As an average of
the 12 closely agreeing determinations of the soil treated with arsenic
there was obtained 0.59 mgm. of water-soluble phosphorus, while the
untreated soils yielded 0.52 mgm. This is a slightly greater quantity
in the arsenic-treated soil than in the untreated, which is probably
due to the fact that more of the phosphorus had been changed in the bodv
of the soil organisms to nucleoproteins or phosphoproteins. That this is
the correct interpretation is shown by the results obtained from the
remaining samples. Twelve of these samples, six with and six without
arsenic, were digested for six hours with 100 c. c. of 12 per cent hydro-
chloric acid and the phosphorus determined in the filtrate. The other
samples were ignited and the phosphorus extracted by the 12 per cent
hydrochloric acid determined. The average of the results thus obtained
is given in the tabular form below :
Samples not ignited:
Soil with arsenic 105. 6 mgm. of phosphorus.
Soil without arsenic 100. o mgm. of phosphorus.
Excess of acid-soluble phosphorus in
soil with arsenic 5. 6 mgm. of phosphorus.
Samples ignited:
Soil with arsenic 107. 7 mgm. of phosphorus.
Soil without arsenic 100. 8 mgm. of phosphorus.
Excess of acid-soluble phosphorus in
soil with arsenic 6. 9 nigm. of phosphorus.
This would give by the Schmoeger method 2.10 mgm. of organic phos-
phorus in the arsenic- treated soil, while in the untreated soil there was
4o6 Journal of Agricultural Research voi. vi, no. n
only 0.80 mgm. of organic phosphorus. This excess of organic phosphorus
could not have come from the water-soluble phosphorus, as there was a
difference of only 0.07 mgm. in the two soils; hence, it must be concluded
that the arsenic increases the solubility of the phosphorus. This, how-
ever, may be due either to a direct interchange between the insoluble
phosphorus of the soil and the arsenic or to its action upon bacteria, which
causes them to become more active in growth and formation of various
acids which act upon the insoluble phosphates of the soil, rendering them
soluble.
GENEIL\L CONSIDERATIONS
The data reported prove conclusively that the arsenical compounds,
with the single exception of Paris green, stimulate the nitrogen-fixing
organisms of the soil and that this influence varies qualitatively but not
quantitatively with the various soils. The results also bring out the fact
that both the anion and the cation of the compounds have a marked
influence upon the growth of the organisms. With some compounds
both the anion and cation act as stimulants, while with others one stimu-
lates and the other is markedly toxic. It is likely that little or no influ-
ence is exerted upon the nitrogen-gathering organisms by the sodium (Lip-
man and Sharp, 1912), and that the stimulating influence noted with
dilute solutions and the toxic influence exerted with more concentrated
solutions are due entirely to the arsenic. It is quite likely that the
stimulating influence which Riviere and Bailhache (19 13) have found
sodium arsenate to have upon wheat and oats is an indirect effect which is
exerted upon the bacterial flora of the soil and which in turn influences the
yield of the various grains.
Both the anion and cation undoubtedly act as stimulants in the lead
arsenate. Stoklasa (191 3) has shown that lead when present in soil
stimulates the growth of higher plants. This he (191 1) ascribes to the
catalytic action of these elements on the chlorophyll. The results herein
reported, together with those previously published (Greaves, 1913a),
indicate that it is due to the influence of the compounds upon the biologi-
cal transformation of the nitrogen in the soil. The fact that the lead
plays no small part in the stimulating influence is borne out by the work
of Lipman and Burgess (1914), who found lead to stimulate nitrifying
organisms.
Paris green is toxic to the nitrogen-fixing organisms in the lowest con-
centration tested. This is due to the copper and not to the arsenic, as it
is well known that the copper ion is a strong poison to many of the lower
plants. Brenchley (191 4) found it to be toxic to higher plants when
present in water to the extent of i part in 4,000,000,000. Although
Russell (191 2, p. 47) states that it is not as toxic in soil as in
water, Darbishire and Russell (1905) found it to be toxic in soils, and
they failed to get a stimulating influence with it. Montemartini (1911)
June 12, 1916 Influence of Arsenic upon Soil Organisms 407
has noted a stimulation with copper sulphate when used in dilute solu-
tions. This, however, may have been due to the anion and not to the
cation, as sulphates do stimulate plants by their action on insoluble
constituents of the soil (Greaves, 1910, p. 298). The same interpreta-
tion could be placed upon the results obtained by Lipman and Wilson
(191 3) and also those reported by Voelcker (19 13), in which they noted a
stimulation with copper salts. Clark and Gage (1906) have found that
very dilute solutions of copper have an invigorating influence upon bac-
terial activity. In order that the stimulation may be noted the copper
must be present in small quantities. Jackson (1905) found that i part of
copper sulphate in 50,000 parts of water killed Bacillus coli and B.
typhosus. Kellerman and Beckwith (1907) found that the common
saprophytic bacteria are more resistant to copper than is B. coli. There
is considerable evidence (Lipman and Burgess, 1914; Greaves, 1913a, p.
8) that copper stimulates the ammonifying and nitrifying organisms
of the soil, but these results show the nitrogen-fixing organisms of the
soil to be very sensitive to copper, and if it does act as a stimulant it
must be in extremely dilute solutions. The toxicity of the copper in the
Paris green is great enough in the dilution of 10 parts in 1,000,000 to
offset the great stimulating influence of the arsenic in combination with it.
The very marked stimulating influence noted where the arsenic
trisulphid is used is very probably due to both the arsenic and the
sulphur. Demolon (191 3) attributed much of the fertilizing action of
sulphur to its action upon bacteria, and Vogel (1914) found that sulphur
decidedly increased the activity of the nitrogen-fixing organisms. The
results which Russell and Hutchinson (1913, p. 173) obtained with calcium
sulphid are interesting in this connection. They found that after 30
days there were five times as many organisms in the soil to which calcium
sulphid had been added as in the untreated soil, and the yield of ammonia
and nitrates in this time was one-third greater in the treated soil than
in the untreated soil. This, in turn, reacts upon the crop harvested, as
shown by Shedd (1914, p. 595).
The first part of the curve (fig. i) for the zinc arsenite nearly coincides
with that of the sodium arsenate, but the zinc arsenite stimulates in
greater concentrations than does the sodium arsenate. This is partly
due to the difference in solubility of the two compounds, but there is
another factor which enters, and that is that the zinc also acts as a stimu-
lant. Latham (1909) found that small quantities of zinc stimulated
algge. The same results have been obtained by Silberberg (1909) in
working with higher plants. Ehrenberg (1910) concludes that zinc salts
are always toxic when the action is simply on the plant, but that they
may lead to increased growth through some indirect action on the soil.
He found that zinc stimulated plant growth in soils, but when the soil
was sterilized the zinc became toxic. Lipman and Burgess (1914, p. 133)
4o8
Journal of Agricultural Research
Vol. VI, No. II
have shown that it does stimulate the nitrifying organisms and that the
influence is shown by the yield obtained from such soils (Lipman and
Wilson, 1 91 3). The great variation in the results reported by the vari-
ous investigators for zinc, arsenic, and lead is probably due to the fact
that it modifies the bacterial flora of the soil, and when heated soil or
water cultures are used a different result is noted. This, however, is not
the only factor which enters, for these results show a marked difference
in soil and in water. The lead arsenate stimulates the nitrogen-fixing
organisms when placed in soils but becomes very toxic to the same
organisms when placed in nutritive solutions.
The difference is due in part to the adsorption of the soil, but in this
case we would have to attribute it to the silica compounds of the soil, for
the nitrogen-fixing organisms are stimulated by arsenic in quartz sand
zo
1^ /o
I
si
i
A
'^s
1^
y
^^^-...NJ
NJ
_=
^^«
s
^
^
\
^
■^
^
-y
£yr^/./^/v^ 7-/0 A/
— — '
20
^O
GO &0 /OO /20
/^o
/<5-0 /&0
Fig. 2.
-Graph showing the effect of aeration on the nitrogen-fixing activity of soil containing compounds
of arsenic.
free from organic colloids. In this case the arsenic becomes concen-
trated at the surface, layers of the silica leaving the inner part of the
water film comparatively free from arsenic, in which the micro-organisms
multiply and carry on their metabolic processes. This being the case,
one should, and probably could, find a water solution weak enough to
stimulate bacteria. A great difference, however, between the solution
and the sand-culture method is the greater aeration in the latter than in
the former. That the aeration of a cultural medium does play a great
part in determining the activity of the nitrogen-fixing powers of a soil is
very strikingly brought out in figure 2. The graphs in this figure are
made from the data given in Tables IV, V, and X.
It is remarkable how the aeration of the soil or the filtering of the soil
extract can prevent the great loss of nitrogen which is noted at first in
the unaerated soil. This can not be attributed directly to the denitrifying
organisms; otherwise it would not be removed by filtration. The graphs
June 12. 1916 Influence of Arsenic upon Soil Organisms
409
also bring out the fact that the addition of arsenic and the filtering of
the soil only shift for the time the equilibrium within the soil, and later
it tends to regain its old equilibrium. This is a condition which coincides
well with what one would expect if the limiting element were some other
microscopic forms of life. The filter would not separate them quanti-
tatively, and it is possible that the arsenic has only a selective influence.
Later, many of the organisms become accustomed to its presence; or,
what is more likely, the arsenic becomes fixed (McGeorge, 191 5) within
the soil.
That this limiting factor is a thermolabile body is brought out more
clearly in figure 3, which is made from the data reported in Table XL
/oo%
&59
Fig. 3. — Graph showing the effect of heat on the nitrogen-fixing power of soil treated and not treated
with arsenic.
The quantity of nitrogen fixed by the unheated soil receiving no arsenic
has been taken as 100 per cent, and the heated soil with and without
arsenic is compared with this.
The heating of the soil extract to 50° C. for 15 minutes has exactly
the same influence measured in terms of nitrogen fixed as does 0.0728
gram of lead arsenate. The stimulating influence of heat is noted even
in the presence of arsenic and reaches its maximum effect in the absence
of arsenic at 60°, while in the presence of arsenic at 65° above these
temperatures there is a decline in the nitrogen fixed. But even the soil
inoculated with solutions which had been heated to a temperature of 85°
fixed nitrogen ; or at least there is more nitrogen accumulated in such soil
than in that inoculated with the untreated soil solution. The results indi-
cate that many of the organisms which take part in the gathering of nitro-
4IO Journal of Agricultural Research voi. vi, no. h
gen in this soil are very resistant to heat. It is also significant that the
greatest stimulating influence is exerted in soil which had been inoculated
with solutions heated just above what Cunningham and Lohnis (1914)
found to be the thermal death point of soil protozoa.
The data presented in this paper, together with these presented in
former publications, make it possible to compare the sensitiveness of the
ammonifying, nitrifying, and nitrogen-fixing organisms toward the
various arsenical compounds. Figure 4 represents the percentage of
activity of the various classes of organisms in the presence of 400
p. p. m, of arsenic in the form of the various arsenical compounds. The
untreated soil has been taken in every case as 100. The ammonifying
organisms are retarded more by the lead arsenate than the nitrogen-
fixing or nitrifying organisms. The latter two are influenced in nearly
the same way by this concentration of lead arsenate. All three types
of organisms are influenced in the same order by the arsenic trisulphid,
while with the zinc arsenite the nitrogen-fixing and nitrifying organ-
isms act about normally in concentrations of 400 p. p. m. of arsenic,
but the ammonifiers are greatly depressed. Paris green stimulates the
nitrifiers, but greatly depresses the other types of organisms. The
results, with the exception of copper, show that the nitrifying and
nitrogen-fixing organisms are very similar.
In figure 5 are shown graphically the quantities of arsenic in the form
of various arsenicals which are required by the different organisms to
give the greatest stimulation.
It has been shown that stimulation within a specific group of organisms
varies with the quantity of water-soluble arsenic and the stimulating
influence of the electropositive ion associated with the arsenic. But
when we examine stimulation by these substances with diff'erent groups
of organisms, we find a marked difference which can not be attributed
to solubility but must be due to a physiological difference existing
in the various organisms; for instance, the nitrogen-fixing organisms
require 200 p. p. m. of arsenic in the form of lead arsenate for the
greatest stimulation, w^hile the nitrifiers and ammonifiers require much
smaller quantities. For maximum stimulation with arsenic trisulphid
the nitrogen-fixing organisms require the greatest concentration, fol-
owed by the nitrifying and ammonifying organisms in the order given.
Zinc arsenite, on the other hand, has to be present in large quantities
for a maximum stimulation of the nitrifying organisms, while very small
quantities give a maximum stimulation with the other two groups of
organisms. Practically the same order is followed by the organisms in
the presence of sodium arsenate and Paris green, there being, however,
this significant difference, that neither the ammonifiers nor the nitrogen-
fixing organisms are stimulated in any concentration by the presence of
copper, and it is quite possible that the same holds for the nitrifying
June 12, 1916
Influence of A rsenic upon Soil Organisms
411
/SO
/^o
/30
/SO
^O
^O
^O
/o
I
/^/i'S£'A//^rS T/?/SULf=>H/D /^/FS^A^/r£- (F/FE^'Ay
Fig. 4. — Effect of various arsenic compounds in the ratio of 400 parts of the compound to 1,000,000 parts
of soil on the activity of various soil organisms.
412
Journal of Agricultural Research
Vol. VI, No. II
organism. This set of organisms are, however, more resistant to copper
than are others, and what we have occurring is a suppression of other
types which feed on nitrates, thus permitting a greater accumulation
eoo
Z^y^/?
/9/?S£-A//C
ZZ/VC'
SOO/CAf
/=i4/f'/S
Al/i'S^/v/^rf T/f/S(//.^/y/D y^/f'S£-^/rs ^/T'S£^/v,4r^ t^/?£^£-A/
Fig. 5. — Graph showing parts per million of various arsenic compounds in the soil at which the greatest
stimulation occurred.
of nitrates under these conditions. While not so likely in the other
cases, the same possibility does arise. This, however, can be answered
definitely only by further experiments.
June 12, 1910 Influence of Arsenic upon Soil Organisms 413
SUMMARY
Arsenic, when applied to a soil in the form of lead arsenate, sodium
arsenate, arsenic trisulphid, or zinc arsenite, stimulates the nitrogen-fixing
powers of the soil. This stimulation is greatest when lead arsenate is
applied and least when zinc arsenite is applied. Paris green did not stimu-
late in any of the concentrations. This compound becomes very toxic
when the concentration reaches 120 p. p. m. The toxicity of this com-
pound is due to the copper and not to the arsenic contained in it.
Sodium arsenate became toxic when a concentration of 40 p. p. m.
of arsenic was added, and when 250 p. p. m. were added it entirely
stopped nitrogen fixation. Lead arsenate was not toxic even at a con-
centration of 400 p. p. m. of arsenic. The toxicity of arsenic trisulphid
and zinc arsenite was very small at this concentration.
The stimulation noted when arsenic is added to a soil is not due to any
inherent peculiarity of the soil used, for soils which vary greatly in
physical and chemical properties had their nitrogen-fixing powers greatly
increased when arsenic was applied to them. Soils high in organic matter
fixed as much nitrogen in the presence of arsenic and in the absence of
mannite as they did in the presence of mannite and absence of arsenic.
The stimulation is greatest when the water-soluble arsenic content of the
soil is about 10 p. p. m. This quantity exceeds that found in most
soils, so it is likely that in agricultural practice arsenic will stimulate
and not retard bacterial activity in the soil.
Only one type of Azotobacter was isolated which was stimulated by
arsenic, and in this case the stimulation was due to the organism utilizing
more economically in the presence of arsenic its source of carbon than it
did in the absence of arsenic. The arsenic compounds do not act as a
source of energy to the organisms. The main part of the stimulation
noted in the soil with its mixed flora is undoubtedly due to the arsenic
inhibiting injurious species.
A quantity of arsenic which acts as a stimulant to bacteria when placed
in soil may become very toxic when tested by the Remy-solution method.
Arsenic can not replace phosphorus in the vital process of the nitrogen-
fixing organisms, but it can in some manner liberate the phosphorus from
its insoluble compounds. This may be either a direct or an indirect
action.
Arsenic stimulates the cellulose ferments, and these in turn react upon
the activity of the nitrogen-fixing organisms.
The nitrogen-fixing powers of soil extract, of filtered soil extract, and
soil dried for some time are only slightly stimulated by arsenic, showing
that arsenic acts mainly by the removal of a thermolabile body which
occurs in the soil.
414 Journal of Agricultural Research voi. vi, No. ix
LITERATURE CITED
ASHBY, S. F.
1907. Some observations on "nitrification." /n Jour. Agr. Sci., v. 2, pt. i, p. 52-67.
Brenchley, Winifred E.
1914. Inorganic Plant Poisons and Stimulants, no p., 19 fig. Cambridge.
Clark, H. W., and Gage, S. DeM.
1906. On the bactericidal action of copper. In Jour. Infect. Diseases, Sup. 2, p.
175-204.
Cunningham, Andrew, and Lohnis, Felix.
1914. Studies on soil protozoa. I. The growth of protozoa on various media and the
effect of heat on active and encysted forms. In Centbl. Bakt. [etc.], Abt.
2, Bd. 39, No. 23/25, p. 596-610.
Darbishire, F. v., and Russell, E. J.
1907. Oxidation in soils, and its relation to productiveness. II. The influence of
partial sterilisation. In Jour. Agr. Sci., v. 2, pt. 3, p. 305-326, 3 fig.
Demolon, a.
1913. Recherches sur Taction fertilisante du soufre. In Compt. Rend. Acad. Sci.
[Paris], t. 156, no. 9, p. 725-728.
EhrenbErg, Paul.
1910. Wirkung des Zinks bei Vegetationsversuchen. Zugleich Beitrage zur Am-
moniakfrage II. In Landw. Vers. Stat., Bd. 72, Heft 1/2, p. 15-142, pi. 1-6.
Fred, E. B.
1911. IJber die Beschleunigung der Lebenstatigkeit hohrerund niederer Pflanzen
durch kleine Giftmengen. In Centbl. Bakt. [etc.], Abt. 2, Bd. 31, No.
5/10, p. 185-245, I fig.
Greaves, J. E.
1910. Effects of soluble salts on insoluble phosphates. In Jour. Biol. Chem., v. 7,
no. 4, p. 2S7-307.
1913a. The influence of arsenic upon the biological transformation of nitrogen in
soils. In Biochem. Bui., v. 3, no. 9, p. 2-16.
1913b. The occurrence of arsenic in soils. In Biochem. Bui., v. 2, no. 8, p. 519-523.
1913c. Some factors influencing ammonification and nitrification in soils. I. In-
fluence of arsenic, hi Centbl. Bakt. [etc.], Bd. 39, No. 20/22, p. 542-560.
1913d. Some factors influencing the quantitative determination of arsenic in soils.
In Jour. Amer. Chem. Soc, v. 35, no. 2, p. 150-156, i fig.
19 14. A study of the bacterial activities of virgin and cultivated soils. In Centbl.
Bakt. [etc.], Abt. 2, Bd. 41, No. 11/17, p. 444-459.
GUNTHER, E.
1897. Beitrag zur mineralischen Nahrung der Pilze. Inaugiu-al-Dissertation.
Erlangen. (Abstract.) In Bot. Ztg., Abt. 2, Jahrg. 55, No. 24, p. 379.
Harden, Arthur.
191 1. Alcoholic fermentation. 128 p., 8 fig. London, New York. Bibliography,
p. 115-126.
and Young, W. J.
1906. Influence of sodium arsenate on the fermentation of glucose by 3^east-juice.
(Preliminary notice.) In Proc. Chem. Soc. [London], v. 22, no. 315,
p. 283-284.
June 12, 1916 Influetwe of Arsenic upon Soil Organisms 415
Hoffmann, C, and Hammer, B. W.
1910. Some factors concerned in the fixation of nitrogen by Azotobacter. Wis.
Agr. Exp. Sta. Research Bui. 12, p. 155-172, 2 fig.
Jackson, D. D.
1905. Purification of water by copper sulphate. (Abstract.) In Rev. Amer.
Chem. Research, v. 11, no. 12, p. 675 (Jour. Amer. Chem. See, v. 27,
no. 12). 1905. Original article in Municipal Engin., V. 29, no. 4, p. 245-246.
1905. Not seen.
JONSSON, B.
1896. On the influence of arsenic on the germination of seeds. (Abstract.) In
Exp. Sta. Rec, v. 8, no. 3, p. 232-233. 1896. Original article in Kgl.
Landt. Akad. Handl., bd. 35, p. 95-112. 1896. Not seen.
Kanda, Masayasu.
1904. Studien iiber die Reizwirkimg einiger Metallsalze auf das Wachstum hoherer
Pflanzen. Jour. Col. Sci. Imp. Univ. Tokyo, v. 19, art. 13, 39 p., i pi.
Kellerman, K. F., and Beckwith, T. D.
1907. The effect of copper upon water bacteria. In U. S. Dept. Agr. Bur. Plant
Indus. Bui. 100, p. 57-71.
McBeth, I. G., Scales, F. M., and Smith, N. R.
1913. Identification and classification of cellulose-dissolving bacteria. In Centbl.
Bakt. [etc.], Abt. 2, Bd. 39, No. 20/22, p. 502-522, 2 pi.
KoPELOFF, Nicholas, Lint, H. C, and Coleman, D. A.
1915. Separation of soil protozoa. In Jour. Agr. Research, v. 5, no. 3, p. 137-140.
Literature cited, p. 139-140.
Lafar, Franz.
191 1. Tecknical Mycology, v. i. London. Cites Molisch, p. 37.
Latham, M. E.
1909. Nitrogen assimilation of Sterigmatocystis nigra and the effect of chemical
stimulation. In Btil. Torrey Bot. Club, v. 36, no. 5, p. 235-244.
LrPMAN, C. B., and Burgess, P. S.
1914. The effect of copper, zinc, iron, and lead salts on ammonification and nitrifica-
tion in soils. In Univ. Cal. Pub. Agr. Sci., v. i, no. 6, p. 127-139.
and Sharp, L. T.
1912. Toxic effects of "alkali salts" in soils on soil bacteria. III. Nitrogen fixa-
tion. In Centbl. Bakt. [etc.], Abt. 2, Bd. 35, no. 25, p. 647-655, i fig.
and Wilson, F. H.
1913. Toxic inorganic salts and acids as affecting plant growth. (Preliminary com-
munication.) In Bot. Gaz., v. 55, no. 6, p. 409-420.
McGeorgE, W. T.
1915. Fate and effect of arsenic applied as a spray for weeds. In Jour. Agr. Re-
search, V. 5, no. II, p. 459-463.
Mh^leR, N. H. J.
1914. Agricultural chemistry and vegetable physiology. In Ann. Rpts. Prog.
Chem., V. 10, 1913, p. 211-232.
Montemartini, Luigi.
loii. The stimulating effect of sulphate of manganese and sulphate of copper on
plants. (Abstract.) In Intemat. Inst. Agr. [Rome], Bui. Bur. Agr. Intel,
and Plant Diseases, year 2, No. 11/12, p. 2467. 1911. Original article in
Staz. Sper. Agr. Ital. v. 44, fasc. 7, p. 564-571. 1911. Not seen.
NoBBE, F., Baessler, p., and Will, H.
1884. Untersuchung uber die Giftwirkung des Arsen, Blei, und Zink in pflanzli-
chen Organismus. In Landw. Vers. Stat., Bd. 30, p. 381-419, pi. 2.
37768°— 16 3
41 6 Journal of Agricultural Research voi. vi, no. n
Riviere, Gustave, and BailhachE, Gabriel.
1913. De 1 'influence des substances catalytiques. In Jour. Soc. Nat. Hort. France,
s. 4, 1. 14, p. 782-788.
Russell, E. J.
1905. Oxidation in soils and its connection with fertility. In Jotir. Agr. Sci., v. i,
pt. 3, p. 261-279, 2 fig.
1912. Soil Conditions and Plant Growth. 168 p. London, New York. A selected
bibliography, p. 154-156.
■ and Hutchinson, H. B.
1913. The effect of partial sterilization of soils on the production of plant food. II.
The limitation of bacterial numbers in normal soils and its consequences.
In Jour. Agr. Sci., v. 5, pt. 2, p. 152-221, 7 fig.
Shedd, O. M.
1914. The relation of sulfur to soil fertility. Ky. Agr. Exp. Sta. Bui. 188, p. 593-630.
Silberberg, Berenice.
1909. Stimulation of storage tissues of higher plants by zinc sulphate. In Bui.
Torrey Bot. Club, v. 36, no. 9, p. 489-500, 4 fig.
Stoklasa, Jules.
1897. De la substitution de I'acide arsdnique a I'acide phosphorique dans la nutri-
tion de la plante. In Ann. Agron., t. 23, no. 10, p. 471-477.
1911. Catalytic fertilizers for sugar beets. (Abstract.) In Exp. Sta. Rec, v. 26,
p. 225, 1912; Chem. Ztg., Bd. 35, No. 86, Repert., p. 361, 1911. Original
article in Bl. Zuckerriibenbau, Bd. 18, No. 11, p. 193-197, 1911. Not seen.
1913. De 1 'influence de 1 'uranium et du plomb sur la vdg6tation. In Compt. Rend.
Acad. Sci. [Paris], t. 156, no. 2, p. 153-155.
VOELCKER, J. A.
1913. Pot-culture experiments, 1913. In Jovr. Roy. Agr. Soc. England, v, 74, p.
411-422, 10 pi.
VOGEL, J.
1914. Die Einwirkung von Schwefel auf die bakteriellen Leistungen des Bodens.
In Centbl. Bakt. [etc.], Abt. 2, Bd. 40, no. 1/8, p. 60-83.
TRANSMISSION AND CONTROL OF BACTERIAL WILT
OF CUCURBITS '
By Frederick V. Rand, Assistant Pathologist, and Ella M. A. Eni^ows, Scientific
Assistant, Laboratory of Plant Pathology, Bureau of Plant Industry
WILT TRANSMISSION
That the striped cucumber beetle (Diabrotica vittata Fab.) is a direct
carrier of the bacterial-wilt organism {Bacillus tracheiphilus) from infected
to healthy cucurbits was shown several years ago by Smith.^ He also
expressed the conviction that it was the most important, if not the only,
summer carrier, and stated the possibility of its serving also as the winter
carrier of the disease. Observation and experiment by the senior writer '
during the last two seasons have abundantly confirmed the implication
of the striped cucumber beetle as a summer carrier and have brought out
strong proof that this insect is not only the principal summer carrier but
also the winter carrier of the wilt organism. The twelve-spotted cucumber
beetle {D. duodecim punctata L.) must be included with the striped cucum-
ber beetle at least as an important summer carrier of the disease.
INSECT TRANSMISSION
Relative to cucumber beetles as winter carriers, several direct cold-
storage tests have been carried out by the writers in Washington. During
the summer and fall of 191 5 hundreds of beetles were collected and
placed in cold storage at temperatures ranging from 4° to 10° C. These
early experiments were conducted partly with a view to determining
the proper conditions of feeding prior to storage and the temperature
and humidity most favorable to hibernation in storage. The optimum
environment for hibernation varies for different insects, and it is neces-
sary to work out this problem for each species. Consequently in these
preliminary tests the greater portion of the beetles placed in cold storage
was lost. Infection experiments with the few surviving beetles gave the
results here detailed.
Experiment i. — Several striped cucumber beetles were collected in October,
1914, and fed about two weeks on cucumber vines {Cucumis sativus) wilting as a
result of natural infection with B. tracheiphilus. After six weeks' hibernation in cold
storage the five surviving beetles were caged with a young squash plant on which
' Some of the details of the field experiments at East Marion, N. Y., were carried out by Mr. Wayland
C. Brown, of the Bureau of Plant Industry. The land used in these experiments was furnished by Messrs.
J. H. Douglass and G. S. Nowell, of East Marion.
^ Smith, Erwin F. Bacteria in relation to plant diseases, v. 2, p. 215. Washington, D. C, 1911.
A conspectus of bacterial diseases of plants. In Ann. Mo. Bot. Card., v. 2, no. 1/2, p. 390. 19x5.
3 Rand, F. V. Dissemination of bacterial wilt of cucurbits. In Jour. Agr. Research, v. s, no. 6, p.
257-260, pi. 24. 191S.
Journal of Agricultural Research, Vol. VI, No. 11
Dept. of Agriculture, Washington, D. C. June 12, 1916
eb G— 83
(417)
41 8 Journal of Agricultural Research voi. vi, no. n
they were allowed to feed for 1 1 days. Observation after two weeks showed unmis-
takable signs of incipient wilt around some of the beetle injuries on the leaves — that is,
a lighter dull green and slight flaccidity of the tissues. With the expectation that
the wilt would extend throughout the leaves the pouring of plates was deferred.
However, these incipient infection areas dried up without spreading further, and
consequently it was impossible to obtain cultures. That B. tracheiphilus was present
in the wilted vines fed to these beetles was shown by the subsequent inoculation of
cucumbers, cantaloupes, and squashes with cultures obtained from these wilted
vines (strains R230 and R235). Numerous inoculations with these two strains have
shown them to be virulent upon cucumbers and cantaloupes, but inoculations on
several varieties of squash have given nothing more than incipient infection.
Experiment 2. — On October 25, 19 15, striped cucumber beetles were collected at
Giesboro Point, D. C, in a squash field where bacterial wilt was very prevalent.
These beetles were fed for three days on plants which were wilting as a result of inocu-
lation with pure cultures of B. tracheiphilus. They were then placed in small boxes
provided with screened covers, and held in the ice compartment of a refrigerator at a
temperatiu-e of about 10° C. for five weeks and four days. At the end of this time
(Dec. 6) the beetles were removed and placed in cages containing young cucumber
plants. Four to six beetles were placed in each of the six cages used, each cage con-
taining three young plants. After being allowed to feed on these plants for 10 days
the beetles were removed and the plants kept in one of the Department greenhouses
where there had been no cucurbit wilt since the preceding spring and where no cucur-
bitaceous insects were present.
On December 17 leaves injured by the beetles on three of these plants were wilted.
Microscopic examination showed bacteria present in great number in the vessels of the
petioles, and poiired plates from the wilted leaves and petioles gave pure cultures of
the wilt organism (strain R313). Needle-prick inoculations from these cultures again
gave typical wilt on cucumber plants. On December 24 a gnawed leaf on a fourth
plant was found wilting, and was removed from the plant. Enormous numbers of
bacteria were present in the vascular tissues, and cultures (strains R31S and Eni26)
isolated therefrom gave also successful infection when pricked into the leaves of
young cucumber plants. From the portion of petiole remaining the wilt gradually
extended throughout the plant, which finally collapsed. On January 4 another
plant was found entirely wilted. The gnawed leaf which had wilted first, and from
which the wilt had spread throughout the plant, was photographed and preserved.
Cultures (strain Eni24) and paraffin sections (En36) were made from the petiole of
this leaf. The organism isolated gave typical infections when inoculated into cucum-
ber plants.
Experiment 3. — Another lot of D. vittata collected in the squash field referred to
in experiment 2 was fed for three days on old wilting stems of squash (C. maxima) col-
lected in the same field. After keeping these beetles in storage for two months under
the same conditions as in experiment 2, they were removed and caged for five days
with 12 young cucumber plants. Although these plants were under observation for
over two months no wilt appeared in any of them.
Experiment 4. — On December 16, 1915, five specimens of D. vittata and four of
D. duodecimpunctata hibernating under natural conditions in the squash field at
Giesboro Point, D. C, were sifted from the siu-face soil and taken to the greenhouse.
The striped and spotted beetles were placed at once in separate cages, each containing
three young cucumber plants. Although the beetles fed freely on these plants, the
results of this experiment were negative.
The negative results in experiment 3 possibly may be explained by
the fact that the wilted plants fed to the beetles were old, ripe squash
junei2, i9i6 Bacterial Wilt of Cucurbits 419
vines which had been diseased for a long time. Doubtless few living
organisms were present, since great difficulty was experienced in obtain-
ing cultures of B. tracheiphilus from this field (strains En 102 and Em 10).
The beetles used in experiment 4 were collected when hibernating in a
field where wilt was known to have occurred, but it is evidently not pos-
sible to determine whether they had fed upon wilted plants. On the
other hand, it is not reasonable to assume that all beetles which have
fed upon wilted plants would necessarily be able to carry infection
on their mouth parts for any great length of time. Experiments i
and 2 show that at least in some cases the striped beetles may
carry the wilt organism for at least five or six weeks and still be able
to infect healthy plants. This, in connection with the field experiments
previously published,^ seems to establish beyond doubt that D. mttata
is a winter carrier of the cucurbit organism.^ Experiments with other
species of insects have thus far given negative results, as here detailed.
In each of seven tests carried out with the common squash bug {Anasa
irisiis DeG.) during the summer and fall of 191 5 in field and greenhouse,
two to six of these insects were fed for one to three days on wilted cucum-
ber leaves and petioles and then inclosed with several healthy cucumber
plants. After feeding on these plants for one to two days the bugs
were removed and the plants kept under observation for three to four
weeks. No wilt appeared in any of these plants, but no absolute con-
clusion can be drawn from the negative results of so small a series of
tests.
The twelve-spotted (or squash) lady beetle (Epilachna borealis Fab.)
was very scarce in eastern Long Island during the season of 191 5, but
two tests with it similar to those outlined above gave negative results.
The melon aphis (Aphis gossypii Glov.) and the flea beetle (Crepidodera
cucumeris) apparently do not serve as wilt carriers. This has been
shown by the negative results from transfer of insects fed upon wilted
plants to healthy cucumber plants in insect-proof cages (three tests),
and b}^ the fact that no wilt developed during the season in cucumber
plants grown in 48 large screened cages (East Marion, Eong Island, N. Y.,
191 5), although numerous wilted plants occurred around all of these
cages, and aphids and flea beetles had free access through the meshes
of wire netting and were abundant both outside and inside the cages.
In only 2 out of 50 cages did wilt appear and in these cases striped
cucumber beetles had gained access or had been purposely introduced,
and the disease had started from points gnawed by the beetles.
' Rand, F. V. Op. cit.
2 Wild cucurbits may be eliminated as possible carriers of bacterial wilt so far as the experiments at
East Marion are concerned. Personal observations, together with those of Bumham and Latham
(Bumham, Stewart H., and Latham, Roy A. The flora of the town of Southold, Long Island and Gar-
diner's Island. In Torreya, vol. 14, nos. 11-12, 1914), and a search through the herbaria of the Xcw York
and Brooklyn Botanical Gardens, have established beyond doubt that no wild Cucurbitaceae occur within
10 to 15 miles of the experimental plots.
420 Journal of Agricultural Research Voi. vi, no. n
In each of eight direct summer field tests, one to five striped cucumber
beetles were fed for one to three days on wilting cucumber leaves and
petioles and then at once caged up with several healthy young cucumber
plants. In six out of these eight tests bacterial wilt appeared in one to
two weeks and only on plants gnawed by the beetles.
In the two fields (East Marion, Long Island, N. Y.) where spray tests
were carried out during the season of 191 5 the prevalence of bacterial wilt
closely followed that of the striped cucumber beetle. Throughout the
season careful and frequent observation failed to disclose a single case of
wilt which had not evidently started in a part of the plant injured by
cucumber beetles (PI. LUI). In these two fields no wilt had appeared
up to the ist of July. A few cases were observed on July 3, while
the greatest number of cases was found during the last 10 days of the
month. Practically no new cases of wilt appeared after the 30th of
July. The first striped cucumber beetles of the season were seen from
June 15 to 17. In field i the first beetles were found on June 17 between
cages 14 and 15.^ On July 3 there were only seven cases of wilt in the
whole field, and six of these occurred near or about where these beetles
had been collected. The beetles were most numerous between June 24
and July 8, in fact so numerous that in order to save the plants from
entire destruction an application of a proprietary dust insecticide (con-
taining lime, Paris green, etc.) was made upon the unsprayed plots.
Thus, for a few days, or until new growth appeared on the vines, there
were no untreated cucumber plants in these two fields upon which the
beetles could feed. From this date on, the beetles began to disappear
from these fields. In the variety-test block and commercial fields in
the vicinity the plants were younger and for the most part were untreated.
In fact, most commercial plantings were just breaking through the
ground on July 10. Such fields present an attractive feeding ground for
the beetles. In the two experimental fields there were only a few beetles
present on July 15, and they were exceedingly scarce after July 30.
When it is remembered that under field conditions usually one to three
weeks elapse between time of infection and the appearance of wilting in
the plants, it will be seen that the rise and fall in the number of plants
with bacterial wilt closely follows the rise and fall in the number of beetles
(fig. I).
The two fields just discussed had been planted to cucumbers the pre-
ceding season. About a quarter of a mile from field i a cucurbit variety
test block was located. This land had not been plowed for several years.
Although separated only by slightly rolling, plowed land from field i,
where striped cucumber beetles appeared on June 17, no beetles appeared
here until about the end of the first week in July. This was just after
* These beetles were used in the cage transmission tests recorded in a former paper (Rand, F. V., op. cit.)
and mentioned in a preceding paragraph.
Juae 12, 1916
Bacterial Wilt of Ciicurbits
421
PERCENTAGE OF BEETLES (MAX.IOOJ^)
^U3 O ^^ 00 CM (O
00 do *o -^ CO —
O ^ A CO r.
RAIN FALL (In Inches).
■*'
^rf?
^ jj
/
/I
<
^S
•|\
<f''
•0
)
^
<'
\
/
//
<
?^
*,
\
/
/
i
i:^
y
\
/i
^^
r
**•-
/
t 1
' 1
1
1
/
■{
^'(
ft^
r*
/
\
^ /
y ■*
1
,
V
.^'
7^'
^s
^{i
s
Y"
-*/,
} /
/ V
^
z
§5^
•V.o
a/
^'
t
y
I 1
^
s.^
1
...„^
\
(
\
I /
I f
( 1
c<
A
>
/
^fc
\ •
''■'^.>.
-^
>-b
.i
r
/ .
{'-
/'•
i
\
^
P'
♦.
f>\
^
N^
V •
c
<
>
,.5
■''^
>
V V '
ii
- <'
••*'
\V.]
^^
'••g
•
(
^
<
^
jji
1
k
\^
>\
•*
^''^
Jr^
<!
\
? ^
..*
'^^?
^
'^
<
f'
/
^"^^
\
>^Vv
'•**'° ■
ca
==f
^
:
i
Fig. I. — Comparison of the amount of wilt with striped-beetle prevalence and with meteorological
phenomena in three fields. East Marion, Long Island, N. Y., season of 1915.
422 Journal of Agricultural Research voi. vi.no. n
they had begun to disappear from field i. In the variety test the first
scattered cases of wilt were observed on July 17, whereas in field i the
first cases were noted on July 3. The largest number of cases of wilt in
the variety block were found between August 15 and 30, and the striped
cucumber beetles were most numerous here during the last part of July.
Again, allowing for the necessary time between infection and actual
wilting, it will be noted that here also there is a direct relation between
the number of wilt infections and the number of beetles present (fig. i).
The graphs (fig. i) show the daily relation between meteorolo-
gical conditions, the number of beetles present, and the number of
wilted plants in the three fields from June 10 to August 31. In these
graphs there is shown a definite relation between the beetle and the wilt
curves, but no relation between the latter and the meteorological curves.
The meteorological instruments from which the data were obtained for
this graph were kept in a United States Weather Bureau instrument
shelter at ground level, so that the environment would correspond as
nearly as possible to that of the cucumber plants (PI. LIV, fig. 4).
Reference should be made to the fact that in taking notes the total
number of plants showing bacterial wilt was recorded at each date of
observation. This number included not only the new cases but also
cases holding over from the preceding observation. Ordinarily the
older the plant at the time of infection the longer the interval between
infection and death. This explains the apparently too great interval
between the maxima of the beetle and wilt curves. If it had been the
original intention to represent graphically the relation between the
prevalence of the beetles and the occurrence of wilt, the data would
have been obtained in a form better suited to this method. It was only
after tabulating the results of the field observations that the very striking
parallel was noted. Obviously it would be impossible to enumerate
absolutely the beetles present in a field; hence, the percentages used in
the graphs are based partly on actual counts and partly on careful esti-
mates made throughout the season. In the curves, 100 per cent repre-
sents the maximum number of striped cucumber beetles present at any
one time.
Attention should be drawn to the fact that although there was a
difference of only three days in planting time between field i and field 3,
the beetles appeared between two and three weeks earlier in field i,
which had been planted to cucumbers the preceding season. This
would suggest that these insects hibernated in or near the old cucumber
field and that they did not leave this field the following spring as long
as young and tender plants remained for them to feed upon. A similar
tendency of both striped and twelve-spotted cucumber beetles to hiber-
nate in old cucurbit fields was observed by the writers near Congress
Heights, D. C. The first frosts occurred in these fields during the first
part of October. About the middle of December, 191 5, soil sif tings to
June 12. 1916 Bacterial Wilt of Cucurbits . 423
a depth of 7 inches were made at numerous points over this squash field.
Considerable numbers of dormant beetles were found under clods, old
vines, mummied squashes, and around the bases of old squash stems
just beneath the soil surface. No beetles were found below the first 2
inches and most of them were found at a depth of less than i inch.
SOriv TRANSMISSION
In the experiments of 191 5 at East Marion, Long Island, N. Y.,^ bacterial
wilt was not transmitted to the plants from the soil, although in the
same fields during the preceding season the crop had been largely
destroyed by this disease. In a large number of greenhouse inocula-
tions into one of two or more cucurbit plants in a single pot (seven
experiments, including in all 126 pots), none but the inoculated plants
ever took the disease, although the latter wilted to the ground, and the
pots were kept under obser\'ation from one to three months. The
house was free from cucurbitaceous insects.
In addition to these observations and experiments relative to soil
transmission three series of direct soil inoculations were made:
Series of March 18, 1915. — Thirty-two Arlington White Spine forcing
cucumber plants 4 to 5 inches high, transplanted March 9 and not dis-
turbed from that date until the date of inoculation, were inoculated as
follows :
Eight cucumber plants not root-pruned and the same number of plants
root-pruned were inoculated with strain R230 by pouring on the soil beef-
bouillon cultures 6 days old. Sixteen plants were inoculated in the
same way with strain R235. Sixteen plants were root-pruned and the
soil moistened with tap water only, these plants being held as checks.
The cultures used were tested as to virulence by needle-puncture inocu-
lations into the leaves of several cucumber plants of the same age and
variety.
All plants inoculated by needle puncture promptly wilted.
On April i the 16 inoculated plants which had been root-pruned showed
two cases of wilt. No wilt was evident in the 16 non-root-pnmed plants
at this date.
On April 12, among the 16 root-pruned plants there were 10 wilted
and among the 16 non-root-pruned there were 2 wilted. The 16 check
plants (root-pruned) showed no signs of wilt.
Isolations were made from all plants showing infection from the
soil, and these cultures produced wilt promptly upon inoculation into
leaves of healthy plants.
The experiment was continued for two months from the date of
inoculation, but no further cases of wilt appeared.
Series of March 31, 191 6. — Sixty Chicago Pickling cucumbers
planted January 28, 191 6 (transplanted once), in pots in the greenhouse
> Rand, F. V., 1915. Op. cit.
424 Journal of Agricultural Research voi. vi, No. n
were inoculated by pouring on the soil tap-water suspensions of B. irach-
eiphilus from beef-agar slants 6 days old. Of these 60 plants 24 were
root-pruned, and the remaining 36 were left uninjured. In this experi-
ment 26 strains, isolated from squash, cucumber, and cantaloupe
(Cucumis melo), were used, and each culture was proved to be virulent
by needle-puncture inoculation into the leaves of healthy cucumbers
of the same age and variety. The virulence tests were made about 30
minutes before these agar slants were used for the soil inoculations.
The plants were under daily observation, and there were no signs of
wilt until April 1 1 , when one of the root-pruned plants wilted. Between
this date and April 19, 6 of the 24 root-pruned plants (25 per cent) and
8 of the 36 uninjured plants (22 per cent) wilted. Examination of the
stems and main roots showed the typical stringy slime in the vascular
system, and cultures from these roots proved the presence of B. irachei-
philus.
It will be seen that in this test the percentage of wilt was about the
same in the root-pruned plants and in those not root-pruned. However,
too much weight can not be given to the results of this experiment, since
the cucumbers had been recently transplanted and examination of the
roots showed considerable eelworm injury. These wounds might, of
course, afford entrance for the wilt organism.
SEED TRANSMISSION
Ripe cucumber fruits were collected from wilted vines at Malone and
Constable, N. Y., on September 23, 1914. Five of the fruits from Malone
and one from Constable showed on cutting an abundance of the sticky
white ooze characteristic of this bacterial wilt, and microscopic examina-
tion revealed enormous numbers of typical bacteria in the vascular
system. The seeds were carefully preserved, and three months later
were planted in the greenhouse. Good germination resulted, and after
three months' growth no signs of wilt had occurred in any of the plants.
In July, 191 5, a large White Spine cucumber fruit almost full grown
was inoculated from a pure culture of the bacterial wilt organism. The
fruit became infected and the wilt extended gradually throughout the
whole vine to which it was attached. Seeds from this fruit were pre-
served, and six months later a part of them were planted in the green-
house. Several plants came up and were under observation for four
weeks, but no wilt occurred. A portion of the seeds remaining were
used in cultural tests. The seeds were steriUzed in the usual way with
mercuric-chlorid solution, the seed coats removed under aseptic condi-
tions, and the embryos crushed in sterile bouillon from which plates were
poured. No clouding of the bouillon subsequently occurred, and no
growth from the plates.
On August 29, 1914, a ripe cantaloupe was collected from a wilted
vine near Albany, N. Y. The vascular elements of the cantaloupe con-
junei2, i9i6 Bacterial Wilt of Cucurbits 425
tained an abundance of the typical stringy ooze which microscopic exami-
nation showed to consist entirely of characteristic bacteria. Seed ger-
mination and cultural tests similar to those described for the cucumber
gave negative results.
During the latter part of September, 19 14, ripe Hubbard squashes
were collected from wilted vines at IMedina, Malone, and Constable, N. Y.
These squashes upon examination showed the same evidence of bacterial
wilt as did the cucumbers and cantaloupes referred to above, and in
addition a pure-culture isolation of B. tracheiphilus was made from the
Medina squash, which subsequently gave typical infections when inoc-
ulated into healthy cucumber and squash plants. Seed germination and
cultural tests from the seeds gave the same negative results as in cucumbers
and cantaloupes.
STOMATAL INFECTIONS
Two inoculation tests with cucumber and one with cantaloupe were made
during the summer of 191 5, using sterile- water suspensions of the wilt
organism (strain R230) . The plants were put into tight inoculating cages,
and the plants and walls of the cages sprayed with tap water. Two hours
later the plants were inoculated by spraying with a very cloudy suspen-
sion of bacteria from 7-day-old agar slants. Check plants were inocu-
lated by needle punctures from the same cultures. All the plants were
left in the cages tightly closed for 24 hours, and semiopen for two days
longer. The punctured checks wilted promptly, but no infection occurred
in the uninjured sprayed plants, although they were kept under observa-
tion for two months. Another test was made in March, 191 6. Three
young and three older cucumber plants and four young squash plants
were inclosed in a dampened inoculation chamber and sprayed with a
tap-water suspension of B. tracheiphilus (strain En58 isolated from
squash) from a 6-day-old beef-agar slant. Three hours later the plants
were again sprayed with this bacterial suspension. This culture was at
the same time tested by needle-puncture inoculations into the leaves of
two cucumber and two squash plants of the same varieties. The sprayed
plants were left in the inoculating chamber in a saturated atmosphere for
three days, after which they were held under ordinary greenhouse condi-
tions. After two months no infection had appeared in the sprayed unin-
jured plants, although the plants inoculated from the same culture by
needle punctures all developed typical wilt within one week after inocu-
lation.
A fifth trial was made in April, 1916, using five young and five older
cucumber plants. All aerial parts were thoroughly sprayed mth a tap-
water suspension of the wilt organism from a beef-agar slant 6 days old
(strain Ensy). This culture was tested by needle-puncture inocula-
tions into cucumbers of the same age and variety. The latter inocula-
tions resulted in typical wilt, but the uninjured plants sprayed with the
bacterial suspension had shown no signs of infection after five weeks.
426 Journal of Agricultural Research voi. vi.no. n
In these five direct tests stomatal infection did not occur, thus con-
firming the observational data during the past two seasons and Dr.
Smith's earlier observations and experiments. In hundreds of field and
greenhouse observations the stems and leaves of wilted and healthy plants
were closely intertwined, exposing in many cases uninjured healthy parts
to direct contact with cut and broken infected surfaces. Even here the
disease was in no case transmitted.
DISCUSSION OF OBSERVATIONS
The field observations of the senior writer during the last two seasons,
covering the territory from the District of Columbia to eastern L-ong
Island, northward to the Canadian Provinces of Quebec and Ontario,
and westward to Michigan, Wisconsin, and Indiana, have abundantly
confirmed the experimental evidence outlined above that the striped
cucumber beetle and probably also the twelve-spotted cucumber beetle
are the principal if not the only carriers of bacterial wilt of cucurbits.
It has been suggested that the larvae of cucumber beetles may also serve
as a means of dissemination, but from their habits it would appear that
the only possible way in which they could bring about infection is by
carrying the organism from the soil into their burrows in the cucumber
stems. This appears highly improbable. However, the data at hand
do not warrant any definite statement.
Mechanical injuries, such as those resulting from storms, cultivation,
etc., and injuries from flea beetles, aphids, and squash bugs have been
closely watched in the experimental fields and cages described elsewhere,
but no evidence has been obtained of any relation to bacterial wilt.
WILT CONTROL
The problem of control therefore resolves itself into (i) the finding
or developing of cucurbit varieties resistant to bacterial wilt, (2) spray-
ing the plants with a bactericide, or (3) eliminating the beetles through
poisons or repellants.
VARIETY TESTS
Early in the spring of 191 5 a preliminary test was made with eight
varieties of cucumber planted in pots in one of the department green-
houses. Several plants of each variety were inoculated by needle punc-
tures in the leaves from 6-day-old agar-slant cultures of a single strain
of B. iracheiphilus. All the inoculated plants contracted the disease
and no difference in rapidity of wilting appeared — that is, individual
plants of the same variety showed as great differences in rate of wilting
as appeared among the different varieties.
In the variety-test block previouslv mentioned (East Marion, Long
Island, N. Y.) 32 varieties of cucumber, 39 varieties of cantaloupe, and
25 varieties of squash were planted on June 10, 191 5. From 8 to 20 or
more hills were given to each variety, 12 being the usual number. Most
of the cucumber and squash varieties gave fair to good stands, but the
cantaloupes were planted in an exceedingly light sandy soil infested
with witch grass, and in consequence of this the seed either did not come
June 12, 1916
Bacterial Wilt of Cticurbits
427
up at all or gave a very poor stand of plants. Only seven of the canta-
loupe varieties were in such location and condition as to be included in
a summary of results. It was intended at first to inoculate artificially
at least one plant of each variety with the wilt organism, in order that
all varieties might have an equal chance, but the disease soon became so
general over the experimental block that it was thought unnecessary to
interfere with its natural spread.
Careful observations were made throughout the season and the num-
ber of wilted plants in each variety was noted. Table I gives the per-
centage of wilted plants for each variety during the season.
Table I. — Percentage of wilt in different varieties of cucumbers, squashes, and cantaloupes
at East Marion, Long Island, N. Y., season of igi^
CUCUMBER
Variety.
West India Gherkin
Rollistons Telegraph
Emerald
Cool and Crisp
Vaughans Prolific
Lemon
Westfield Chicago Pickling. . . .
Snows Fancy Pickling
Davis Perfect (regular stockj . .
Davis Perfect (selected stock) .
Noas Forcing
Extra Early Long White Spine
Boston Forcing White Spine . . .
Improved Jersey Pickling
Boston Pickling (U. S. 18589).
Rockyford Klondyke
Japanese Climbing
Percent-
age of
wilt.
30
2,i
40
40
44
45
50
50
50
66
66
66
66
66
66
Variety.
Improved Long Green
Fordhooks Famous
Vaughans XXX Pickling
Cumberland Pickling
Fordhook Pickling
Early Cyclone
Improved White Spine
Arlington White Spine
Arlington White Spine (U. S,
19300)
Improved Long Green (U. S,
18590
Early Cluster
Serpent or Snake
Carters Model
New Centiuy
Grand Rapids Forcing
Percent-
age of
wilt.
66
70
71
75
77
77
77
80
83
83
83
88
90
100
100
SQUASH
Mammoth White Bush
Early White Bush (U. S. 19339)
Vaughans Giant Summer Crook
neck
Early White Bush
Early White Bush (U. S. 19340) ,
Mammoth Yellow
Giant V/hite Summer
Straight Neck
Bush Fordhook
Fordhook
Yellow Bush
Summer Crookneck
Improved Hubbard
Pikes Peak
Delicata
Essex Hybrid
Delicious
Faxons Brazilian
Chicago Market Hubbard.
Orange Marrow
Red or Gold Hubbard
Marblehead
Golden Bronze
Boston Marrow
Vegetable Marrow
42
42
50
50
60
70
75
87
88
100
100
100
100
CANTALOUPE
Emerald Green (U. S. 19352). . . .
Landreths Early Citron
Rockyford (V. S. 19319)
Rockyford (regular stock)
Netted Gem Rockyford (selected
stock)
Burrell's Gem (U. S. 19312).
Burrell's Gem (U. S. 19348).
Vegetable Peach
Oval Netted Gem
25
25
28
66
428 Journal of Agricultural Research voi. vi, no. h
None of the 30 varieties of cucumber were free from wilt, the diseased
plants in each variety ranging from 30 to 100 per cent. In the 7 varie-
ties of cantaloupe exposed to infection, the wilt ranged from 9 to 66
per cent. Of the 24 varieties of squash, 2 remained free from wilt
throughout the season, and in the remaining 22 varieties the disease
occurred in 10 to 100 per cent of the plants. Little hope of finding a
high degree of resistance is to be noted in the cucumber record. A
considerable difference in percentages of wilt is found, but whether
this will persist from year to year or is merely accidental can be ascer-
tained only by further trials in different localities and seasons. A
greater promise of resistance was evidenced by the squash varieties.
In his experiments Dr. Krwin F. Smith, Bureau of Plant Industry,
obtained infection in squashes with B. iracheiphilus obtained from
wilted squash plants, but with cultures obtained from cucumbers squash
infections were rare or where they did occur failed to extend beyond the
immediate vicinity of inoculation.
Experiments relative to the infection of squash plants by means of
cultures of B. iracheiphilus obtained from cucumbers, cantaloupes, and
squashes are not yet completed. However, up to the present time, 15
strains from cucumber, i from cantaloupe, and 7 from squash have been
tested by inoculation into these three hosts. All the strains have proved
infectious on cucumber and cantaloupe. Of the 15 cucumber strains
inoculated into the Yellow Crookneck and Early White Bush squashes,
7 have given no infection, 2 (En66 and En68) have given doubtful
signs of incipient wilt, 4 (En68, Eniop, R305, and R307) primary wilt
(not extending beyond the inoculated leaves), and 2 (R308 and En 108)
wilt involving the entire plant. The single cantaloupe strain in most
cases failed to infect squash. In those cases where infection did occur,
the signs did not extend far beyond the inoculation punctures. All the
squash strains were infectious to squash, varietal differences, however,
being evident.
Among the common cultivated cucurbits cucumbers appear to be the
most susceptible, and following them in succession should be placed
cantaloupes, squashes, and pumpkins, with watermelons (Citrullus vul-
garis) as most resistant. So far as the writers know, bacterial wilt has
been reported but once as occurring naturally upon watermelons, and
this case was reported without accompanying proof.^ The ordinary
watermelon wilt is caused by a species of Fusarium.
Summarizing the season's work upon cucurbit varieties, together with
the general field obser\^ations of the senior writer, it may be stated
that there is little hope of controlling the disease in the cucumber through
host resistance to the parasite. The cantaloupe and squash, especially
1 Selby, A. D. Certain troublesome diseases of tomatoes and cucurbits. In Ann. Rpt. Columbus Hort.
Soc. 1896, p. 113. 1897.
juneis. i9i6 BacteYial Wilt of Cucurbits 429
the latter, showed a considerably greater evidence of resistance. For
these plants, therefore, this method of control is at least worthy of
further investigation, but up to the present time the observations and
experiments do not justify definite conclusions.
SPRAYING
In two fields situated near the variety-test block a series of spraying
experiments was carried out in 191 5 upon the Fordhook Famous cucum-
ber, planted on June 5 and 7, and Woodruffs Hybrid cucumber, planted on
June I. The relation between the striped cucumber beetle and wilt in
these two fields has already been detailed (p. 420 and fig. i). The relative
merits of Bordeaux mixture alone, Bordeaux mixture with lead arsenate,
and lead arsenate alone were tried out by spraying different plots with
each of these three mixtures on a succession of dates, beginning June 25
and continuing at intervals of 5 to 10 days thereafter (fig. 2).
To determine the best time for treatment, the first application of the
Bordeaux-mixture-lead-arsenate combination was made upon different
plots at successive dates. The first application was made on June 25, just
as the first true leaves had opened on the cucumber plants, and at each
succeeding application a new plot was added. In every case a check
plot was left between the two successively sprayed plots. In field 2
each plot consisted of three parallel rows, each row 21 feet long. In
field I the plots were about twice this size. The first three applications
of Bordeaux mixture were made with a weak suspension (2:2: 50) in
order not to injure the young plants, but in the later treatments the
strength was gradually increased to the 4 : 6 : 50 formula. In all cases
where lead arsenate was used it was applied at the rate of 2 pounds to
50 gallons of liquid. No appreciable injury from any of the spray
mixtures was observed.
The relative amount of control effected in field i at different dates of
appHcation is graphically shown in figure 3.
The spray treatments were conducted as follows: Plot i (fig. 3a)
received its first application of Bordeaux and lead arsenate on June 25,
and additional sprayings at intervals of 5 to 10 days throughout the
season. In plots 2, 3, and 4 (fig. 3, b, c, and d), the first applications
were made on July 6, 14, and 19, respectively, and further sprayings were
made at intervals of 5 to 10 days as in plot i. It will be noted that most
of the infections had occurred before the third treatment, July 14, for in
plot 3 and its corresponding check the number of wilted plants was about
the same. In the first two plots there was much less wilt than in the cor-
responding unsprayed plots, the first sprayed plot showing by far the
best results. There would be a still greater difference between plot i and
its untreated check were it not for the facts that the latter was only about
three-fourths the size of the sprayed plot and that it received one applica-
430
Journal of Agricultural Research
Vol. VI, No. II
tion of a dust insecticide to prevent the total destruction of the plants
by striped cucumber beetles.
The relative amount of control given by the three kinds of spray mix-
ture tested is shown in figure 2, in which the number of wilted plants in
each sprayed and check plot is given. It will be noted that the lead arse-
32
1:520
FIELD N0.1
Bordeaux and lead
— Borcfesi/x A/o/7e
Lead A/one
Fig. 2. — Comparison of relative wilt control of Bordeaux mixture plus lead arsenate, Bordeaux mixture
alone, and lead arsenate alone in field i. East Marion, Long Island, season of 1915.
nate and Bordeaux mixture combined gave better results than either used
alone.
The results obtained in field 2 are corroborative of the data graphically
shown for field i (fig. 2 and 3), but the control effected was not quite so
striking, since the plants were nearly a week older than those in field i at
the time of the first spraying. Furthermore, the stand was poor in some
June 12, 1916
Bacterial Wilt of Cucurbits
431
I
A
/
P ■
1
f
h
V
1
'
1
/
/
/
/
^
>
■^
L
f
l2
Kj
^i
y
^
^ ^^
7
-A
__JSlI::
__Jl
_/
A
C
\
N
\
,4
p
1
\
\
\
^
f
\
V
/
\
y
r
>/
f
/
7
1 5 9 13 (7 21 25 29 2 6 10
JULY AUGUST JULY AUGUST
Fig. 3. — Curves showing relative wilt control of Bordeaux mixture and lead arsenate with date of first
application as a variant in field i. East Marion, Long Island, season of 1913.
37768°— 16 4
/?
V
^
^
,
A,
•V,
9
r^
N
'3
'/,
4
5
r
\
'/i
^
i
r
i
I
/
/
1 !
) S
1
} 1
7 2
\ 1
5 2
9 I
. e
10
432 Journal of Agricultural Research voi. vr, No. h
parts of the field and the beetles appeared here a few days earlier than in
field I .
Field observations indicate that the results obtained were due to a bac-
tericidal and repellent action of the Bordeaux mixture and lead-arsenate
combination, and probably in part to an insecticidal action by the latter
ingredient. The beetles were less frequent on the sprayed than on the
unsprayed plots, and among the sprayed plants injured by beetles there
was apparently a smaller percentage of infection resulting than among
similar unsprayed plants. That is, the control effected by the Bordeaux
mixture alone was apparently due to its repellent and bactericidal action,
and that by the lead arsenate alone to its repellent and insecticidal action,
while the more complete control by the two mixtures together was due to
a combination of their bactericidal, repellent, and insecticidal properties.
The bactericidal action of Bordeaux mixture has been further investi-
gated in a series of six greenhouse tests, in which sprayed and unsprayed
leaves of potted plants were inoculated in as nearly an identical manner
as possible by needle punctures from cultures of the same strain of organism.
The spray used was 2 13 150 Bordeaux, and this was allowed to dry
thoroughly on the leaves before inoculating. In most cases the plants
were not inoculated until 24 hours after spraying.
In the first test, December 2, 191 5, three weeks after planting, seven
unsprayed and seven sprayed Chicago Pickling cucumber plants were
inoculated from i -week-old beef-agar slant cultures. After 15 days
the unsprayed plants showed 100 per cent of infection, and the sprayed
plants 29 per cent.
In the same way and at the same time a test was carried out on three
varieties of cantaloupe — Rockyford, Sweet Air, and Baltimore Nutmeg.
Thirty-five inoculations were made into unsprayed plants and 37 into
sprayed plants. There was no apparent difference in susceptibility
among the three varieties used. Of these inoculations the unsprayed
gave 95 per cent of infection and the sprayed leaves 46 per cent.
In a third test (Jan. 8, 191 6), Chicago Pickling cucumbers planted
November 13, 1915, were used. In this test 36 unsprayed and 37 sprayed
plants were inoculated with the wilt organism, using agar slants 9 days
old. After 19 days it was found that 92 per cent of the unsprayed and
35 per cent of the sprayed cucumbers had contracted the disease.
A further trial was made (Jan. 19, 191 6) with 19 Chicago Pickling
cucumbers planted October 29, 191 5. In the case of these older cucum-
bers unsprayed and sprayed leaves on the same plant and as nearly of the
same age and appearance as possible were used for inoculation. Both
sprayed and unspra}'ed leaves had been dusted with flowers of sulphur
for the control of powdery mildew, and this treatment, together with
the age of the plants, considerably reduced the infection. However,
even here the unsprayed leaves gave 63 per cent and the sprayed leaves
1 1 per cent of infection.
junei2, i9i6 Bacterial Wilt of Cucurbits 433
Two more tests (Jan. 19, 191 6) were made with Baltimore Nutmeg
cantaloupes planted November 13, 191 5, using a bacterial strain of low
virulence (strain R311) and one of high virulence (strain R304). The
cultures of these two strains used for inoculation were beef-agar slants
10 days old. With the former strain 10 unsprayed and 9 sprayed plants
were inoculated, and these gave, respectively, 40 and 11 per cent of
infection. With the highly virulent strain 16 unsprayed and 17 sprayed
plants were inoculated. These gave, respectively, 94 and 24 per cent
infection.
Remarks : It will be seen that in all cases the presence of Bordeaux
mixture on the leaves greatly reduced infection, and an average of the
six trials gives 80.6 per cent of infection in the unsprayed against 26 per
cent of infection in the sprayed plants. These results can scarcely be
considered as accidental, and they strongly confirm field observations
regarding the bactericidal effect of Bordeaux mixture. Furthermore, the
natural mode of inoculation is considered identical with the method used
in these tests, for in the one case the organism is pricked into the leaf
tissues by the mouth parts of the cucumber beetle and in the other case
by the inoculating needle.
Wet and Dry Inoculations
On January 8, 1915, an experiment w^as conducted to determine the
effect of wet and dry inoculations into sprayed and unsprayed plants.
In this test 68 cucumber plants were used. The inoculations were all
made in a uniform manner by needle punctures into the two youngest,
fully opened leaves of each plant. Of these plants 34 were sprayed with
Bordeaux mixture and 17 were inoculated before the Bordeaux mixture
had dried. The remaining 17 were inoculated about 2 hours later when
the Bordeaux mixture was thoroughly dry. At the same time 34
unsprayed plants were inoculated, 17 while dry and 17 immediately after
sprinkling with tap water. All of the plants were shaded from the sun
until the following day. At the end of 19 days after inoculation 95 per
cent of the unsprayed plants inoculated when wet had contracted the
wilt and 88 per cent of those inoculated when dry. In the sprayed
plants there was 33 per cent of infection among those inoculated before
drying and 36 per cent among those inoculated after drying.
As will be seen, the percentage relations between infection in wet and
dry leaves vary inversely in the sprayed and unsprayed plants. The
difference is small, but it occurs in the direction to be expected from
known facts concerning conditions favorable to infection. In the
absence of bactericidal substances a moist leaf surface presents a better
environment for infection by the bacteria; but when a bactericide is
present which is effective in solution the maximum effect occurs in the
presence of water. This is exactly the result obtained in the experiment
under discussion.
434 Journal of Agricultural Research voi. \^, no. n
SUMMARY
(i) In fields where wilt had largely destroyed the cucumber crop dur-
ing the preceding season the disease did not appear in 1915 on cucumbers
in 48 beetle-proof cages. On the other hand, wilt was very prevalent in
those fields on all sides of the cages. In a large number of greenhouse
tests where one out of two plants in a pot was inoculated and wilted to
the ground the second plant in no case contracted the wilt. The inocu-
lations by means of bacterial suspensions poured on the soil around
potted cucumber plants showed a small but varying percentage of wilt.
Root injuries were found in most of these cases of root infection. Ap-
parently infection does not enter the uninjured root system from the soil.
(2) In all cases seeds from diseased fruits failed to produce diseased
plants, and cultures from such seeds in no case gave the wilt organism,
but further tests should be made.
(3) In the tests made stomatal infection did not occur.
(4) The experiments thus far completed show that cucumber beetles
{Diahrotica spp.) are the most important, if not the only, summer car-
riers of the wilt organism {Bacillus tracheiphilus) and that at least one
species (D. vittata) is capable of carrying the wilt over winter and infecting
the spring planting of cucumbers. In the tests by the writers the squash
bug {Anasa tristis), the flea beetle (Crepidodera ciicutneris), the melon
aphis {Aphis gassy pii), and the twelve-spotted lady-beetle {Epilachna
borealis) have failed to transmit the disease.
(5) In the field experiments during one season with many different
varieties of cucurbits, the greatest difference in resistance was shown by
varieties of squash, in which the percentage of infection varied from o to
100. The varieties of cucumber and cantaloupe, while showing some
difference in their susceptibility to the wilt, give much less promise of
control by varietal resistance.
(6) In the spraying experiments of 1915 wilt was effectively con-
trolled by early treatments with a combination of B.ordeaux mixture
and arsenate of lead. Plots sprayed with either mixture alone showed
much less wilt than unsprayed plots, but control was not as complete as
where the two were used together. Both field observations and green-
house experiments indicate that the wilt control is effected through the
bactericidal action of Bordeaux mixture, the insecticidal action of
arsenate of lead, and the repellent action of both against the cucumber
beetles.
(7) Inasmuch as it has been definitely proven that the striped cucum-
ber beetle {D. vittata), and also the twelve-spotted cucumber beetle (£>.
duodecimpunctata) , are the most active carriers of the bacterial wilt, it
becomes necessary to control the insects in order to prevent the disease.
This phase of the work will be actively undertaken in cooperation with
the Bureau of Entomology during the coming season.
PLATE LIII
Two wilted cucumber plants which contracted bacterial wilt at beetle gnawings of
the leaves marked x. Three healthy, uninjiu-ed plants in same hill are also shown.
From field i, East Marion, Long Island, N. Y., July 19, 1915-
Bacterial Wilt of Cucurbits
Plate LI 1 1
Journal of Agricultural Research
Vol. VI, No. II
Bacterial Wilt of Cucurbits
Plate LIV
V«
¥«#^'
Journal of Agricultural Research
Vol. VI, No. 11
PLATE LIV
Plots in field i, East Marion, Long Island, N. Y., 1915. The poor stand In figures 2
and 3 was due entirely to bacterial wilt.
Fig. I. — Plot sprayed with Bordeaux mixttire and lead arsenate, beginning June
25. Photographed September 20, 1915.
Fig. 2. — Plot sprayed with Bordeaux mixture and lead arsenate, beginning July
19, after most of the striped-beetle injury had occiured. Photographed September
20, 1915.
Fig. 3. — Plot sprayed with Bordeaux mixtiu-e and lead arsenate, beginning July
27. Practically no beetle injmy occiured after this date. Photographed September
20, 1915.
Fig. 4. — General view of field, showing cages and meteorological-instrument shelter.
Photographed July 10, 1915.
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, $3.00 Per Year
'i . '-t I -
Vol. VI JUNE 19, 1916 No.' 12
JOURNAL OP
RESEARCH
CONTENTS
Page
Correlated Characters in Maize Breeding - - -435
G. N. COLLINS
Comparative Study of the Amount of Food Eaten by Para-
sitized and Nonparasitized Larvse of Cirphis unipuncta - 455
DANIEL G. TOWER
Aleyrodidae, or White Flies Attacking the Orange, with De-
scriptions of Three New Species of Econamic Importance 459
A. L. QUAINTANCE and A. C. BAKER
DEPARTMENT OF AGRICULXUEE
■VVA.SHINGTON, D.C.
WAeHINVroN : OOVSaNMENT PRINTIOKi OfFIQE : 1818
rmmB
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICUETURE, WITH THE COOPERATION
OF THE ASSOCIATION OF AMERICAN AGRICUE-
TURAE COEEEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPAKTMENT
FOR THE ASSOCIATION
KARL F. KELIvERMAN, Chairman RAYMOND PEARL
Physiologist and Assistant Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Siaiions
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomolozv
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
DirectOtT, Institute of Ani^nal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural Experiment Station of
the University of Minnesota
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 Experiment Stations ' should be
addressed to Raymond Pearl, Journal of Agricultural Research, Orouo, Maine.
JOlfflSL OF AGRKIIIIRAL RESEARCH
DEPARTMENT OF AGRICULTURE
Vol. VI Washington, D. C, June 19, 191 6 No. 12
CORRELATED CHARACTERS IN MAIZE BREEDING
By G. N. CoLUNS,
Botanist, Office of Acclimatization and Adaptation of Crop Plants, Bureau of Plant
Industry
INTRODUeXION
The study of correlations as an aid to plant breeding was at one time
thought to be full of promise, but in recent years little use has been made
of correlations by practical workers. From this fact it might appear
that the early hopes were unwarranted, and that correlation is a factor
of little or no importance. It must be conceded that the elaborate
calculations of correlation coefficients have in few instances proved of
value to the practical breeder, yet it must be admitted on reflection that
nearly all successful breeding has in reality been made possible by the
fact that correlations exist.
In plant breeding the improvement and preservation of varieties has
largely resulted from the ability of the breeder to recognize desirable
types, and the existence of definite types is in itself a manifestation of the
correlation of characters. The existence of types must mean that there
are many individuals that present approximately the same combination
of characters, and this is exactly what correlation implies. The charac-
teristics of the desired type are recognized by the breeder even though they
may not be formulated, and varieties are seldom established by selection
confined to a single character. If the study of correlations has appeared
to have little bearing on plant breeding, it must be that we have been
studying the wrong characters or studying them in the wrong way.
In the improvement of maize varieties (Zea mays), as with other plants,
the recognition of types has been an important factor. The selection,
however, has been almost entirely confined to the ear. In a field of any
commercial variety it is easy to recognize differences in the plants, but
even after long familiarity with the variety the plants refuse to be classi-
fied into distinct groups. This difficulty in recognizing types among
maize plants greatly increases the difficulty of breeding this crop.
The lack of recognizable types in maize is very different from the con-
dition that obtains, for example, in cotton (Gossypium spp) . With cotton.
Journal of Agricultural Research, Vol. VI, No. la
Dept. of Agriculture, Washington, D. C. June 19, 1916
ee Cr— 84
(435)
436 Journal of Agricultural Research voi. vi, no. u
skilled breeders are able to detect deviation from type even in the early
stages of development and the practice of roguing can proceed with cer-
tainty. It appears that when a cotton plant deviates from type it
deviates in a more or less definite way and in many particulars, or, in
other words, there are a number of coherent or correlated characters.
It seemed desirable to determine whether the difficulty in recognizing
types in maize is due to a lack of familiarity with the plants or whether
there is a fundamental difference between the heredity of maize and that,
for example, of cotton.
In the seed characters of maize a definite correlation has been found
between the color of the aleurone and the texture of the endosperm (Col-
lins and Kempton, 1913). Correlations have also been noted between the
color of the silk and the color of the anthers (Webber, 1906), and between
the color of the seed and the color of the cob, dwarfness and broad leaves,
and between stamens in the ear and club-shaped tassels (Emerson, 191 1).
There was, therefore, abundant reason for suspecting that the difficulty
of recognizing types among maize plants might be due to a lack of suffi-
cient discrimination, and it was with the idea that correlations were the
rule rather than the exception that the present experiment was under-
taken. Contrary to expectation, the results give evidence that for the
varieties and characters studied there is almost a complete absence of
genetic correlations.
CLASSIFICATION OF CORRELATIONS
Correlations may be classified in a great variety of ways and with almost
any degree of refinement. As with any classification of organic activities,
no particular grouping can be made to serve all purposes, for it is necessary
to divide the subject in different planes.
For purposes of the present discussion correlations, or the mutual
relations of characters, are divided into three main groups, to which the
names "physical," "physiological," and "genetic" may be applied.
Physical correlations are those in which the relation is obviously
causal. In many instances correlations of this kind are little more than
different names for the same phenomenon, or parts of the same phe-
nomenon, as when increased weight is correlated with increased height.
In physical language one of the characters would be described as a
function of the other.
Physiological correlations are those where both characters are
the result of the same physiological tendency, as when long intemodes
in the main stalk are correlated with long internodes in the branches.
This may be looked upon as a general tendency to elongated growth
that is manifested in different parts of the plant.
Genetic correlations cover the large residue of correlations, the
nature and causes of which are questions of controversy, but which are
June 19, 1916 Correlated Characters in Maize Breeding 437
associated with the method or mechanism of heredity. An example of
this type of correlation is shown in the association of yellow petals and
deeply lobed leaves in Egyptian X Upland cotton hybrids.
This classification differs from those proposed by Webber (1906) and
East (1908) chiefly in placing physical correlations outside the pale of
biological correlations. Most of those correlations classed by Webber as
morphological would here be considered as physical. This distinction
is made because it seems to the writer that the relation between length
and weight, for example, is inherent in the properties of matter and is not
a biological phenomenon. Certainly a relation of this kind would be
found in stones or any inanimate objects selected at random.
Since physiological functions are always directly or indirectly induced
by or at least associated with environmental stimuli, Webber's environ-
mental and physiological correlations are here combined. That the
examples of physiological correlations cited by Webber are reverse or
negative correlations need not confuse the issue, since by simply stating
the relation in other terms the correlations can be made to appear as
positive.
The distinction between physiological and genetic correlations may
not always be easy to apply, and the apparent need of it may disappear
entirely with a more complete knowledge of inheritance and methods
of growth. For the present, however, the distinction will be useful even
if physiological correlations are confined to pure lines or asexually
propagated stocks where differences in inheritance can be eliminated.
To ascribe the long intemodes of the main stem and branches to the
activity of a single determiner or gene is hardly less futile than to offer
the same explanation for the correlation between the length and weight
of inanimate objects. If the one is inherent in the properties of matter,
the other is inherent in the properties of plants.
All examples of genetic correlation are exceptions to the third law of
Mendel, which implies that characters are redistributed in the perjugate
generations of a hybrid in accordance with the laws of chance. Con-
versely, all instances in which Mendelian ratios, other than the 3 to i
ratio of a monohybrid, are followed with exactness demonstrate the
action of this third law and the absence of correlations among the factors
which make up the characters. It should be kept in mind, however, that
multiple hybrid ratios have seldom been determined with any great degree
of accuracy, so that correlations, unless of a pronounced type, would
escape detection.
The significant factor in genetic correlations is the grouping of the
characters in the ancestry and not the inherent properties of the charac-
ters themselves. Thus, when colored aleurone and horny endosperm
are found to be correlated in the progeny of a hybrid, involving colored
and white aleurone and homy and waxy endosperm, it does not indicate
438 Journal of Agricultural Research voi. vi, no. 12
any attraction between colored aleurone and horny endosperm, but
rather that one of the parents had colored aleurone and horny endos-
perm, while the other parent had white aleurone and waxy endosperm.
This tendency fcr parental combinations to reappear has been called
"coherence," and, so far as known, all genetic correlations thus far re-
corded are of this nature.
Many investigations have been devoted to correlations in agricultural
plants, but unless the special class of correlations covered by coherence
is kept in mind the results are likely to be disappointing to the breeder.
Cylindrical ears of maize may be correlated with high yield in one popu-
lation and the opposite result be reached in another case, depending on
whether these characters were introduced into the population under
investigation from the same parent or from different parents.
There are doubtless many physiological correlations that may be de-
tected by elaborate measurements, but unless the observations are con-
fined to asexually propagated groups or to those of which the ancestry
has been carefully studied, there will always remain the uncertainty
whether there is an inherent physiological relation between the develop-
ment of the two characters or whether the correlation is the result of
ancestral combinations. The distinction is not without practical impor-
tance, for a physiological correlation can not be reversed by any direct
means at the disposal of the breeder — that is, without evoking mutation
or some form of evolutionary change — while, if the correlation is genetic,
the relation between the characters may usually be reversed by a few
generations of selection in the desired direction.
Two principal theories have been advanced to explain genetic correla-
tions. These are the theory of reduplication (Bateson, Saunders, and
Punnett, 1906) and the theory of linkage developed by Morgan and his
students (1915) from studies of the fruit fly Drosophila ampelophila.
Both of these theories deal with characters which are alternative, both
having been derived from the study of Mendelian inheritance.
With the idea that continuous inheritance is to be looked upon as a com-
plicated form of alternative inheritance, it should be interesting to
learn what light the study of genetic correlations between characters
that are blended in inheritance may throw on the theories of reduplica-
tion and linkage. The experiments described below constitute a prelim-
inary attempt to extend the study of genetic correlations to characters
that are continuously inherited.
METHODS OF DISTHNTGUISHING BETWEEN PHYSIOLOGICAL AND
GENERIC CORRELATIONS
To determine with certainty that a given correlation is physiological
and not genetic, it would be necessary to demonstrate the existence of
the correlation in material where all the individuals possessed the same
June 19, 1916 Correlated Characters in Maize Breeding 439
hereditary tendencies with respect to the characters studied. The-
oretically this is only possible in asexually propagated groups. Ap-
proximately pure lines may be obtained where self-pollination is possible,
so that if correlations are found they may with assurance be considered
physiological. In maize, however, even to approximate pure lines
produces such abnormal conditions that some other method of study
must be sought.
Even in maize it would seem that the question might be approached
by comparing the degree of correlation in types or varieties having a
relatively restricted ancestry with that observed in the perjugate genera-
tions of hybrids between two contrasting forms.
An equally satisfactory method is to compare the degree of correla-
tion in the first or conjugate generation of a hybrid with that of the per-
jugate generations. Where the conjugate generation is all descended
from a single cross, the gametic differences should be no greater than self-
pollinated progenies of the parents.
Unfortunately in our experiment the number of first-generation indi-
viduals w'as too limited to detect any but relatively large correlations.
Wherever data were available, additional evidence has been presented
from the behavior of the original varieties. Although a large number
of plants of both parent varieties have been grown and measured, the
data have been secured in different localities and in different years, a
fact that renders many of the measurements unavailable for these studies.
DESCRIPTION OF MATERIAL
The hybrid that afforded the data for the present paper was a cross
between Waxy Chinese and Esperanza, two varieties of maize separated
by a number of definitely contrasted characters. The hybrid was made
at Lanham, Md., in 1908.
The peculiarities of the Waxy Chinese variety (PI. LV-LVI) have
been described elsewhere (Collins, 1909).
The particular plant used as female parent of the hybrid was grown
from the original seed imported from China. The individual notes taken
in 1 908 give the following details :
Height, 167 cm. Length of fifth leaf from the top, S>7, cm. Width oF fifth leaf, 9 cm.
Leaf sheath smooth. Nodes above the ear, 4. Suckers, o. Plant rather open, but
distinctly one-sided.
The Esperanza variety belongs to a peculiar type of maize that appears
to be confined to the table-lands of Mexico, the Zea hiria of Bonafous
(1829). This variety was obtained in 1906 from Esperanza, Pueblo,
Mexico, by Mr. H. Pittier, of the Bureau of Plant Industry (PI. LVIII
and LIX).
The plant that was the male parent of the hybrid was raised from seed
grown at Lanham, Md., in 1907. Regarding the 1907 plants, the notes
440
Journal of Agricultural Research
Vol. VI, No. 12
State that all were typical of the hairy Mexican type, ranging from 150 to
210 cm. in height. The notes recorded for the 1908 plant used in making
the hybrid state that it was typical of the variety except for a general
shortening of the intemodes. It was 105 cm. high, had three tassel
branches, four nodes above the ear, and the fifth leaf from the top
measured 83 by 14 cm.
Sixteen first-generation plants were grown in 1909. Three pure-seed
ears that provided seed for the second generation are designated as No.
1,2, and 3. Four plants entered into the parentage of these three ears.
No. I and 2 were reciprocals. No. i resulted from pollinating plant 225
by plant 226. No. 2 by pollinating plant 226 by plant 225. No. 3 was
the result of pollinating plant 262 by plant 263. The ears on all three of
the first-generation plants that produced ears 1,2, and 3 showed the usual
mixture of waxy and horny seeds that result from crossing the Waxy
Chinese and a corneous or homy variety. The notes taken on the four
first-generation plants are presented in Table I.
Table I. — Description of four first-generation maize plants grown in IQOQ
Plant No.
Height cm.
Number of tassel branches ,
Nodes above the ear
Length of fifth leaf cm .
Width of fifth leaf mm.
Exsert of tassel
Arrangement of leaves
222
20
5
14
(>2,
(^)
228
5
13
76
(^)
Q>)
212
18
5
14
86
o
263
230
17
5
13-5
a Exserted.
6 Scorpioid.
c Neither monostichous nor scorpioid.
The final planting was made in 191 4. The remnant of seed from the
original hybrid ear was planted and furnished 31 first-generation plants.
Six rows of approximately 30 plants each were secured of second-gen-
eration plants, one row from waxy, and one from the horny seeds of each
of the first-generation ears.
The means of the characters measured are given in Table II, and the
coefficients of variation in Table III.
June 19, 1916
Correlated Characters in Maize Breeding
441
s
T1
>.
a
a
fiija
1
•d
•t
S
■w
>.
a
X
s&
O <U
PhXI
2 *^
1 t -^ ^ "* o
-H-H-H-H-H-H-H4^4^-H4^-H-H-H-H-H-H-H-H-H-H4^-H^^
O 00 00 a>0 »0 O ^M w wtnO t^vitOt^W O Ov t^'O 00 f^
^o
io(*)*^oo fO'^f^M O\00 M c^w ^O\a0 «o Tfr-o m t-i o
1 00 p^ O* --I- <
4^-H-H-H-H-H-H-H4^-H-H-+^-H-H-H-H-H-W-H-H-H4^-H-H
O »OMt>-vO «o 0>" fO'^'!t« »nc^O'*<^*J^M MOO 0»>'-)
-H-H4^-H-H-H-H-H-H-H-H-H-H-H4^-H-H-+^-H4^-H-H-+^-H
o 00
-H-H41-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-H-HHH-H
0000 O fOt^-O O 000 "100 '•tOOvOOO M O OvOOO O t^O
^ TO ^00 fO <
-H-+^-H-H-t^-H-H-H-H-H : -H -H -H -H -H -H -H -H -H -H -H -H -H
» -(t t'. >-» ONOO * 00
) C>00 •- M
i-2
r* (^ M o 0> '
^ O 00 ^ PO O ** ►
-H-H-H-H-H-H-H-H-H-H-H-H-H4^-H41-H-W-H-H-H-H-H-H
O O\00 iH 0\ i^ O O\c<t-o0 'to OMO ^O MiocO w cor*
TO r-O -^ 0\ ^ OvX 'tM
0» « O -^00 O >
O t^ ^ 't >-• 00 -^O O
-H-H-ti-H-H-H-H-H-H-W-H-H-H-tl-H-H-W-H-H-H-H-H-H-H
O O »^ !-• f*500 O r»00 M loOiO^O tor*m"*cO« O^t**^
^00 ^0\N TfrtOiW t \ri<Xi
§§:
Eg
=« — -S „ « a
cd at
•Co
js.i:
■M O
SiSi
•a
■ a
m2-
"3 11
UoJItli'-JSOMtlMCirtoOI
o o
■yJ3j3
5t:-j3j2
rt hii *-• y r! r3
— ««-a """m^
u I- o ^ O t.
o o
335!
a ^^
u I- « ^ " t.
XI J2 :3 .ti v XI
■5SS3:i3a-^S3o3333S3
««4 It u u
OTJ 1* 4*
1,--X2X3
"tj ii B B
a a 3 3
0 o ~
u Lm O
01 11 ^
X2^t;
BBS!
3 3 X
442
Journal of Agricultural Research
Vol. VI, No. la
Table III . — Coefficien t of variation of characters in first- and second-generatio n maize p lants
Character.
Height cm..
Length of branching
space cm. .
Length of central spike,
cm
Nmnber of branches
Number of secondaries . .
Number of nodes above
ear
Length of longest leaf,
cm
Width of longest leaf.
Ratio of length to width.
Number of nodes above
longest leaf
Total number of nodes . .
Number of sheaths with
hairs
Number of sheaths encir-
cled by hairs
Length of hairs. . . .mm. .
Density of spikelets
Length of gliunes. .mm. .
Number of erect blades . .
Angleof tassel axis. ..( °) . .
One-sidedness
Number of rows, upper
ear
Number of rows, lower
ear
Ntunber of husks, upper
ear
Number of husks, lower
ear
First
genera-
tion.
. o± o.
•3i !■
5± I-
3± 2-
4± 4.
7 14- 7± 1-5
8 20. 4± 1.9
19. 7± 2.0
27. o± 2.7
40. 4± 4. 2
4± I.
4± •
.o± 2.
•3± •
l± I.
2± .1
I± I- 2
4± 2.
3± 2.
7± I-
A± 6.
2±
6± 6
8.9
7± I.-
6± 2.(
i± I.,
3± 2-<
Second generation.
Ear I.
Plants
from
waxy
seed.
17. 4± 1-7
8.i± .7
10. 7± i.o
ii.o± 1.0
18. 4± 1.7
10. 9± I.I
26. 2± 2.5
[70. o±42.o
19. 7± 1.9
20. 2± 1.9
9-I± .9
68. 7± 9.8
98. 2 ±16. s
48. 8± 5.7
20. 7± 2. s
IS- S± 2.S
24. 6± 2.6
23. i± 3.6
Plants
from
horny
seed.
i3-o± I.
19. 6± I.
lS-2± I-
25. 2± 2.
42. 4± 4.
i7-3± I-
8.2± .
iS'li I-
10. 7± I.
I7-S± 1-6
19- 8± I.
83.4±ii-
17. 2± I.
36- 2± 3.
6.8± .
66. 9± 9.
66. 7 ± 7.
56. S± 7-
17. 2± I.
19- 9± 3.
20. 6± 2.
14. 4± 2.
Ear 2.
Plants
from
waxy
seed.
i4-5± 1-4
15- 8± 1.5
13- 8± 1.4
17- 5± 1-7
43- 6± 4-7
9. 6± 1.0
6.7± -8
10. 4± 1.0
10. 2± 1.0
14.2+ 1.4
6.8± .7
29. 4± 3.2
169. 0±42.O
l8.s± 1-9
29- 7± 3- I
10. 2± 1.0
60. 6±io. I
74-4±io.4
49. 6± 6.6
I3.8i 1.8
12. 4i 1.6
18. 8± 2.0
II. 6± 1.8
Plants
from
homy
seed.
19. o± I.
18. 3± 1.7
I7.5± 1.6
22-0± 2.1
41. 9± 4.6
l4-3± 1-4
9. 7± 1.0
8.2± .7
10. s±
2i.8± 1.9
9-i± .9
26. 4± 2-5
I32.0±25.O
10. 4± 1.0
19. 8± 1.9
lo.5± 1.0
75. o± 9.8
89. 5 ±13. 1
52. 6± 6.4
13- 7± 1-4
15. o± 2.0
20. 2 ± 2.1
i9-5± 2-4
Ear 3.
Plants
from
waxy
seed.
14. 9± !•'
22. o± 2.
II. 6± 1.,
22. o± 2.
62. o± 9.
8.3± .'
8.7± ..
12. S± I.,
12. 9± I..
18. 8 ± 2.(
8.o± i.(
23-3± 2.5
15-8
o±78.
6± 2.
6± 2.
i± I.
2± 8.
9±I2.
4± S-
7± I-
8± I.
3± I-
± 2.
Plants
from
horny
seed.
.o± 1.8
.6± 2.2
• 7± I-O
■i± 3-3
. 2±II.6
. 4± I.I
.4± 1.0
.4± 1-9
•7± 1-3
• i± 1.4
•7± -7
19. i± 2.1
o± so. o
3± 2.3
8± 2.8
S± 1-7
3± 7-5
o±i8.6
8± 6.3
3± 1-9
o± 1.9
7± 2.6
o± 2.5
A comparison of these tables shows that the first-generation plants
exceeded the second-generation plants in height, length of branching
space, length of central spike, length and width of longest leaf, number
of nodes above the longest leaf, number of leaf sheaths with hairs,
and number of single-ranked blades. The second-generation plants
exceeded the first-generation plants in the number of tassel branches.
In the other characters there was no significant difference between the
means of the first and second generation plants.
The first-generation plants were distinctly less variable than the
second-generation plants in height, length, and width of longest leaf,
number of nodes above the longest leaf, total number of nodes, and
number of leaf sheaths with hairs. The first-generation plants were more
variable with respect to the length of the tuberculate hairs and density
of spikelets.
The least variable character measured was the length of the longest
leaf. Thctotal number of nodes was also comparatively, uniform. The
very high coefficient of variation for the number of sheaths encircled by
hairs results in part from the alternative nature of this character.
June 19, 1916 Correlated Characters in Maize Breeding 443
In the progeny of the reciprocal ears i and 2 there are no really signifi-
cant differences. The progeny of ear 3, however, which descended from
entirely different first-generation plants, shows a number of differences
from the remainder of the second-generation plants.
Although the average height of the plants from all these ears is prac-
tically the same, the progeny of ear 3 shows smaller values for a number
of other dimensional characters. The number of branches, primary and
secondary, length of leaf, total nodes, length of glumes, and number of
rows of grains are all slightly lower. With the exception of length of
leaf and length of glumes, these differences might be interpreted as
indicating a more pronounced development of the Esperanza characters.
The same may be said of the exsert, which is higher in ear 3. In the
development of tuberculate hairs, on the other hand, the progeny of ear
3 was decidedly more like the Chinese variety.
In addition to the measurements given in Tables II and III, there are a
number of differences that deserv^e to be more fully discussed.
HAIRS ON THE LEAF SHEATH
Perhaps the most striking difference between the varieties is the cov-
ering of the leaf sheaths. In the Chinese variety the leaf sheaths are
similar to those of the ordinary types of maize. The surface is smooth,
except for fine spicules, which occur especially over the fibrovascular
bundles. The spaces between the fibrovascular bundles are crossed by
numerous diagonal ridges or cross veins irregularly arranged and usually
discontinuous at the fibrovascular bundles. These cross veins with the
fibrovascular bundles cover the surface of the sheath with a coarse
reticulum.
In the Esperanza variety the cross veins of the sheaths are absent or
confined to the seedling leaves, and the spaces between the bundles are
occupied by tubercles, each bearing a long hair (PI. LVIII). These
tuberculate hairs are absent from the sheath of the first six to eight
leaves of the seedling. They appear abruptly and may cover the entire
surface of the first sheath on which they appear. The hairs are from
3 to 5 mm. long, and the tubercle is approximately }4 mm. wide and of
the same height.
In the Waxy Chinese variety tuberculate hairs are completely absent
(PI. IvVI, fig. 2). As in all varieties, there is a small area closely confined
to the throat of the sheath that is clothed with lopg hairs. It is not
clear whether these hairs are homologous to the tuberculate hairs of the
Esperanza variety or not. Even considering these hairs at the throat
of the leaf sheath in the Waxy Chinese variety as of the same type, the
two varieties are completely separated, \^dth not even an approach to
overlapping forms. In the hybrid and its progeny three methods of
measuring the degree of hairiness were employed:
(i) By recording the total number of nodes with hairy sheaths.
444 Journal of Agricultural Research voi. vi, no. 12
(2) By recording the number of nodes with hairs completely encircling
the sheath. In the pure Esperanza maize this usually occurred at the
lowest node on which hairs were borne ; or at most there was a difference
of only one of two nodes. In the hybrid plants there were usually a
number of sheaths with tuberculate hairs at the side, but with a narrow
smooth strip at the back over the midrib.
(3) By recording the length of the longest tuberculate hairs. In all
hybrid plants of both the first and second generation tuberculate hairs
were present, there being no plant that resembled pure Waxy Chinese
plants in this particular. The length of the hairs varied, however, in
different plants, thus affording another measure of the extent to which
hairs were developed.
TASSEL CHARACTERS
In the nature of the tassel the two varieties are hardly less distinct
than in the covering of the leaf sheath. The Waxy Chinese variety
has many branches, 15 to 30 primary branches in normally developed
plants, with numerous secondaries. The Esperanza (PI. LVII) seldom
has more than 5 branches and in many plants the tassel is simple, con-
sisting only of a large central spike. Associated with the difference
in the number of branches is a corresponding difference in length of the
axis or "branching space," the distance from the lowest to the upper-
most branch.
In the Esperanza variety the glumes vary from 10 to 16 mm. in length
with a mean of 1 1.7 ±0.14. In the Waxy Chinese variety the range is
from 7 to 12 mm., with a mean of 9.2 ±0.09. All of the above characters
were directly measured or counted.
The typical arrangement of the spikelets is also different in the two
varieties. In the Waxy Chinese the arrangement on the branches is
similar to that in most of the better known varieties of maize. The
spikelets are paired, one pediceled and one sessile, the pairs alternating
on the sides of the branch. In the Esperanza maize when branches
occur the spikelets are nearly all sessile and are borne in clusters of
from 2 to 5. They are also disposed on all faces of the branch instead
of being confined to the sides. The arrangement of spikelets and gen-
eral appearance of the branches in the Esperanza is similar to the arrange-
ment on the central spike. One result of these differences in arrange-
ment of spikelets is a greater crowding of spikelets in the Esperanza.
As a measure of this difference the number of spikelets in the last 10
cm. of the lowest tassel branch were counted. This number is referred
to as the "density of the spikelets."
TASSEL EXSERT
In the Waxy Chinese variety the base of the tassel is frequently
inclosed in the uppermost leaf sheath. In the Esperanza variety the
lowest branch of the tassel is usually well above the uppermost leaf
June 19, 1916 Correlated Characters in Maize Breeding 445
sheath in the mature plant. Differences in this particular were recorded
by measuring the distance from the top of the uppermost sheath to the
origin of the lowest tassel branch, the measurement being expressed as
a minus quantity when the base of the branch was included in the sheath.
This character is especially subject to environmental changes. Un-
favorable conditions, such as drought occurring late in the season, will
prevent the elongation of the upper intemodes to such an extent that
all varieties may show a minus exsert. Comparisons must therefore
be confined to plants grown in a single season in the same locality.
The range as recorded for Waxy Chinese grown at different times is
from — 14 cm. to 7 cm., with the mean at — 1.31 ±0.3. In Esperanza the
range is from —3 cm. to 18 cm., with the mean at 6.o7±o.5.
NUMBER OF ERECT LEAF BLADES
In the Waxy Chinese variety the upper leaf blades are held erect
instead of diverging. In ordinary varieties which the Esperanza resem-
bles with respect to this character the upper leaf blades make approxi-
mately a right angle with the axis (PI. LV, LVII). As a measure of this
character the number of erect leaf blades was recorded. For example,
if the two uppermost leaves were erect and the third leaf was the
first to exhibit an angle, the plant was classed as 2, with respect to this
character.
Recorded in this way there would be some overlapping in the parent
varieties, since in some Waxy Chinese plants even the uppermost leaf
shows an appreciable angle. In reality, however, the two types are dis-
tinct, for in the Esperanza not only is the uppermost leaf never erect, but
it is seldom borne at less than a right angle with the stalk.
ANGLE OF TASSEL AXIS
In the Esperanza variety the tassel is always erect. In the Waxy Chi-
nese plant the tassel is usually curved or declined (PI. LV, LVII). This
character is variable in the Chinese, some plants having the tassel per-
fectly erect. The tendency, however, to an inclined tassel, as it appears
in the hybrid, may properly be ascribed entirely to the Chinese variety,
no similar tendency ever having been observed in any Esperanza plant.
The character was measured by estimating the angle which the branch-
ing space, or that portion of the axis of the tassel between the lowest and
highest branch, made with the main stalk. In the pure Waxy Chinese
variety this character appears definitely associated or physiologically
correlated with the following character of *'one-sidedness."
ONE-SIDEDNESS
One of the most striking peculiarities of the Waxy Chinese variety of
maize is the displacement of the leaf blades from the usual distichous
arrangement, with the result that a number of the upper leaf blades are
446 Journal of Agricultural Research voi. \i. no. 12
borne on one side of the plant instead of alternately on opposite sides of
the culm (PI. LVI, fig. i). Like the angle of the tassel, this character
is not universally present in the Waxy Chinese plant, but, on the other
hand, no tendency of this kind has ever been observed in the Esperanza
variety.
When one-sided plants occur in the hybrid generations, it is therefore
reasonable to assume that the character was derived from the Chinese
parent. Measurements of these characters in the hybrid plants were
made by recording the number of monostichous or single-ranked leaves,
A recapitulation of the more definitely contrasting characters of the
two parent varieties is here presented in parallel columns :
Esperanza variety
Homy endosperm.
Branching space short.
Tassel erect.
Spikelets in clusters.
Glumes long.
Leaf sheaths with tuberculate hairs.
Upper leaf blades horizontal.
Upper leaf blades distichous.
Waxy Chinese variety
Waxy endosperm.
Branching space long.
Tassel curved.
Spikelets in pairs.
Glumes short.
Leaf sheaths without tuberculate hairs.
Upper leaf blades erect.
Upper leaf blades monostichous.
If the characters of maize were subject to coherence, the second genera-
tion of a cross between two such diverse and long-established types as
Esperanza and Waxy Chinese would seem a most favorable opportunity
for its manifestation.
In the whole series of second-generation plants there were none that
even approximately represented either parent variety ; nor did the plants
fall into recognizable groups. With respect to the individual characters,
the parental forms reappeared or were even intensified in some instances,
but an almost complete and chance reassortment of the characters seems
the rule. If the characters were completely independent, a reappearance
of the parental types could not, of course, be e;s:pected, for, treating the
characters as alternative and allowing for only 10 characters, a plant
possessing all the characters of either parent could not be expected oftener
than once in 10 billion plants. Although the characters themselves,
with few exceptions, were non-Mendelian in the sense that they were not
alternative, the results conformed to the Mendelian law of recombination.
Examples of the combination of characters from the two parent varieties
are shown in Plates LIX to LXIII.
Endosperm texture was the only strictly alternative character noted.
The number of erect leaves and angle of tassel, while not alternative in
the sense of falling into definite groups without intermediates, do, how-
ever, approach a Mendelian form of inheritance. The distribution,
instead of approximating a normal frequency curve, was distinctly
bimodal with respect to these characters. A similar tendency is ap-
parent in the first-generation plants. In connection with this evidence
June 19. 1916 Correlated Characters in Maize Breeding 447
of segregation in the first generation, it should be recalled that neither
of these characters, which belong to the Waxy Chinese variety, is univer-
sally present in the plants of that variety, and the parent plant may
have been heterozygous. There is also a less pronounced indication in the
second-generation plants that one-sideness is Mendelian in its inheritance.
CORRELATIONS
Eleven of the characters most definitely contrasted in the parents were
selected and the correlation coefficients between all possible combina-
tions were calculated for both the first and second-generation plants.
The results are shown in Table IV. The correlations are so stated that a
positive, or plus, correlation indicates a correlation between the char-
acters derived from the same parent; in other words, a coherence. For
example, the Waxy Chinese variety has a large number of tassel branches
and no tuberculate hairs, while the Esperanza variety has a small number
of tassel branches and well-developed tuberculate hairs. In expressing
the relation between these two characters, when a large number of tassel
branches is found associated with short tuberculate hairs, the correlation
is recorded as positive.
Since ears i and 2 were reciprocals and no significant differences were
found between their progenies, the observed values were used directly
in calculating the coefficients of correlation. Where the mean progeny
of ear 3 differed from the mean of the combined progenies of ears i and 2
with respect to any character, all measurements in the progeny of ear i
were multiplied by the percentage difference between the means before
combining the populations in a correlation table.
The combined progenies of the three first-generation ears numbered
183 individuals. Complete notes could not be taken on all the plants,
so that the number of individuals entering into the different correlation
tables was reduced to from 125 to 150. Assuming all correlations that
are more than 3.5 times the probable error to be worthy of consideration,
an examination of Table III shows that 20 of the 55 character pairs
fall into this class. ^ With three exceptions the coefficient for the charac-
ter pairs of this group is 0.2 or larger. Of these 20 character pairs that
may be held to show definite correlations in the second generation, 17
are positive — that is, in the nature of coherences — and 3 are negative.
All but 5 of the 20 are, however, open to the suspicion of being physio-
logical correlations, since they do not differ materially from the corre-
lations shown for the same characters in the first generation.
The 5 character pairs that show most evidence of genetic correlation
are given in Table V, Even here there are no very striking differences
between the coefficients of the first and second generations, and it is by
no means impossible that even here the differences may be due to chance.
' These coe£5cients are printed in bold-face type in Table IV.
448
Journal of Agricultural Research
Vol. VI, No. 13
Table IV. — Correlation coefficients
Characters.
First
generation.
Second generation.
Small exsert of tassel and
Long branching space and
Large number of tassel branches and
Large number of erect leaf blades and
High degree of one-sidedness and
long branching space
large number of tassel branches
large number of erect blades
high degree of one-sidedness
large angle of tassel axis
small number of sheaths with hairs. . . .
short hairs
low density of spikelets
short glumes
waxy endosperm
large number of tassel branches
large number of erect leaf blades
high degree of one-sidedness
large angle of tassel axis
■ small number of sheaths with hairs. . .
short hairs
low density ol spikelets
short glumes
waxy endosperm
large number of erect blades
high degree of one-sidedness
large angle of tassel axis
small number of sheaths with hairs
short hairs
low density of spikelets
short glumes
waxy endosperm
high degree of one-sidedness
large angle tassel axis
small number of sheaths with hairs
short hairs
low density of spikelets
short glumes
waxy endosperm
large angle of tassel axis
small number of sheaths with hairs. . . .
short hairs
low density of spikelets
short glumes
waxy endosperm
Ismail number of sheaths with hairs
short hairs
low density of spikelets
short glumes
waxy endosperm
ilong hairs
low density of spikelets
short glumes
waxy endosperm
(low density of spikelets
Short hairs andsshort glumes
(waxy endosperm
Low density of spikelets and{^\-;,|^Xfp
Short glumes and waxy endosperm .
endosperm .
Coef. P.E.^
o. 27 0
'3
•30
13
. 22
14
. 00
IS
.18
14
— . 21
13
—•39
12
— . 10
14
— . 24
13
•50
09
— . 22
14
— . 14
13
. 00
13
. 00
12
— . 12
12
•32
II
— .01
12
— . II
14
.27
12
•38
II
-•30
II
•25
12
•45
10
•07
12
•17
14
.29
13
— . 21
14
•17
14
-.18
14
•31
13
.67
07
—•15
13
•IS
13
•36
12
-.08
13
—•23
II
. 01
12
. 02
13
•47 •
10
—•OS
12
— . 01
12
. 21
II
. 12
12
-.05 .
12
—.16 .
12
Coef.
0.345
.287
.346
.353
.411
— . 109
— . ISO
.oiS
—.036
— . 069
.442
. 202
.234
. 170
. 091
• 056
.198
-.188
.242
.222
.221
.243
— . 091
— . 021
.214
—.165
. 104
.487
.513
—.118
—.081
Coef.
P.E.^ P. E.
0.053 6.5
.055
.059
.056
.050
•059
. 060
•059
.059
. 070
.045
.061
.057
•05s
.054
. 056
.054
.053
.066
.063
.058
.054
.056
•057 -A
.054 4.0
.054 3.0
.067 1.6
.050 9.7
. 047 10. 9
. 064 I. 8
.043
.'S82
. 069
.068
.030
.017
.118
.203
. 007
•123
• 043
• 037
.193
•134
.009
. 192
.027
5.2
5.9
6.3
8.2
2. o
2-5
•3
.6
3.7
3.6
3.7
3.5
3.8
4.5
1.6
.063
.063
.065
1-3
1-3
040 14. 5
I. 2
I. I
•5
.059
. 060
.074
.053
.056
•055
. 056
.079
.053
.058
•05s
.067
•055
•055
. o6g
.052
.o58
.059
1.6
3.8
3.6
2-3
4.5
1 P. E.= probable error.
Table V. — Character pairs exhibiting genetic correlations
Character pair.
Coefficient of correlation.
Difference
between first and
second genera-
tions.
Difference-T-
First generation.
Second generation.
error.
Small exsert of tassel and
one-sidedness
0. oo±o. 15
-. 22± . 14
-. I4± . 13
— . ii± . 14
. i7± . 14
0^353 ±0^056
. 202 ± . 061
•234± -057
. 222 ± . 063
. 4S7± . 050
0. 353±o. 160
.422± . 153
.374± • 142
.332± . 143
.3i7± . 149
2. 2
Branching space and num-
ber of erect blades
Branching space and one-
sidedness
2.8
2.6
Number of tassel branches
and number erect
2-3
2. I
Number of erect blades and
one-sidedness
June 19, 1916 Correlated Characters in Maize Breeding 449
Owing to the small number of first-generation individuals and the
consequent uncertainty that attaches to correlation coefficients in that
generation, it is, on the other hand, possible that other correlations
shown in the second-generation plants are really genetic. From this
point of view, it should be noted, however, that 18 of the second-genera-
tion correlations are negative.
The possibility of a reduction of physiological correlations must also
be considered. The existence of a significant positive correlation in the
first generation is taken to indicate a physiological correlation between
the characters. With such characters as branching space and the number
of branches, the relation is obvious; indeed this relation might almost
be classed as physical, since as the branching space approaches zero the
number of branches must necessarily become less. There would also
appear to be a necessary relation between one-sidedness and angle of the
tassel axis, for a perfectly erect tassel could scarcely occur with a high
degree of one-sidedness. Where correlations of this nature are lowered
in the second generation, it would seem necessary to assume that this
reduction is brought about by a tendency for the characters from different
parents to reappear in the same individual, thus reducing the normal
physiological correlation that exists between the characters.
The following are two such character pairs :
First Second Differ- D.
generation generation ence P.E.
One-sidedness and low density
of spikelets o. 36±o. 12 c. c30zto. 059 o. 330±o. 13 2. 5
Large angle of tassel axis and
short glumes 47± .10 •C43± .056 •42'j± .114 3.7
It has been mentioned that with respect to both the number of erect
leaf blades and the angle of tassel axis there was a tendency for the plants
to fall into two groups. This raised a doubt as to the applicability of the
customary "product-moment" method of calculating the correlation
coefficient where these characters were involved. This group of correla-
tions was therefore recalculated, using Pearson's biserial correlation
coefficient (Pearson, 1909). Slightly different values were obtained, but
no additional significant correlations were brought to light.
In the second generation the waxy and homy seed were planted
separately, thus affording an opportunity for observing whether the
plants from seeds having the waxy endosperm characteristic of the Waxy
Chinese variety showed any preponderance of other Chinese characters.
No consistent differences were apparent in the general appearance of the
rows from the waxy and homy seeds. There was such great individual
diversity, however, that comparison was difficult. Analysis of the
measurements showed little more. The only character that showed a
measurable correlation with endosperm texture was the degree to which
tuberculate hairs were developed on the leaf sheaths.
450 Journal of Agricultural Research voi. vr, ko. «
Since endosperm texture is strictly alternative, while all other charac-
ters were expressed in varying degrees, the method for calculating the
correlation coefficients was necessarily different for this group of character
pairs. In calculating the correlations with endosperm texture Pear-
son's (1909) method for calculating a biserial correlation, together with
Soper's (19 14) formula for the probable error, were used. With a strictly
alternative character such as endosperm texture, it would seem impos-
sible to distinguish physiological from genetic correlations. Since one
variety always has waxy and the other always has homy endosperm, to
detect correlations with this character in the parent varieties seems out
of the question. Likewise, as a result of the dominance of the homy
endosperm, the seeds from which the first-generation plants were grown
were all homy, and there was no opportunity to determine correlations
with endosperm texture among first-generation plants.
At the time of planting it was, of course, impossible to distinguish
between the seed that were pure for the homy character and those that
were heterozygous. An examination of the open-pollinated ears produced
by the second-generation plants grown from homy seeds made such a
separation possible. All ears that produced any waxy seeds must have
grown from heterozygous seeds. No correlations sufficiently large to
be detected in the small number of individuals available were found
between these two classes and other contrasting characters.
It may be urged that the absence of coherence in the progeny of such
a diverse hybrid as the one here discussed may not prove that there is a
similar lack of coherence among crossbred individuals within the
variety. All maize varieties are, however, of such mixed ancestry that
they are in effect hybrid progenies, and even if an exhaustive study of
the inheritance of the characters of a narrow-bred variety should show
the existence of coherence the results would be beside the point from a
practical standpoint, for to maintain a satisfactory degree of vigor in
maize a condition of mixed ancestry must be retained.
INTENSIFICATION OF CHARACTERS
The present hybrid affords an interesting sample of an intensified
character. One of the peculiarities of the Waxy Chinese variety is the
scorpioid top. In plants which exhibit this character the leaf blades of
the upper nodes are monostichous and erect, and the tassel is curved to
one side. The curvdng of the tassel w^as originally interpreted as a direct
result of the monostichous arrangement and erect blades. The manner
in which this complex of characters reappears in the hybrid with the
Esperanza variety shows that, although always associated in pure
Chinese maize, they are separable and each may be inherited independ-
ently of the others. The curved tassel supposed to be merely the result
of the other characters may not only occur alone — that is, in plants with
June iQ, 1916 Correlated Characters in Maize Breeding 45 1
distichous leaf blades all of which make an angle with the main axis —
but the extent of the cunning is much greater in some of the hybrid plants
than has ever been observed in pure Waxy Chinese plants. The angle
of the tassel axis had not been recorded for Waxy Chinese plants before
the season of 191 5, but thousands of individuals have been obser\^ed, and
it can be definitely stated that no plant showed a tassel inclined as much
as 90° from the perpendicular.
In 148 hybrid plants of the second generation of the hybrid there were
12 plants with the axis of the tassel inclined from the perpendicular by
more than 100° and 5 plants having the angle of the tassel axis recorded
as more than 145°. The phenomenon is not due to any weakness of the
culm, as examples of more than 180° show (PI. LXII) ; in fact, the upper
part of the culm is particularly thick and rigid, a characteristic of the
Chinese parent.
The positiveness of the character was well shown in some of the plants
where the curving of the culm caused it to break through the upper leaf
sheaths. In such plants the pendent tassels very strongly suggested the
"goose neck" of certain sorghum varieties. A plant of this type is
shown in Plate LXIII.
CONCLUSIONS
Two principal methods of breeding may be distinguished, depending
on the manner in which selection is applied :
(i) Selection may be directed toward the isolation and propagation
of desirable types of individuals. The new type may occur as an aber-
rant individual or as a recognizably distinct strain within the variety, but
in either case it is differentiated from the stock by many characters.
(2) Selection is directed to variations of the individual characters
regarding which improvement is desired.
With most crop plants the method of selecting types has been by far
the most productive, but in the improvement of maize, this method has
figured very little. Selection has been by characters instead of by types.
Why the isolation of types of plants has not been a factor in the im-
provement of maize has not been clear. Though diversities in plant
characters are obvious and striking, few breeders have been able to dis-
tinguish well-defined types of plants within commercial varieties.
If recognizable types exist it must mean that groups of characters tend
to appear together; in other words, the characters are correlated. The
extent to which obvious characters are correlated is therefore proposed
as a measure of this tendency toward the persistence of types. In the
progeny of a hybrid between two very different maize varieties the results
here reported show that the characters studied, instead of forming cohe-
rent groups, are almost completely independent in inheritance.
37769°— 16 2
452 Journal of Agricultural Research Voi. vi, No. la
By attempting to measure the extent to which types persist by means
of correlation coefficients, it is necessary to distinguish different kinds of
correlations. For this purpose correlations are here classified as physical,
physiological, and genetic. A method is also proposed by which physio-
logical and genetic correlations may be distinguished.
The case studied was a hybrid between two extreme types that must
have been completely isolated from very remote times. The large num-
ber of well defined characters which differentiate the varieties rendered
this material exceptionally favorable for the study of coherence, by
which is meant the tendency for characters associated in one of the
parents of a hybrid to remain together in the later generation of the
hybrid.
For the study of correlations 1 1 characters were selected in which the
parent varieties showed little or no overlapping. The correlation coeffi-
cients of all the combinations were calculated, and of the 55 possible
combinations 20 were found to exhibit significant correlations. In all
but 5 of these, however, the correlations are believed to be physiological
rather than genetic. In no instance was there a correlation between two
characters closer than 0.5, a fact which in itself offers an explanation of
the difficulty of recognizing types in maize.
This lack of coherence of characters in maize, taken with the fact that
to maintain a satisfactory degree of vigor a diversified ancestry must be
maintained, would appear to make the method of isolating types inap-
plicable to this plant. As an offset to the limitation thus imposed, advan-
tage may be taken of the facility with which desirable characters derived
from different parents can be combined.
LITERATURE CITED
Bateson, William, Saunders, Edith R., and Punnett, R. C.
1906. Experimental studies in the physiology of heredity. In Rpts. to Evolution
Com. Roy. See. [London], no. 3, 53 p., 4 fig.
BONAFOUS, Mathieu.
1829. Note siu: une nouvelle esp^ce de mals. In Ann. Sci. Nat., t. 17, p. 156-158,
pi. 8.
CoixiNS, G. N.
1909. A new type of Indian com from China. U. S. Dept. Agr. Btir. Plant Indus.
Bui. 161, 30 p., 2 pi.
and Kempton, J. H.
1913. Inheritance of waxy endosperm in hybrids of Chinese maize. In IV Conf.
Intemat. G6netique Paris, Compt. Rend, et Raps., 1911, p. 347-357.
East, E. M.
1908. Organic correlations. In Amer. Breeders' Assoc. Rpt., v. 4, p. 332-343.
Emerson, R. A.
191 1. Genetic correlation and spurious allelomorphism in maize. In Nebr. Agr.
Exp. Sta. 24th Ann Rpt. 1910, p. 59-90, 9 fig.
Morgan, T. H., Sturtevant, A. H., Muller, H. J., and Bridges, C. B.
191 5. The Mechanism of Mendelian Heredity. 262 p., illus. New York. Bibli-
ography, p. 237-256.
June 19, 1916 Correlated Characters in Maize Breeding 453
Pearson, Karl,
1909. On a new method of determinmg correlation between a measured character A ,
and a character B, of which only the percentage of cases wherein B exceeds
(or falls short of) a given intensity is recorded for each grade of A. In
Biometrika, v. 7, pt. 1/2, p. 96-105.
SoPER, H. E.
1914. On the probable error of the bi-serial expression for the correlation coefficient.
In Biometrika, v, 10, pt. 2/3, p. 384-390.
Webber, H. J.
1906. Correlation of characters in plant breeding. In Amer. Breeders' Assoc. Rpt.,
V. 2, p. 73-83, pi. I.
PIATE LV
Typical plant of the Waxy Chinese variety o£ maize, showing numerous tassel
branches, erect leaf blades, one-sidedness, and curved tassel.
(454)
Correlated Characters in Maize Breeding
Plate LV
Journal of Agricultural Research
Vol. VI, No. 12
Correlated Characters in Maize Breeding
Plate LVI
I
Journal oV Agricultural Research
Vol. VI, No. 12
PLATE LVI
Fig. I. — Uppermost kaf sheaths of Chinese maize plant, showing the one-sided
arrangement of leaf blades and absence of hairs. Natural size.
Fig. 2. — Leaf sheath, of the Waxy Chinese variety of maize, showing the transverse
lines and absence of hairs. Compare with Plate LX. Natural size.
PLATE LVII
A plant of the Esperanza variety of maize, showing the drooping leaves, few tassel
branches, and elongated intemodes characteristic of the variety.
Correlated Characters in Maize Breeding
Plate LVIi
Journal of Agricultural Rt
v\..; VI, r-i-. i::
Correlated Characters in Maize Breeding
Plate LVIII
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LVIII
Leaf sheaths of the Esperanza variety of maize, showing the maximum develop-
ment of tuberculate hairs. Compare with Plate LVI. Natural size.
PLATE LIX
A leaf sheath of a second-generation hybrid maize plant. This plant represents the
maximum length of hairs. They are even longer than any thus far observed in the
Esperanza variety. Compare with Plate LX. Natural size.
Correlated Characters in Maize Breeding
Plate LIX
Journal of Agricultural Research
Vol. VI, No. 12
Correlated Characters in Maize Breeding
Plate LX
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LX
A first-generation plant of Chinese XEsperanza maize hybrid. Measured by the
jiumber of sheaths with hairs, this was the most hair>' plant in the first generation.
Combined with this Esperanza character is an accentuation of the Chinese character
of a scorpioid top.
PLATE LXI
A second-generation plant of a Chinese XEsperanza maize hybrid. This plant
showed a maximum development of the Esperanza character of hairiness combined
with the erect crowded leaf blades and deflexed tassel of the Chinese variety.
Correlated Characters in Maize Breeding
Plate LXi
Journal of Agricultural Researcti
Vol. VI, No. 12
Correlated Characters in Maize Breeding
Plate LXII
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LXII ,
A second-generation plant of a Chinese XEsperanza maize hybrid. An extreme
example of the scorpioid top ; the angle was recorded as 190°.
PLATE LXIII
A second-generation plant of a maize hybrid, showing the "goose-neck" character
tliat appeared for the first time in this hybrid. This plant showed few Esperanza
characters. Although the plant is one-sided, it shows that the displacement of the
tassel is not the result of crowding by the leaf blades.
Correlated Characters in Maiie Breeding
Plate LXIIl
Journai of Agricultural Research
Vol. VI, No. 12
COMPARATIVE STUDY OF THE AMOUNT OF FOOD EATEN
BY PARASITIZED AND NONPARASITIZED hARVJE OF
CIRPHIS UNIPUNCTA
By Daniel G. Tower,
Scientific Assistant, Cereal and Forage Insect Investigations, Bureau of Entomology
INTRODUCTION
The aim of an experiment which was conducted at the United States
Entomological Laboratory in West La Fayette, Ind., during the summer
of 1 91 5, vras to determine whether lar\'ae of the army worm (Heliophila)
Cirphis unipuncta Haworth, when attacked by an internal parasite,
Apanielcs militaris Say, ate less, as much as, or were stimulated to eat
more than when nonparasitized ; and as a sequence, to determine whether
this or a similar parasitism is directly beneficial in the generation para-
sitized or only indirectly, resulting in subsequent smaller generations.
Although only 9 of the 25 parasitized larvae with which the experiment
was started lived until the emergence of the parasites, the others dying
soon after oviposition took place, the records of these 9 lar\^ae are suf&-
ciently definite to satisfy the purpose of the experiment.
The excellent work of Mr. J. J. Davis and Mr. A. F. Satterthwait ^ in
determining the total amount of food eaten by healthy larvae of C. uni-
puncta under different feeding conditions has been used to compare with
the amount of food eaten by parasitized larvae.
The results of the experiments have been drawn up in tabular form to
show the life of the host larvae from the time they were oviposited in
until their death coincident with the emergence of the parasite and the
life history of the parasite in relation to its host (Table I).^
EXPERIMENTAL METHODS
The parasites were induced to oviposit in the host larva while confined
in test tubes into which a larv^a was introduced and left until recognized
as a host and parasitized. Often this occurred immediately, and three
or four ovipositions might take place before the larv^a could be removed.
In other cases it would be some minutes before the parasite could be
induced to oviposit.
These parasitized larvae were confined separately in large vials, placed
in the shade in a well-aired room, and fed pieces of mature corn leaves,
conveniently cut out so as to measure i square inch each.
In order to obtain unfertilized females, individual cocoons were placed
in gelatin medicine capsules previous to the emergence of the adults, the
sex being easily determined through the transparent gelatin, when the
adults emerged.
' Data as yet unpublished; may appear in a later issue of this Journal.
2 The author was ably assisted in the care and feeding of the larvae by Mr. H. J. Hart, who was tem-
porary assistant at the laboratory during the summer of 1915.
Journal of Agricultural Research, Vol. VI. No. 13
Department of Agriculture, Washington, D. C. June 19, 1916
eg (455) K-36
456
Journal of Agricultural Research
Vol. VI, No. 13
9
.-ft.
u
"7^
pq
<
00
-^
0 M
1^ IJ
1 •«•
00 o\ c
3 1
■0
<r!
t " ■» !->
J3
J3
J-°
p-«' •"
i
>0 >< H vO 0
,c
00
00
00 S
fi.^ j: «
w
%t
"
"
" 0
*^
"£ « " "
•^
vi
1b
•°a
■^i
i-oB'^
"a^'
B'^ai-^a-^B
■^B
^
"
"
"
" "
' "
a
0 «j-
.
.
E
CD i;
•d
H
s8S
N 0
Sn
J3 -
00
a s ^
•0
•0 Bj3
*i 0
•0 '^
i 1
•d
0
•^■" •?■
TJ
Oi
Oi
o.
M
00 «
Oi
_a
00
T) -^
0
t^ u
■) -^
00 0
•» -^
a
t/J l/l CD
J3
00
8S^
00
J •« i
0) o fe
.
,*J <^
.
.
.
.
"
•dg
-osi
-^a
^g-D
g-d
a-^a -aa
•6
p
M
"
►?*?►?
t
•d
ts
■^s
.-? a-?
a ':
i'H •s:a-?a
si
fc OS
at .
fc si
•^ cs
« :
"^ a ^ «■ S <-■
•0.2
2s
^00
00-0
.00
00 ■
0 :
■^ 0
«i
":-i
^i
41
^^i|
^.0
0 .
a
w
w
C/2
w w
w xn tn
w 8
0,0
•"si
•^a
•> a -^ g
• .
tJ »
Uoo
d j^ aoo
as
a a
ti.2
. M ^ . • bc 3
^
0
•0
0
•q
is ^3
u 3 ui 3
3 >
ea a*j a r
a 3 0
30 30
fi M
<
-«!
<
<J <
•isoq qoB3
Oi
c^
00
vo c
rO
't 00 00
ni sstusBjed JO jaqnmj^
^
>^
t
2
^ CO 0>
j3 n.ti D
_a
ge of
whe
paras
merg
5B
d
d
0 d
d
0 d d
>1 M
•q
T3
•q T)
•0
•q rq -q
< ''
S
•paSjania ajiscjEd naqM
s .*
« ^
•O ■» WJ
isoq JO pcaq jo ^^plJ^
"^ t^
->
^ ^H
to <^ to
•nisptsB
;j ^
r^
£, J?
0
M 3- 0>
-aed jaijE ua^Ba aSeqcj
naoa jo ^unoniB iVjox
Oi
4 6
•a- ■o ov
^ s
'
ja
ja
i ■. :
§§-s
s
i
* ; ;
■|2g
!8
N
' '■ J3
flj trt y
^ •§
1
.D
<
•q -q 2
: : +
o
*^
"
"
" ;
•suopisodiAO JO Jaqnm^j
+
t
-
+
>->
+
E
a
a
a a
a
a a a
^ s
d
d
d
d d
d
d d d
^.2
^S)
r^
0
0 ";>
°
2 00
n r^ ri
nl o
01 ^
"•5
H4
M
w
« «
«
W MM
g
U
M
bio
u bi
ti
u ei b«
3
3
3
3 3
3
3 3 3
<;
<
<
< <
<5
< < <
4-1 , •
CS
: • -.
SJ-- B
2 >•"
.a ot3
2
.a
.2
6
d
•q S
0
0 d 0
Kj3 g
^
•q
•0
T3
•q -q -q
<^a
£
s
?
1
•ni pa^isodiAO naqM
S 'f
„
„
« 00
r^
00 °0 *^ .rf
isoq JO pEaq jo q^pi^
_<Ji-
"
"
-5 M
"
^ MM
-
•OM ^naunoadxa
0
"
M
0
a
s
« «
S'
June 19, 1916 Feeding Parasitized LarvcB of Cirphis unipuncta 457
^ . — — t _^
LIFE CYCLE OF THE PARASITE
The biology of A. militaris has already been studied and the results
published.^
Oviposition took place with great rapidity and apparently anywhere in
the host, attempts even being made by the parasite to oviposit in the
head. The largest number of eggs inserted at one time, according to
the observations herein recorded, was 154 for 3 ovipositions, averaging
51 + each (Table I, Experiment 23). The two endoparasitic stages and
the egg stage required an average of 14 days, 1 1 }i minutes, while the time
spent by the third larval stage and the pupal stage in the cocoon averaged
9 days, 8 hours, and 45 minutes, and the average for the total life cycle
was 23 days, 12 hours, and 26 minutes.
The parasitic larva leaves its host by means of an individual exit
hole cut through the muscles and epidermis by its mandibles. As the
larvge squeeze through the holes they molt their second larval skins, and
when about two-thirds of their way out commence to spin their cocoons.
After the cocoon is spun and previous to pupation, the accumulated
wastes are passed, being deposited at one end of the cocoon. Shortly
following this the larva pupates and the last larval skin is pushed to the
same end of the cocoon.
The adult issues, after kicking off its pupal skin, by cutting off a cap-
like portion at one end of the cocoon, cleans itself, and at the same time
passes a quantity of waste. It is now ready for copulation, oviposition,
or feeding, as the case may be. In this respect it was found that females
were at once ready to oviposit following emergence and previous to feed-
ing or copulation, and that the progeny from such females were all males.
Hence it is seen that unfertilized females give rise parthenogenetically
to a generation of males.
CONCLUSIONS
In using the data compiled by Davis and Satterthwait on the amount
of food eaten by healthy larvae of C. unipuncta, for comparison with the
amount eaten by parasitized lar\^ae, it will only be necessary to use the
feeding records for the last three instars in one series of their experi-
ments, this being the one in which the larvae were confined in lantern-
globe cages. These records were selected in preference to those obtained
by keeping the larvae in large vials, because in the former case a larger
number of records were obtained, although in the latter case the averages
of the feeding records for the same periods run higher.
Larvae 10, 11, 15, and 16 were newly molted fifth-stage specimens
when oviposited in, and they ate 16.21, 12.16, 11.97, and 14.50 square
inches of com foliage, respectively, during their last two stages previous
to the emergence of the parasites, which is a much smaller amount than
' Tower, D. G. Biology of Apanteles militaris. In Jour. Agr. Research, v. s, no. la, p. 493-308, i fig.,
pi. 50. 191S.
458 Journal of Agricultural Research voi. vi, no. 12
I :
the average of 33.6 square inches eaten by 20 nonparasitized larvae dur-
ing the same stages. Larv^ae 19, 22, 23, 24, and 25 were partially de-
veloped fourth-stage specimens when oviposited in, and they ate, during
the remainder of their life, which lasted until the parasites emerged from
them some time during the last or sixth stage, 20.63, 17-36, 21.41, 17.64,
and 17.99 square inches of corn leaf, respectively, as compared with the
average of ;54-77 square inches eaten by 20 nonparasitized larvae during
the last three stages. (See Table I.)
From these results it will be seen that parasitized larv^ae ate approxi-
mately half as much as unparasitized larvae during the same periods,
and it seems conclusive, even from these few records, that parasitism by
A. viilitaris is directly beneficial in the generation attacked. From the
results obtained it might seem as though larvae oviposited in at an earlier
date would eat more before being killed, but the time spent in the host
by the parasites seems to be fairly constant, and this was also noticed in
a larger number of cases in former experiments with A . militaris. Hence,
it is believed that in such cases the larvae would have only approxi-
mately the same amount of time for feeding, and a larger portion of this
period would occur during the earlier stages, when a much smaller
amount of food is eaten, so that the amount eaten would be less than
the normal for unparasitized lar\'-ae.
ALEYRODIDAE, OR WHITE FLIES ATTACKING THE
OIL\NGE, WITH DESCRIPTIONS OF THREE NEW
SPECIES OF ECONOMIC IMPORTANCE
By A. L. QuAiNTANCE, Entomologist in Charge of Deciduous Fruit Insect Investiga-
tiotis, and A. C. Baker, Entomological Assistant, Bureau of Entomology
Thirteen species of so-called white flies are recorded in literature as
infesting Citrus plants in different parts of the world. Eight of these
are present in Florida, four of them being native to the United States
and four having been introduced. The native forms have thus far been
of little economic importance, whereas two of the introduced species
are first-class Citrus pests. The remaining two introduced forms,
although recently established on the orange {Citrus aurantiaca), have
already attracted attention by reason of their injuries. Our knowledge
of the remaining five species of Citrus white flies, while meager, indicates
that these, in their range of distribution, are abundant and destruc-
tive and would in all probability prove to be very undesirable immi-
grants. The new forms treated herein must be classed in the same
category, especially Aleurocanthus woglumi, which, although previously
named, is here technically described for the first time. This last species,
of oriental origin, has already found its way to Jamaica and the Bahamas,
where it infests the orange to a serious extent.
The present paper brings together the essential information concern-
ing the distribution and food plants of the white flies which attack Citrus
plants and describes three new species of economic importance.
Aleurocanthus citricolus (Newstead)
Aleurodes citricola Newst., 1911, in Mitt. Zool. Mus. Berlin, Bd. 5, Heft 2, p. 173.'
This species is known only from the original description. It was
taken at Dar es Salaam, German East Africa, on Citrus sp. in 1902.
The immature stages occurred in large, overcrowded colonies, appearing
to the unaided eye as patches of a sootlike deposit on the lower surface
of the leaves. This is one of the spiny forms and bears a general resem-
blance to A. woglumi (fig. 2, A-J, Pi. LXIV, LXV).
Aleurocanthus citriperdus, n. sp.
This insect (fig. i) was taken by "Mr. R. S. Woglum, of the Bureau of
Entomology, in several localities in the Orient, as follows: Royal Bo-
tanic Gardens, Ceylon, on an unknown tree, October, 19 10; Lahore,
India, on Citrus sp., July, 191 1; Buitenzorg, Java, on orange, January,
191 1 ; Sandan Glaya, Java, on Citrus sp., January, 191 1. It is reported
' All bibliographic citations in synonymy are civcn in full in "Literature cited," pp. 471-472.
Journal of Agricultural Research, Vol. VI, No. 12
Dept. of Agriculture, Washington, D. C. June 19, 1916
el K-37
37769°— 16 3 (459)
460
Journal of Agricultural Research
Vol. VI, No. 12
Fig. z.~A.leurocanthus cilriberdus: A, Pupa case; B, egg; C, polygonal markings of egg; D, vasiform orifice
of pupa case; E, spine from dorsum of pupa case; F, margin of pupa case; G, genitalia of adult male; H,
forewing of male; /, antenna of pupa case; /, leg of pupa case; K, L, marginal teeth, much enlarged; M,
central swollen spine f i-om dorsal area.
June 19, 1 9 1 6 A leyrodidae 461
as occurring abundantly on species of Citrus and is regarded as of con-
siderable economic importance.
Pupa case (fig. i, A). — Length 1.36 mm.; width 0.96 mm.; shape elliptical to
oval, broadest across the third abdominal segment, narrowest cephalad. Dorsum
with a moderate central abdominal ridge on which the abdominal segments are not
distinctly marked off, though they may be distinguished. Submarginal area some-
what flat; suture separating the thorax and abdomen quite distinct; stuface appearing
somewhat granular or faintly corrugated , an appearance which may be due to difference
in pigmentation. Dorsum with numerous heavy spines (fig. i , E) which after clearing
remain black at the tips, but are otherwise a clear greenish yellow. These are arranged
as follows: On the submarginal area is a more or less even row of usually 32 spines.
This row is composed of two series alternating with one another. The one is made up
of spines averaging about 0.288 mm., and the other of spines averaging 0.192 mm. in
length. Near the medio-dorsal abdominal line there are three pairs of spines, one
pair situated about 0.19 mm. anterior to the vasiform orifice and the others on the
cephalic part of the abdominal region. The spines of the pair on the first abdominal
segment are somewhat more widely separated than those of the other two pairs. Six
other pairs of spines are present on the abdomen. Five of these pairs are short,
about o.oS mm. long, and form an even subdorsal row on each side, the rows thus
formed diverging on the cephalic part of the abdomen. The remaining pair is com-
posed of much longer spines, situated about 0.29 mm. from the thoracic suture and
about the same distance from the lateral margin of the case. On the thorax there is a
subdorsal row of four spines on each side (fig. i, M) and near the medio-dorsal line
another pair of spines is present. Just anterior to the vasiform orifice a pair of tubercled
setae is situated, and another pair is present on the medio-caudal portion of case. The
margin of the case (fig. i, F) is dentate, the teeth (fig. i, K, L) being rather fine and
acute. A distance of 0.16 mm. is occupied by twelve of the teeth. At the base of
the teeth small clear areas are found, and some distance in from the margin a row of
elliptical areas, possibly glands, are present. These appear to be on the under siu-face
of the case, while on the submarginal dorsal region, scattered between the margin
and the insertion of the spines, are small dark pores. The vasiform orifice is situated
on a tubercle which forms the caudal portion of the medio-dorsal ridge. It is sub-
circular in outline, tending to cordate. The operculum is somewhat similar in shape
and obscures the lingula. The color of cleared specimens under the microscope is a
light brown, with the margin and the borders of the dorsal ridge darker.
On the leaf the cases are shining black. There is little or no dorsal secretion, but
a short , w^hite, waxy marginal fringe is present. The rods forming this fringe are not
distinct, but are more or less frayed and give a woolly appearance to tlie outer edgeof
the fringe. In some specimens, however, this woolly appearance is not evident,
but the wax forms a series of marginal plates. WTien the pupae are removed from the
leaf, their former position is marked by the white oval wax ring which remains
attached to the leaf. The larvae present a similar appearance on the leaf, but are
brown instead of black.
Adult male. — Length 0.96 mm.; general color brownish, shaded with dusky.
Vertex rotmded, with a longitudinal median ridge, color dark brown; ocelli clear;
compound eyes Vandyke, constricted; antennae absent in the specimens at hand;
labium tipped with dusky; thorax shaded with dusky. Forewings 0.88 mm, long
by 0.35 mm. wide, marked with dark bluish gray, as indicated in fig. i, H. Veins
olive color; radial sector bent caudad at 0.4 mm. from the distal end. Hind wings
0.64 mm. long and 0.25 mm. wide at widest part; color uniform dusky, vein olive
color. Legs with the femora and the proximal half of the tibiae dusky, the remainder
of the tibiae and the tarsi greenish yellow. Fore femora o. 19 mm. ; fore tibia; 0.23 mm. ;
fore tarsi, proximal segment 0.08 mm., distal 0.064 mm.; middle femora 0.24 mm..
462
Journal of Agricultural Research
Vol. VI, No. n
^>^'
Fig. 2.—AleuTocanthus -woglumi: A, egg; B, polygonal markings of egg; C, pupa case; D, margin of
pupa case; E, vasiform orifice of pupa case; F, forewing of adult female; G, same, showing variation
in markings; H, costal margin at base of wing of female; /, forewing of male; /, male genitalia.
June 19, 19 16 Aleyrodidae 463
hind tibiae 0.36 mm., hind tarsus, proximal 0.112 mm., distal 0.72 mm. Claws normal,
with a hairy central paronychium; genital segment dark brown, 0.112 mm., broad at
the insertion of the claspers. These latter are dark-brown, becoming lighter at their
distal tips. They are 0.128 mm. long and each about 0.03 mm. at the shoulder near
the base. They are acute at the tips, curved inward, and armed on the inner margin
with a number of prominent spines (fig. x, G). A few small hairs are scattered here
and there, situated on small tubercles. The penis is as long as the claspers, somewhat
bulbous at the base, greenish yellow, and slightly curved upward.
Adult FEm.'U.e. — Unknown.
Described from adult males in balsam mounts and numerous pupa cases in balsam
mounts and dry upon the foliage.
Type. — Cat. No. 19099, U. S. National Museum.
Aleurocanthus woglumi Ashby.'
AleuTocanlhus woglumi Quain., Ashby, 1915, in Ann. Rpt. Dept. Agr. Jamaica, 1914/15, p. 31.
Aleurocanthus woglumi Quain., Ashby, 1915, in Bui. Dept. Agr. Jamaica, n. s. v. 2, no. 8, p. 322.
Specimens of this species (fig. 2; PI. LXIV, LXV), w^hich may be
called the "spiny Citrus white fly," were first received by the Bureau of
Entomology on June 16, 1 910, from Dr. E. W. Berger, the material coming
from India from H. Maxwell-Lefroy. Specimens were also received in
1910 from Mr. George Compere, who had collected the insect in the Philip-
pine Islands. During 1910 and 191 1 Mr. R. S. Woglum, in the course
of his search for parasites of the orange white fly (Dialeurodes citri
Ashm.), found this insect common and widely distributed on orange in
India and Ceylon, and it has subsequently been received from that
region from Mr. A. Rutherford.
Our first knowledge of its presence in the Western Hemisphere came
with the receipt of specimens from Col. C. Kitchener, Half Way (King-
ston), Jamaica, on November 27, 191 3. Additional material was
received during 1914 from Jamaica from Col. Kitchener and from Prof.
S. F. Ashby, Microbiologist of the Jamaica Department of Agriculture.
Under date of February 5, 1916, specimens were submitted by Mr. P.
Cardin, Entomologist of the Cuba Agricultural Experiment Station,
for verification of determination made by Prof. Ashby. On February 7,
1 91 6, a large lot of orange leaves infested with A. woglumi was received
from Mr. L. J. K. Brace, Nassau, New Province, Bahamas, who states:
Certain orchards in this island at least have been very much affected with this
insect, all of the leaves being so much infested on their undersurfaces that they present
a black appearance, not only killing the trees but causing some persons to attempt to
stop the mischief by cuttiTig down the trees, though the yoimig shoots become again
covered * * *. I have no doubt that the planters' exchange have introduced
this pest from the East. Plants have been for some time obtained by individuals
here from the Jamaican establishment and also from Florida.
Prof. Ashby thinks the insect was introduced into Jamaica on the
mango during the last 20 years. In that island it has become very
1 Aleurccani'.itis woglumi, the writers' mJiauscript name for this species, vvus furnished to Prof. Ashby.
According to the International Code, his descriptive remarks, as cited, make him the author of the
species.
464
Journal of Agricultural Research
Vol. VI, No. 12
prominent, infesting the leaves of all species of Citrus on the lowland
plains. Honeydew is excreted in small amounts, which is followed by
the development of sooty fungi, but not to the extent that is true of
certain other white flies and scale insects.
The present known distribution and food plants are shown in Table I.
Tabi,E I. — Present known distribution and food plants of Aleurocanthus woglumi
Date.
Quaint-
ance
No.
Locality.
Host plant.
Collector.
June 16, 1910
1910
1910
Oct., 1910. . .
Do
(?)
Nov., igio.. .
Do
Do
Do
Dec., 1910. . .
June, 191 1. . .
Sept., 1911 . .
Aug., 1913...
Sept., I9I3..
Nov., 1913. . .
Feb., 1914. . .
May, 1914. . .
Feb., 1916...
Do
5264
6763
6764
6750
6744
6553
6556
6564
6557
6560
6528
8021
8012
8753
8922
12066
12067
India
Manila, P.I
do
Royal Bot. Gardens, Cey-
lon.
do
India
Lahore, India
Gujranwala, India
Lahore, India '
do
Kalimpong, Sikkim, India.
Lahore, India
Nagpur. C. P., India
Peradeniya, Ceylon
do
Half Way, Jamaica
do
Kingston, Jamaica
Orange
....do
....do
Capparis rozburghi .
Guantanamo, Cuba
Nassau, N. P., Bahama .
Capparis pedunculosus . . . .
Unknown tree
Citrus sp
....do
....do
....do
....do
Citrus sp. and Morus sp. . ,
(?)
Salacia reticulata
Kurrimia zeylanica
Orange
do
Citrus sp.; Guiacum of-
ficinale; Cestrutn noc-
turnum L.
Orange
..do
Maxwell- Lefroy.
George Compere.
Do.
R. S. Woglum.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
A. Rutherford.
Do.
Col. C. Kitchener.
Do.
S. F. Ashby.
P. Cardin.
L. J. K. Brace.
Egg (fig. 2, A). — Size, 0.208 mm. by 0.08 mm.; shape elliptical, curved, with the
stalk short and attached some distance from the base. Color yellowish, surface appar-
ently without reticulations in some cases and with them in others, which is no doubt
due to the structure being destroyed in boiling. When they are present (fig. 2, B)
they average 0.006 mm. in diameter.
Larva. — Larvae are present in the material at hand, but they are in too poor a
condition for accurate description. They are browTi in color and armed with numerous
long spines.
Pupa case (fig. 2, C). — Size variable in the different lots of material, averaging
1.4 by 0.89 mm.; shape regularly elliptical, with the dorsum considerably arched or
rounded; the median ridge high, but not markedly distinct from the dorsal area,
excepting near the caudal portion of the abdomen and at the vasiform orifice, which
is elevated into a more or less prominent tubercle. Color dense black, so much so
that it is almost impossible, even after prolonged boiling, to make out details. When
the denser dorsal portion of the case is removed the ventral part appears under the
microscope as dark brown and more or less irregularly mottled. Submarginal area
with usually 20 spines forming a ring. These vary consideably in length, but the
caudal pair is nearly always the longest. The spines are ctu-v^ed outward. A pair
of hairs is present on the caudal margin caudad of the vasiform orifice. The spines
on the dorsum are small excepting two pairs on the abdomen and three pairs on the
thorax. Their number and arrangement are shown in the figure. The vasiform
orifice (fig. 2, E) is prominent, being on a tubercle, but is small. It is somewhat
triangular in shape, tending to circular. The operculum almost entirely fills the
orifice obscuring the lingula — all but a very small portion at the tip. Cephalad of
the orifice a pair of minute setae is situated one on each side. The margin of the case
juneip. i9i6 Aleyrodidae 465
is dentate, the teeth large and bluntly rounded (fig. 2, D). The inner spaces are not
acute, but often squarely truncate. A space of 0.1 mm. is occupied by six or seven
teeth. On this feature alone the case is easily separable from those of the other species.
At the base of the teeth, forming a ring around the case, is a series of minute, clear,
porelike areas. On the leaf the case is jet black with the dorsum somewhat arched
and the abdominal segments marked, but not distinctly separated. On the margin
all around is a narrow cottony lateral wax fringe. This sometimes extends mesad,
irregularly covering the submarginal area, but dorsal secretion is usually absent.
Adult female. — Length from vertex to tip of ovipositor, 1.12 mm.; color brovm,
under the microscope a deep wine color with darker shadings on head, thorax, and
tip of abdomen. The specimens at hand are somewhat imperfect and it is difficult
to make out the structure. The vertex seems to be rounded and possessed of a slight
median ridge. The eyes are very dark brown. The antennae are absent from the
specimens at hand. Labium yellowish, tipped with black. Legs yellowish, shaded
on femora with dusky. The femora and tibise of the hind legs are considerably darker
than the others; length of hind femora 0.288 mm.; hind tibiae 0.432 mm. The tarsi
have the proximal segment o.i mm. and the distal 0.06 mm. The proximal segment
is armed on its distal extremity with one large spine and several smaller ones; the
foot is normal, with the paronychium straight and hairy. Theforewings (figs. 2, F, G)
are 1.268 mm. long and 0.76 mm. wide at the widest part. The radial sector is heavy,
yellowish brown in color, and much curved. The cubitus is very fine, long and
slightly ciu^ed, that portion of the wing below it forming a more or less distinct lobe.
In color the wing is a deep smoky, excepting as follows: A line following the cubitus,
and a rather large spot near its distal extremity are colorless. A line following the
radial sector from its distal extremity to almost its median curve, and another crossing
it almost at right angles are colorless. This gives the appearance of a white cross on
a dark backgrotind. In some wings the marking is not so evident, but there is one
curved colorless line angling across the wing a short distance above and parallel with
the radial sector. The border of this white line seems more heavily shaded than the
remainder of the wing. The margin of the wing (fig. 2, //) is armed with a series of
rather prominent teeth directed toward the distal extremity of the wing. Each one
of these is armed with one prominent spine and usually three smaller ones. The
margin formed by these teeth and a line along their bases is bright wine red. The
hind wing is uniform smoky, with the vein yellowish brown.
Adult male. — Much smaller than the female, measioring only about 0.79 mm.
from vertex to tip of claspers. The specimens are in poor condition, the antennae
are absent, and it is impossible to make out the structure with certainty. The color
is a yellowish or a reddish brown. The hind femora, 0.24 mm. and the hind tibia, 0.4
mm. in length. They are marked as in the female. The claspers (fig. 2, J) are 0.126
mm. long. Near their distal ends there are a number of jagged teeth and they are armed
with a number of long slightly ciu-ved hairs, those near the tip being the longest.
The penis is as long as the claspers, yellowish, and almost straight.
Described from females, males, and pupa cases in balsam mounts and pupa cases
and eggs on the leaves.
Aleiuocanthus spiniferus (Quaintance)
AleuTodes spinifera Quain., 1903. in Canad. Ent., v. 35, no. 3. P- 63.
Collected on CHrus sp. and rose by Mr. C. L. Marlatt, of the Bureau of
Entomology, at Garalt, Java, on December 7, 1901; also taken on
orange at Macao, South China, by Mr. R. S. Woglum, in February,
1911.
466 Journal of Agricultural Research voi. vi, No. 13
Aleurolobus marlatti (Quaintance)
AleuTodes marlatti Quain., 1903, in Canad. Eat., v. 35. no- 3. P- 6r.
This species (Pi. LXVI, fig. 3) was collected by Mr. C. L. Marlatt
on May 17, 1901, at Kumomoto, Japan, on orange; also by Mr. R. S.
Woglum on Citrus sp. and Morus sp. at lyahore, India; also collected by
Mr. Woglum on Ficus sp. in the Royal Botanic Gardens, Ceylon; on
an unknown tree in the Botanic Gardens, Buitenzorg, Java. This
insect has also been received by the Bureau of Entomology from Mr.
S. I. Kuwana, collected at Fukuoka, Japan. Mr. Kuwana states that
this same species has been collected on Rivkin Island. One lot of infested
orange leaves is also in the Bureau collection from Tokyo, Japan.
Aleurothrixus floccosus (Maskell)
AleuTodes floccosa Mask., 1896, in Trans, and Proc. N. Zeal. Inst., v. 2S (n. s. v. 11). 1895, P- 432-
Aleurodes horridus Hempel, 1899, in Psyche, v. 8, no. 280, p. 394.
This species (fig. 3, //) is based on material from Jamaica on lignum-
vitas (Guaiacum officinale?) and was first recorded on orange by Cockerell
(1902) ^ from Mexico. The insect has several color phases, ranging from
clear yellow, the typical and more abundant form, to individuals with
the dorsum striped with dark brown, or with the dorsal disk dark
brown and the submarginal area yellow, etc.
Hempel's A. horridus from Brazil on guava (Psidium guajava) is
apparently the same as A. floccosus. This latter differs from A. howardi
only in the absence of a comb of teeth on the caudal margin of the
vasiform orifice (fig. 3, H). Both A. floccosus and A. howardi are almost
always present together on the same leaf and their food plants and
distribution are practically identical. A. floccosus is common in the
islands of the West Indies and also occurs in Florida, Mexico, British
Guiana, Brazil, Argentina, Canal Zone, Chile, and Paraguay. In addi-
tion to the orange, lime, grapefruit, etc., A. floccosus has been taken
on the sea-grape (Coccoloba uvifera), Plumeria sp., Baccharis genisielloides,
guava, a coarse grass, and a climbing vine.
Aleurothrizus howardi (Quaintance)
Aleyrodes howardi Quain., 1907, U. S. Dept. Agr. Bur. Ent. [Bui.] 12, pt. s. Tech. Set., p. 91.
This species (fig. 3, E, J; PI. LXVI I) occurs on the same host plant
and has the same distribution as A. floccosus. It was apparently first
found in Florida by Prof. P. H. Rolfs at Miami on sea-grape, September
25, 1900, and therefore gained a foothold in that State some years pre-
vious to its discovery by Dr. E. A. Back.
Aleurothrizus ported, n. sp.
This species (fig. 3, A-D, F, G, I, K, L; PI. LXVIII) has been received
only from Chile and Brazil. The first collection was sent by Prof. T. D. A.
* Bibliographic citations in parentheses refer to " Literature dted," pp. 471-472.
Jtuie 19, 1916
Aleyrodidae
467
Fig 3 -Aleurolhrixus Pnrieri. A. howardi, and A.flyccosus: A, Aleurolhrixus porieri: Larva, first instar.
B A /.or/m; Caudal spine of pupa case. C. A . />or/en; Clasper of male. D.A. PoTlert: U^g. E. A. how-
ardi ■ Caudal spine. F. A . Porieri: Pupa case. G. A . Porteri: Forewing of adult. H.A. ftoccosus: Vast-
form orifice of pupa case. /, A. p.irieri: Vasiform orifice of pupa case. /, A. howard,: Vasiform onfice
of pupa case. K, A. porteri: Margin of pupa case. L, A. porieri: Margm of early larva.
468
Journal of Agricultural Research
Vol. VI, No. 12
Cockerell on June 7, 1895, who received the material from Mr. Lataste,
under the name phalaenoides.
In a letter to the senior author in January, 1905, Cockerell suggested
that Lataste supposed the species to be Blanchard's phalaenoides. Since
that time we have shown that phalaenoides Blanchard is a species of
Aleurodicus. Table II records the distribution and food plants of the
specimens of A . porteri in the collection of the Bureau of Entomology.
Table II. — Distribution and food plants of A leurothrixus porteri in the collection of
the Bureau of Entomology
Date.
Collector.
Host.
Locality.
Feb. 13, 1894.
May 14, 1894.
Mar. J4, 1895.
Feb. 4, 1896..
Apr. 1, 1899..
Oct., 1904. . .
Oct. 25, 1904.
Jtine 20, 191 2
Mar., 1913. . .
Jan. 5, 1914..
(?)
Aug. 8, 1915.
M. Lataste
do
do
Edward Reed. . .
D. G. Fairchild.
M. J. Rivera. . . .
....do
Prof. Carlos E. Porter .
....do
Popenoe and Dorsett . .
Prof. Carlos E. Porter.
....do
Orange,
do.
(?).
Orange
Solanaceous plant
Orange?
Schinus dependens Or-
tega.
Schinus molle ,
Orange
Jaboticaba
Lippia citriodora Kunth.
Myrtus
4062 . .
4063 . .
4064. .
4065..
351- •■
3214. ■
12022.
8726. .
8820..
12004.
12024.
12062.
Santiago, Chile.
Do.
Chile.
Ransagua, Chile.
Villa del Mar, Chile.
San Bernardo, Chile.
Santiago, Chile.
Chile.
Santiago, Chile.
Rio de Janeiro, Brazil.
Santiago, Chile.
Do.
Of this material, Quaintance No. 351 is chosen for the type.
Larva, first stage (fig. 3, A). — Size 0.352 by 0.208 mm. Shape elongate elliptical;
abdomen with a moderately distinct keel, the caudal extremity of which projects to
the vasiform orifice; dorsum armed with foiu- pairs of stout straight spines; margin
very minutely serrate and armed on its caudal part with a pair of long curved spines
and the remainder of the margin with 11 pairs of minute spines; antennae straight,
not quite as thick as the dorsal spines and extending slightly beyond the margin;
vasiform orifice almost completely filled by the operculum; color under the micro-
scope pale brown.
Pupa case. — Size 0.88 by 0.502 mm.; shape elliptic, some specimens slightly
broader across the thorax than across the abdomen; dorsum somewhat elevated, the
abdomen with a distinct keel; incisions between marginal wax tubes shallow; vasi-
form orifice (fig. 3, /) small, elevated, operculum filling about half of the orifice and
obsciuing the lingula; spines latero-cephalad of the vasiform orifice and those on the
caudal margin of case short, stout, and somewhat vasiform (fig. 3, 5); those on the
medio dorsum long; other characters very similar to those of A . floccosus. Color vary-
ing from yellow to dark brown and with flocculent wax asm A. floccosus.
Adult male. — Color yellow, eyes dark brown; clasper rather short (fig. 3, C), its
spur acute and not armed within with lobes; a few prominent spines present; length
0.08 mm.; length of insect from vertex to tip of claspers 0.88 mm.; forewing 1.04 mm.
long, without markings, but often uniformly clouded with dusky.
Adult female. — Similar to male in color; length 1.12 mm.; forewing 1.28 mm.
The adults in the collection are poorly preserved, and it is impossible to describe
them in detail.
Described from larvae, pupa cases, and adults in balsam mounts and pupa cases
upon foliage.
Type. — Cat. No. 20171, U. S. National Museum.
juneig, i9i6 AleyYodidoe 469
Bemisia giffardi (Kotinsky)
Aleyrodes giffardi Kotin., 1907, m Bd. Com. Agr. and Forest. Hawaii Div. Ent. Bui. 2, p. 94.
This insect is reported present on Citrus trees in several gardens in
Honolulu, where it is stated to be so abundant that the foliage of the
trees becomes blackened by the sooty fungus growing on the exuded
honeydew. Mr. Kotinsky beUeves that the insect has been introduced
into Hawaii, and this opinion is strengthened by its discovery in collec-
tions of material made by Mr. Woglum at Lahore, India, in 191 1. The
host, however, was an unknown tree.
Dialeurodes citri (Ashmead) ^
Aleyrodes citri Riley and Howard, 1893, in Insect Life, v. 5, no. 4, p. 219.
AleuTodes eugeniae, var. ouran/ji Mask., 1896, in Trans, and Proc. N. Zeal. Inst., v. 28 (n. s. v. 11). 1895,
P- 431-
Aleyrodes aurantii Ckll., 1903, in Fla. Agr. Exp. Sta. Bui. 67, p. 663.
This is the destructive Citrus white fly of Florida, where it has been
known since about 1880 (PI. LXVI, fig. i). It is rather generally dis-
tributed over the orange-growing regions of the Gulf States and is common
on chinaberry and Cape jasmine considerably north of the Citrus belt.
It is also recorded from Colorado, Illinois, and the District of Columbia,
where it is probably confined to conserv^atories. This insect was dis-
covered in California in 1907 and serious attempts were made to efifect
its eradication. It is still present in one locality (Marysville) , where it
is now so widespread and abundant that its eradication is considered to
be impracticable (Weldon, 191 5).
Dialeurodes citri is undoubtedly of oriental origin. It has been re-
ceived from numerous localities in India, Ceylon, Japan, China, etc.
According to Kirkaldy it is present in Chile, Mexico, and Brazil. In
addition to Citrus plants, the insect in Florida infests numerous others
as Melia azederach, Gardenia jasminoides, Ligustrum spp., Diospyros kaki,
Diospyros virginiana, Syringa sp., Cofjea arahica, Ficus nitida, etc.
This and nearly related species are very generally parasitized in the
Orient by certain four-winged flies, which are in that region apparently
effective checks on their undue increase.
Dialeurodes citrifolii (Morgan)
Aleyrodes citrifolHyioTgan, 1893, La. Agr. Exp. Sta. Spec. Bnl., p. 70.
Aleyrodes nubifera Berger, 1909, Fla. Agr. Exp. Sta. Bui. 97, p. 67.
Aleyrodes nubiferaMoT. and Back, 1911, U. S. Dept. Agr. Bur. Ent. Bui. 92, p. 86.
This species, long confused with D. citri, may be readily distinguished
from that species by the reticulate eggs, character of the tracheal folds
of the pupa case, and the smoky patch on front wings of the adults. The
insect is known from North CaroUna, Mississippi, Louisiana, California,
1 This species was first fully described by Riley and Howard in Insect Life, as cited, but had earlier
been named and briefly described in The Florida Dispatch, November, 1885. by W. H. Ashmead, who,
according to the rules of the International Code, must be known as the author of the species.
470 Journal of Agricultural Research voi. vi, no. la
and Florida. While not as important as D. ciiri, it is nevertheless
decidedly noxious. It is also known to occur in Cuba and Mexico. No
specimens of this insect were found in the Woglum collection of white
flies from India, Ceylon, and other points in the East visited by him.
By reason of its affinities, D. citrifolii is, however, almost surely oriental
in origin.
This species, with one exception, is known to attack only Citrus plants.
It was found on Ficus nitida growing in greenhouses at Audubon Park,
New Orleans, La.
Paraleyrodes perseae (Quaintance)
AleuTodes perseae Quain., 1900, U. S. Dept. Agr. Div. Eat. [Bui] 8, Tech. Ser., p. 32.
Paraleyrodes perseae Quain. and Baker, 1913. U. S. Dept. Agr. Bur. Ent. [Bui.] 27, pt. 1, Tech. Ser., p. 82,
This species is known only from Florida, where it is frequently found
on orange, though never in destructive numbers thus far. It also feeds
upon Persea, the avocado (Persea americana) , and doubtfully on per-
simmon (Diospyros spp.) . Several species of the genus are common in the
West Indies, perseae being the only one known from the United States.
Trialeurodes floridensis (Quaintance)
Aleurodes floridensis Quain., 1900, U. S. Dept. Agr. Bur. Ent. [Bui.) 8, Tech. Ser., p. 26.
T. floridensis has thus far been recorded by the Bureau of Entomology
only from Florida, where it is rather generally distributed. It infests
avocado, guava, Annona squamosa, and the orange. While often
very numerous on guava and avocado, it is at present of no importance
on orange.
Trialeurodes vitrinellus (Cockerell)
Aleyrodes vitrinellus Ckll., 1903, in Ent. News, v. 14, no. 7, p. 241.
The type of this species is from Mexico on orange. Apparently the
same insect has been taken in southern California on oak. Its injuries
to orange in Mexico are probably not great.
Tetraleurodes mori (Quaintance)
Aleurodes mori Quain., 1899, in Canad. Ent., v. 31, no. 1, p. i.
This indigenous species (PI. LXIX, fig. 2) is widely distributed over
the eastern United States and occurs on a large variety of plants, as
mulberry, sycamore, maple, dogwood, hackberry, persimmon, holly,
mountain laurel, etc. It has been found several times on orange, but
not as yet in injurious numbers. That it may become troublesome under
certain conditions, however, will be evident from the discussion relative
to T. mori, var. arizonensis, which follows:
juneig. i9!6 Aleyrodidoe 471
TetraleuTodes mori, var. arizonensis (Cockerell)
Aleyrodes mori, var. arizonensis C^W., 1903, in Fla. Agr. Exp. Sta. Bui. 67, p. 666.
Aleurodes mori CkH., 1900, in Sd. Gossip, n. s. v. 6, no. 72, p. 366.
Described from specimens taken in Arizona on orange (PI. LXIX) . The
variety T. mori arizonensis is stated to differ from the typical T. mori in
having the wings white marked with black without any red. An exami-
nation of the type specimens after mounting shows the presence of red
markings on wings exactly as in T. mori, and we are unable to distinguish
any characters in support of its status as a variety. On different occa-
sions the Bureau of Entomology has received from Mexico an aleyrodid
seriously infesting the orange (PI. LXIX) which we are unable to distin-
guish in the immature stages from T. mori, and this species is considered by
Cockerell to be identical with his variety T. mori arizonensis. While the
variety, in our judgment, is invalid, we retain the name to designate a
race of T. mori which, in Mexico, for some reason breeds abundantly on
orange and is a pest of importance. T. mori arizonensis occurs only on
orange in Mexico so far as bureau records indicate. It was first collected
in 1894 by Dr. C. H. T. Townsend at Guadalajara and San Luis, and
again by Townsend in 1902 at Zapotlan. Two lots of material were
received from Prof. A. L. Herrera in 1905, without statement as to
locality.
LITERATURE CITED
ASHBY, S. F.
191 5. Notes on diseases of cultivated crops observed in 1913-14. Black fly or
"black scale." In Bui. Dept. Agr. Jamaica, n. s. v. 2, no. 8, p. 321-322.
191 5. Report of the microbiologist. Citrus diseases. In Ann. Rpt. Dept. Agr.
Jamaica, 1914-15, p. 31.
Berger, E. W.
1909. White-fly studies in 1908. Fla. Agr. Exp. vSta. Bui. 97, p. 39-71, 19 fig.
Cockerell, T. D. A.
1900. Economic entomology in Arizona. In Sci. Gossip, n. s. v. 6, no. 72, p. 366-
367-
1902. A synopsis of the Ale>Todidas of Mexico. In Mem. y Rev. Soc. Cient.
"Antonio Alzate," t. 18, p. 203-208, 3 fig.
1903. Aleyrodes (Trialeurodes) vitrinellus Ckll. In Ent. News, v. 14, no. 7, p. 241.
1903. White fly. (Aleyrodes citri) and its allies. In Fla. Agr. Exp. Sta. Bui. 67,
p. 662-666.
Hempel, Adolph.
1899. Descriptions of three new species of Aleurodidae from Brazil. In Psyche,
V. 8, no. 280, p. 394-395-
KoTiNSKY, Jacob.
1907. Aleyrodidae of Hawaii and Fiji, with descriptions of new species. In Bd.
Com. Agr. and Forest. Hawaii, Div. Ent. Bui. 2, p. 93-103, pi. i.
Maskell, W. M.
1896. Contributions toward a monograph of the Aleurodidas, a family of Hemiptera-
Homoptera. In Trans, and Proc. N. Zeal. Inst., v. 28 (n. s. v. 11), 1895,
p. 411-448, pi. 24-35.
472 Journal of Agricultural Research voi. vi, no. 12
Morgan, H. A.
1893. The orange and other citrus fruits, from seed to market, with insects bene-
ficial and injurious, with remedies for the latter. La. Agr. Exp. Sta. Spec.
Bui. no, p. 36, fig., 3 pi. (2 col.).
Morrill, A. W., and Back, E. A.
191 1. White flies injurious to citrus in Florida. U. S. Dept. Agr. Bur. Ent. Bui. 92,
109 p., 19 fig., 10 pi. (i col.).
Newstead, Robert.
191 1. On a collection of Coccidae and Aleurodidae, chiefly African, in the collection
of the Berlin Zoological Museum. In Mitt. Zool. Mus. Berlin, Bd. 5, Heft
2, p. 155-174.
QUAINTANCE, A. L.
1899. New, or little known, Aleurodidae-I. In Canad. Ent., v. 31, no. i, p. 1-4,
illus.
1900. Contributions toward a monograph of the American Aleurodidae. In U. S.
Dept. Agr. Div. Ent. [Bui.] 8, Tech. Ser., p. 1H34, 8 pi.
1903. New oriental Aleurodidae. In Canad. Ent., v. 35, no. 3, p. 61-64.
1907. The more important Aleyrodidae infesting economic plants, with description
of a new species infesting the orange. U. S. Dept. Agr. Bur. Ent. [Bui.]
12, pt. 5, Tech. Ser., p. 89-94, fig. 23-24, pi. 7.
and Baker, A. C.
1913. Classification of the Aleyrodidae. Pt. i. U. S. Dept. Agr. Bur. Ent. [Bui.]
27, pt. I, Tech. Ser., 93 p., 11 fig., 34 pi.
[Riley, C. V., and Howard, L. O.]
1893. The orange Aleyrodes. In Insect Life, v. 5, no. 4, p. 219-226.
Weldon, G. p.
1915. White fly at Mar>'sville. In Mo. Bui. State Com. Hort. [Cal.], v. 4, no. 8,
p. 386-388.
PLATR LXIV
A leurocanthus woglumi: Eggs, larvae, and pupa cases on orange leaves.
Aleyrodidae
Plate LXIV
^^^S$
• - ■ 9J_ ,rf« V * ^ , /.
jO*
it^^JSiE^^^'-
r*^?5^:'^f;
Journal of Agricultural Research
Vol. VI, No. 12
Aleyrodidae
Plate LXV
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LXV
A leur acanthus woglumi:
Fig. I. — Colony on an orange leaf.
Fig. 2.— Eggs and pupa cases, greatly enlarged
37769°— 16 4
PLATE LXVI
Fig. I. — Dialeurodes citri: Pupae, much enlarged.
Fig. 2. — Male and female adults of an aleyrodid.
Fig. 3. — Aleurolobus marlatti, much enlarged.
Aleyrodidae
Plate LXVI
'■ -»,
Journal of Agricultural Research
^■*
^<^
Vol. VI, No. 12
Aleyrodidae
Plate LXVII
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LXVII
Aleurothrixus howardi: Larvae and pupa cases on an orange leaf, enlarged.
PLATE LXVIII
Aleurothrixu^ porteri: Larvae and pupa cases on Myrtus sp., enlarged.
Aleyrodidae
Plate LXVIII
Journal of Agricultural Research
V._.,. v'l, No. 12
Aleyrodidae
Plate LXIX
Journal of Agricultural Research
Vol. VI, No. 12
PLATE LXIX
Fig. i.—Tetraleurodes mod, var arizonensis: Larvae and pupa cases on an orange
leaf, enlarged.
Fig. 2.—Tetraleurodes mori: Pupa cases on a mulberry leaf, much enlarged.
ADDITIONAL COPIES
OF THIS PUBLICATION MAT BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
25 CENTS PER COPY
Subscription Price, $3.00 Per Year
Wh'^-'-0::
Vol. VI JUNE^ 26, 1916 No. 13
J
^GRICUUrURAlv
CONTENTS
Page
Relative Water Requirement of Corn and the Sorghums - 473
SDW3N C. MILLER
Availability of Mineral Phosphates for Plant Nutrition - 485
\Y. L. BURLISON
DEPARTMENT OF AGRICUETURE
■WA-SHINGTON, DX^
•H1N<JT0N i aOVESNMENT PRINTINO OFFICE : 1S1
PUBLISHED BY AUTHORITY OF THE SECRETARY
OF AGRICULTURE, WITH THE COOPERATION
I OF THE ASSOCIATION OF AMERICAN AGRICUL-
TURAL COLLEGES AND EXPERIMENT STATIONS
EDITORIAL COMMITTEE
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELIvERMAN, Chairman RAYMOND PEARL
I'hyiiologisl a?id Assistafil Chief, Bureau
of Plant Industry
EDWIN W. AELEN
Chief, Office of Experivunt Stations
CHARLES L. MARLATT
Assistant Chief, Bureau of Entomology
Biologist, Maine Agricultural Experiment
' St at i 071
H. P. ARMSBY
Director, Institute of Animal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Patlwlogist , and Assistant
Dean, Agricultural Experiment Station of
the University of Miymesota
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 Experiment Stations should be
addressed to Rayraond Pearl, Journal of Agricultural Research, Orono, Maine.
JOfflALOFAGEIOlTlALlSEMCH
DEPARTMENT OF AGRICULTURE
Vol. VI
Washington, D. C, June 26, 1916
No. 13
RELATIVE WATER REQUIREMENT OF CORN AND
THE SORGHUMS
By Edwin C. Miller '^BW Vu*
Assistant Plant Physiologist, Department of Botany, Kansas Agricultural Experiment ""T AMC.*
Station <iA»l>P.N
INTRODUCTION
During the summers of 1914 and 1915 a physiological study of the
water relations of corn and the nonsaccharin sorghums was made at the
State Branch Experiment Station at Garden City, Kans. In connection
with other experiments it was thought advisable to determine the water
requirement of several varieties of these plants. The term "water
requirement," as used in this paper, means the ratio of the weight of the
water absorbed by the plant to the weight of the dry matter produced.
EXPERIMENTAL METHODS
CLIMATIC DATA
The instruments for recording the climatic conditions consisted of a
hydrograph, a thermograph, maximum and minimum thermometers
placed in a standard shelter 4 feet from the ground, a rain gauge, an
evaporation tank, and an anemometer which measured the wind velocity
at a height of 2 feet.
A portion of the weather records for the two seasons averaged for five-
day periods is shown in Table I. These show that the climatic con-
ditions for the two seasons were in marked contrast. The summer of
1 91 5 was much cooler than that of 1914 and the rainfall for the months
of May, June, July, August, and September in 191 5 was approximately
three times that for the same months in 191 4. The evaporation during
5-day periods is shown graphically in figure i.
The evaporation for each of the growing months with but one excep-
tion was much higher in 191 4 than in 191 5.
CULTURAL METHODS
The plants were grown in large metal cans made from 2 2 -gauge gal-
vanized iron. These cans were 24 inches in height with a diameter of
15 inches, and under the conditions of these experiments contained about
Journal of Agricultural Research,
Dept. of Agriculture, Washington, D. C.
(473)
Vol. VI. No. 13
June 26, 1916
Kans. — s
474
Journal of Agricultural Research
Vol. VI, No. 13
no kgm. of soil. Forty of these cans were used in 1914 and 60 in
191 5. The upper foot of field soil was worked through a sieve with a
5^-inch mesh and then thoroughly tamped in the cans. The soil was in
good tilth, and for both seasons the moisture content ranged from 20
to 21 per cent (dry basis). It had a moisture equivalent of 24 and a
wilting coefficient of 13, as calculated by the formula of Briggs and
Shantz.^
The cans were provided with metal lids which were sealed with ordi-
nary binding tape (PI. LXXII, fig. 4). This was made waterproof by
V^K J6/A/S JuLi" August September
Fig. I. — Curves of the evaporation at Garden City, Kans., for the growing period of 1915.
giving it a heavy coat of varnish after it was in position. Three 2-inch
holes equidistant from one another were made near the periphery of
each lid to accomodate the plants. The seeds were planted in the cans
and the young plants gradually thinned to the desired number. Three
corn plants were grown in each can, both in 1914 and 191 5. Six sorghum,
plants were grown in each can in 1914, but in 191 5 the number was re-
duced to three plants to each can. In order that the plants might be
as nearly as possible under the same climatic conditions during the
growing season, the seeds of all the plants used were sowed on the same
date. These were planted on May 26 in 1914, and on May 22 in 1915.
1 Briggs, L. J., and Shantz, H. L. The wilting coefficient for different plants and its indirect determina-
tion. U. S. Dept. Agr. Bur. Plant Indus. Bui. 230, 83 p., 9 fig., 2 pi. 1912.
June 26, 1016 Water Requirement of Corn and Sorghum
475
Table I. — Summary of the climatic conditions at Garden City, Kans.,for IQ14 and 1915
Period (inclusive).
Air temperature (°F.).
1914.
May:
I to 5
6 to 10
II to 15
16 to 20
21 to 25
25 to 31
Jime:
I to 5
6 to 10
II to 15 j 76
16 to 20
2 I to 2 5
26 to 30
July:
I to 5
6 to 10
II to 15
16 to 20
2 I to 2 5
26 to 31
August:
I to 5
6 to 10
II to 15
16 to 20
2 I to 2 5
25 to 31
September:
I to 5
6 to 10
II to 15
16 to 20
21 to 25
26 to 30
May.
I to
6 to
II to
16 to
20 to
25 to
June:
I to
6 to
II to
16 to
21 to
26 to
July:
I to
6 to
1915-
Average 0
-
Means.
Maxi-
Mini-
mums.
mums.
58
68
47
65
78
51
53
61
44
62
68
55
72
84
59
69
79
57
76
87
65
77
8q
64
76
88
63
76
8q
62
82
94
69
1 "r
/ /
94
59
74
85
62
11
91
60
86
99
6q
76
87
62
81
94
6s
«3
98
66
77
93
65
77
91
62
77
91
62
82
99
64
11
91
61
13
87
60
11
94
60
79
90
64
75
89
58
77
90
60
63
80
44
67
86
51
53
65
38
56
^9
44
71
87
55
46
55
39
67
78
57
■ 55
65
47
65
75
58
64
78
52
66
78
53
71
85
61
69
79
58
72
84
59
66
77
55
76
90
60
Maxi-
mum.
92
79
90
92
91
96
99
98
103
94
93
103
lOI
98
102
95
95
95
102
99
Mini-
mum.
44
41
38
50
57
49
62
51
59
58
64
SI
53
53
64
58
64
64
Precipi-
tation.
94
54
103
55
102
59
96
48
97
56
85
37
90
47
76
31
81
32
94
46
68
32
90
44
72
39
81
55
86
36
87
50
95
56
91
56
88
57
83
49
96
54
Inches.
I. 40
• 19
. 20
19
61
39
04
15
. 21
. 10
Trace.
61 1 .38
56 I Trace.
58 I .19
62 .06
03
79
Evapora-
tion.
Inches.
0- 953
1.484
1- 135
•596
1.584
1.294
1.432
I. 728
I. 520
1.409
I. 991
1.862
I. 200
1.440
1.822
I. 416
1-451
2.074
1-477
1.792
1.474
1-959
1-745
1-563
1-739
I. 501
1-653
1.390
1-343
1.740
I. 187
•985
1-857
1-324
I. 069
I. 169
-738
1.386
1.490
1.48s
I. 181
I. 419
I- 451
1-732
Wind
velocity
per hour.
Miles.
9.0
II. 8
10. 9
13.6
10. 2
6.9
13.0
15.2
9-3
6.4
9-9
7-7
6.1
4-7
5-7
7-7
5-7
5-7
6.1
8.0
7.0
8.2
7-5
7-4
7-5
8.6
II. 4
7-6
6.4
11. I
10. o
7-7
10.8
12. 2
8.6
8.1
8.7
8.6
8.0
8.8
8-5
7-1
8.6
476
Journal of Agricultural Research
Vol. VI, No. 13
Table I . — Sumviary of the climatic conditions at Garden City, Kans. ,for IQI4 and igi^ —
Continued
Period (inclusive).
1915-
July — Continued.
II to 15
16 to 20
2 1 to 2 5
25to3i
August :
I to 5
6 to 10
II to IS
16 to 20
2 1 to 2 5
25 to 31
September:
I to 5
6 to 10
II to IS
16 to 20
2 1 to 2 s
25 to 30
Air temperature (°F).
Average of —
72
74
75
69
70
72
61
70
63
68
69
71
69
66
56
Maxt- Mini-
mums, naums.
97
84
85
74
83
80
83
80
81
77
83
81
84
82
76
67
Maxi-
mum.
67
62
61
64
56
60
61
61
60
50
55
56
60
55
58
lOI
96
91
90
90
94
86
85
87
91
97
87
84
78
Mini-
mum.
64
56
56
62
51
56
59
57
57
40
51
54
53
39
50
44
Precipi-
tation.
Inches.
O. 06
•15
•13
.24
.90
5- II
. 10
• 03
.46
.82
Evapora-
tion.
Trace.
. 20
I. 00
•25
Inches.
1-743
1.407
1-397
1.528
1. 012
.860
•927
.790
1.018
1-313
1.424
I. 029
-983
I. 072
.864
•66s
Wind
velocity
per hour.
Miles.
6.7
7.0
5-5
6.8
7-4
6.3
7-2
5-2
18.2
4-4
The holes in the lids were made water-tight by using a mixture of
approximately 16 parts by weight of beeswax to i part of Venetian
turpentine. Under ordinary conditions the young seedlings of the corn
and sorghum can readily penetrate this wax. After the plants had
emerged through the wax, it was replaced by a mixture containing a
much smaller amount of Venetian turpentine, in order to secure a seal
that would remain firm around the plants during the hot summer weather.
The lids of the cans were given a heavy coat of white paint and were
then covered with a layer of burlap in order to protect them from ex-
cessive heat. The water lost by the plants was replaced every 48 hours
by the method used by Briggs and Shantz * in their extensive work on
the water requirement of plants.
It was thought advisable to determine the water requirement based on
the dry weight of both the aerial portions and the roots of the plants.
The water requirement was obtained in this manner for Pride of Saline
corn, Blackhull kafir, Dwarf milo, and Dwarf Blackhull kafii. The
method used in the isolation of the root systems of these plants has been
previously reported by the writer int his Journal.^
1 Briggs, L. J., and Shantz, H. L. The water requirement of plants. I. Investigations in the Great
Plains in 1910 and 1911. U. S. Dept. Agr. Bur. Plant Indus. Bui. 284, 49 p., 2 fig., 11 pi. 1913.
The water requirement of plants. II. A review of the literature. U.S. Dept. Agr. Bur.
Plant Indus. Bui. 285, 96 p., 5 fig. 1913.
The relative water requirement of plants. In Jour. Agr. Research, v. •?, no. i, p. 1-64,
ifig., pi. 1-7. 1914-
2 Miller, E. C. A comparative study of the root systems and leaf areas of corn and the sorghums. In
Jour. Agr. Research, v. 6, no. 9, p. 311-332. 1916.
June 26, 1916 Water Requirement of Corn and Sorghum
477
SCREENED INCLOSURE
The plants were grown in a screened shelter in order to protect them
from the hailstorms and severe winds that are frequent in this region.
The inclosure was 20 feet square and had a flat top 10 feet from the
ground. It consisted of a framework of 2 by 4 inch studding spaced
3 feet apart and covered on both the top and sides by a wire netting
with a X-inch mesh. Cheesecloth w^as placed around the sides of the
inclosure to a height of 4X f^^t from the ground. This was held in
position by poultry netting tacked over the outside (PI. LXX, fig. i).
The bottom of the inclosure was provided with a smooth, rigid floor
made of matched pine lumber. The cans were placed in three double
rows running north and south inside the inclosure, with a space of 2 feet
between each row. The height of the floor was such that the upper sur-
face of the cans came to within i^ feet of the top of the cheesecloth.
The rate of evaporation inside and outside the shelter was determined
by two Livingston * porous-cup atmometers. These were renewed every
three or four weeks. They were connected with burettes which were
graduated to o.i c. c, and readings were made twice each day. The
atmometer outside the inclosure was placed at a distance of 2 feet from
the ground in the center of a plot that was planted to com. The atmom-
eter in the inclosure was placed a few inches above the upper surface
of the cans during the early part of the growing season and 2 feet above
their tops when the plants had reached 3 feet in height. The monthly
evaporation for the two seasons from the porous-cup atmometers,
having a coefficient of 0.74 is given in Table II.
Table II. — Monthly evaporation (in cubic centimeters) inside and outside the screened
inclosure for igi4 and igi^
Period.
Ratio.
1914
June 10 to July 10
July 10 to Aug. 10
Aug. 10 to Sept. 10
1915-
June 10 to July 10
July 10 to Aug. 10
Aug. 10 to Sept. 10
The rate of evaporation within the inclosure as measured by the
porous-cup atmometers, was only approximately two-thirds as high as
that in the field. Briggs and Shantz ^ found that plants grown in such
a shelter had a water requirement approximately 20 per cent lower than
' Livingston, B. E. The Relation of Desert Plants to Soil Moisture and to Evaporation. 78 p., illus.,
Washington, D. C, 1906. (CarneKie Inst. Washincton, Pub. 50.) Literature cited, p. 77-7H.
Operation of the porous-cup atmometer. In Plant World, v. 13, no. 5, p. 111-119. 1910.
» Briggs, L. J., and Shantz, H. L., 1913- Op. cit.
478
Journal of Agricultural Research
\o\. VI. No. 13
plants of the same kind grown in the open. The relative water requir-
ment, however, is probably affected little, if at all, by the shading due to
an inclosure of this kind, and it offers the only scientific method for
studying the relative transpiration of plants under the severe climatic
conditions experienced in this region.
WEIGHING THE CANS
Each can was placed on a small wooden platform, which was provided
with a screw eye at either end and mounted on four iron castors. By
means of an iron rod, hooked at one end and bent into a handhold at the
other, the cans could be moved easily wherever desired (PI. LXX, fig. 2).
The cans were pulled over a track made of pine flooring to a small scale
house located 12 feet from the shelter and were weighed every 48 hours
on platform scales that were sensitive to 50 gm. (Pi. LXX, fig. i). In
this manner two men could easily weigh the 60 cans in less than i>^
hours.
EXPERIMENTAL DATA
CORN
Four varieties of com were grown in 19 14 and three varieties in 191 5.
The results for the two years are shown in Tables III and IV.
Table III. — Water requirement of Pride of Saline corn at Garden City, Kans., in 1914
and 191 5
Number
of
plants.
Dry
matter,
including
roots.
Dry matter,
without
root
(stem and
leaves).
Water
transpired.
Water requirement
based on —
Period of growth.
Pot
No.
Total dry
weight,
including
roots.
Total dry
weight,
excluding
roots (stem
and leaves).
1914.
May 26 to Aug. 22.
13
14
15
.16
3
3
3
3
Gm.
164.3
169.8
147.0
180. I
Gm.
150. 6
153-9
131- 4
163- 7
Kgm.
53-5
63-3
61.7
61. 4
325-8
373- I
420. 0
341.0
355-4
411. 7
469.9
375-3
365±i5
403 ± 18
17
18
19
20
21
22
23
24
25
.26
3
3
3
3
3
3
3
3
3
3
236. I
285.6
260. 4
230.6
244-3
260. 0
205. 6
252-5
234-4
202. 4
211. 2
228.3
165.4
178.6
180.5
154- 4
56-9
64. 2
63-4
55-3
58.1
59-2
46.6
46. 0
46. 7
40. 6
1915-
May22 to Aug.25.
241. I
225. I
243-7
240. 0
238.1
227.9
276. 0
254.6
270.7
273-4
275-5
259.6
282. 2
257-9
259.0
263.0
Mean
236±3
267±2
June 26, 1916 Water Requirement of Corn and Sorghum
479
Table IV. — Water requirement of Sherrods White Dent, Chinese, and hybrid corn at
Garden City, Kans. , in IQI4 and IQI5
Variety and period of growth.
Pot No.
Number of
plants.
Dry matter,
excluding
roots
(stem and
leaves).
Water
transpired.
Water re-
quirement.
1914.
Sherrods White Dent, May 26 to
Aug. 22
I 19
3
3
3
Gm.
142.4
143-7
Kgm.
54-7
50-7
60.8
410. 8
356-3
423-3
Mean
396±i6
r 20
I 21
3
3
136. I
157-3
58-1
64- 5
Chinese, May 26 to Aug. 22
427. I
410.3
Mean
4i8±7
22
23
24
I 25
3
3
3
3
120. I
143-3
142. 6
155-0
40. 2
50-5
54-4
51-9
Hybrid F3 Hsg,^ May 26 to Aug. 22
335-3
361.8
381.4
342.0
Mean
3SS±8
27
28
29
I 30
3
3
3
3
145-7
150-4
145- 2
120. 5
42.8
43-7
41. 0
39-8
1915-
Sherrods White Dent, May 22 to
Aug. 18
293-7
291. 0
282. 9
330.9
299±8
43
44
45
46
I 47
3
3
3
3
3
239-7
125. 6
137-7
248.5
249-3
54- I
33-3
36-3
58.0
60. 7
Hybrid F4 Hjg,^ May 22 to Aug. 25
225-9
265.7
264. 0
233-4
243-7
Mean
246±6
o This hybrid has the following origin: The female parent was a plant belonging to the Fi generation of a
cross between Sherrods White Dent corn 9 and white Chinese com cf . The male parent was a plant of the
variety known as Esperanza (Mexican corn). The cross was made on the breeding grounds of the Depart-
ment of Botany of the Kansas Experiment Station in 1910.
Four cans of Pride of Saline com were grown in 1914 and ten in 191 5.
These plants varied in mature height from 5 to 6 feet, but produced no
ears during either season. The plants grew from May 26 to August 22
in 1914, and from May 22 to August 25 in 1915. The water requirement
of Pride of Saline corn, based on the total dry matter, including the roots,
was found to be 365 ± 15 in 1914 and 236 ±3 in 1915. The water require-
ment, based on the total dry matter of the aerial parts of the plants, was
403d: 18 and 267±2 for the years 1914 and 1915, respectively (PI.
LXXII, fig. 2).
Sherrods White Dent corn was grown in three cans in 19 14 and in
four cans in 191 5. In 1 914 the seeds were planted on May 26 and the
480 Journal of Agricultural Research voi. vi. no. 13
plants were harvested on August 22, while in 191 5 they were planted on
May 22 and harvested on August 18. The water requirement of this
variety of com, based on the total dry matter of the aerial parts, was
found to be 396±i6 in 1914 and 299±8 in 1915.
In 1 914 two cans were planted to white Chinese com. The growing
season of these plants was from May 26 to August 22. The water
requirement, based on the dry weight of the aerial parts, was 418 ±7.
In 1 914 four cans were planted to the F3 generation of a segregate of
a hybrid corn developed by the Department of Botany of the Kansas
Experiment Station. Five cans of the F4 generation of this hybrid were
grown in 191 5. Its water requirement, based on the total dry matter
of the aerial parts, was 355 ±8 and 246 ±6, for the years 1914 and 1915,
respectively.
SORGHUMS
Dwarf milo and Blackhull kafir were the only sorghums grown in 1 914.
In addition to these two varieties, dwarf black-hulled white kafir, f eterita,
and Sudan grass were grown in 191 5. The results for the two seasons
are shown in Tables V and VI.
Six cans of Dwarf milo were planted in 1914 and eight cans in 1915.
The plants in the former year reached a height of 3 feet, and during the
latter year they stood 4>^ feet high (PI. lyXXI, fig. i). The growing
season was from May 26 to August 22 in 1914, and from May 22 to Sep-
tember 3 in 1 91 5. The water requirement, based on the total dry matter,
including the roots, was found to be 319 ±5 in the former year and 228 ±3
in the latter. The water requirement, based on the total dry matter of
the aerial parts, was 340 ±5 and 244 ±3 for the years 19 14 and 191 5,
respectively. The water requirement, based on the production of grain,
was 1,022 ±100 in 1914 and 5o8±6 in 1915.
Blackhull kafir was grown in six cans in 1914 and in eight cans in 1915.
The seed was planted on May 26 and the plants were harvested on Septem-
ber 3 in 1914, while in 1915 the growing period was from May 22 to Sep-
tember 18. The plants reached a height of 6 feet in each of the growing
seasons (PI. LXXII,fig. 3). The water requirement, based on the total
dry matter, including the roots, was 305 ±6 in 1914 and 204±2 in 1915,
while the water requirement, based on the total dry weight of the aerial
parts, was 325 ±7 for the former year and 2 17 ±2 for the latter. The
water requirement, based on the production of grain, was 1,178 ±45 in
1914 and 696± 19 in 1915.
June 26, 1916 Water Requirement of Corn and Sorghum
481
Table V. — Water requirement of Dwarf milo and Blackhull kafir at Garden City, Kans.,
in I 91 4 and 191 5
DWARF MILO
6
'A
1
"a
t-i
1
a
1
i-T
«
6
>.
1
t-.
1
a
s
>.
Q
s
0
.Si
•d
a,
"a
a
a
u
Water requirement based on —
Period of growth.
.a S
M 0
>..H
•51
r
1°
•01
■jo's
a
■5
0
1
•0
1
1914.
May 26 to
August 22. .
I
2
3
4
5
I 6
6
6
6
I
6
Gm.
199.0
172. 2
186.8
196.4
173-7
169. 2
Gm.
187.3
161. 5
173-7
184.4
161. 7
159-7
71.8
40.4
45-0
79-3
58.8
68.5
Gm.
115- 5
121. I
128.9
105. I
102. 9
91. 2
Kgm.
61. 4
55-8
65-1
60.6
51.6
55-1
308.9
324.2
348.7
308.7
297-5
326.0
328.2
345-6
374-6
328.7
319-9
345-3
855-3
1,381.9
I, 447- 7
764-4
879.0
805. 2
532-2
461. 0
505.4
576.8
502.3
604.8
Mean . .
3i9±5
340 ±5
1022 ±100
53o±i5
I
2
3
\
6
7
. 8
214. 6
226. 4
231.4
223.3
233-3
217. 6
230-5
225.8
103. I
114. 6
105. 6
102. 0
109. 6
107. 6
114. 7
108.3
III. 5
III. 8
125.8
121. 3
123.7
no. 0
115-8
"7-5
1915-
May 22 to Sep-
tember 3 . . .
3
3
3
3
3
3
228. I
239.1
245-4
245-9
248.3
231.6
247-3
240.8
51-5
50-5
55-3
56.1
55-6
54-7
60. 4
55-3
225.7
211. 5
225. 7
228. 2
224. I
236-3
244.4
229.8
239-9
223-3
239-3
251-3
238.5
251-5
262. 2
245- I
499-5
437-2
524-5
549-1
507-7
508.7
526.7
511. 0
461.8
452.3
440.3
462. 6
449.8
497-6
521.8
471.0
Mean . .
228±3
244±3
5o8±6j 469^7
BLACKHULL KAFIR
I9I4.
f 7
4
234-4
217.9
54-5
163.4
75-0
320.0
344-2
1,376.4
459-1
8
5
247.0
234. I
66. 7
167.4
68. I
276. 0
291. 2
I, 022. I
407.3
May 26 to Sep-
9
5
226.8
212. 6
5.5- 5
157- I
67.7
298.7
318.7
I, 220. 9
431-3
tember 3 . . .
10
6
233- 3
219. 5
60.5
159. 0
78.1
335- 0
356-0
I, 291. 9
491-5
II
4
186. 5
175- 6
52.0
123.6
58.6
314-5
334-1
I, 126. 7
474.6
I 12
6
278.1
257.3
77-3
180. 0
79-8
287. 0
310.0
1,032.4
443-2
Mean . .
305 ±6
325±7
I, i78±45
45i±7
f 9
3
360.7
341-7
126. 7
215-0
74.8
1915-
207.5
219. I
590.9
348.2
10
2
233-9
219.3
72.1
147.2
49- 2
210. 7
224.7
683.6
334.8
II
3
324. 7
299.7
92.4
207- 3
67.8
208. 9
226.3
734.3
327-3
May 22 to Sep-
12
3
311- 3
287.8
81.2
206. 6
64-5
207.3
224. 2
794.8
312.3
tember 18. .
13
3
325-0
310. 3
97-2
213. I
67.9
208. 9
218.8
698.5
318.6
14
3
363- 8
342.8
89.6
253-2
70.9
194.9
206.8
791-5
280. 0
15
3
353- 9
333-8
"4-3
219-5
70. 6
199-5
211. 5
617.7
307.6
I 16
3
368.9
354-2
109. 6
244.6
72.0
195-2
203.3
657-3
294-5
Mean . .
204 ±2
2I7±2
696±i9
3i5±5
482
Journal of Agricultural Research
Vol. VI, No. 13
XablE VI. — Water requirement of Dwarf Blackhull kafir, feterita, and Sudan grass at
Garden City, Kans., in IQIS
6
I
"H.
"0
S
.0
8
u
g
'•3
_c
u
S
>■
Q
1
<u
1
c
"a
0
>
1
T3
C
«
a
•2
"a.
a
a
u
OS
Water requirement based on —
Plant and period
of growth.
.5
1°
>..s
0
p bH
d
'3
0
>
a
ClJ
s
1915-
Kafir, Dwarf
Blackhull,
May 22 to
September
II
[31
32
32,
34
^ 35
3
3
3
2
3
Gm.
265.7
235-4
273.2
179.2
247.1
Gm.
249-7
221.8
257-8
168.8
230.2
Gm.
107.0
88.4
"9-9
71.7
95-1
Gm.
142.7
133-4
137-9
97-1
135- I
Kgm.
56-4
48.3
56-7
37-5
50.0
212.3
205-3
207.3
209.7
202.3
225.9
217.9
220. 2
222. 6
217. 2
527-3
546.8
473-4
524.1
525-8
395-4
362.3
411. 6
387.0
370- I
Mean . .
207 ±2
22I±2
5i9±8
385±6
f 36
37
38
39
I 40
3
3
3
2
3
175-6
204.7
158.8
143- I
182.6
59-6
66.0
55- 7
42.0
52.4
116. 0
138.7
103.1
lOI. I
130.2
42. 6
49-4
41.8
35-5
45-3
Feterita, May
22 to Sep-
tember 6 . . ,
242.7
242. I
263.3
248.2
248.3
715-2
748.4
750.8
845-7
865.2
367-5
365-1
405.6
351-3
348.2
Mean . .
249 ±2
785±24
367±6
41
42
43
5
5
5
'
186.4
173.6
176.4
33-2
28.4
42.5
153-2
145-2
133-9
52-5
50-7
60. 5
Sudan grass.
May 22 to
September
14
281.6
292.5
343-3
1,581-3
I, 788. 3
1,425.2
342.6
349-7
452-3
Mean . .
3o6±i5
i,598±76
38i±28
Dwarf Blackhull kafir was grown only in 191 5. The growing season
for these plants was from May 22 to September 11. The water require-
ment, based on the total dry matter, including the roots, was 207 ±2,
and based on the total dry weight of the aerial portions, was 221 ±2.
The water requirement, based on the production of grain, was 5i9±8
(PI. LXXI, fig. 2).
Feterita was grown in five cans in 191 5. The seed was planted on May
22 and the plants were harvested on September 6. The water require-
ment, based on the total dry matter of the aerial parts, was 249^2,
while the water requirement, based on the seed production was 785 ±24
(PI. LXXI, fig. 3).
Three cans were planted to Sudan grass in 1915. These plants reached
a height of 6 feet during the growing period from May 22 to September
14 (PI. LXXI I, fig. i). The water requirement, based on the dry weight
of the aerial parts, was 306 ±15 and, based on the production of grain,
was 1 598 ±76.
June 26, 1916 Water Requirement of Corn and Sorghum
483
SUMMARY
The water requirement was determined for four varieties of com and
two varieties of sorghum in 1914 and for three varieties of com and five
varieties of sorghum in 191 5.
The plants were grown in large sealed galvanized-iron cans which
contained approximately 1 10 kgm. of soil. The soil had a wilting coeffi-
cient of 13, and under the conditions of the experiment it had a moisture
content of 20 to 21 per cent (dry basis). This moisture content was
kept approximately constant by replacing every 48 hours the water that
had been lost by transpiration.
Three plants of com were grown in each can during both seasons.
Six sorghum plants were grown to each can in 1914, but in 191 5 the
number of plants was reduced to three plants to a can.
The plants were grown in a screened inclosure in order to protect them
from the hailstorms and severe winds that are prevalent in western
Kansas. The rate of evaporation in such a shelter was found to be only
two-thirds as high as under field conditions.
The season of 191 5 was cooler and more humid, and the rate of evapora-
tion much lower than in 191 4. As a consequence the water requirement
of the former year was only about 66 per cent of that of the latter year.
A summary of the water requirement for the two seasons is given in
Table VII.
Table VII. — Summary of tJie water requirement of the varieties of corn and sorghum
grown at Garden City, Kans., in IQ14 and igi^
Plant and period of growth.
1914.
Corn:
Pride of Saline, May 26 to August
Sherrods White Dent, May 26 to
August 22
Hybrid F3 H58, May 26 to August
22
Chinese, May 26 to August 22
Kafir :
BlackhuU, May 26 to September 3
MiLO:
Dwarf, May 26 to August 22
1915-
Corn:
Pride of Saline, May 22 to August
25
Sherrods White Dent, May 22 to
August 18
Hybrid F4 Hjg, May 22 to August
25
Water requirement based on —
Dry matter,
including
roots.
365±i5
305 ± 6
3i9± 5
236± 3
Dry matter,
excluding
roots.
403 ±18
396±l6
355± 8
4i8± 7
32S± 7
340 ± 5
267± 2
299 ± 8
246 ± 6
Grain.
i.iySdb 45
I, 022 ±100
Stem and
leaves.
403 ±18
396±i6
355± 8
4i8± 7
45i± 7
53o±i5
267 ± 2
299 ± 8
246 ± 6
484
Journal of Agricultural Research
Vol. VI, No. 13
Table VII. — Summary of the water requirem-ent of the varieties of corn and sorghum
grown at Garden City, Kans., in igi4 and 1915 — Continued
Water requirement based on-
Plant and pieriod of growth.
Dry matter,
including
roots.
Dry matter,
excluding
roots.
Grain.
Stem and
leaves.
1915-
Kafir:
BlackhuUjMay 22 to September 18.
Dwarf Blackhull, May 22 to Sep-
tember II
204 ± 2
207 ± 2
228± 3
2I7± 2
22I± 2
244± 3
249 ± 2
3o6±is
696± 19
5i9± 8
508 ± 6
785 ± 24
I, 598± 76
z-^s^ s
385 ± 6
469 ± 7
367 ± 6
38i±28
Mao:
Dwarf, May 22 to September 3. . . .
Feterita:
May 22 to September 6
Sudan grass: May 22 to September 14.
Using the water requirement of Blackhull kafir as i , the water require-
ment of the plants grown in 1914 would be as follows: Dwarf milo 1.04,
hybrid com 1.09, Sherrods White Dent corn 1.22, and Pride of Saline
com 1.24. In 1 915, if the water requirement of Blackhull kafir be con-
sidered as I , the water requirement of Dwarf Blackhull kafir would be
1.02; Dwarf milo, 1.12; feterita, 1.14; hybrid corn, 1.17; Pride of Saline
com, 1.23; Sherrods White Dent com, 1.37; and Sudan grass, 1.41.
PLATE LXX
Fig. I. — General view of the screened inclosure and the scale house.
Fig. 2. — Method of moving the cans.
Fig. 3. — General view of the plant shelter and the surrounding country at Garden
City, Kans.
Water Requirement of Corn and Sorghum
Plate LXX
Journal of Agricultural R:
Vol. VI, No. 13
Water Requirement of Corn and Sorghum
Plate LXXI
Journal of Agricultural Research
Vol. VI, No. 13
PLATE LXXI
Fig. I. — Dwarf milo, grown May 22 to vSeptember 3, 1915. Water requirement based
on total dry matter, including roots, 228±3. Based on dry matter, excluding roots,
244±3. Average of 8 cans.
Fig. 2. — Dwarf BlackhuU katir, grown May 22 to September 11, 1915. Water
requirement based on total dry matter, including roots, 207^2. Based on total dry
matter, excluding roots, 221 ±2. Average of 5 cans.
Fig. 3. — Feterita, grown May 22 to September 6, 1915. Water requirement based
on total dry matter, excluding roots, 249±2. Average of 5 cans.
PLATE LXXII
Fig. I. — Sudan grass, grown May 22 to September 14, 1915. Water requirement
based on total dry matter, excluding roots, 3o6±i5. Average of 3 cans.
Fig. 2. — Pride of Saline com, grown May 22 to August 25, 1915. Water requirement
based on total dry matter, including roots, 236±3. Based on total dry matter, ex-
cluding roots, 267 ±2. Average of 10 cans.
Fig. 3. — Blackhull kafir, grown May 22 to September 18, 1915. Water require-
ment based on total dry matter, including roots, 204±2. Based on total dry matter,
excluding roots, 2 17 ±2. Average of 8 cans.
Fig. 4. — Method of sealing the lids with tape and the wax seal around the plants.
Water Requirement of Corn and Sorghunn
Plate LXXII
Journal of Agricultural Research
Vol. VI. No. 13
AVAILABILITY OF MINERAL PHOSPHATES FOR PLANT
NUTRITION 1
By W. L. BuRLisoN,
Associate Professor, Crop Production, Agricultural College, and Associate Chief, Crop
Production, Illinois Agricultural Experiment Station^
INTRODUCTION
Phosphorus is the key to permanent systems of agriculture for a large
portion of the common soils of the com belt. These soils contain, as an
average, 5,000 pounds of nitrogen, 1,200 pounds of phosphorus, and
35,000 pounds of potassium for the surface soil to the depth of 6^ inches.
If the land were producing com at the rate of 100 bushels per acre, the
nitrogen would be sufficient for 50 crops, the phosphorus for 70 crops, and
the potassium for about 1,842 crops. The nitrogen supply can be main-
tained by the growth and judicious management of leguminous crops.
Potassium is present in quantities adequate for many years. With
phosphorus the problem is different. This element can not be gathered
from the soil air by legumes; nor is it one of unlimited supply. When
once removed, phosphorus must be returned to the land in crop residues,
in farm manures, or in commercial fertilizers which contain phosphorus.
Since the introduction of commercial fertilizers, more or less discussion
has been carried on concerning the value of insoluble mineral phosphates
as a source of phosphorus for the nutrition of plants. In Europe (28,
p. 329)^ the highest authorities on agricultural problems have discouraged
the use of insoluble phosphates, while in America scientists and practical
men have disagreed. Investigations which have been conducted on the
use of insoluble minerals are by no means conclusive. Therefore it is
the purpose of the work reported in the following pages to throw more light
on this question, which is of so great economic importance and scientific
significance. The subject matter will be presented according to the fol-
lowing divisions:
I. Review of literature regarding the availability of phosphate minerals.
II. The aviailability of phosphoms in Tennessee brown rock phosphate
for wheat {TrUicum vulgare), oats (Avena saliva), rye (Secale cereale),
barley {Hordeum sativum hexasiichon) , cowpeas (Vigna caijang), soybeans
{Glycine hispida), timothy (Phleum pratense), red clover {Trifolium
praiense), and alfalfa (Medicago saliva).
* This paper is submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in Agronomy in the Graduate School of the University of Illinois, 1915.
* The author wishes to express his appreciation for the suggestions and encouragement tendered by
Dr. C. G. Hopkins and Dr. A. L. Whiting, of the Illinois Experiment Station.
* Reference is made by number to "Literature cited," p. 513-514.
Journal of Agricultural Research, Vol. VI, No. 13
Dept. of Agriculture, Washington, D. C. June 26, 1916
ef , III.-3
37770°— 16 2 (485)
486 Journal of Agricultural Research voi. vi, no. 13
III. A comparative study of the productive powers of six mineral
phosphates for farm crops.
IV. The influence of fermenting dextrose and crop residues on the
availability of phosphorus in finely ground rock phosphate.
V. The influence of the size of particles on the availabiHty of phos-
phorus in mineral phosphates.
REVIEW OF LITERATURE
The availability of mineral phosphates for plant nutrition has been
under investigation at various institutions for more than half a century.
Among the earlier scientists who attempted to determine the availability
of the phosphorus in mineral phosphates was Dyer (4), who found that
undissolved phosphate produced better returns than dissolved phosphate
for swedes and oats. Frear (5) studied the comparative value of various
phosphorus carriers for farm crops. Finely ground bone and reverted
phosphate produced the largest number of mature stalks of corn,
and finely ground bone, the highest yield of ears. Superphosphate
and certain mineral or raw phosphates were put in field trials by
Johnson (9), and for corn, dissolved bone black was superior to others
tested. Bishop (i) grew soybeans in pot cultures and concluded that
concentrated phosphate and acid phosphate were more desirable than
Florida soft rock and iron and aluminum phosphate. Equivalent
amounts of different carriers of phosphate were employed by Hess (7)
in a 4-year rotation of corn, oats, wheat, and grass. Finely ground bone
gave the highest yields of wheat, with raw rock second. Ground bone
was most effective for corn, while for oats insoluble ground bone seemed
to be satisfactory. South Carolina rock was very useful for clover.
Jordan (10) conducted two experiments at the Maine Station with
different forms of phosphate. In the first experiment the minerals were
applied in equal quantities. For the first two years the acid phosphate
gave the highest returns, but later bone meal took the lead. Raw rock
was only about half as productive as the other two. In the second trial
equal money values of phosphates were applied; and the author points
out that, with but one exception, the raw rock gave larger returns than
acid phosphate. The work of Jordan, previously mentioned, was con-
tinued by Merrill (15), who used pure sand cultures in the greenhouse.
Two facts are clear from Merrill's work. First, plants differ widely in
their power to assimilate phosphorus from different phosphates. Second,
turnips and rutabagas gave almost as good results with raw rock phos-
phate as with acid phosphate. Later, at the New York Station, Jordan
(11) continued the work which he had begun at the Maine Station. His
results are in accord with the work previously reported by himself and
Merrill.
june26, i9i6 Mineral PliospJiates and Plant Nutrition 487
In 1890 Goessman (2) outlined what has since become a most extensive
investigation, concerning the availability of phosphate minerals. In
reporting on this work Brooks says (3, p. 104) that —
It is possible to produce profitable crops of most kinds by liberal use of natural
phosphates, and in a long series of years there might be a considerable money saving
in depending, at least in part, upon these rather than upon the higher-priced dissolved
phosphates.
Results from a second series of experiments begun in Massachusetts in
1897, along the same line as that outlined by Goessman, indicate that
phosphatic slag was "exceedingly available for crops, but the Florida
soft phosphate was very inferior. For certain crops, South Carolina
rock gave surprisingly good returns * * *."
Prianishnikov (20) states that lupins and peas have a very marked
ability to obtain phosphorus from natural phosphate, while wheat and
oats must be assisted by the solvent powers of the soil or they can not
produce normal crops. Schloessing (22) concludes from his experiments
that it is not necessary that phosphate should be in a state of solution,
since the roots of plants are able to dissolve the phosphorus compounds
without the intervention of water.
Patterson (18) reports results, based on a study of various phosphates,
which indicate that reverted phosphate gave the highest average yield
for corn, wheat, and hay. South Carolina rock phosphate produced
slightly better yields than bone black, and Florida soft rock phosphate
was quite available for wheat. Wheeler and Adams (30, 31, 32) found
raw phosphate profitable for peas, oats, crimson clover, and Japanese
millet when used on unlimed land ; but for fiat turnips, beets, and cabbage
it gave poor yields. They are of the opinion that rock phosphate is likely
to be most useful when applied to moist soils rich in organic matter,
where legumes, com, and "possibly wheat and oats are to be grown."
Thome (24, 25), of Ohio, in 1897 inaugurated a very extensive study
of the comparative value of raw rock phosphate and acid phosphate used
in conjunction with manure. Where, in computing the yields of com,
wheat, and clover, he took the average of all the unfertilized plots as a
basis for comparison, he reports (24, p. 18) —
By this method of calculation the average increase on Plots 2 [floats plus yard
manure] and 3 [floats plus stall manure] combined is found to be practically the same
as that on Plots 5 [acid phosphate plus yard manure] and 6 [acid phosphate plus stall
manure] combined, but when the larger cost of the acid phosphate is deducted the net
gain is a little greater on Plots 2 and 3 [with raw phosphate].
By another method of computing the increase he obtains results less
favorable to raw phosphate.
Truog (27) has demonstrated rather clearly that fann crops are vari-
able in their ability to secure phosphorus from difTercnt sources. Nine of
the ten crops tested by him made a better growth on aluminum phos-
488 Journal of Agricultural Research voi. vi, no. 13
phate than on calcium phosphate, and "six made better growth on iron
phosphate than on calcium phosphate."
Under the direction of Hopkins (8), the Illinois Experiment Station is
conducting probably the most extensive investigation of any in the
world on the use of rock phosphate. Some of the most interesting
results were obtained from a field near Galesburg, Knox County, 111., on
brown silt loam prairie soil.
Phosphorus applied in fine-ground natural rock phosphate in part as top dressing,
and with no adequate provision for decaying organic matter, paid only 47 per cent on
the investment as an average of the first three years. But it should be kept in mind
that the word investment is here used in its proper sense, for the phosphorus that was
removed in the increase produced was less than 2 per cent of the amoim.t applied,
and that removed in the total crops less than one-third. During the last six years,
however, the phosphorus has paid 130 per cent on the investment, even though two-
thirds of the application remains to positively enrich the soil (8, p. 15).
Newman (16) investigated the use of floats with and without cotton-
seed meal. He found a marked increase in availability where organic
matter was used in conjunction with the mineral phosphate. Later
experiments by Newman and Clayton (17) confirmed the above results.
Lupton (13) continued the work of Newman, but used acid phosphate as
a check on the raw rock phosphate, both with and without organic matter.
His results are also in accord with Newman's earlier experiments. Where
floats were mixed with cottonseed meal and allowed to ferment, the data
seemed to show that the fermentation of the material had very little, if
any, influence on the availability of the phosphate. Pfeiffer and Thur-
man (19) found no beneficial results from composting raw rock phosphate
with decaying organic matter. In Canada (23) fermenting manures
were found to have only slightly solvent action on composted rock
phosphate.
Hartwell and Pember (6) mixed fresh cow manure and floats and
allowed them to ferment. They feel that there was practically no
increase in the availability of phosphorus in the floats. McDowell (14)
also found no increase in the availability of phosphate in finely ground
rock phosphate by composting the mineral with cow and horse manure.
Tottingham and Hoffmann (26), following the same line of investigation
as that which McDowell observed, actually found a decrease in water-
soluble phosphorus, but the results were similar with acid phosphate.
Krober (12) was unable to find any increase in availability of mineral
phosphates by composting with sawdust and allowing fermentation to
proceed. Truog (27) believes that fermented manure has a slightly
solvent action on crude phosphate. He also points out that a uniform
distribution of the phosphate in the soil will give much better results
than that poorly distributed.
Krober (12) shows that the acid-forming bacteria and yeasts are of
great value in rendering some of the phosphorus in insoluble phosphate
June 26, 1916 Mineral Phosphates and Plant Nutriiioii 489
available. He makes the statement that carbon dioxid was more active
than other acids in this respect.
The degree of fineness plays an important part in the availability of
the crude phosphates. Jordan (11) proves this quite conclusively.
He procured better results from the phosphates which were ground to
an impalpable powder. Analysis of the plants showed an increase in
the proportion of dry matter to phosphorus as the size of the particles
decreased. Voelcker (29) in some of the earliest work says that the
efficiency of insoluble calcium phosphate depends upon the minuteness
of division; the finer the particles the more energetic will be its action.
EXPERIMENTAL WORK
MEDIUM FOR PLANT GROWTH
Pure white sand was. used throughout these experiments as a medium
for plant growth. For most of the work this material was leached with
a dilute solution of hydrochloric acid for three days to insure the removal
of plant food. The sand was then washed with distilled water until
there was no trace of acid in the drainage solution. Next it was placed
on clean paper until dry, when it was sifted, in order that foreign particles
might be removed. Samples were collected for a phosphorus determina-
tion from each lot of sand washed, but in no case during the progress of
the study was the slightest trace of phosphorus detected.
POTS
Two sizes of pots were used in this investigation. Wlien it was neces-
sary to grow the crop to maturity, the small glass battery jars, approxi-
mately 6 inches in diameter and 8 inches in height, proved very satis-
factory; but when a grain crop was desired, the 4-gallon stone pots were
more suitable. All jars were supplied with adequate drainage.
For the cultures grown in the winter the pots were covered with a
coat of black paint, but for the summer series a white coat was placed
over the black. The black paint prevented the growth of algge and the
white had a tendency to keep the temperature from becoming excessive
within the jars. This precaution was clearly justified, for upon several
occasions there was a difference of 5° to 10° in temperature between the
black and white pots.
KINDS OF CROPS GROWN
Wheat, oats, rye, barley, timothy, cowpeas, soybeans, clover, and
alfalfa — nine common crops that are cultivated on Illinois farms — were
grown under various treatments for this investigation. High-grade
seed from the previous season's crop was selected for planting, and
in all cases the grains were treated with a solution of formalin to
prevent smut.
490 Journal of Agricultural Research voi. vi, no. 13
In planting the seed special care was exercised in order to obtain a
perfect stand, and in only a few instances was there a failure to get the
proper number of plants for each pot. It seems in keeping with accurate
methods of research to plant more seeds per pot than would be required
for a perfect stand if they all germinated. It is safer to remove the
extra plants than to transplant or reseed, and the plants are more likely
to be uniform if it is possible to make some choice in thinning them down.
An exact record was kept of the number of seeds planted, and all those
which failed to germinate were dug out.
For inoculating the legumes, nodules from the same crop as the plant
to be infected were crushed and placed in i liter of distilled water, and
10 c. c. of this solution were applied to the zone nearest the seed. If the
nodules were not available, 300 gm. of soil from a field where the respec-
tive legumes had been grown were well shaken with 500 c. c. of water,
filtered, and 10 c. c. of this solution were applied in the same manner as
indicated above.
PLANT FOODS
The first appHcation of plant food was made when the crops were
planted, the others at intervals of two weeks. The plant foods were
made up in the following manner:
Nitrogen : Dissolved 80 gm. of ammonium nitrate, 50 gm. of potassium
sulphate, and 20 gm. of magnesium sulphate each in 2,500 c. c. of dis-
tilled water, and o.i gm. of ferric chlorid in 250 c. c. of distilled water.
A standard application of these plant foods was 10 c. c. of each of the
first three and i c. c. of the last diluted as desired. In no case was the
solution applied in a concentrated form.
MOISTURE SUPPLY
Throughout the first period of these experiments, the water content of
the sand was maintained at 14 per cent by weighing the jar each week.
This phase of the details became so burdensome that it was omitted.
The method was not accurate, at least during the latter period of growth,
because of the irregularity in plant development due to different treat-
ments. Some pots gave off more than 10 times the quantities transpired
from others. vSatisfactory results were obtained by watering the pots
when they needed a supply of moisture and no difficulty was experienced
in determining the point where the water content of the sand was below
normal.
Whenever weather conditions would permit, the pot cultures were
placed on trucks and removed to the cage out of doors.
TIME OF HARVESTING AND HANDLING THE CROP
The time of harvest was governed largely by the condition of the
experiment. However, in most instances the same factors which control
the time of harvest in general farm practice held true here. The grain
June 26, 1916
Mineral Phosphates and Plant Nutrition
491
crops developed to full maturity, while the clover and alfalfa were cut
for hay. Cowpeas and soybeans grown during the winter months were
cut for hay, but those planted in the spring produced a seed crop.
Complete data on time of blooming, time of heading, number of plants,
number of stems, and height of plants were collected for a comparison
which might be of value in interpreting results, although such records
will be omitted from, this paper. The total weight of grain and straw,
together with photographs, will suffice for drawing conclusions.
After harvesting the pot cultures, the materials were suspended in
cheesecloth bags from the roof of the greenhouse for a period of two
weeks. This was sufficient time for the product to come to a constant
air-dried condition. Usually two weighings at an interval of two days
were made as a check to insure accurate results.
ANALYSIS
The plants were first cut fine and then ground in a steel mill until the
particles would pass a sieve of 80 meshes to the inch. Next, the materials
were thoroughly mixed and samples taken for analytical purposes.
The method for the determination of phosphorus was essentially the
Pemberton outline, with slight modifications.
Two gm. of the sample ^ were weighed out and moistened with calcium
acetate. The sample was then dried in an electric oven and afterwards
transferred to a muffle and there remained until the product was burned
to a white ash. The ash was taken up with 5 c. c. of nitric acid and
heated on a water bath for several minutes. It was necessary to filter
to remove any silica present. From this point the regular procedure
followed in the volumetric method was observed.
The mineral phosphates used in this investigation represent six types
from different sections of the United States and Canada. The total
phosphorus and the phosphorus soluble in citric acid are reported in
Table I.
Table I. — Total phosphorus and citric-acid-soluble phosphorus in various kinds of rock
phosphate ^
Kinds of phosphate.
Tennessee brownti rock pHosphate . . .
Tennessee blue rock phosphate . . . .
Utah rock phosphate
South Carolina land rock phosphate
Florida soft rock phosphate
Canadian apatite
Phosphorus.
Total.
12.75
13.40
13.81
13-75
13.98
". 75
Soluble in
citric acid.
9.92
10. 29
8.66
6.89
IO-55
5-57
1 Two gm. was satisfactory for straw and hay, but for the grain }i cm. was sufficient.
' Four gm. of each of the miner;;! i)liosph;ites were placed in a i-liter llask aiul then 1,000 c. c. of a 0.2 per
cent solution of citric acid was poured on the ground rock, where it remained for 48 hours with occasional
shaking. Then some of the solution was filtered and 100 c. c. of the filtrate taken for analysis.
492
Journal of Agricultural Research
Vol. VI, No. 13
AVAILABILITY OF THE PHOSPHORUS IN TENNESSEE BROWN ROCK
PHOSPHATE
This series comprises a study of the ability of different crops to secure
phosphorus for growth from Tennessee brown rock phosphate without
the aid of decaying organic matter. The literature indicates rather
clearly that crops differ widely in this respect, but there is but very little
direct information from trials conducted under controlled conditions
where sand was used as a substitute for soil. The suggestion has been
made, also, that there is slight increase in the yield with large applica-
tion of phosphate. The object of the series reported in Tables II to VI
is to present new information on these two important points.
The pots used were the large, glazed 4-gallon jars into which could be
placed 22,000 gm. of sand (PI. LXXIII, LXXIV, LXXV). In this case
the sand was not leached with dilute acid, but was washed for several
days with distilled water. The rock phosphate was ground sufficiently
finetopass through a sieve of 100 meshes to the inch. On March 20, 191 4,
the pots were seeded; and after the plants had made satisfactory growth,
they were thinned to 15 to each jar.
Table II. — Dry matter and phosphorus content of plant products from wheat and oats
SERIES ia; spring wheat harvested on JUNE 29, 1914a
Phos-
phate
added.
Grain.
Straw.
Phosphorus.
Pot
No
Grain.
straw.
Grain.
Straw.
Total
in grain
and
straw.
Percent-
age re-
moved.
I. . .
Gm.
0
0
II
II
22
22
66
66
220
220
Gm.
0
0
I. 0
1.4
4. I
4.0
12.7
12.8
17-5
16.8
Gm.
6.0
6-5
16. 0
19. I
21. 9
20. 6
39-3
35-9
42.9
40.4
Per cent.
Per cent.
Mgm.
Mgm.
Mgm.
2. . .
3---
4...
6...
7...
8...
9...
10. .
0, 260
0.038
3-64
7. 26
10. 90
0. 78
•257
. 029
10.31
6.06
16.37
.58
.240
. 019
30.72
6.64
37-36
.44
•335
. 026
56.20
10.30
66.50
.24
SERIES IB; SIXTY-DAY OATS HARVESTED ON JUNE 7, I914&
II . .
0
0
II
II
22
22
66
66
220
220
0
0
4-7
4-5
7.6
7.2
10. 9
12. I
16.8
14-3
6.0
6. I
10. 0
10.7
13.6
14.0
18.6
15-4
22. 9
20. 0
12. .
13- •
14..
I5--
t6
•035
2. 14
2. 14
0. 184
. 189
.038
.032
8.28
14.36
4. II
4-35
12.38
18. 72
.88
.67
17..
18..
19..
20. .
. 229
.038
27. 71
S85
33-56
.40
•354
•059
50. 62
11.80
62. 42
. 22
o Seed planted in each pot in series lA contained 0.46 per cent of phosphorus. Fifteen seeds contained
1.7 mgm. of phosphorus.
0 Seed planted in each pot in series iB contained 0.35 per cent of phosphorus. Fifteen seeds contained
1.29 mgm. of phosphorus.
June 26. 1916 Mineral Phosphates and Plant Nutrition
493
Table III. — Dry matter and phosphorus content of plant products from timothy and
red-clover hay
SERIES IE; TIMOTHY HARVESTED ON JULY 21, SEPT. 26, AND NOV. 25, I914
Phos-
. phate
added.
Crop.
Phosphorus.
Pot No
First
cutting.
Second
cutting.
Third
cutting.
First
cutting.
Second
cutting.
First
cutting.
Second
cutting.
Total,
two
crops.
Percent-
age re-
moved.
41
Gm.
0
0
II
II
22
22
66
66
220
220
Gm.
0-5
.8
II. 8
12.8
17.2
17. I
27-5
28.0
31.6
27.7
Gm.
I. 0
.6
8.6
7-1
10. 0
9.6
26. 0
25. 6
28.0
30.8
Gm.
1.8
1-5
4-3
4-5
6.0
6.0
7-9
8.0
8.2
8.7
Per cent.
Mgm.
Per (cni.
^Igjn.
^Igm.
/I2
A^
44.
45
46
0. 067
0. 102
II. 52
10. 15
21. 67
0.77
47
48
49
50
a. 126
. 170
37.80
49.98
87.78
•31
SERIES IH; RED-CLOVER HAY HARVESTED ON JULY 20, SEPT. 26, DEC. 25, 19 14
22
22
66
66
220
220
. 2
. I
. I
. I
?,-S
3-4
2.8
3-5
12. 2
8.4
12.8
8.7
23.0
^3-2,
24. I
13-0
36. 5
oio. 0
37- «
14.2
041
169
055
215
72- 55
6-73
78.29
9.967
150.839
36
54
o The phosphorus content of average timothy hay is 0.09 per cent.
6 Attacked hy worms.
c The phosphorus content of average red-clover hay is about 0.21 per cent.
Table IV. — Dry matter and phosphorus content of plant products from, cowpeas and
soybeans
SERIES IP; COWPEAS HARVESTED ON JULY 6, I914
Pot No.
51-
52-
53-
54-
55-
56.
57-
58.
59-
60.
Phos-
phate
added.
Gm.
22
22
66
66
220
220
Gm.
o
<3-3
o. 7
o
1.4
o. 7
11. 7
12. I
12. I
14. I
Straw.
Gm.
2.8
2. 7
4-3
4.9
7.6
6.7
23.8
27.6
22.5
30-9
Phosphorus
Grain.
Per cent.
a. 29;
Per cent.
0.073
• °9;
.09-
128
Mgm.
3-82
55-'o8'
41.86
Straw.
Mgm.
1.97
4.66
7-37
27.92
39-40
Total in
grain and
straw.
Mgm.
1.97
4.66
II. 19
62.99
81.26
Percent-
age re-
moved.
O. 40
■75
29
o Seed planted in each pot contained 0.434 per cent of phosphorus. Fifteen cowpea seeds coataincd 11.7
mgm. of phosphorus.
494
Journal of Agricultural Research
Vol. VI, No. ij
Table IV. — Dry matter and phosphorus content of plant products from coupeas and
soybeans — Continued .
SERIES IG; SOYBEANS HARVESTED ON JUNE lO, I914
Pot No.
61.
62.
63-
64.
65-
66.
67.
68.
69.
70.
Grain.
Straw.
Phosphorus.
Phos-
phate
added.
Grain.
Straw.
Grain.
Straw.
Total in
grain and
straw.
Percent,
age re-
moved.
Gm..
Gin.
I. 0
1. 0
2. 0
2.8
3-5
2.4
2.9
3-4
4-7
4. 2
Gm.
9.0
8.3
9.2
10. 2
13-5
10.8
14.9
15-3
IS- 2
13-4
Per cent.
Percent.
Mgtn.
Mgm.
Mgm.
0
0.360
0. 058
3.60
A- 11
8.37
II
•359
•045
10. 06
4-54
14. 60
I. 04
22
66
.448
. 062
10. 76
6.64
17.40
.62
66
220
•449
. 061
i5^25
9^38
24.63
.29
220
a. 448
.088
18.83
11.83
30. 66
. II
"■ Seed planted in each pot contained 0.6 per cent of phosphorus. Fifteen soybean seeds contained 11.9
mgm. of phosphorus.
Table V. — Dry matter and phosphorus content of plant products from, alfalfa harvested
on fune 4, fuly 18, Sept. 26, and Nov. 11, 1914 — series ll
Phos-
phate
added.
First
cut-
ting.
Second
cut-
ting.
Third
cut-
ting.
Fourth
cut-
ting.
Phosphorus. a
Pot No.
First
cut-
ting.
Sec-
ond
cut-
ting.
Third
cut-
ting.
First
cut-
ting.
Sec-
ond
cut-
ting.
Third
cut-
ting.
Total,
three
cut-
tings.
Per-
centage
re-
moved
from
pots.
81
Gm.
0
0
II
II
22
22
66
66
220
220
Gm.
0-3
•3
5-5
7-1
13.0
II. 0
13^6
13-6
18.6
17- 1
Gm.
0. 2
. 2
8.9
9.0
13.0
II. 0
13^9
12.8
19.9
IS- 6
Gm.
o-S
.6
13.0
12.7
" 10. 0
18.6
16.5
20. 0
16. 0
Gm.
0.4
•4
6.0
6.5
S-o
S^S
10. 2
10. I
10.7
10. 0
Per ct.
Per cl.
Per ct.
Mgm.
Mgm.
Mgm.
Mgm.
Per ct.
82
83
84
8s
86
0.17
0. 19
0. 18
22
2S
22
69.9
2.49
87
88
89
90
. lO
.26
.28
17
SI
SS
124.6
•44
a Alfalfa hay contains 0.0172 per cent of phosphorus. ^ Attacked by worms.
Table VI. — Dry m,atter produced by spring rye and barley — series iC
Rye.
Pot No.
Barley.
Pot No.
Phosphate
added.
Grain.
Straw.
Phosphate
added.
Grain.
Straw.
21
22
23
24
25
26
27
28
29
30
Gm.
0
0
II
II
22
22
66
66
220
220
C)
oooooooooo
1.9
1.8
10. 9
11. 6
22. 2
21. 0
37^o
38.2
45^i
42.0
31
32
33
34
35
36
37
38
39
40
Gm.
0
0
II
IX
22
22
66
66
220
220
Gm.
0
0
2.6
1-3
2. 0
8^5
5-0
14-2
20.9
16.5
Gm.
8
7
20
16
17
22
23
23
34
28
7
2
3
4
8
3
3
4
8
0
June 26. 1916 Mineral Phosphates and Plant Nutrition 495
Probably the most striking point shown by Tables II to VI is the gradual
increase in the yield of both grain and straw from wheat, oats, and barley
and in the hay from rye and timothy. In all cases larger applications of
phosphorus gave higher returns, though not always in the same degree.
The grain yield of wheat is especially interesting. Eleven gm. of
Tennessee brown rock phosphate produced 1.2 gm. of grain, while double
this application produced 4.05 gm., or almost four times the yields from
the light-application pots. Pots 7 and 8, which received 6 times as
much phosQ^orus as pots 3 and 4, produced approximately 1 1 times as
much wheat. Pots 9 and 10 received 20 times as much phosphorus as
pots 3 and 4, but gave in return only about 14 times as much grain.
Scarcely more evidence is necessary to show that wheat is able to take
its phosphorus supply from Tennessee brown rock phosphate. It is also
evident that the rate of yield is to a certain degree dependent upon the
rate of application of the fertilizer. In the case of the heavy application,
there were indications that the size of the pot was a limiting factor.
Oats responded more uniformly to the phosphate application than did
wheat. The average yield of grain for pots 13 and 14 was 4.6 gm.; pots
15 and 16, which received double the quantity of phosphorus supplied
to pots 13 and 14, yielded less than twice the amount of grain. For the
highest application there is still a larger difference in the phosphorus
applied and the crop produced, due, no doubt, to the limited size of the
pot. The yield of straw followed about the same rate of increase as the
grain.
Spring rye was not able to endure the heat of the summer days, and
at the time of harvest growth had almost ceased without producing a
single grain. The hay yield shows a gradual increase in dry matter
as the application of phosphate rock was increased.
The yields from barley are not so consistent as those reported for
wheat and oats. However, in all probability the same uniformity
would have resulted had the crop not been attacked by smut. Although
pots 34, 35, 37, 38, and 39 were badly affected, there was a gradual
increase in grain and straw as the application of phosphorus increased.
A yield of even 18 bushels for barley is not altogether unsatisfactory.
The data on timothy are no less interesting than .those on the growth
of the cereals, because of the opportunity to study the yields of the
various cuttings. Timothy displays the same tendency to produce
larger returns for greater quantities of phosphorus applied to the sand.
For each pot there was a gradual decrease from the first to the last
cutting, although the drop was less abrupt between the first and second
than between the third and fourth cuttings.
Contrary to what might be expected the legumes respond to phosphate
treatment no better than do the cereals. Perhaps on the whole this
latter group produced larger gains than the former.
496 Journal of Agricultural Research voi. \^, no. 13
The results from the cowpeas show some points of particular interest.
There was scarcely any seed produced for the pots to which 1 1 and 22 gm.
of raw rock had been applied, but there was a decided increase for the
pots which received 66-gm. applications. The next treatment, which
was 220 gm. per pot, showed a slight increase, approximately 3 bushels
per acre. For the cowpea hay the results are very similar to the seed
yield. There is not a very marked increase in the hay production until
the larger applications are made. The pots which received 66 gm.
produced nearly as much hay as the pots which received 22c^gm. of rock
phosphate.
Cowpeas do not give results that correspond with those from soy-
beans. In the first place, the no-treatment pots produced a significant
quantity of soy-bean seed, the yield on the acre basis amounting to 2.64
bushels, while the returns from the pots receiving the largest applica-
tion just about quadrupled those from the former. The ratios for the
yields of hay are about the same as for the grain. The yields for both
seed and hay in the case of soybeans are unsatisfactory, which is not
true of the cowpeas. It would seem that the latter legume utilizes rock
phosphate better than soybeans.
To the practical agriculturist the returns from red clover will prove of
considerable interest. It will be observed that the lowest treatment, 1 1
gm. per pot, produced hay at the rate of 772 pounds per acre. With
double the application a little less than the former yield is recorded.
When the lowest application is increased to six times the original amount,
the yield of hay is increased about three times. The largest application,
which was 20 times that of the lowest, produced practically 10 times as
much hay as the first treatment. The above figures are for the first
cutting only.
For the second harvest the relative yields of the 22- and 66-gm. treat-
ments are more satisfactory than for the first cutting. It will be observed
that the yield of the pots with 11 -gm. applications and those with the
220-gm. applications hold the same relation for the second cutting as for
the first. No direct comparison for the third cutting should be made,
because pot 79, just previous to cutting, was attacked during a single
night by a large cutworm which did considerable damage to the growing
crop. It is true, however, thai there had not been as much difference
in the growth on the high-treatment pot as had been observed earlier in
the season. The total yield for three cuttings for the heaviest application
is large, but it can hardly be said that the pots which received 22 gm.
of rock phosphate produced unprofitable yields.
Because of its extensive root system alfalfa would be expected to
produce greater yields than clover. However, the difference in this
experiment is not so marked. From four cuttings of alfalfa the yield of
hay from the lowest treatment was 5,451 pounds, as against 1,819 pounds
June 26, 1916 Mineral Phosphates and Plant Nutrition 497
of clover from the same treatment for three cuttings. For the next
higher treatment the comparison is 6,426 pounds of alfalfa to 4,674 pounds
of clover. The yields are approximately the same for the third applica-
tion, but for the heavy treatment the clover almost doubles the yield
from the alfalfa. Special attention is called to Plates LXXIII, LXXIV,
and LXXV.
In drawing conclusions from an investigation of this kind the actual
growth of the plant must be regarded as a most significant factor. How-
ever, an analytical study of the crops harv^ested can not fail to be of great
value. Since phosphorus is the element with which this paper chiefly
concerns itself, quantitative determinations were confined to that
substance.
The determinations show that in practically all cases phosphorus is the
limiting element in production. In every instance the dry matter
increased as the phosphorus content of the pot was increased; also the
quantity of phosphorus assimilated increased as the dry matter increased.
The percentage of phosphorus in the plant in the majority of cases
increased as the application of raw rock grew larger. This is especially
noticeable in the hay crop. The most notable exceptions were observed
in wheat and oat straw. There is no definite relation in the quantity of
phosphorus applied and the percentage assimilated by the crop. There
was a slight tendency in the grain for the percentage removed to decrease
as the application was increased, but for the legumes this ratio does not
hold. As high as 2.49 per cent of the phosphorus supplied in raw rock
phosphate was removed in one season's growth of alfalfa.
COMPARATIVE STUDY OF THE PRODUCTIVE POWERS OF SIX MINERAL
PHOSPHATES
The results from Tennessee brown rock phosphate proved so interesting
that it was planned to determine the comparative value of mineral phos-
phates from the various mines of America, For this purpose Tennessee
brown rock phosphate, Tennessee blue rock phosphate. South Carolina
land rock phosphate, Utah rock phosphate, Canadian apatite, and Florida
soft rock phosphate were selected.
The materials were ground so that all particles would pass through a
sieve with 100 meshes to the inch and were applied in quantities which
contained equal amounts of phosphorus for a given set of pots. Clover,
oats, and cowpeas were grown with these different phosphates.
Because of limited space the small battery jars into which could be
placed conveniently 4,800 gm. were selected for this rather extensive
trial. Without crowding, eight plants per pot could be grown
(PI. IvXXVI). Table VH gives the quantity of the phosphate applied
and the yields of the crops in question. The planting was done on
October 3, 191 4, and the crops of clover were harvested on March 5
and April 9, 1915, while the oats were cut on February 5, 191 5.
498
Journal of Agricultural Research
Vol. VI. No. 13
Table VII. — Dry viatter produced by different kinds of mineral phosphates — series 2
Kind of phosphate added.
None
Do
Tennessee bro^vn rock . .
Do
Do
Do
Do
Do
Canadian apatite
Do
Do
Do
Do
Do
South Carolina land rock
Do
Do
Do
Do
Do
Utah rock
Do
Do
Do
Do
Do
Tennessee blue rock
Do
Do
Do
Do
Do
Florida soft rock
Do
Do
Do
Do
Do
Red clover.
19
20
22
2Z
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
Quantity
of phos-
phate.
Gtn.
O
I. 81
I. 81
3.62
3. 62
10.86
10.86
I. 81
I. 81
3.62
3. 62
10.86
10.86
1.68
3-3^
3-3^
10. 07
10. 07
I. 67
1.67
3-34
3-34
10. 01
10. 01
I. 72
I. 72
3-44
3-44
10.33
^0.33
1.65
1.65
3-3°
3-3°
9.90
9.90
First
cutting.
Gm.
2.7
I. o
4-3
4-3
7- I
6.1
3-3
■9
I. 2
1. I
o
1.8
3-4
3-3
o
2. 2
2.9
•9
•3
7.0
5-6
5-0
3-5
6. I
4.4
5-8
5-8
Second
cutting.
Gm.
O
1.9
•5
3-3
3-8
4.8
4.9
1-3
1.8
o
4.9
5-0
3-0
o
3-7
4-7
I- 5
1.8
6.7
7.6
5- I
5- I
5-0
5-9
Sixty-Day oats.
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
Phos-
phate
added.
Gm.
O
O
3.62
3. 62
10.86
10.86
I. 81
3.62
3. 62
10.86
10.86
1.68
1.68
3-3(>
3-36
10. 07
10. 07
1. 67
1.67
3-38
3-38
10. 01
10. 01
I. 72
1.72
3-44
3-44
IO-33
IO-33
1.65
1.65
3-3°
3-30
9.90
9.90
Yield of
straw.
Gm.
I. I
I. O
4.4
3-3
5-7
4. 2
6.9
6.7
I. I
I. I
1. I
1-3
2-3
1.4
2. 2
2. o
1-3
2. 2
3-4
2. o
2. o
1.4
I. 2
1.9
I. 2
1-5
1.6
1.4
30
3-3
3-3
3-4
1-5
I. 2
1.9
1.9
3-0
3-2
In the foregoing series the greatest contrast is shown by the clover in its
response to Tennessee brown rock phosphate and Canadian apatite. With
brown rock the yield advanced rapidly with each increase in the amount
of phosphate applied; but apatite, even with repeated plantings, failed to
produce growth. South Carolina land rock phosphate proved better than
apatite, but the growth for this treatment was very irregular. Utah
phosphate excelled the South Carolina land rock phosphate. Except for
June 26. 1916 Mineral Phosphates and Plant Nutrition
499
the lowest treatment, Tennessee blue phosphate gave fairly satisfactory
yields. Florida phosphate for the three treatments gave almost as good
returns as the Tennessee brown rock. Attention is called to the com-
parative yields of the Florida rock for the lowest and highest treatments.
In this case a smaller quantity of the soft phosphate gave almost as large
returns as the greater supply.
Table VIII.-
-Dry matter produced by different kinds of mineral phosphate in red clover
and Sixty-Day oats — series 3
Kind of phosphate added.
None
Do
Tennessee brown rock .
Do
Do
Do
Do
Do
Do
Do
Canadian apatite
Do
Do
Do
Do
Do
Do
Do
Utah rock
Do
Do
Do
Do
Do
Do
Do
South Carolina land rock
Do
Do
Do
Do
Do
Do
Do
Tennessee blue rock . . .
Do
Do
Do
Do
Do
Do
Do
Red clover.
Pot
No.
13
14
15
16
17
18
19
20
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Phos-
phate.
Gni.
O
O
II
22
22
66
66
220
220
22
22
66
66
220
220
10. II
10. II
20. 22
20. 22
60. 66
60.66
202. 20
202. 20
10. 16
10. 16
20.32
20.32
60. 96
60. 96
203. 20
203. 20
10. 42
10. 42
2G. 84
20. 84
62. 52
62. 52
208. 40
208. 40
Yield
of hay.
Gm.
O. I
I. c
1. o
2. O
2. O
3-9
4. 1
3-8
4.0
Sixty-Day oats.
Pot
No.
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
79
80
81
82
83
Phos-
phate.
Gm.
o
22
22
66
66
220
220
II
II
22
22
66
66
220
220
10. II
10. II
20. 22
20. 22
60.66
60.66
202. 20
202. 20
10. 16
10. 16
20.32
20.32
60. 96
60. 96
203. 20
203. 20
10. 42
10. 42
20. 84
20. 84
62. 52
62. 52
208. 40
208 40
Grain.
Gm.
1.4
3-2
3-0
5-3
5-2
3-8
5-0
Straw.
Gm.
500
Journal of Agricultural Research
Vol. VI, No. 13
Under greenhouse conditions it was extremely difficult to secure a seed
crop of oats during the winter months; hence, the differences of produc-
tive power of the various phosphates must be measured by the yields of
straw. In a general way the results obtained in this manner are in har-
mony with those reported for clover. The brown rock excelled the other
phosphates in the production of hay; blue phosphate ranks second; and
where apatite was applied it will be observed that the plants made very
little growth. Plate LXXVI indicates greater difference in the growth
of clover than the dry weight of the top.
The above data indicate that there was an increase in yield as the
quantity of phosphorus was increased. The question naturally arises
as to the point at which larger applications of rock phosphate failed to
produce greater returns. In order to answer this query, the following
results are inserted (Table VIII) : The lowest treatment in the table is
about the same as the highest application in series two. This set of pot
cultures was planted on August 27, 1914, and harvested on December
4, 1914.
By referring to the clover in Table VIII, a comparison of the yields
shows nothing particularly in favor of excessive quantities of rock phos-
phate. One point, however, is of interest, and that is that the oats pro-
duced a seed crop on the land with the heavy application of brown rock.
The hay on the other pots was scarcely more than could be produced by
the phosphorus in the seeds planted.
Table IX. — Dry matter produced by various kinds of mineral phosphates in cowpeas,
series j
Kind of phosphate
added.
None
Do
Tennessee
brovvn rock . .
Do
Do
Do
Do
Do
Do
Do
Canadian apa-
tite
Do
Do
Do
Do
Do
Do
Do
Utah rock
Do
Do
Pot
No.
19
20
Phos-
phate.
Gm.
O
O
II
22
22
66
66
220
220
II
II
22
22
66
66
220
220
10. II
10. II
20. 22
Yield of
hay.
Gm.
0,9
I.
4-
4-
II.
10.
17-
13-
Kind of phosphate
added.
Utah rock
Do
Do
Do
Do
South Carolina
land rock . . .
Do
Do
Do
Do
Do
Do
Do
Tennessee blue
rock
Do
Do
Do
Do
Do
Do
Do
Pot
No.
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Phos-
phate.
Gm.
20. 22
60.66
60.66
202. 20
202. 20
10. 16
10. 16
20.32
20.32
60. 96
60. 96
203. 20
203. 20
ID. 42
10. 42
20. 84
20. 84
62. 52
62.52
208. 40
208. 40
Yield of
hay.
Gm.
2. 2
1.4
1.4
•9
•9
4.0
3-5
4-7
S-o
3-9
4.2
1.9
2. I
June 26, 1916
Mineral Phosphates and Plant Nutrition
501
Soon after the clover was harvested in series 3, these pots v\^ere seeded
to cowpeas. Cowpeas were planted to determine the ability of this
legume to utilize the phosphorus contained in mineral phosphates. The
cultures were seeded January 24, 191 4, and harvested April 5, 1914
(Table IX).
The results secured for series 4 are in accord with those from the
clover and oats grown on the pots with large applications. The pots to
which had been added brown rock phosphate produced a good return of
cowpea hay after having given satisfactory yields of clover.
The data presented in the previous tables show conclusively that cer-
tain species of plants have the power to obtain phosphorus from brown
rock phosphate, but how they acquire this element is the problem of
vital concern. Do they secure their phosphorus without indirect aid
and what influence do other plant foods applied in a soluble form exert
on the phosphorus compounds?
It will be remembered that the sand cultures were maintained at a
moisture content of 14 per cent. The plant food application, the infusion,
and the water added when the seeds were planted constituted the first
moisture supply; or, in other words, all these solutions brought the
water content up to 14 per cent. In most of the cases five applications
of plant food were sufficient to produce a crop of clover or oats.
To estimate the influence of water and plant-food solutions on the
solubility of the phosphates, quantities of raw rock which correspond to
the smallest application (1.81 gm.), soluble plant food equivalent to
five applications, and water sufficient to bring the supply of the solu-
tion to the same amount that was necessary to bring the moisture con-
tent to 14 per cent, or 672 c. c, were placed in a i -liter flask and shaken
each day for three months. The soluble phosphorus was then deter-
mined with the results shown in Table X.
Table X.
-The influence of soluble plant foods on the solubility of the phosphorus in
mineral phosphates
Material applied
and pot No.
Water only:
I
Water and solu-
ble plant food :
Kind of phosphate.
Tennessee brown rock ,
.do.
Tennessee blue rock ,
Canadian apatite
South Carolina land rock .
Utah rock
Florida soft rock
37770'
-16—3
Quantity of
Phosphorus
phosphate.
dissolved.
Gm.
Mgm.
I. 81
0. 25
I. 81
■33
1.72
•05
I. 81
•05
1.68
.056
1.67
• 14
1.65
.28
502 Journal of Agricultural Research voi. vi, no. 13
The solutions dissolve very little of the phosphorus from the insoluble
phosphate.
Brown rock phosphate and Florida soft rock phosphate gave the best
results vath clover, but the former was very much better suited for oats
than the latter. There is a slight indication that phosphates which are
more soluble in water are more easily assimilated by plants.
THE INFLUENCE OF FERMENTING DEXTROSE AND CROP RESIDUES
ON THE AVAILABILITY OF PHOSPHORUS IN FINELY GROUND ROCK
PHOSPHATE
Though the data are not conclusive, a large number of field experiments
conducted in America show that raw phosphate, when applied in con-
junction with organic matter, produces very appreciable increases in
crop yields. The work which follows is an effort to determine the in-
fluence of decaying substances on the availability of the phosphorus in
crude phosphate rock. Dextrose was employed because it ferments
rapidly under greenhouse conditions. Crop residues are also included
in this section, but owing to the slow growth of crops through the winter
months it will not be possible to do more than to make a preliminary
report on this phase of the problem.
Throughout the study included in this division, the glass battery jars
were utilized with success and the same quantity of sand employed as
previously noted — namely, 4,800 gra. per pot. For all the cultures grown
in the dextrose section, the sand was leached with dilute hydrochloric
acid.
The first series reported below was outlined primarily to secure data
on the value of rock phosphate alone and in conjunction with dextrose
for rye and clover. It will be observed that the applications of the rock
phosphate and the dextrose were made on the percentage basis. In
order to hasten fermentation, an infusion from a rich soil was a part of
the treatment. This series was planted on April 12, 191 3, and harvested
on August 19, 1 91 3.
Since dextrose applied at the rate of 48 gm. per pot injured the rye and
destroyed the clover, a point of importance to decide was what quantity
v/ould not injure plant development, but would assist in the liberation of
phosphorus. With this point in mind, series 6 was planned. The
planting was done on June 21, 191 3, and the crop harvested on December
I, 1913. (See Table XI.)
The dextrose in series 5 had no beneficial influence. If the average of
pots 7, 8, and 9 is compared with the results from either set of pots i, 2,
and 3 or pots 4, 5, and 6, it will be evident that the dextrose is harmful.
Clover failed to make growth where the dextrose was added, but did
fairly well on the pots which received rock phosphate alone.
June 26, 1916
Mineral Phosphates and Plant Nutrition
503
The data in Table XI show that dextrose .fails to be of any particular
advantage for rendering phosphorus available for the growth of rye and
clover. Even small quantities of this material killed clover.
Table XI. — Dry matter prodticed by Tennessee brown rock phosphate and dextrose in
growing spring rye and red clover
SERIES 5
Rye.
Red clover.
Pot
No.
Phosphate
added.
Dextrose
added.
Infusion
added.
Hay
yield.
Pot
No.
Phosphate
added.
Dextrose
added.
Infusion
added
Hay
yield.
I
2
3
4
5
6
7
8
9
Gm.
48
48
48
48
48
48
48
48
48
Gm.
48
48
48
48
48
48
0
0
0
C.c.
0
0
0
20
20
20
0
0
0
Gm.
22. 9
33-9
3^-3
35-3
32.3
22. 0
27. 2
32-5
36.0
19...
20. . .
21. . .
22 . . .
23...
24.. .
Gm.
48
48
48
48
48
48
Gm.
48
48
48
0
0
0
C.c.
20
20
20
0
0
0
Gm.
Oo
ao
a 0
3-0
3-8
2.9
SERIES 6
3-
4-
5-
6.
7-
8.
9-
10.
48
48
48
48
48
48
48
48
48
48
4.8
4.8
14.4
14.4
24. 0
24. 0
48.0
48.0
0
0
20
20
20
20
20
20
20
20
20
20
14
17
17
16
16
20
14
II
18
17
6
9
5
0
8
7
S
6
7
7
17..
18..
19..
20. .
21. .
22. .
23- •
24..
25--
26..
27..
28..
48
48
48
48
48
48
48
48
48
48
0
0
4.8
4.8
14.4
14.4
24. 0
24. 0
48.0
48.0
0
0
0
0
20
20
20
20
20
20
20
20
20
20
20
20
ft 20
&20
^20
*20
^20
^20
b 20
b 20
3-
2.
o
o
o The clover in pots 19, 20, and 21 was dead on June 29. 1913.
6 The clover in pots 17 to 24, inclusive, was dead in less than i month after planting.
Rye and clover were replaced in series 7 (Table XII) by cowpeas, with
the feeling that the latter crop might respond more readily to various
treatments (PI. LXXVII). The cowpeas v/ere planted on July 4, 191 3,
and harv^ested on October 2, 191 3.
The cowpeas grown in series 7 show clearly that so small a quantity of
dextrose as 4.8 per cent was injurious to plant growth. Where dextrose
was applied, smaller quantities of phosphorus were assimilated, due, no
doubt, to the injury of the plant by the acids formed from decomposing
dextrose. However, the percentage of phosphorus increased as the
quantity of the fermentable substance was increased.
504
Journal of Agricultural Research
Vol. VI, No. 13
Table XII. — Dry matter and phosphorus content of plant products of cowpea'i from pot
cultures, with tire addition of Tennessee brown rock phosphate and dextrose; series 7
Phosphate
added.
Dextrose
added.
Infusion
added.
Hay
yield.
Phosphorus content.
Pot No.
Hay.
Hay
yield.
Removed
from pot.
I
Gm.
48
48
48
48
48
48
48
48
0
0
48
48
0
0
Gm.
4.8
4.8
14. 4
14. 4
24. 0
24. 0
48. 0
48. 0
48. 0
48.0
0
0
0
0
C.c.
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Gm.
25. 0
22.3
II. 9
8.8
ao
a 0
a 0
ao
rtO
a 0
29.9
29. I
4.0
3-0
Per cent.
0.319
Mom..
79-75
Per cent.
1-30
3
4
.381
33-33
•54
5
6
7
8
9
12
.286
85-51
X. 40
13
b. 128
3-84
14
<» The plants on pots s to 10, inclusive, were all dead by Aug. 9, 1913.
b Five cowpea seeds were planted in each pot. These contained 3.92 tngm. of phosphorus.
Rye, clover, and cowpeas failed to thrive wherever the smallest quan-
tity of dextrose was present. There is but little doubt that this destruc-
tive influence is due to the decomposition of dextrose. If this conclusion
be true, a liberal use of calcium carbonate should neutralize the acids
developed, and a normal growth of the plants should result. Series 8
(Table XIII) was designed for determining what influence calcium car-
bonate would have in stimulating plant growth by producing an alkaline
medium and to ascertain Avhether calcium served as a food.
Table XIII. — Dry matter produced in spring rye by Tennessee brown rock phosphate with
the addition of dextrose and calcium carbonate — series 8
p
N
ot Phosphate
0. added.
Dextrose
added."
Calcium
carbonate
added.
Hay i
yield.
Pot
No.
Phosphate
added.
Dextrose
added."
Calcium
carbonate
added.
Hay
yield.
Gm.
Gm.
Gm.
Gm.
Gm.
Gin.
Gm.
Gin.
I
48
48
10
7.2
13- •
0
0
0
°-l
2
48
48
10
9-3
14..
0
0
0
. 6
^
48
48
0
5-2
15- -
48
0
0
11. 2
4
48
48
0
3-5
16..
48
0
0
12. I
5
48
48
10
7.0
17..
48
4.8
10
10.7
6
43
48
10
7.0
18..
48
4.8
10
10. 0
7
0
0
0
. I
19..
48
4.8
0
9.6
8
0
0
0
. I
20. .
48
4.8
0
9.0
0
48
48
0
7.8
21 . .
0
0
0
•4
10
48
48
0
8.9
22 . .
0
0
0
• 3
II
0
0
0
. I
23--
48
48
0
6. q
12
0
0
0
. I
24. .
48
48
0
6.0
a Pots s and 6 were leached and the leachings placed on pots 7 and 8. Pots 9 and 10 were leached and
the leachings placed on pots 11 and 12. Pots 13 and 14 received all plant food but phosphorus. Pots 21
and 22 received nothing. Pots 23 and 24 were leached and drainage water taken for analytical purposes.
June 26, 1916
Mineral Phosphates and Plant Nutrition
505
Series 8 shows that dextrose in conjunction with calcium carbonate
did not give as good results as raw rock phosphate alone, and that
10 gm. of calcium carbonate was not sufficient to nullify the harmful
influence of the dextrose.
Series 9 (Table XIV), wnich follows, is just the same as series 8
except that cowpeas are substituted for rye, the object being to deter-
mine the relative response of rye and cowpeas to the different treatments.
Table XIV. — Dry matter produced in cowpeas by Tennessee brown rock phosphate with
the addition of dextrose and calcium carbonate — series Q
Pot
No.
Phosphate
added.
Dextrose
added.
Calcium
carbonate
added.
Hay
yield.
Pot
No.
Phosphate
added.
Dextrose
added.
Calcium
carbonate
added.
Hay
yield.
Gm.
Gm.
Gm.
Gm.
Gm.
Gm.
Gm..
Gm.
iF..
48
48
10
5-5
13F..
0
0
0
3-3
2F..
48
48
10
5-3
14F..
0
0
0
3-°
.3i^^•
48
48
0
5-2
isF..
48
0
0
12.4
4K..
48
48
0
4.0
16F..
48
0
0
14- 5
s^'- ■
48
0 48
10
5-3
17F..
48
4.8
10
8.1
6F..
48
48
10
6.1
18F..
48
4.8
10
10. I
7F..
0
0
0
3-8
19F..
48
4.8
0
II. I
8F..
0
0
0
3-8
20F..
48
4.8
0
II. 8
qF..
48
48
0
4.2
21F..
0
0
0
3-9
loF. .
48
48
0
3-0
22F..
0
0
0
4-S
iiF..
0
0
0
3-1
23F..
48
48
0
4.0
12F. .
0
0
0
2-3
24F..
48
48
0
4.0
a See note to Table XIII.
Series 9 shows that brown rock phosphate, dextrose, and a limited
supply of calcium carbonate failed to give as good results with cowpeas
as raw phosphate alone. For further comparison see Plate LXXVIII.
Thus far it has not seemed necessary to use calcium carbonate alone,
because it was thought that the plants would get enough calcium, for
full growth from the phosphate, however, in order to avoid criticism at
this point calcium carbonate was added to certain pots in the following
series. The quantity of this compound was increased to 48 gm. per pot,
which is almost five times as much as the application in the preceding
series.
By making a comparison of the pots which received raw rock phos-
phate alone and those which received raw rock and calcium carbonate
very little difference in the yield is observed, only 0.6 gm. more in favor
of the addition of the lime compound. There is no strong evidence in
Table XV to show that the omission of calcium was a mistake. Where
lime was applied with rock phosphate and dextrose, the injury by dex-
trose reported earlier was nullified by the application of lime (PI.
LXXIX and LXXX).
Attention is called to the percentage of phosphorus in the cowpea hay
grown in the pots which received soluble phosphorus.
5o6
Journal oj Agricultural Research
Vol. VI, No. 13
Table XV. — Dry matter produced in cowpeas by Tennessee brown rock phosphate with
the addition of dextrose and calcium carbonate — series lo
Phosphate
added.
Dextrose
added.
Calcium
added.
Hay yield.
Phosphate content.
Pot No.
Hay.
Percentage
removed
from pot.
iG
Gm.
0
0
0
0
48
48
48
48
48
48
48
48
0
0
48
48
(a)
(a)
0
0
648
48
48
48
Gm.
0
0
0
0
0
0
0
0
0
0
48
48
48
48
48
48
0
0
0
0
0
0
48
48
Gm.
10
10
0
0
0
0
0
0
10
10
0
0
48
48
10
10
0
0
0
0
0
0
48
48
Gm.
3-0
3-5
2.8
3-0
6. I
6.3
5-3
6.8
7-4
6.2
4.9
4.6
3-2
3-3
6.8
7.0
4.9
4.8
3-0
3-0
7.0
6.8
6.9
6.3
Per cent.
0. Ill
Mgm.
3-37
2G
3G...
4G...
5G...
6G...
7G...
8G...
9G...
loG...
iiG
.119
3-57
. 246
15-55
0.25
.236
.156
•239
16.05
II. 52
II. 71
•52
.19
.19
12G
13G...
14G...
15G...
16G...
17G...
18G
•054
1-73
. 122
8.54
.14
.660
31.68
19G...
20G...
21G
. 104
5-19
22G...
23G...
24G...
.388
24. 06
•39
•115
7-25
. 12
a Soluble phosphate.
b In pots 21 and 22 potassium chlorid was substituted for potassium sulphate.
Under the conditions of this experiment", fermenting dextrose was a
failure in bringing about the liberation of phosphorus. Since the use
of crop residues is a common farm practice for supplying organic matter,
which is said to aid in the liberation of phosphorus, the series next re-
ported was planned with timothy hay and clover substitutes for dextrose.
Timothy and clover cultures on which data are reported in Table III
are used for this phase of the problem. Of the duplicate pots the hay
from one was taken for analytical study, while the product of the other
was ground and returned as organic matter. This series (Table XVI)
shows the original treatment with the quantity of air-dried hay turned
under. The contents of the pots to which organic matter was added
were turned out and the ground material thoroughly incorporated with
the sand on December 3, 1914. On January 23, 1915, the pots were
planted to the respective crops. They were harvested on April 17, 191 5.
The organic matter with phosphate in the above series gave larger
returns in most cases than where the phosphate was alone. This increase
is probably due to the liberation of phosphorus by the decaying residues
or the organic phosphorus in the crop residues themselves.
June 26, 1916 Mineral Phosphates and Plant Nutrition
507
Table XVI. — Dry matter produced in timothy and red clover by Tennessee brown rock
phosphate and crop residues — series 11
Timothy.
Pot No.
41a
42.
43-
44.
45-
46,,
47-
48.,
49-
so-
Phos-
phate
added.
Gm.
22
22
66
66
220
220
Organic
matter
added.
Gm.
O
2.9
o
24.4
o
32- 7
o
61.6
o
67. 2
Hay
yield.
Gm.
0. 25
•05
. 02
. 20
•30
1. 10
3- 40
10. 70
8. 70
11. 70
Red clover.
72
73
74
75
76
77
78
Phos-
phate
added.
Organic
matter
added.
Gm.
Gm.
0
0
0
•55
II
0
II
10.5
22
0
22
66
30-5
0
66
57-9
Hay
yield.
Gm.
O. 02
. 10
.40
. 10
2. 70
6. 40
4. 70
12.50
" See series i, Tables II to VI.
INFLUENCE OF SIZE OF PARTICLES ON THE AVAILABILITY OF PHOS-
PHORUS IN MINERAL PHOSPHATES
The degree of fineness of rock phosphate particles has been held by
many irui^estigators to be an important factor in the availability of
mineral phosphates. Dr. Jordan, of the New York Experiment Station,
showed rather conclusivel}' that plants supplied with very finelv ground
rock phosphate contained more phosphorus and produced a greater
quantity of dry matter than those supplied with the coarser grades. For
the purpose of determining a comparative value of the same rock when
ground very fine to that left in particles of a larger size, series 12 (Table
XVII) was begun. As a check on the rock which was obtained from the
Mount Pleasant mills some lump rock from the same source was secured
and ground. These results are reported along with the data on the
influence of the size of particles on the availability.
Table XVII. — Relation of size of phosphate particles to the availability of phosphorus by
Sixty-Day oats harvested on fuly 10, IQIS — series 12
Pot
No.
Phos-
phate
added.
Fineness.
Grain
yield.
Straw
yield.
Pot
No.
Phos-
phate
added.
Fineness.
Grain
yield.
straw
yield.
I
Gm.
0
0
2.6
2.6
2.6
2.6
2.6
Gm.
0
0
3-25
2.05
3- 70
4. 70
5. 00
Gm.
I. 20
I. 60
6.15
5-35
6.60
9-30
8. 70
10. .
17..
18..
19..
20. .
21 . .
22 . .
Gin.
2.6
2.6
2.6
2.6
2.6
2.6
2.6
200 degrees or
over
Gm.
4. 00
5- 90
7. 00
5- 80
7.70
8.65
7- 15
Gm..
6.3
2 . .
80 to 100 de-
grees.
5--
80 to 100 de-
grees
6 80
do
10. 05
11. 05
11-35
6..
I-
do
100 to 200 de-
grees
100 to 200 de-
grees
do
8..
9..
do
200 degrees or
over
200 degrees or
over
do
5o8 Journal of Agricultural Research vci. vi. no. 13
Pots 5 to 10, inclusive, received the ground rock phosphate as it was
obtained from the mills. The degree of fineness varied from that passing
a sieve 80 to 100 meshes to the inch to that which would go through a
sieve of 200 meshes to the inch. Pots 17 to 22, inclusive, received the
ground phosphate which was shipped in the lump form and afterward
ground to the same degree of fineness as that ground at the mill.
There is a tendency for the dry matter to increase as the degree of
fineness increases. The phosphate received from the mill in lump form
was sHghtly better than that sent to us in a ground condition.
DISCUSSION
Under the conditions of these experiments a fairly large portion of the
phosphorus in brown rock phosphate was available for plant growth.
The quantity was variable, depending upon the crops and the circum-
stances attending the full development of the plant. The data show
only a very small amount of phosphorus soluble in water and plant food
solutions. It is clear that other factors which might bring about avail-
ability must be considered. The sand cultures contained very little
organic matter; hence, these sHght fermentable substances should not be
considered. There is nothing left but the plant for our examination and
there is abundant proof that the plant itself is a significant item. Since
plants excrete large quantities of carbonic acid, there is but little question
that this substance plays the primary roll in the liberation of phosphorus.
The reactions with carbon dioxid which occur when tricalcium phos-
phate is put into sand cultures of the kind described in these pages may
be shown in the following manner:
(I)
(2)
(3)
(A)
Gaseous
(4)
^HaCOgi
=^H''+HC03<=
In solution
(5) _
^H+++CO,
(B) Ca3(P04)2^Ca3(P04)2^3Ca++ + 2PO,
Solid In solution.
When A and B are mixed, the following equilibria develop:
(6)
(C) PO4+H+ HPO4 (The ion of reverted phosphate)
■ (7)
HP04 4-H+<=^H2P04 (The ion of soluble phosphate)
(8)
HjPO^-hH+^^HgPO^
June 26, 1916 Mineral Phosphates and Plant Nutrition 509
Equations A and B make it evident that the hydrogen ion concentra-
tion for the various acids will determine the course of the reactions ren-
dering the rock phosphate available. The hydrogen ion concentration
is made up of two factors — namely, the concentration and the strength
of the acid. Obviously under the conditions of these experiments satu-
rated solutions of rock phosphate and carbonic acid are employed. The
relative insolubility of the rock phosphate tends to decrease greatly the
concentration of the H+ from either 6, 7, or 8. The relatively greater
solubility of the calcium bicarbonate, since it furnishes HCO3, would
also tend to decrease the H+ concentration from carbonic acid, but this
factor of common ion effect is of far less importance upon the concen-
tration of the H+ from H.COg than the solubility of the tricalcium phos-
phate upon equations 6, 7, and 8, especially since the Ca++ from the
Ca(HC03)2 is removed by plants.
Assuming equivalent or unit concentrations of the substances HjCOg,
H3PO4, H2PO4, and HPO4 are present— that is, eliminating the factor of
concentration of the substances producing the H — the relative strength
of these acids is given by their ionization constants, thus:
(i)« H2C034=^H'-+HC03 Kai«° 3.0X10-
(6) HP04^P0,+ H+ Ka^8° 3.6X10-13
^—7
(7) H2PO,i=^HP04 + H+ Ka^^" 1.95 X IO-
CS) H3PO,<=^H^4 + H"" Ka^^o 1.1X10-2
The mass law for monobasic acids (HAc) has the form
(Cone H+) (Cone Ac)^ ^.^^^ ^^^ ^^.^^ ^^ equations i, 6, and
Cone HAc
7 are weak acids (Ka<io-^), the mass law assumes the form
Ka=(Conc H+) (Cone Ac) = (Cone H+)2, because the concentration of
HAc is practically unity. The concentrations of H+ for these equations
at 1 8° Care for:
(i) V3X io-' = 5.5X 10-^ for C(i>H+
(6) V3-6X 10-^^ = 6X10-' for C(^)H+
(7) Vi-95X io-' = 4-4X 10-* for 0')H+
For the first hydrogen of H3PO4 the above expression can not be
used, since the amount H3PO4 compared to its ions is small rather than
C0C2
large. Here the mass law must be used in its true form, K = ^_^, where
C is equal to the initial concentration of H3PO4 and oc degrees of ioniza-
o See equations, page 508.
5IO Journal of Agricultural Research voi. yi, No. 13
tion. For the purpose cc is taken equal to 90 per cent, from which the
concentration of H is calculated thus:
Ka^^° = , where (C) = Ka^^ ^ = concentration of H.
I — oc oc
r TT*/W8WT+x i.iXio~2(o.i) i.iXio"^ ^ _3
. • . concentration of H+(C(^>H+) — -^ — -' = — — =1X10 ^ =
o.ooi.
It is seen that only the first H of H3PO4 can furnish a. greater concen-
tration of H+ than H2CO3 for equivalent concentrations. In the actual
experiment the concentration of H3PO4 is much less than that of H2CO3.
However, the availability of the rock phosphate by means of H2CO3 is
not conditioned by the liberation of free H3PO4 according to equation 8.
Equation 6 or 7 is driven in the direction to remove H+, would render
the tricalcium phosphate more available, but a reaction between ions pro-
ceeds if a lesser ionized product be formed. Calculations of the H+ con-
centration for equations i , 6, and 7 shows that for equivalent concentra-
tions the H+ from carbonic acid is greatly in excess of the H+ concen-
tration for equations 6 and 7. So if equations i, 6, and 7 are present
simultaneously HPO4 and H2PO4 of equations 6 and 7 would be formed
by the union of H+ of H2CO3 with PO4 and HPO4, respectively, thus
causing more Ca3(P04)2 to dissolve to reestablish the equilibria for
equations 6 and 7. It is a fact, however, that a greater concentration
of H2CO3 is present than any of the ionizing substances, as HPO4, HjPO^,
or H3PO4. This would increase the rate of availability of the trical-
cium phosphate.
These calculations are borne out by the fact that more Ca3(P04)2
dissolved in water containing H2CO3 than in pure water. Seidel's
solubility tables state that i liter of water saturated with H2CO3 dis-
solves 0.15 to 0.30 gm. of Ca3(P04)2 at 25°, while i liter of pure water
dissolved only o.oi to o.io gm. of Ca3(P04)2 at 25°.
Reactions 6 and 7 may be shown in the nonionic form as follows:
Ca3(P04)2-f 2H2C03t;Ca2H2(P04)2 + CaH2(C03)2 or
Ca,(P04)2+2HX03!i;Ca2(HP04(2 + Ca(HC03)2
Ca2H2(P04)2 4-2H2C03;i!CaH4(P04)2+CaH2(C03)2
i i
or or
Ca(H2P04)2 Ca(HC03)2
In the first equation calcium is found in a form readily assimilated by
plants, and in the second the monocalcium phosphate is in a very assimil-
able form. On this equation we have based our belief that there is no
necessity for applying lime to sand cultures to which had previously been
June 26. 1916 Mineral Phosphates and Plant Nutrition 511
added raw rock phosphate. When the calcium bicarbonate and mono-
calcium phosphate are both removed from the medium of growth by
plants, the reaction is driven rapidly to the right. Mass relationship in
a mixture of this kind confirms such an interpretation as the one presented
above. Our first assumption, that plants should get their calcium from
rock phosphate in the same manner that they get their phosphorus, is
supported at several points in this work. This must be so, since the
calcium is furnished by the calcium salt of phosphoric acid or by the
bicarbonate. There was no greater growth when calcium carbonate was
added than where raw rock alone was used. In fact, the growth might
be even less, since calcium carbonate might furnish a greater concentra-
tion of Ca(HC03)2 or HCO3, which might decrease the concentration of
H from equation i , thus decreasing the rate of the availability of rock
phosphate.
The most marked feature of the investigation is the difference of the
availability of the various minerals. The fact that the crop yields
increase as the application of the brown rock phosphate was increased
indicates that a portion of the phosphorus was readily assimilated while
the plants were young, and that by the time these plants became well
established they were able to utilize the more insoluble form. If we are
to assume that a part of the phosphorus is of animal origin, this position
probably is more tenable, or on the other hand, through long years of
weathering the compound had been so changed that a portion was more
easily taken up by plants than before weathering began.
There is an indication that the crops grown first took up the more
available phosphorus and that the second crop made very slow growth
because the more soluble phosphorus was removed by the first crop and
nothing left but the rather insoluble for later crops. These points have
proof from the cowpeas on the large application series and the clover on
the crop residue series.
Brown rock phosphate and Florida soft rock phosphate lead the others
in supplying available phosphorus for plant nutrition, especially for clover.
The brown rock phosphate leads for all the crops. These two phos-
phates gave the largest quantity of phosphorus soluble in water and
plant-food solutions. The results indicate a relation in solubility in
plant-food solution and the availability for plants.
The difference in the assimilation of these phosphates can not be
attributed to the degree of fineness of the particles, since they were all
ground, so that the entire sample passed through a sieve of 100 meshes
to the inch. If the degree of fineness influenced the results, the differ-
ences then come from the size of particles, which were smaller than those
found in commercial phosphates.
The variation in the agricultural value of the six mineral phosphates
studied is difficult to explain. Their productive powers seemed not to
512 Journal of AgriciiIturQl Research voi. vi, ko. ij
have any direct relation to the amount of phosphorus which they con-
tained. Brown rock, which had the smallest amount of phosphorus,
produced the most satisfactory yields. The differences must be attrib-
uted to modes of formation and weathering since the minerals were laid
down.
SUMMARY
(i) Phosphorus in rock phosphate can be assimilated by farm crops
in sand cultures under greenhouse conditions, even in the absence of
decaying residues.
(2) Crop residues, when employed in conjunction with brown rock
phosphates, were beneficial.
(3) Tennessee brown rock phosphate, Florida soft rock phosphate, and
Tennessee blue rock phosphate in the heavier applications proved superior
to South Carolina land rock phosphate, Utah rock phosphate, and
Canadian apatite, for oats, clover, and cowpeas when grown in sand.
(4) The phosphorus in brown rock phosphate and Florida soft rock
phosphate was more soluble in water and in plant-food solutions than
the phosphorus in other mineral phosphates. The superiority of these
two phosphates over the others tested is shown chiefly by the first crop.
(5) Chemical analysis showed that the plant-food solutions applied
did not appreciably modify the results.
(6) The cereals produced as satisfactory yields as the legumes.
(7) The crop yields tended to increase as the application of rock
phosphate increased up to a point where the size of the pots seemed to
be a limiting factor, apatite being the only exception.
(8) The plants obtained their calcium, as well as their phosphorus,
from brown rock phosphates. No better results were secured when
calcium carbonate was applied than when rock phosphate alone was
used.
(9) There was no particular relation between the citric-acid-soluble
phosphorus and the availability of these phosphates for plants.
(10) Dextrose, when used as a fermentable substance, was harmful.
(11) The degree of fineness is a factor which determines to some
extent the availability of rock phosphate, as indicated by the brown
rock.
(12) These investigations extended over a period of 3^^ years, and
embrace results from 700 pot cultures and 400 phosphorus determinations.
june26, i9i6 Mmeral Phosphates and Plant Nutrition 513
LITERATURE CITED
(i) Bishop, W. H.
1894. Report of pot experiments with phosphates. Del. Agr. Exp. Sta. 6th
Ann. Rpt., 1893, p. 193-202, illus.
(2) Brooks, W. P.
1898. Natural phosphates compared with each other and with acid phosphate.
In Mass. Hatch Agr. Exp. Sta. loth Ann. Rpt., [1896/97], p. 14-19.
Reviews work done by Goessmann.
(3) •
1901. The relative value of different phosphates. In Mass. Hatch Agr. Exp.
Sta. 13th Ann. Rpt., [1899/1900], p. 101-105.
(4) Dyer, B.
1884. Comparison of dissolved and undissolved phosphates. (Abstract.)
In Jour. Chem. Soc. [London], v. 46, Abstracts, p. 774. 1884.
Original article in Jour. Roy. Agr. Soc. England, s. 2, v. 20, p. 113-
126. 1884. Not seen.
(5) Frear, William.
1887. Agricultural experiment work for 1885 and 1886. II. Experiments with
different kinds of phosphoric acid. Penn. State, Col. Rpt., 1886, p.
124-131.
(6) HartwEll, B. L., and Pember, F. R.
1912. The effect of cow dung on the availability of rock phosphate. R. I.
Agr. Exp. Sta. Bui. 151, p. 165-174, i pi.
(7) Hess, E. H.
1896. Experiments with soluble, reverted and insoluble phosphoric acid.
Penn. Agr. Exp. Sta. Ann. Rpt., 1895, p. 157-210.
(8) Hopkins, C. G., Mosier, J. G., Pettit, J. H., and Fisher, O. S.
1913. McDonough cotmty soils. 111. Agr. Exp. Sta. Soil Rpt. 7, 46 p., 7 fig.,
2 col. maps.
(9) Johnson, S. W.
1890. Field experiments in 1889. In Conn. Agr. Exp. Sta. Ann. Rpt. 1889,
p. 203-232.
(10) Jordan, W. H.
1895. Field experiments with fertilizers. In Maine Agr. Exp. Sta. Rpt.,
1894, p. 16-32.
(11) .
1913. Studies in plant nutrition. I. N. Y. Agr. Exp. Sta. Bui. 358, 30 p.
(12) Krober, E.
1909. tjber das Loslichwerden der Phosphorsaure aus wasserunloslichen
\'erbindungen imter der Einwirkung von Bakterien und Hefen. In
Joiu". Landw., Bd. 57, Heft i, p. 5-80.
(13) Litton, N. T.
1893. The effect of decomposing organic matter on natiiral phosphates. Ala.
Agr. Exp. Sta. Bui. 48, 80 p.
(14) McDowell, M. S.
1908. Is the phosphoric acid of floats made soluble by rotting manure? Penn.
Agr. Exp. Sta. Ann. Rpt., 1907/0S, p. 175-17S.
(15) Merrill, L. H.
1899. Box experiments with pho.sphoric acid from different sources. Maine
Agr. Exp. Sta. 14th Ann. Rpt., 1898, p. 64-74, 7 pi.
(16) Newman, J. S.
1889. Cotton — experiments with fertilizers. In Ala. Agr. Exp. Sta. Bui. 5,
n. s., p. 3-13.
514 Journal of Agricultural Research voi. vi, no. 13
(17) Newman, J. S., and Clayton, James.
1891. Experiments with cotton — 1890. Ala. Agr. Exp. Sta. Bui. 22, n. s., 29 p.
(18) Patterson, H. J.
1907. Fertilizer experiments with different sources of phosphoric acid. Md.
Agr. Exp. Sta. Bui. 114, p. 1 13-144.
(19) Pfeiffer, Th., and Thurman, H.
1896. Uber das Verhalten einiger Phosphate bei der Kompostierung. In
Landw. Vers. Stat., Bd. 47, Heft 4/5, p. 343-356.
(20) Prianichnikow, D. N.
1899. Etude sur la valeur relative des phosphates mineratix. In Ann. Agron.,
t. 25, no. 4, p. 177-187.
(21) Russell, H. L.
1913. Report of the director. Availability of phosphate to various crops.
Wis. Agr. Exp. Sta. Bui. 240, p. 22-23, fig. 11.
(22) SCHLOESSING, T., jr.
1902. A study of the phosphorus nutrition of plants. (Abstract.) In Exp.
Sta. Rec, v. 14, no. 3, p. 233. 1902. Original article in Compt.
Rend. Acad. Sci. [Paris], t. 134, no. i, p. 53-55. 1902. Not seen.
(23) SCHUTT, F. T.
1897. Report of the chemist. The fermenting of manure with finely ground
mineral phosphate. In Canada Exp. Farms Rpts., 1896, p. 196.
(24) [Thorne, C. E.]
1910. Plans and summary tables of the experiments at the central farm , Wooster,
on the maintenance of soil fertility. Ohio Agr. Exp. Sta. Circ. 104,
20 p.
(25)
1913. Farm manures. 242 p., illus. New York, London.
(26) Tottingham, W., E., Hoffman, C.
1913. Nature of the changes in the solubility and availability of phosphorus in
fermenting mixtures. Wis. Agr. Exp. Sta. Research Bui. 29, p.
273-321, 3 fig-
(27) Truog, E.
191 2. Factors influencing the availability of rock phosphate. Wis. Agr. Exp.
Sta. Research Bui. 20, p. 17-51, 6 fig.
(28) Verband landwirtschaftlicher Versuchs-Stationen im Deutschen Reiche.
1907. Vorlilufige Mitteilung der Beschlusse der XXIV. Hauptversammlung
des Verbandes zu Dresden am 14. September 1907. In Landw. Vers.
Stat., Bd. 67, Heft 5/6, p. 321-329.
(29) VoELCKER, A.
1880. Comparative value of soluble and insoluble phosphates. (Abstract.)
In Jour. Chem. Soc. [London], v. 40, Abstracts, p. 640-641. 1881.
Original article in Jour. Roy. Soc. Agr. Soc. England, s. 2, v. 16, p.
152-159. 1880. Not seen.
(30) Wheeler, H. J., and Adams, G. E.
1900. The needs and treatment of the Warwick Plain and other sandy soils of
Rhode Island. R. I. Agr. Exp. Sta. Bui. 68, p. 159-174.
(31)
(32)
1906. A comparison of nine different phosphates upon limed and unlimed land
with several varieties of plants. R. I. Agr. "Exp. Sta. Bui. 114, p.
117-137-
1907. Continued test of nine different phosphates upon limed and unlimed
land with several varieties of plants. R. I. Agr. Exp. Sta. Bui. 118,
p. 55-86.
PLATE LXXIII
Effect of varying quantities of Tennessee brown rock phosphates on plant growth:
Fig. I. — Spring wheat. (Table II, Series lA.)
Fig. 2. — Sixty-Day oats. (Table IT, series iB.)
Mineral Phosphates and Plant Nutrition
Plate LXXIII
':.v<^f^ii;^..v-*i' '
Journal of Agricultural Research
Vul. VI, No. 13
Mineral Phosphates and Plant Nutrition
Plate LXXIV
Journal of Agricultural Research
Vol. VI, No. 13
PLATE LXXIV
Effect of varying quantities of Tennessee brown rock phosphate on plant growth:
Fig. I.- -Barley. (Table VI.)
Fig. 2. — Timothy. (Table III, series lE.)
37770°— 16 4
PLATE LXXV
Effect of varying quantities of Tennessee brown rock phosphate on plant growth;
Fig. I. — Cowpeas. (Table IV, series iF.)
Fig. 2. — Soybeans. (Table IV, series iG.) Photographed just before cutting.
Fig. 3. — Red clover. (Table III, series iH.)
Fig. 4. — ^Alfalfa. (Table V.) Photographed before first cutting.
Mineral Phosphates and Plant Nutrition
Plate LXXV
Journal of Agricultural Researcl
Vol. VI, No. 13
Mineral Phosphates and Plant Nutrition
Plate LXXVI
Journal of Agricultural Research
Vol. VI, No. 13
PLATE LXXVI
Effect of different kinds of mineral phosphate applied in different quantities for
red clover. (Table VII.) Photograped just before first cutting.
PLATE LXXVII
Gjwpeas, showing the comparative effect of Tennessee brown rock phosphate alone
and in combination with dextrose. (Table XII.)
Mineral Phosphates and Plant Nutrition
Plate LXXVIi
Journal of Agricultural Research
Vol. VI, No. 13
Mineral Phosphates and Plant Nutrition
Plate LXXVIll
I 'v„ Uext
■ No DeM
i <V<> K R
I No RR
lOOrni CaCo"
1 Leachlngs
Leached
^ Irom SHJ
Journal of Agricultural Research
Vol. VI, No. 13
PLATE LXXVIII
GDwpeas, showing the comparison of their growth when treated with Tennessee
brown rock phosphate, phosphate and dextrose, and phosphate, dextrose, and calcium
carbonate. (Table XIV.) Photographed just before harvesting.
PLATE LXXIX
Effect of different substances on the gro"'/Vth of cowpeas:
Fig. I. — Growth after the addition of varying quantities of raw rock. (Table XV.)
Fig. 2 . — Growth after-the addition of dextrose and soluble phosphate. (Table XV.)
Mineral Phosphates and Plant Nutrition
Plate LXXIX
Journal of Agricultural Researoi
Mineral Phosphates and Plant Nutrition
Plate LXXX
< 0 1^ K IP
10 um-iCnCx!
Journal of Agricultural Research
Vol. VI, No. 13
PLATE LXXX
Effect of various substances and combmations on the growth of cowpeas:
Fig. I. — Effect of adding lime, phosphate rock, dextrose and lime, and phosphate
rock, dextrose, and lime to the soil. (Table XV.)
Fig. 2. — Effect of adding nothing, lime, phosphate rock, and phosphate rock and
lime to the soil. (Table XV.)
ADDITIONAL COPIES
OF THIS PUBLICATION MAY BE PROCURED FROM
THE SUPERIXTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT
20 CENTS PER COPY
Subscription Price, $3 Per Year
A
New York Botanical Garden Librar
3 5185 00263 4036