x^.
J
y,T" ^'-^£
JOURNAL OF
AGRICULTURAL
RESEARCH
Volume XII
JANUARY 7— MARCH 25, 191 8
BOTANICAL
PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINGTON, D. C.
V/. \3l
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Physiologist and Associate Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Entomologist and Assistant Chief, Bureau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist, and Assistant
Dean, Agricultural ExperiTnent 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.
* Dr. Pearl has undertaken special work in connection with the war emergency ;
therefore, until ftirther notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
II
CONTENTS
Effect of Time of Digestion on the Hydrolysis of Casein in the Page
Presence of Starch. J. S. McHargue i
Behavior of Sweet Potatoes in the Ground. Heinrich Has-
SELBRING 9
Studies in Soil Reaction as Indicated by the Hydrogen Electrode.
J. K. Plummer 19
Pure Cultures of Wood-Rotting Fungi on Artificial Media. W. H.
Long and R. M. Harsch 33
Gossypol, the Toxic Substance in Cottonseed. W. A. Withers
and Frank E. Carruth 83
Fruit-Fly Parasitism in Hawaii During 191 6. C. E. PembERTOn
and H. F. Willard .• 103
Irrigation Experiments on Apple-Spot Diseases. Charles
Brooks and D. F. Fisher 109
Relation of Carbon Dioxid to Soil Reaction as Measured by the
Hydrogen Electrode. D. R. Hoagland and L. T. Sharp. . . . 139
A Study of the Plow Bottom and Its Action Upon the Furrow
SHce. E. A. White i49
Influence of Nitrates on Nitrogen-Assimilating Bacteria. T. L.
Hills 183
New-Place Effect in Maize. G. N. Collins 231
Relation of the Variability of Yields of Fruit Trees to the Accu-
racy of Field Trials. L. D. Batchelor and H. S. Reed 245
Interrelations of Fruit-Fly Parasites in Hawaii. C. E. Pember-
TON and H. F. Willard 285
Water Extractions of Soils as Criteria of their Crop-Producing
Power. John S. Burd 297
Effect of Season and Crop Growth in Modifying the Soil Extract.
Guy R. Stewart 311
The Freezing-Point Method as an Index of Variations in the Soil
Solution Due to Season and Crop Growth. D. R. Hoagland. . 369
Efficacy of Some Anthelmintics. Maurice C. Hall and Win-
throp D. Foster 397
Tobacco Wildfire. Frederick A. Wolf and A. C. Foster .... 449
Gipsy-Moth Larvae as Agents in the Dissemination of the White-
Pine Blister-Rust. G. Flippo Gravatt and G. B. Posey 459
Influence of Carbonates of Magnesium and Calcium on Bacteria of
CC Certain Wisconsin Soils. H. L. Fulmer 463
■<
IV Journal of Agricultural Research voi.xn
Humus in Mulched Basins, Relation of Humus Content to Orange
Production, and Effect of Mulches on Orange Production. Page
Charles A. Jensen 505
Relation of Kinds and Varieties of Grain to Hessian-Fly Injury.
James W. McColloch and S. C. Salmon 519
Wilt Diseases of Okra and the Verticillium-Wilt Problem. C. W.
Carpenter 529
Winter Cycle of Egg Production in the Rhode Island Red Breed
of Domestic Fowl. H. D. GoodalE 547
Digestion of Starch by the Young Calf. R. H. Shaw, T. E. Wood-
ward, and R. P. Norton 575
Toxicity of Volatile Organic Compounds to Insect Eggs. Wil-
liam Moore and Samuel A. Graham 579
Corn-Stover Silage. J. M. Sherman and S. L. Bechdel 589
Weevils Which Affect the Irish Potato, Sweet Potato, and Yam.
W. Dwight Pierce 601
Sterility in the Strawberry, W. D. Valleau 613
Effect of Nitrifying Bacteria on- the Solubility of Tricalcium
Phosphate. W. P. KellEY 671
Respiration of Stored Wheat. C. H. Bailey and A. M. Gurjar. . 685
Effects of Mistletoe on Young Conifers. James R. Weir 715
Determination of Fatty Acids in Butter Fat. E. B. Holland and
J. P. Buckley, jr 719
Index 733
ERRATA AND AUTHORS' EMENDATIONS
Page 60, Table XIII, heading, " Folyporus dryophilus" should read " Polyporus dryophilus."
Page 105, Table I, " Peach {Prunus persica)" should read " Peach {Amygdalus persica)."
Page 331, line 8, "clay" should read "clay loam."
Page 335, hne 5 from bottom, "(fig. 3-6)" should read "(fig. 8-20)."
Page 361, line 18 from bottom, omit " 10."
Page 366, citation 27, " 1917" should read " 1918."
Pages 364-368, in citations 2, 16, 29, 30, 56, and 57, omit " Not seen,"
Pages 394-395, citations 3 and 8, " 1917" should read " 1918."
Page 429, Une 19 from bottom, "or against tapeworms" should read "and against tapeworms."
Page 430, line 6, "instability" should read "irritability."
Page455, line 5, "stroke" should read "streak."
ILLUSTRATIONS
PLATES
GossYPOL THE Toxic Substance in Cottonseed
Plate I. Effect of feeding cottonseed feeds to pigs: A. — Pig 3, showing con- page
dition on the ninety-fourth day on a feed containing ether-extracted cot-
tonseed kernels. B . — Pig 3 , showing condition on the fiftieth day. C. — Pig
4, showing condition on the twenty-seventh day on a feed containing vita-
mines. D. — Pig I, showing condition on the fiftieth day on a feed contain-
ing cottonseed meal. See figure G. E. — Pig 2, showing condition on the
twenty-seventh day on a feed containing gossypol. F. — Pig 3, showing
condition on the twenty -seventh day on a feed containing ether-extracted
cottonseed kernels. G.^Pig i, showing condition on the fiftieth day on a
feed containing cottonseed meal 102
Irrigation Experiments on Apple-Spot Diseases
Plate 2. A. — Early stage of bitter-pit on Northern Spy apple. B. — Cross
section of the apple shown in A 138
Plate 3. A. — Late stage of bitter-pit on Rhode Island Greening apple. B.—
Internal bro^\^ling accompanying bitter-pit. C— Jonathan-spot on Jona-
than apple. D.— Early stage of drouthspots on a Winesap apple from
Wenatchee, Wash. E. — Late stage of drouthspots on a Winesap apple.
F. — Cross section of the apple sho\vn in E 138
Plate 4. A. — An apple orchard showing the furrow system of irrigation em-
ployed in the experimental work at Wenatchee, Wash. B. — ^Jonathan
apple tree showing the effects of drouth, Wenatchee, Wash 138
Plate 5. A. — Cork on Yellow Newtown apple from Hood River, Greg. B. —
Cross section of the apple shown in A. C. — White Pearm.ain apple showing
the severity of the 1915 drouth at Wenatchee, Wash. D. — Cork, or "dr\--
rot, " on a King apple. E. — Blister on an Esopus apple from Entiat, Wash.
F. — An extreme ca;se of Yorkspot on a York Imperial apple. G. — Cross
section of the apple shown in F 138
A Study of the Plow Bottom and its Action Upon the Furrow Slice
Plate 6. A. — A plow bottom with two sets of straight lines. B. — A plow bot-
tom, the surface of which is composed of each of two surfaces. C. — A plow
bottom similar to B, but vnth the surfaces merging into each other farther
back on the moldboard. D. — A plow bottom, the surface of which does not
contain an infinite set of straight lines 182
Plate 7. A. — A plow bottom with a convex surface which has two sets of
straight lines. B. — Instrument for measuring the space coordinates of any
point of the plow bottom. C. — A sod plow showing the furrow slice turned
by it 182
Plate 8. A.— Rows of wooden pins driven into the sod for estimating the
stretch of the furrow slice. B. — Furrow slice showing the position of the
pins when on the moldboard 182
Plate 9. A. — Plow showing attachment used to obtain the x, y, and 2 coordi-
nates of points in the furrow slice. B. — Moldboard showing the paths of
five soil particles. C. — Measurement of the angle Ny by use of a protractor
and a plumb bob 182
(V)
VI Journal of Agricultural Research voi.xii
Interrelations ok Fruit-Fly Parasites in Hawaii
Plate io. Diachasvm tryoni: A. — Freshly hatched larva with its mandibles Page
actually embedded in the body of a newly hatched but dead larva of Opius
humilis. B. — Newly hatched larva with its mandibles closed, showing
ventral serosal material surrounding the body and the two gill-like appen-
dages on the first body segment 296
Plate h. Diachasma tryoni: A. — Lateral view of larva in the second instar,
showing particularly well the f atbody of the host recently taken in as food .
B. — Lateral view of a 2 -day-old lai-va engorged with food and about to molt,
showing the enlarged and stiffened body 296
Plate 12. Opiu? humilis: A, B. — Dead larva in first instar; killed by first-
stage larva of Diachasma tryoni, showing cut on body made by the attacking
larva and mandibles extended in final death struggle. C— Dead larva in
first instar; killed by first-stage larva of Diachasma tryoni. D. — Dead larva
in first instar; badly lacerated and distorted by attack of first-stage larva of
Diachasma tryoni 296
Plate 13. Opius humilis: A. — Dead larva in first instar, with body shriveled
and twisted through attack by first-instar larva of Diachasma tryoni. B. —
Dead larva in first instar; killed by first-instar larva of Diachasma tryoni.
C. — Dead larva in first instar; killed by first-stage larva of Diachasvia tryoni.
D. — Healthy, living larva in first instar. E. — Healthy, uninjured, living
larva in first instar 296
Effect of Season and Crop Growth in Modifying the Soil Extract
Plate 14. A. — General views of soil containers. B. — Bins for storage of sur-
plus soil 368
Tobacco Wildfire
Plate 15. Bacterium tabacum: A. — Tobacco leaf, four days after artificial
inoculation, showing chlorosis and lesions. B. — Natural infection with
brown lesions bordered by tissues of a water-soaked appearance 458
Plate 16. Bacterium tabacum: A. — Natural infection. Lesions are large and
concentrically zonate. B. ^Numerous confluent lesions on one side of
the midrib have resulted in distortion of the leaf. C. — Almost the entire
leaf is involved and a portion of the rotted tissues have fallen out. Natural
infection 458
Wilt Diseases of Okra and the Verticillium-Wilt Problem
Plate A. Fusarium vasinfectum on vegetable media: 1-3. — Growth on
steamed potato. Both potato cultures show pionnotes. 2, 4. — Growth on
rice. Cultures i and 2 were grown in a strong north light; 3 and 4 in a
subdued light 546
Plate 17. A-H. — Verticillium albo-airum: A. — Simple conidiophores and
conidia. B. — Same showing, respectively, the collection of the conidia
on the sterigma in irregular aggregations in dry air, and in water drops in
humid air. C. — Verticillate conidiophores bearing one and three whorls,
or virtels, of branches, respectively. D. — Verticillate conidiophore having
conidial heads, from humid environment — that is, moisttu-e drops in which
the conidia float as in figure B. E. — Mycelium of V. albo-atrum, in the
vascular ducts of an okra plant inoculated with this fungus. F, H. — Ger-
minating conidia. G. — Swollen, sclerotia-like mycelium.
I-M. — Fusarium vasinfectum: I. — Terminal, intercalary and conidial chlamy-
dospores. K. — Germinating macroconidium. L. — F. vasinfectum from
okra-wilt. M. — F. vasinfectum from cotton-wilt 546
Jan. 7-Mar. 25. i9i8 I llustvationS VII
Plate 18. Longitudinal section of an okra plant naturally infected with Page
V erticillium albo-atrum, showing the typical appearance 546
Plate 19. Verticillium albo-atrum: Two-weeks-old colony on potato agar,
showing the concentric rings of black sclerotial bodies 546
Plate 20. Solanuvi melongena, showing effect of wilt: A. — Qjntrol plant of
the same age as the wilted plant (B). B. — Wilted plant photographed two
months after inoculation at the hypocotyl with Verticillium albo-atrum
isolated from wilted eggplant 546
Plate 21. Abelmoschus esculentus , showing effect of wilt: A. — Control plant.
B. — Wilted plant photographed two weeks after inoculation at the hypo-
cotyl with a pure culture of Verticillium albo-atrum 546
Plate 22. Abelmoschus esculentus, showing effect of wilt: A. — Wilted plant
inoculated with Verticillium, albo-atrwrn. B. — Control plant of the same
age as wilted plant. Both plants were photographed two months after
the wilted plant had been inoculated 546
Plate 23. Abelmoschus esculentus, showing the effect of wilt as a result of
inoculation with Fusarium vasinfectum isolated from okra- wilt 546
Plate 24. Gossypium herbaceum (Columbia variety): Control plants 35 days
old 546
Plate 25. Gossypium herbaceum (Columbia variety), showing effect of wilt:
Wilting plants photographed 15 days after inoculation at the hypocotyl
with Fusarium vasinfectum, isolated from wilting cotton plants 546
Plate 26. Gossypium herbaceum (Columbia variety), showing eft'ect of wilt:
Wilting plants photographed 15 days after inoculation at the hypocotyl
with Fusarium vasinfectum, isolated from wilting okra 546
Plate 27. Abelmoschus esculentus, showing the characteristic symptoms of
the wilt produced by V erticilliurn albo-atrum 546
Weevils Which Affect Irish Potato, Sweet Potato, and Yam
Plate 28. Trypopremnon sanfordi: Adult from Cuzco, Peru. A. — Dorsal
view. B. — Face of same. C. — Side view of thorax and head. D. — Ven-
tral view of adult 612
Plate 29. Trypopremnon latithorax: Larva from La Paz, Bolivia. A. — Pro-
thoracic spiracle. B. — ^Larva, lateral view. C. — Lateral view of head.
D. — Right side view of apex of labium. E. — Corresponding hair on left
side. F. — Maxillary palpiger and palpus, lateral view. G. — Face 612
Plate 30. Trypopremnon latithorax: Pupa from La Paz, Bolivia. A. — Dorsal
vitfw. B. — Ventral vievv^. C. — Enlarged sketch of eighth, ninth, and
tenth abdominal segments 612
Plate 31. Species of the genus Cylas: A.—Cylasforviicarius elegantulus from
Honolulu, Hawaii, side view of head and thorax. B. — Cylas iurcipennis
from Sumatra, side view of head and thorax. C. — Cylas brunncus from
East Africa, dorsal view of thorax. D. — Cylas brunneus, side view of head
and thorax. E. — Cylas brunneus ventral view of thorax. F. — Cylas
femoralis, side view of head and thorax 612
Plate 32. Sweet-potato and yam weevils: A. — Cylas formicarius elegantulus,
female, from, sweet potatoes. New Orleans, La. B. — Same, head of male.
C. — Euscepes batatae, from sweet potatoes, Hawaii. D. — Same, side view
of head. E.^ — Palaeopus dioscoreae, from yams {Dioscorea batatas), Jamaica.
F. Same, side view of head 612
Plate 33. Pups of sweet-potato weevils: A. — Euscepes batatae, Barbados,
venter. B. — Same, latero-vcntral view of fifth to tenth segments. C. —
Same, dorsal view. D. — Same, venter of seventh to tenth segments. E. —
Cylas formicarius elegantulus, Victoria, Texas, ventral view of sixth to
tenth segments. F. — Same, ventral view. G. — Same, latero- ventral
view. H. — Same, dorsal view 612
VIII Journal of Agricultural Research voi.xii
Plate 34. Larvae of su-eet-potato weevils: A. — Cylas formicarius elegantulus, page
Victoria, Texas, lateral view. B. — Same, dorsum of head. C. — Same,
face. D. — Same, side of head. E. — Eusccpes batatce, Barbados, dorsum
of head. F. — .Same, face. G. — Same, side of head. H. — Same, lateral
view of larva 612
Steriutv IX THE Strawberry
Plate B. Minnesotii 3: i. — Pollen mother cell previous to synapsis. 2. —
Presynapsis in the pollen mother cell showing loops extending out from
synaptic mass. 3. — Two loops and portion of a loop extending from the
presynaptic mass. Same stage as figure 2. 4. — Synapsis in a pollen
mother cell. 5. — A postsynaptic stage. G.^Open spireme stage. 7. — A
presegmentation stage of the spireme. 8. — A portion of a bivalent spireme
thread of the same stage as figure 7. 9. — Segmentation of the bivalent
spireme into chromosome pairs. 10. — A portion of the bivalent spireme
during segmentation. 11, 12. — Chromosome pairs during the contraction
period following segmentation. 13. — Individual chromosome pairs show-
ing various figures commonly formed diu-ing contraction. 14. — Diakenesis
in the pollen mother cell. 15. — Multipolar spindle stage of pollen mother
cell. 16. — -Early anaphase of the heterotypic division 670
Plate C. i. — Late anaphase of the heterotypic division. 2. — ^Chroraosomes
on the equatorial plate of the homeotypic division. 3. — A portion of an
anther in the tetrad stage, showing the microspores embedded in the
gelatin-like sheath. 4. — A tetrad at the same stage as those shown in figure
3. 5. — A microspore shortly after liberation from the tetrad. 6. — A
liberated microspore in which growth has commenced. 7. — A later stage
than figure 6, showing the slight thickening of the wall and the irregularities
due to grov^lh of the wall. 8. — Microspore growth completed previous to
division of the microspore nucleus. 9. — A section through a microspore
nucleus in prophase showing the continuous univalent spireme. 10. —
Another section of the same nucleus, showing the first stages of the disappear-
ance of the nucleolus. 11. — Metaphase of the division of the microspore
nucleus. 12. — F. virginiana. Anaphase in the division of the microspore
nucleus. 13. — Telophase of the division of the microspore nucleus. 14. —
A later stage than figure 13 in which the generative cell has been definitely
cut off. 15. — A young pollen grain shortly after division, showing an in-
crease in cytoplasm content. 16. — End view of a pollen grain showing the
pattern of the laminate layers shown in figure 15 and Plate D, figures i, 6,
and 15 670
Plate D. i. — Nearly mature pollen grain. 2. — Mature pollen grain. 3, 4,
5, 7. — Various types of degenerate microspores from anthers bearing micro-
spores of the stage shown in Plate C, figure 6. 6. — An aborting microspore
from an anther containing half-grown microspores. 8. — An aborting micro-
spore of the same type as that shown in figure 6 from an anther containing
nearly full-grown microspores as in Plate C, figure 8. 9, 11. — Microspores
of the same types and same age as figures 6 and 8, in which degeneration
has proceeded farther. 10. — An aborted microspore from an anther con-
taining microspores of the stage sho\\'Ti in Plate C, figure 8. 12. — An early
stage of degeneration in a full-grown i-nucleate microspore. 13.— An
early stage of degeneration in a full-grown i-nucleate microspore. 14. —
An aborting microspore containing an abnormally small amount of light
staining cytoplasm; from an anther containing i- and 2-nucleate micro-
spores. 15. — An aborted microspore from an anther containing i- and
2-nucleate microspores. 16. — An aborted microspore containing very
scant cytoplasm. The nucleus has completely degenerated and degenera-
tion of the cytoplasm has begun 670
Jan. 7-Mar. 25, 1918 IllustratlOnS IX
Plate E. i. — A slightly more advanced stage of the condition shown in Page
Plate D, figure 16. 2. — An early stage in the abortion of a full-grown i-
nulceate microspore. 3. — An early stage of abortion directly following
microspore division. 4. — A full-grown i-nucleate microspore containing
ver)' scant light-staining cytoplasm; from an anther containing i- and
2-nucleate microspores. 5. — Another type of degeneration of a full-grown
i-nucleate microspore. 6. — An aborted microspore found among i- and
2-nucleate m.icrospores. 7. — A later stage of the type of degeneration shown
in Plate D, figure 13; from an anther containing microspores of the stage of
development shown by Plate C, figure 15. 8. — Degeneration of the gen-
erative cell shortly after division. 9, 10. — Common types of aborted
microspores found with mature pollen. 11. — An aborted microspore, of
the same type as that shown in figure 7. 12. — A pollen grain showing
abortion of the generative cell and an abnormal vacuolate condition of the
cytoplasm. 13. — A later stage of the type of degeneration shown in
figure 8 670
Plate 35. A. — Tertiary flower of the pistillate variety, Minnesota ioi7XPro-
gressive— 13-40, showing prominent staminodia. B, C. — Primary and
secondary flowers of the perfect variety, Minnesota 1017X Progressive—
9-24; B showing intermediate and C perfect anthers. D, E, F. — Two
primary and a secondary flower of the perfect variety, Minnesota 1017X
Progressive — 2-55, showing pistillate, intermediate, and perfect types of
flowers. G, H, I, J. — Flowers from the perfect variety, Minnesota 1017X
Progressive— 32-1 670
Plate 36. A, B, C, D. — Cross sections of two loculi of staminodia of the pistil-
late varieties. Crescent, Columbia, Minnesota ioi7XProgressive— 11-59,
and Seedling 140, respectively. E. — Degeneration of the tetrads in an inter-
mediate anther of Fragaria virginiana. F, G. — Later stages of the con-
dition shown in figure E. H. — A portion of an intermediate anther from
the first flower of Minnesota 3 670
Effects of Mistletoe on Young Conifers
Plate 37. A. — Pseudotsuga taxifolia infected with Razoumofskya douglasii.
B. — Effect of an inoculation with Razoumofskya campylopoda on the height
growth of 6-year-old Pintis jeffreyi 718
38327°— 19 2
TEXT FIGURES
Behavior of vSweet Potatoes in the Ground
Fig. I. Graphs showing changes in composition of Big Stem sweet potatoes page
during the latter part of the season, from September 18 to November
27, and the minimum temperatures at the United States Weather
Bureau Observatory at Washington, D. C, some 20 miles distant,
during that period 16
GossYPOL, the Toxic Substance in Cottonseed
Fig. I . Graphs of the growth of pigs qi
2 . Graphs of the gains per week of pigs gi
3 . Graphs of the growth of pigs 94
Irrigation Experiments on Apple-Spot Diseases
Fig. I. Diagram showing the soil-moistvtre conditions in irrigated plats of Gano
apples 113
2. Diagram showing the soil-moisture conditions in irrigated plats of
Grimes apples 114
3. Diagram showing the amount of bitter-pit on Grimes apples 116
4. Diagram showing soil-moisture conditions in irrigated plats of Grimes
apples 117
5. Diagram showing the amount of bitter-pit on Grimes apples 118
6. Diagram showing the relation of the amount of bitter-pit to the size of
apples 119
7. Diagram showing the soil-moisture conditions in irrigated plots of Jon-
athan apples in 1915 121
8. Diagram showing the amount of bitter-pit on Jonathan apples in 1915. . 123
9. Diagram showing the soil-moistiu-e conditions on plots of Jonathan
apples in 1916 124
10. Diagram showing the amount of bitter-pit on Jonathan apples inigib. 125
A Study of the Plow Bottom and Its Action upon the Furrow Slice
Fig. I. Diagram giving the generatrices, directrices, and equations of surfaces
of historical plow bottoms 150
2-10. Graphs of the development of plow bottoms 152, 155, 157, 159, 160
11-14. Graphs of the motion of soil particles in plowing 163, 164, 166
15. JeflFerson's plow bottom 173
16. Lambruschini 's plow bottom 175
17-18. Small's plow bottom 175, 176
19-2 1 . Stephen 's plow bottom 177
22-25. Rahm's plow bottom 178, 179
26. Knox's plow bottom 179
Relation of the Variability of Yields of Fruit Trees to the Accuracy of
Field Trials
Fig. I. Diagram showing the individual tree yield of the navel-orange grove . 252
2. Diagram showing the individual tree yield of the navel-orange grove
(Antelope Heights) 254
(XI)
XII Journal of Agricultural Research voi.xii
Fig. 3. Diagrani showing the individual tree jdeld of the Valencia orange page
grove 255
4. Diagram showing the individual tree yield of the Eureka lemon grove . 255
5. Diagram showing the individual tree yield of the seedling walnut
orchard 256
6. Diagram showing the individual tree yield of the Jonathan apple
orchard 257
7. Graphs of tlie reduction of the coefficient of variability by increasing
the number of adjacent trees to the plot 261
8. Graphs of the reduction of the coefficient of variability by increasing
the number of trees to the plot 262
9. Graphs of production, 32-tree plot, navel oranges (Arlington) 267
10. Curve of yields of individual trees, navel orange (Arlington) 268
11. Curve of yield of individual trees, navel orange (Antelope Heights). . 269
Water Extractions of Soils as Criteria of Their Crop-Producing Power
Fig. I. Graphs showing soils arranged with reference to yield and important
characters 307
Effect of Season and Crop Growth in Modifying the Soil Extract
Fig. I. Design of soil containers 323
2. Diagram of the arrangement of the soil containers 324
3. Graphs of the nutrients extracted from soil 4 by varying the ratios of
soil to water 336
4. Graphs of the nutrients extracted from soil 5 by varying the ratios of
soil to water 336
5. Graphs of the nutrients extracted from soil 8 by varying the ratios of
soil to water. Calculated to parts per million of dry soil 337
6. Graphs of the nutrients extracted from soil 10 by var}-ing the ratios of
soil to water 337
7. Graphs of the yield of grain in 1915 and 1916, expressed as a percentage
of tlie maximum yield 342
8. Graphs of the seasonal studies of the water extract of soil 1, Yolo silty
clay loam 343
9. Graphs of the seasonal studies of the water extract of soil 2, Yolo silt
clay loam 344
10. Graphs of the seasonal studies of the water extract of soil 3 , Yolo silty
clay loam 345
11. Graphs of the seasonal studies of the water extract of soil 4, Yolo silty
clay loam 346
12. Graphs of the seasonal studies of the water extract of soil 5, Yolo silty
clay loam 347
13. Graphs of the seasonal studies of the water extract of soil 6, Yolo clay
loam 348
14. Graphs of the seasonal studies of the water extract of soil 7 , Hanford fine
sandy loam 349
15. Graphs of the seasonal studies of the water extract of soil 8, Fresno fine
sandy loam 350
16. Graphs of the seasonal studies of the water extract of soil 9, Kimball
fine sandy loam 351
17. Graphs of the seasonal studies of the water extract of soil 10, Tejunga
fine sandy loam 352
18. Graphs of the seasonal studies of the water extract of soil 11, jVIadera
fine sandy loam 353
Jail. 7-Mar. 25, 1918
Illustrations xiii
Fig. 19. Graphs of the seasonal studies of water extract of soil 12, Arnold fine Page
sandy loam 3 54
20. Graphs of the seasonal studies of the water extract of soil 14, Standish
fine sandy loam 355
21. Graphs of the growth of crops in height, season of 1916 357
22. Graphs of the daily studies of the water extract of soils lA and iB, sea-
son of 1916 359
23. Graphs of the daily studies of the water extract of soils 8A and 8B,
July, 1916 360
24. Graphs of the daily studies of the water extract of soils 8A and SB,
August, 1916 360
The FreEzing-Point Method as an Index of Variations in the Soil Solution
Due to Season and Crop Growth
Fig. I. Graphs of the depressions of the freezing point in soils i and 2, with
and without crop 372
2. Graphs of the depressions of the freezing point in soils 3 and 4, with
and \\nthout crop 373
3. Graphs of the depressions of the freezing point in soil 5, with and with-
out crop 374
4. Graphs of the depressions of the freezing point in soil 6, with and with-
out crop 375
5. Graphs of the depressions of the freezing point in soils 7 and 8, with and
without crop 37°
6. Graphs of the depressions of the freezing point in soils 9 and 10, with and
without crop 377
7. Graphs of the depressions of the freezing point in soils 11 and 12, with
and without crop _ 37°
8. Graphs of the depressions of the freezing point in soil 14, with and
without crop 37°
9. Graphs showing the results of successive extractions of soils 5 and 8 . . . 392
Efficacy of Some Anthelmintics
Fig. I. Apparatus, with control, for administering copper-sulphate solution to
sheep 407
Tobacco Wildfire
Fig. I. Parenchyma cells from the margin of a lesion showing Bacterium taba-
cum in the intercellular spaces and within the cells 454
2. a, Flagellaof Bacterium tabacum stained by Morrey's method; b, Bad.
tabacum from bouillon stained with carbol-fuchsin, showing arrange-
ment of the elements 454
Influence of Carbonates of Magnesium and Calcium on Bacteria of Certain
Wisconsin Soils
Fig. I. Diagram showing the influence of calcium carbonate and limestone
on the number of bacteria in Colby silt loam 47^
2. Diagram showing the influence of calcium carbonate and monocal-
cium phosphate on the number of bacteria in Colby silt loam 472
3. Diagram showing the influence of calcium carbonate and limestone on
the number of bacteria in Plainfield sand 473
4. Diagram showing the influence of calcium carbonate and monocal-
cium phosphate on the number of bacteria in Plainfield sand 474
XIV Journal of Agricultural Research voi.xii
Fig. 5. Diagram showing the influence of magnesium carbonate on the number Page
of bacteria in Colby silt loam 475
6. Diagram showing the influence of magnesium carbonate on the number
of bacteria in Plainfield sand 477
7. Diagram showing the influence of the carbonates and chlorids of mag-
nesium and calcium on the number of bacteria in Colby silt loam
soil 481
8. Diagram showing the influence of dibasic phosphate, monocalcium
phosphate, calcium carbonate, and magnesium carbonate on the
number of bacteria in Colby silt loam 482
9. Diagram showing the influence of calcium carbonate, magnesium car-
bonate, dibasic magnesium phosphate, and monocalcium phosphate
on the number of bacteria in Miami silt loam 483
10. Diagram showing the influence of calcium carbonate, magnesium car-
bonate, limestone, and monocalcium phosphate on nitrate accumu-
lation of Colby silt loam 488
1 1 . Diagram showing the influence of large applications of magnesium car-
bonate on Bacillus azotobacter in sterile Colby silt loam 494
Digestion of Starch by the Young Calf
Fig. I. Bag for receiving feces and harness for supporting it 577
Sterility in the Strawberry
Fig. I. Diagram showing the arrangement of flowers of the strawberry and the
order of blossoming 614
2. Flower diagrams of Fragaria spp., showing stamen arrangement in F.
virginiana and F. americana and many found in cultivated varieties. 61 c
3. Outline camera-lucida drawings of perfect and intermediate anthers and
staminodia of strawberry 617
4. Graphs showing the relation between sepal number and flower position
in the seedling varieties Nos. 373, 968, and 1006 620
Respiration of Stored Wheat
Fig. I. Graph showing the relation of the moisture content of wheat to the rate
of respiration 692
2. Graphs showing the comparative rate of respiration of hard spring, soft
red winter, and soft white winter wheat 694
3. Graphs showing the rate of respiration of shriveled wheat and of plump
wheat of the same class 698
4. Graphs showing the rate of respiration of frosted wheat and sound wheat
of the same class 699
5. Graphs showing the comparative respiratory activity of naturally damp
wheats and of wheats dampened in the laboratory three days before
they were incubated 702
6. Graph showing the relation of temperature to the rate of respiration . . . 705
7. Graph showing the rate of respiration during successive intervals when
the respired carbon dioxid was permitted to accumulate in the mass
of grain 707
Determination of Fatty Acids in Butter Fat:
Fig. I. Apparatus employed in esterification 722
2. Apparatus employed in fractionation 724
Vol. XII JANUARY 7, 1918 No. 1
JOURNAL OF
AGRICULTURAI/
RESEARCH
CONXENXS
Effect of Time of Digestion on the Hydrolysis of Casein in
the Presence of Starch - - - -* • -1
J. S. McEARGUE
( CooMlmtira (zoffi Xentvcky AcrieaMonl Bspetimeitf Sti^
Behavior of Sweet Potatoes in the Ground - - • . 9
HEINRICH HASSELBRING
(Coxrtribotlon trotn BuiMii of Plant Indtistiy)
Studies in Soil Reaction as Indicated by the Hydrogen
Electrode - - - » - - - . . if)
J. K. PLUMMER
CoaWbBden fn»i nMb CcMUaa Acricnltona Bxpwlia^
PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE,
WITH THE COOPERATION OF THE ASSOCUTION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINOTON, D. C.
WMHINOTOM ! OOVtHNMCNT MIMTIMO OTriOC : 1618
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR TBB DEPARTBCSIIT
FOR TBB ASSOCtATlOa
KARL F. KELLERMAN, Chahuian RAYMOND PEARL •
Physiologist and Assodatt Chief, Buttam
of Plant Industry
EDWIN W. AIXEN
Chief, OfficecfExptfimemtStaAm
CHARLES L. MARLATT
Entomdooist and Assislam$ ChUi,
of Entomoloof
Biologist, Maine AtricuUunI E*$m1mml
Station
H. p. ARMSBY
Director, Institute afAnimai SitlrUititi fit
Pennsj4vania State Cotteg*
E. M. FREEMAN
Botanist, Plant Pathologist tnd Attislmit
Dean, Agricultural Experimmi Statiom ef
ihe UnrversUp ofMmmesoU
All correspcmdence regarding articles ttom tbe Department of Apiculture tfkould bt
addressed to Karl F. Kellerman, Journal of Agricultural Research, Washington, D. C
* Dr. Pearl has undertalsxn special work in connection with the war emeigeni^;
therefore, until ftuther notice all correspondence regarding articles from State Ezpeti*
ment Stations should be addressed to fi. P. Armsby, Institute oC Animal Nutdtioaj
State College, Pa.
JOMAL OF AGlilCimiRAL RESEARCH
Vol. XII Washington, D. C, January 7, 191 8 No. i
EFFECT OF TIME OF DIGESTION ON THE HYDROLYSIS
OF CASEIN IN THE PRESENCE OF STARCH '
By J. S. McHarguE,
Chemist, Laboratory of Chemical Research, Kentucky Agricultural Experiment
Station
The Van Slyke ^ method for protein analysis was worked out upon
mixtures of relatively pure amino acids and was not intended to be applied *-'-kt-
directly to crude sources of protein contained in cereals and feeding
stuffs.
Notwithstanding this fact, Grindley, Slater, et al.,^ of the Illinois Experi-
ment Station, published in 191 5 the results of the determination of the
amino acids contained in cottonseed meal, tankage, and alfalfa hay,
applying the Van Slyke method directly to the proteins contained in
these different feeds.
In the same month of 191 5 Nollau,* of this Station, published his results,
obtained by the Van Slyke method, on about 25 different sources of crude
protein contained in various seeds and feeding stuffs.
In December, 1915, Grindley, Slater, et al.,^ published a second paper
on the amino-acid content of various feeds, including wheat, oats, barley,
and sov beans, a number of which had been analyzed by Nollau. The
summary of their second paper in part is as follows :
The results here reported confirm the conclusion previously drawn, namely, that
the Van Slyke method for the determination of the chemical groups characteristic
of the amino acids of proteins can be applied directly to the quantitative determina-
tions of the amino acids of feeding stuffs with at least a fair degree of accuracy.
Tne results which we have obtained for the quantitative determinations of amino
acids in feeding stuffs, on the whole, do not agree well with those recently published
by Nollau. In some determinations the results from the two sources are quite satis-
factory, but in many cases the agreement is far from satisfactory. The lack of con-
• Approved for publication in the Journal of Agricultural Research by A. M. Peter, Acting Director,
Kentucky Agricultural Experiment Station.
^ Van Slyke, D. D. the analysis of proteins by determination of the chemical groups char-
acteristic OF THE DIFFERENT AMINO ACIDS. In Jour. Biol. Chem., v. lo, no. i, p. 15-53. 2 fig. 1911.
' Grindley, H. S., Slater, M. E., et al. the quantitative determination of the amino acids of
FEEDING STUFFS BY THE VAN SLYKE METHOD. In J OUT. Amer. Chem. Soc, V. 37, no. 7, p. 1778-1781;
no. 12, p. 2762-2769. 1915.
* Nollau, E. H. the AMiNO-AaD content of certain commeroal feeding stuffs and other
SOURCES of protein. In Jour. Biol. Chem., v. 21, no. 3, p. 611-614. 1915.
Journal of Agricultural Research, Vol. XII, No. i
Washington, D. C. Jan. 7, 1918
U Key No. Ky.— 6
(I)
2 Journal of Agricultural Research voi. xii, no. i
cordant results is probably due in the main to differences in the details of procediure
in the experimental work.
Hart and Bentley/ of the Wisconsin Experiment Station, comment
unfavorably on the lack of agreement between the results obtained by
Grindley, Slater, et al., and those obtained by Nollau for the amount
of the different amino-acid groups contained in feeding stuffs. They
state that whether accurate determinations of any or all the amino acids
can be secured when the hydrolyzing proteins are in contact with hydro-
lyzing carbohydrates must first be determined before these data can be
accepted as final.
Presumably in order to substantiate the theory in regard to the effect
of hydrolyzing carbohydrates on the different amino-acid groups in pro-
teins. Hart and Sure ^ have published results obtained upon the hy-
drolysis of casein, alone and in the presence of a number of different
carbohydrates. In one of their experiments, 2.4 gm. of casein and 12
gm. of starch were hydrolyzed by boiling in 20 per cent hydrochloric
acid for a period of 48 hours. The result obtained for lysin in this ex-
periment shows that approximately 50 per cent of this amino-acid
group has been changed to some other form of combination. They
summarize their results in part as follows :
The Van Slyke method of protein analysis, applied to casein, hydrolyzed in the
presence of various carbohydrates, brings about a total redistribution of the amino-
acids varying with the nature of the carbohydrate employed. This work on casein
and Gortner's work on fibrin, hydrolyzed in the presence of cellulose, definitely show
the inapplicability of the method of direct hydrolysis for the estimation of amino-
acids in feeding stuSs by Van Slyke 's method. The results so secured will be inac-
curate.
Upon the publication of Hart and Sure's results, it appeared to the
writer that their conclusions were much broader than their experiments
justified. In fact, Hart and Bentley^ make statements which appear to
be merely forecastings rather than conclusions arrived at by experi-
mentation. In order to be able to say positively that the Van Slyke
method for protein analysis can not be applied directly to heterogeneous
mixtures of protein and carbohydrate requires much further experi-
mentation. It is by no means to be taken for granted that results obtain-
ed on a 48-hour digestion will be the same as those carried on for a shorter
length of time.
It therefore occurred to the writer that a duplication of the experiment
of Hart and Sure upon the effect produced on the hydrolysis of casein
by the presence of starch, in which the time of digestion varied, would
afford more conclusive evidence on this subject. Accordingly, five ex-
periments were planned, as follows:
' Hart, E. B., and Bentley, W. H. the character OF the water-soluble nitrogen of some com-
mon FEEDING STUFFS. In Jour. Biol. Chem., v. 22, no. 3, p. 477-483. 1915.
* and Sure, Barnett. The influence of carbohydrates on the accuracy of the van slyke
method in the hydrolysis of casein. In Jour. Biol. Chem., v. 28, no. i, p. 241-249. 1916.
3 Hart, E. B., and BentlEy, W. H. Op. cit.
Jan. 7. 1918 Effect of Time of Digestion on Hydrolysis of Casein 3
Five lo-gm. portions of Hammarsten's casein were weighed out and
transferred to five i -liter round -bottom Jena flasks. Fifty gm, of
cornstarch were then weighed out and added to each of the flasks except
the first, which contained casein alone. Three hundred c. c. of 20 per
cent hydrochloric acid, specific gravity i.ii, were added to each flask.
All the flasks were then heated on the water bath, with frequent shak-
ings, for about two hours. The object of this preliminary heating on
the water bath was to liquefy the starch-casein mixtures, which had
gelatinized upon the addition of the hydrochloric acid. After the starch
had become liquid all the flasks were removed and attached to reflux
condensers and heated to a gentle boil.
Experiments i and 2 were allowed to digest for 12 hours, No. 3 for 15
hours. No. 4 for 24 hours, and No. 5 for 48 hours, each being cut out at
the expiration of its time interval.
After each of the experiments had stood at room temperature for six
or eight hours, they were filtered through paper on a Buchner funnel
and washed practically free of chlorids with hot water. There was no
insoluble residue remaining on the filter from the casein digestion. There
were rather large insoluble carbonaceous residues remaining from each
of the casein-starch mixtures. Each of these was dried at 100° C,
bottled, and set aside for further investigation as to their nitrogen
content.
The filtrates in each of the experiments were concentrated separately
under reduced pressure until practically all of the excess of hydrochloric
acid was removed. The residues were taken up in water and run through
filters into separate flasks of 250-c. c. capacity. After the filters were
washed thoroughly, the contents of each flask were brought up to the
mark with water, and duplicate analyses were carried out by the Van
Slyke method on aliquots from each of these hydrolyzed solutions.
The results obtained are shown in Table I.
From the data in Table I showing the average results obtained upon
casein alone and upon definite mixtures of starch and casein digested at
different intervals of time the following observations may be made.
In all of the experiments there is but slight variation in the ammonia
determinations^ ; the maximum result is obtained in the 48-hour digestion.
The increase in this case is in all probability owing to the change of some
of the amino groups to ammonia compounds, which indicates over-
digestion.
The results for the humin determinations show a diminution in the
1 5-, 24-, and 48-hour digestions over those of the 12-hour digestions.
However, the humin determination in the 12-hour digestion of the
starch-casein mixture agrees well with the humin results obtained on
casein alone.
' Previous to the amraonia determinations the acidity of the hydrolyte, in terms of the calcium-hydrate
suspension, was determined by titration, with phenolphthalein as the indicator. A slight excess of the
calcium-hydrate suspension aoove tne amount necessary to neutralize the acid was always added.
Journal of Agricultural Research
Vol. XII. No. I
13
•^
■^
S
"«
'Si
fO
•c Saab's.
>< o o « u>a
w
'Sib'O
d a u " "i 3
attJ rt u. o J3
oi-S
■a + -« 8 !ii S
a a o " m g
g CI oJ 1-, cjja
S 0)
> bo
12?
b> bo
■5"
■g S 3 °' S S
K o B « o>is
a°fe1
.p 4- 01 u b
•c g a g S
t< o « o>2
> bg
O f~* O "^*0 f^ »^ fO
wr-OO O t-o^O
o o o OS toQO '^r o
r- 6 t> o lo c^ t
wiot*0\r*tH woo
ft ^ c5 * vi5 H ' d>oo 6
D § 6 ' NO M * 6 dco
r*w COM t^mc\^
K^OO O\00 ^M r-vo O
n § o^ * wiO ' CO t^ d
CL S Ov w tovO 00 r^ t
NO r» t^co ^- O 00 o
K-^^NTl-Ol-tl-tOO -^PO
(V § 6 MNCod ' 00 t^ cj«
t^O 00 «o w o
■«t O HI M GO C7»
^ § d w t^oo * 00 o d
-5*
H r» OnOO Ov OvO !>■
" ^dwoco 'oooooo
a,S2
W^Uct fOO\0 M coo ■^
O § 6 M «^o6 ' OM-OO
0\ M O ■" t' O Oi«
Q^ § d M »iod * di t^oo
O « 0( ^ ^ ^00 t^
Q. § 6 M t^ t-- ' CJi «-■<»
i^.t;-3 Si; °-t 2 «
Jan. 7, 1918 Effect of Time of Digestion on Hydrolysis of Casein
The results on humin represent the humin in solution and precipitated
by calcium-hydrate suspension. The high results obtained by Grindley,
Slater, et al. and Hart and Sure for humin nitrogen were made to include
the total nitrogen in the insoluble residue and also the humin in solution.
It has been the experience of the writer that in cases where considerable
insoluble residue was included in the total volume of the hydrolyte, great
difficulty was met with in obtaining uniform aliquots for the total
nitrogen in the solution and also for the aliquot for determination. This
difficulty is avoided by filtering out and washing the insoluble residue.
Then, too, the question arises, Is it fair to consider the nitrogen remaining
in the insoluble residue as humin nitrogen?
The results for arginin show no serious loss in any of the determina-
tions, and the minimum result obtained is only 1.5 per cent below Van
Slyke's result for arginin on casein alone.
The histidin results are practically the same for the two 12-hour
digestions, on casein alone and on the casein-starch mixture. In the
15-hour casein-starch digestion the result for histidin is 0.61 per cent
above that reported in Van Slyke's analysis. In the 24- and 48-hour
digestions there is a loss in histidin nitrogen of considerably more than
50 per cent of that found in the 15-hour digestion. Hence, the results
for histidin in the two last experiments are very significant, indicating
that long periods of digestion of starch and casein bring about a redistri-
bution of the nitrogen in this group. It is to be borne in mind that Hart
and Sure^ obtained similar results on lysin. These writers also report
7.30 per cent as a average for histidin determinations in their experiment.
There is a diminution in the cystin nitrogen of more than 50 per cent in
the 24- and 48-hour digestions. Hart and Sure state that their results for
cystin were so low that they reported the results obtained by Van Slyke
instead.
The results for lysin agree well in the 12- and 15-hour experiments. In
the 24- and 48-hour experiments the results for lysin are high. Lysin
nitrogen is obtained by deducting the sum of histidin, arginin, and cystin
nitrogen from the total nitrogen in the bases; therefore any diminution
in the nitrogen content of either histidin, arginin, or cystin will increase
the results for lysin nitrogen correspondingly.
There is no marked difference between the results obtained in all the
experiments for the amino- and nonamino-nitrogen content in the
filtrates from the bases.
In the footings of the different analyses it is to be noted that the
1 2 -hour digestions give footings more than 2.5 per cent over 100. In
the 15-hour digestion the footing is good, while in the 24- and 48-hour
digestions the footings are 2.75 per cent less than 100, thus indicating
that the 12-hour experiments were probably not completely hydrolyzed;
• Hart, E. B., and Sure, Barnett. Op. cit.
6 Journal of Agricultural Research voi. xii, no. i
whereas the 15-hour digestion was sufficient to bring about complete
hydrolysis and the 24- and 48-hour experiments were overdigested to
the extent that nitrogen was lost.
The insoluble carbonaceous residues which were filtered from the
hydrolyzed solutions were dried at 100° C. and the total nitrogen de-
termined in each.
The insoluble residue from experiment 2, or the 12-hour starch-casein
digestion, contained 1.30 per cent of nitrogen. That from the 15-hour
digestion contained 0.83 per cent of nitrogen. That from the 24-hour
digestion contained 0.80 per cent of nitrogen and that from the 48-hour
digestion contained 0.855 per cent of nitrogen. The results show that
a 15-hour digestion removed as much nitrogen from the insoluble residue
as the 24- and 48-hour digestions.
Two determinations of total nitrogen on a sample of the dry starch
showed an average nitrogen content of 0.05 per cent. The small amount
of nitrogen contained in the starch and the comparatively greater amount
found in the insoluble residues indicate that some nitrogen compound
was absorbed by the latter.
Seven gm. of the dry-carbon residue were weighed out and transferred
to a Claisen flask, 60 c. c. of a 10 per cent calcium-hydrate suspension
added, together with 250 c. c. of distilled water. The apparatus was
connected up as in an ammonia determination and distilled under
reduced pressure at from 40° to 45° C. for 30 minutes. Nine-tenths
c. c. of N/io hydrochloric acid was neutralized by the ammonia evolved,
which shows that the insoluble-carbon residue contained only a trace
of ammonia nitrogen. The insoluble-carbon and calcium-hydrate
precipitate remaining in the Claisen flask was filtered and washed
thoroughly, the filtrate made acid and concentrated under reduced
pressure to about 50 c. c. The concentrate was transferred to a Kjeldahl
flask and the total nitrogen determined in the usual way. The filtrate
contained 0.0032 gm. of nitrogen or 5.3 per cent of the total nitrogen
contained in the insoluble residue. The ammonia nitrogen was 2.1
per cent of the total nitrogen in the carbon residue. It is therefore
evident that a very small percentage of the total nitrogen contained in
the insoluble residue is affected by distilling with calcium-hydrate sus-
pension, which indicates that the nitrogen remaining in the insoluble-
carbon residue after digestion and washing is in what may be considered
an inert form and should not be included in the humin group.
CONCLUSIONS
From the data contained in this paper the following conclusions may
be drawn:
(i) The Van Slyke method for protein analysis, when applied to
mixtures of casein and starch in the proportion of i to 5, and hydrolyzed
Jan. 7. 1918 Effect of Time of Digestion on Hydrolysis of Casein
from 12 to 15 hours with 20 per cent hydrochloric acid gives results for
the amino-acid groups that are comparable with those obtained by
Van Slyke upon casein alone.
(2) A digestion period of more than 15 hours with 20 per cent hydro-
chloric acid on a casein-starch mixture brings about a redistribution of
the nitrogen contained in the histidin and cystin groups.
(3) The insoluble residue obtained from a casein-starch digestion
after being thoroughly washed contains nitrogen, which is not seriously
affected when distilled with calcium-hydrate suspension, very small
amounts being split off as ammonia or remaining in the filtrate. This
indicates that the nitrogen is in an inert form and its estimation should
not be included in the humin determination.
BEHAVIOR OF SWEET POTATOES IN THE GROUND
By Heinrich Hasselbring,
Plant Physiologist, Plant Physiological and Fermentation Investigations,
Bureau of Plant Industry, United States Department of Agriculture
THE PROBLEM
In the course of former investigations^ on the behavior of sweet
potatoes in storage, it was observed that the percentage of starch was
always highest and the percentage of sugar lowest in freshly dug pota-
toes. This observation was more or less incidental, having been made
in the course of experiments whose object was the solution of other prob-
lems. It was therefore not based upon a systematic study of the roots
throughout the latter part of the growing season. Nevertheless the con-
stancy of the condition seemed to justify the conclusion that in the
growing sweet potato the reserve materials exist essentially in the form
of starch, and that the appearance of sugar in considerable quantities is
a phenomenon occuring only in storage or after the destruction of the
leaves.
In order to determine whether these quantitative relations between
the starch content and the sugar content of the sweet potato remain
constant throughout the latter part of the growing season, and to what
extent they are changed by the death of the vines, the carbohydrate
metabolism in Big Stem sweet potatoes was followed from the time the
roots were large enough to furnish the requisite samples until they were
seriously damaged by frost.
The record thus obtained of the condition of the potatoes during this
period may be useful as an aid in determining the time for harvesting
the crop; for it is evident that, if marked changes occur in the roots during
the latter part of the season, the time of harvest will depend upon the
purpose for which they are destined, whether for storage, stock feed, silage,
or, as Keitt has suggested, for the manufacture of starch. In the last
case it is evident that the crop should be harvested when the starch
content is greatest. As a rule, growers are advised to dig sweet potatoes
when they are fully matured or after they have thoroughly ripened.
While these phrases imply the idea that the roots reach a more or less
definite state of ripeness, the characteristics by which this state may be
recognized are not precisely defined. On this matter a record of the
1 Hasselbring, Heinrich, and Hawkins, L. A. physiologicai, changes in sweet potatoes duiuno
STORAGE. In Jour. Agr. Research, v. 3, no. 4, p. 331-342. 191S.
CARBOHYDRATE TRANSFORMATIONS IN SWEET POTATOES. In Jour. Agr. Research, V. s,
no. 13, p. 543-560. 1915.
Journal of Agricultural Research, Vol. XII, No. 1
Washington, D. C. Jan. 7, 1918
In Key No. G— 131
(9)
lO
Journal of Agricultural Research
Vol. XII, No. 1
changes in the roots during the latter part of their growth may also
throw some light.
PREVIOUS INVESTIGATION
The only systematic examination heretofore published on the behavior
of sweet potatoes in the ground is that made in South Carolina by Keitt/
who investigated the behavior of four varieties during 1908 and 1909.
Since his data are not easily summarized his tables recalculated on the
basis of dry matter are given here for purposes of comparison with sub-
sequent data.
Table I. — Percentage composition of sweet potatoes in igo8, according to Keitt
Variety.
Water.
Starch.
Glucose.
Sucrose.
69.23
58.11
1.98
13-58
73- 34
55- 93
5-93
6.83
69.08
56-95
6.08
6.18
68.70
63.64
7-35
4-57
68.75
57.02
4.90
7-52
70. 76
52-94
2.77
14. 16
73-85
50-17
6. 42
10.7s
77-89
51. II
8.68
9-23
70.07
51.69
5-98
1-37
72-53
54.06
6-37
7-35
69.79
60.31
4.04
13-31
71-65
54-14
3-67
14.04
70.30
52.86
5.62
9-87
71.30
63.62
7-39
5-64
65-67
56.22
4.19
6-35
68.40
54.08
3-70
15.98
68.17
52.69
6-57
13.10
67.67
54-38
10.83
6. 22
66. 72
63.91
7-69
1.86
67.74
58-03
i.6i
10. 42
Total
carbohy-
drates.
Aug. 28
Sept. 7
Sept. 18
Sept. 29
Nov. 18
Aug. 28
Sept. 7
Sept. 18
Sept. 29
Nov. 18
Aug. 28
Sept. 7
Sept. 18
Sept. 29
Nov. 18
Aug. 28
Sept. 7
Sept. 18
Sept. 29
Nov. 18
Nancy Hall ,
do
do
do
do
Polo.
.do.
.do.
.do.
.do.
Pvirple yam
do
do
do
do
Fulleton yam
do
do
do
do
73-67
68.69
69. 21
75-56
69-44
69.87
67-34
69. 02
59-04
67.78
77. 66
71-85
68.35
76.65
66.76
73-76
72.36
71-43
73-46
70. 06
' Keitt, T. E. the formation of sugars and starch in the sweet potato. S. C. Agr. Exp. Sta.
Bui. 156, 14 p. 1911.
SWEET potato investigation. S. C. Agr. Exp. sta. Bui. 165, 43 p. 1912.
Jan. 7, 1918 Behavior of Sweet Potatoes in the Ground
II
Table II. — Percentage composition of sweet potatoes in iQog, according to Keitt
Date.
Aug. 31
Sept. 10
Sept. 21
Oct. 10
Oct. 26
Aug. 31
Sept. 10
Sept. 21
Oct. 2
Oct. 10
Oct. 26
Aug. 31
Sept. 10
Sept. 21
Oct. 2
Oct. 10
Oct. 26
Aug. 31
Sept. 10
Sept. 21
Oct. 10
Oct. 26
Variety.
Pumpkin yam.
do
do
do
....do
Purple yam.
do
....do
....do
....do
....do
Polo.
.do.
.do.
.do.
.do.
.do.
Brazilian.
....do...
....do...
....do...
....do...
Water.
72.45
66. 12
74.27
68.34
70. 20
66. 97
64.71
62. 42
61.59
60. 91
63-83
66.81
68.24
69.44
67.98
61. 24
66.08
67-45
64.31
65-51
66. 22
69.97
Starch .
58. 22
57.20
55-03
53-03
59.60
63-73
61. 06
69. 10
69. 72
56.46
71. 00
70.44
61. 52
68.52
64-55
58-38
70. 70
70.78
62. 96
69.99
55-21
65-77
Glucose.
13-03
11.98
6.18
3-95
4-5°
8.27
9. 18
3-70
6.77
2. 76
4. 06
7.28
8.21
2. 61
3-85
4.86
Sucrose.
2. 50
5-73
8.94
9-95
14.70
2.32
5-74
3-98
7. 16
7.70
8.26
Total
carbohy-
drates.
73-75
74.91
70.15
66. 93
78.80
77-51
73-75
77-56
81.67
66. 77
81.56
79-63
77-39
78.77
71.79
67-43
81.64
83.80
75-15
79.76
66.76
78. 89
These figures exhibit considerable fluctuation. In 1908 there was as
a rule a decrease in the percentage of starch from August 28 to September
18, then a sudden increase from September 18 to September 29, and an
equally sudden fall from September 29 to November 18. A frost which
killed the vines occurred on November 6. The Polo variety does not con-
form to the others in its behavior. During the same period all the
varieties show a decrease in sucrose up to September 29, and an increase
from that time until November 18. The glucose shows a gradual increase,
reaching a maximum in the different varieties between September 18
and September 29, after which there is a loss of glucose. During the
next year the figures show even greater irregularity. The percentage of
starch fell and rose alternately between each pair of successive dates.
It is notable that after the frost which killed the vines on October 13,
the starch content of all varieties increased considerately. Hence, there
is one year a loss of starch after the death of the vines and in the following
year an increase. The figures representing the percentages of glucose
and sucrose fluctuate irregularly, but in general it may be said that the
glucose fell throughout the season, while the sucrose increased. It is to
be noted, however, that after the frost both sugars increased. Thus, we
have the remarkable phenomenon of an increase in the percentage of
of total carbohydrates in the roots after the vines had been killed.
12 Journal of Agricultural Research voi.xii.no. i
EXPERIMENTAL WORK
In the present investigation Big Stem sweet potatoes were used. They
were grown at Bell Station, Maryland, in a sandy field having a gentle
slope to the south. Lots of 15 to 20 kgm. were collected each week from
September 18 to November 27. The collections were always made in
the afternoon. The roots were thoroughly washed and stored in a
covered receptacle in the laboratory until the following day, when the
samples were prepared for analysis. In each case moisture, starch, cane
sugar, and reducing sugar were determined in duplicate in five potatoes
of the lot.
At the time of the first digging, September 18, the potatoes were still
small, so that it was difficult to get roots large enough to furnish the
requisite samples for analysis. After two or three weeks there was an
abundance of large roots.. The vines remained green and in a healthy
state until the week of October 9. Heavy frosts during that week,
especially on October 15, killed the leaves but not the stems. By
November 7 the stems which remained green after the leaves had been
killed were slowly dying, and some of the potatoes showed small round
injured spots on the portions near the surface of the ground or projecting
above it. On November 21 many of the roots showed considerable
injury on the exposed ends, but sound potatoes were still abundant. At
the time of the last collection, on November 28, the roots were so ex-
tensively damaged that it was difficult to find enough sound ones for
analysis. After that date the experiment was discontinued.
The results of the analyses are given in Table III. For purposes of
discussion the analytical data have been reduced to the basis of dry
matter in the roots. These results are given in the left-hand part of
the table. Since it may be desirable to have for reference a record of
the actual percentages of the different substances contained in the pota-
toes, the original analytical data based on the fresh weight of the roots
are given in the right-hand part of the table.
Jan. 7, 1918 Behavior of Sweet Potatoes in the Ground
13
Table III. — Percentage composition of Big Stem sweet potatoes in the ground during
the latter part of the growing season
Date.
Sept. 18
Sept. 25
Oct.
Oct.
Oct. 17
Oct. 23
Sweet potato No.
13-
14.
US-
fi6.
17-
18.
19.
20.
23-
24.
Us-
f26.
27.
28.
29.
130 ■
Average .
Average .
Average .
Average . .
Average .
Mois-
ture.
73
Average . . .
73
73
92
24
87
77-43
On the basis of dry matter.
Starch.
71
72
70,
39
39
90
68.51
Cane
sugar.
7-36
" 70
76
17
65
93
52
8.86
Reduc-
ing
sugar
as glu-
cose.
I. 20
1.77
I. 14
1.78
I- 31
1-44
I- 31
.87
I. 19
I. 41
1-49
I. 62
1.68
1-45
2.38
I. 72
1. 81
2.31
2.49
2.23
2. 29
2.23
2-49
3-47
2-99
2.51
2-59
2.81
2.30
3.06
2. 06
3-69
2. 91
2.80
Total
carbo-
hy-
drates.
80
81
76
13
43
80. 18
On the basis of fresh
material.
Starch.
17
19
16
63
39
56
15-47
Cane
sugar.
1. 96
2. II
2. 04
2. 16
2.05
2. 06
2.30
1.97
1-77
1. 92
2. 07
2. 01
30
1.99
2-34
2. 01
2. 17
*2. 00
2. 02
1.80
2. 17
1-97
2. 07
2. II
2.05
1. 69
2. 09
2. 00
Reduc-
ing
sugar
as glu-
cose.
37
ii
45
44
55
61
55
59
82
70
59
63
67
53
69
49
78
Jl
•63
14
Journal of Agricultural Research
Vol. xir, No. I
Table III. — Percentage composition of Big Stem sweet potatoes in the ground during
the latter part of the growing season — Continued
Date.
Oct. 30
Nov.
Nov. 13
Nov. 20
Nov. 27
Sweet potato No.
f3i
32
3Z
34
35
Average
36
37
38
39
40
Average
41
42
43
44
U5
Average
[46
47
48
49
^50
Average
5i----
52
53
54
155
Average
Mois-
ture.
77
75
78
77
78
99
14
78
78.93
On the basis of dry matter.
Starch.
64,
63
55
SO'
65
58
27
73
46. 20
Cane
sugar.
10. 65
11.77
10. 56
14.47
8.76
II. 24
14.65
14- 15
14.44
11.82
13-67
13-75
15.24
18.56
18. 20
20. 08
19-93
lb. 40
20. 99
25-31
19.72
25. 16
24-39
23. II
Reduc-
ing
sugar
as glu-
cose.
2.87
2.81
2. 91
4.24
2.98
3- 16
3-40
2.68
2-95
1.86
2.58
2. 69
2-59
2.99
2-45
3-40
5-03
3-29
3-30
3- 75
2.79
3-41
2. 76
3- 20
28. 10
26.59
26. 61
25-52
23.48
26. 06
5.08
3.60
4.04
3.61
2. 91
3-85
Total
carbo-
hy-
drates.
79
80,
76
77
05
19
96
03
76. II
On the basis of fresh
material.
Starch.
15-07
14-37
14-57
12. 67
15.22
14.38 2
Cane
sugar.
14.97
15-51
14.49
16.44
14.95
15-27
13.09
11-95
13.06
10. 91
11-54
12. 06
10. 04
13-30
9. 61
11.56
II. 31
8.77
10. 22
9. 00
10. 18
10.58
9-75
49
29
5-48
Reduc-
ing
sugar
as glu-
cose.
0.66
.62
•65
.90
.67
70
.81
.64
.69
•47
.61
.64
•57
•65
•56
.72
I. 09
72
.76
•78
.67
•70
•63
71
I. 02
•77
.82
.78
.64
An examination of Table III shows that the sweet potatoes exhibit
some individual fluctuation in composition, but these variations are not
sufficiently great to obscure the seasonal trend. They show, however,
that small deviations in the general contour of the seasonal changes are
to be expected where so small a number of individuals is examined. The
significant changes lie clearly outside the limits of the individual fluctua-
tions. The seasonal changes in the percentage of the various constitu-
ents of the sweet potato as shown by the table are briefly described
below.
Jan. 7, 1918 Behavior of Sweet Potatoes in the Ground 1 5
Moisture. — The moisture content of the roots remained almost
uniform during the period covered by the first three collections, from
September 18 to October 2. After that time there was a gradual increase
in the percentage of moisture until the end of the season. A small fluc-
tuation appears in the lot collected on November 6. The changes in
water content of the roots are therefore fairly regular and uniform. No
such marked fluctuations as those recorded in the tables of Keitt are
evident from these data. The beginning of the accumulation of moisture
in the roots is practically coincident with the destruction of the leaves.
Starch. — The percentage of starch in the dry matter of the roots
shows a slight decrease from September 18 until October 23, varying
during that period between 71.39 per cent and 68.51 per cent. On
October 30, the date of the next collection, the starch content had fallen
to 64.65 per cent. From that date the starch content continued to fall
until the end of the season, when the minimum of 46.20 per cent was
reached. The rapid disappearance of starch follows, somewhat delayed,
upon the death of the leaves.
Cane sugar. — The cane-sugar content remains practically constant
between 7.52 and 8.86 per cent until the time when the percentage of
starch begins to fall rapidly. With the decrease in starch the cane sugar
begins to increase correspondingly until it finally represents 26.06 per
cent of the dry matter of the roots. It should be noted, however, that
in general the changes in cane sugar are inaugurated somewhat later than
those of the starch.
Reducing sugar. — ^The reducing sugar content remains constant at
first and then shows a gradual rise until the final percentage is somewhat
more than double that at the beginning. The increase in reducing sugar
antecedes somewhat the rise in cane sugar. It is noteworthy also that
under these conditions reducing sugar apparently does not accumulate
to the same extent to which it accumulates in sweet potatoes in storage.
ToTAi, carbohydrates. — The total carbohydrate content undergoes
very little change until toward the end of the season, when the roots begin
to show marked injury by frost. At that time an evident loss of carbohy-
drates becomes apparent. The constancy of the total carbohydrate con-
tent of the sweet potato is in marked contrast to the fluctuations observed
by Keitt.
The foregoing facts have been embodied in the curves in figure i. The
curves are based on the averages of the analyses for each week.
CONCLUSIONS
It may be concluded from this investigatiou that the changes occurring
in sweet potatoes in the ground during the later part of the growing
season proceed in a regular and orderly manner. During the later part
of the period of growth the composition of the roots remains remarkably
uniform, and presents no striking or irregular fluctuations. During this
i6
Journal of Agricultural Research
Vol. XII. No. I
period the root is characterized by a high starch content, and a low sugar
content. The changes which occur later are associated with the death
of the vines. Prominent among these changes is the accumulation of
water in the roots as a result of the cessation of transpiration in conse-
13/6. SEPT.
OCTOBER
NOVEMBER
Fig. I. — Graphs showing changes in composition of Big Stem sweet potatoes during the latter part of the
season, from September i8 to November 27, and the minimum temperatures at the United States Weather
Bureau Observatory at Washington, D. C, some 20 miles distant, during that period. The ordinates
indicate percentages in the one case and degrees Fahrenheit in the other.
quence of the destruction of the leaves. With the termination of the
flow of materials from the vines the carbohydrate transformations
characteristic of sweet potatoes in storage are inaugurated. These
changes consist in the transformation of starch into sugars. In point
Jan. 7. 1918 Behavior of Sweet Potatoes in the Ground 1 7
of time the decrease in starch and the increase in reducing sugar precede
somewhat the increase in cane sugar. It appears, therefore, that reducing
sugar is formed first as an intermediate step in the change from starch
to cane sugar. The loss caused by respiration, which is considerable
during the curing process and in storage, is apparently slight in sweet
potatoes in the ground. Appreciable destruction of carbohydrates
appears not to occur under these conditions imtil late in the season when
the roots have been injured by frosts.
The changes here described have a practical bearing on the question
of maturation of sweet potatoes and on the choice of the time of harvest.
Since the carbohydrate relations of the roots in the ground remain practi-
cally unchanged while the vines are uninjured, the roots can not be said
to undergo a definite process of ripening, in the sense of a progressive
transformation of one reserve substance into another, such as the change
of insoluble pectin into soluble pectin derivatives in the peach, or the
transformation of starch into cane sugar and invert sugar in the apple.
Under ordinary conditions the potatoes continue to grow until frost
without reaching any definite state of maturity recognizable by pro-
gressive changes in the reserve materials which they contain. The
changes in storage, which may perhaps be regarded as a process of matura-
tion, do not come in for consideration here, since statements relating to
the degree of maturity of sweet potatoes always refer to the growing
roots. It is evident from these considerations that the choice of time of
harvest is not a matter of maturity of the roots, but is governed by other
factors. The potatoes may safely be kept in the ground until the leaves
have been injured by frost.
Of the changes which occur after the destruction of the leaves,
the accumulation of water in the roots deserves foremost considera-
tion. It can scarcely be doubted that this increased water con-
tent is detrimental to the successful storage of the roots, and causes
them to be more subject to decay than roots of normal water content.
One of the objects of the rather expensive operation of curing is to elimi-
nate a part of the water contained in the roots. As a rule observers agree
that cured sweet potatoes keep better than uncured ones. Only occasion-
ally a statement to the contrary is found. It may therefore safely be
assumed that the increase in the relative proportion of water in the roots
will be detrimental to storage. On this account it is of utmost importance
that the harvesting of sweet potatoes be not long delayed after the leaves
have been killed by frost. The other changes occurring in sweet potatoes
in the ground are essentially the same as the changes occurring in storage.
These changes are therefore in no way detrimental to the crop, since no
appreciable loss of carbohydrates occurs until the roots have been so
severely injured that they have lost their market value.
27804°— 18 2
STUDIES IN SOIL REACTION AS INDICATED BY THE
HYDROGEN ELECTRODE
By J. K. Plummer,
Soil Chemist, Division of Agronomy, North Carolina Agricultural Experiment Station
INTRODUCTION
There has been developed in the past few years a rather voluminous
literature dealing with the subject of soil reaction. One has only to scan
this literature in order to find wide variations of opinion between inves-
tigators as to the cause and nature of soil acidity.
Until recently the various qualitative and quantitative methods in
vogue for indicating soil acidity or lime requirement have not been suffi-
ciently delicate to draw definite conclusions as to the "true reaction'' of
soils. The lack of uniformity and accuracy of methods has undoubtedly
caused such confusion on this subject. By adopting modem methods
for measuring soil reaction^many of the contentions should be obliterated.
Though the hydrogen electrode has been used for some time in indi-
cating changes in reaction, Gillespie (7) ^ was the first to use it on an
extended scale as an indicator of reaction in soils. Sharp and Hoag-
land (12) have since measured the H-ion concentration of a number of
oils in suspension, and have extended this method to studying other
soil phenomena.
The significance of the terms "true acidity," "true alkalinity," and
"true neutrality" need not be defined here, except in so far as an expla-
nation of the method adopted in reporting results obtained. Pure water
dissociates into H and OH ions in equal concentration. The product of
the concentration of these ions in a solution is a constant, approximately
I X 10— 14. When the H ions are present in a concentration greater
than I X 10—7, the solution is acid; the presence of OH ions in greater
concentrations than 1X10—7 results in an alkaline solution. For a
more detailed discussion of this subject, the reader is referred to texts
on electrochemistry.
The investigations herein reported were begun about the time of the
appearance of Sharp and Hoagland's paper, to ascertain if appreciable
differences occurred in the H-ion concentration of soils of humid regions,
especially from those of the Southern States. It would appear that with
the excessive rainfall of this region an accumulation of soluble acids in
the soil would be almost impossible.
1 Reference is made by number (italic) to "Literature cited," p. 30-31.
Journal of Agricultural Research, Vol. XII, No. i
Washington, D. C. Jan. 7, 1918
lo - Key No. N. C— 7
(19)
20 Journal of Agricultural Research voi.xii, no. i
The method has been used for indicating the reaction of the soil film
water. The effects of several fertilizer materials have been studied on
the H-ion concentration of various field soils, receiving applications of
such fertilizers for a number of years. Lastly, the effect of ammonium
sulphate and monocalcium phosphate on the reaction of soil film water
have been investigated.
METHODS OF INVESTIGATION
The apparatus for measuring the H-ion • concentration of soil suspen-
sions and extracts was essentially that described by Hildebrand {8), and
modified by Sharp and Hoagland. Palladium was substituted for plati-
num as the electrode, which was treated in a manner similar to that
described by Findlay (6) for coating electrodes. The supply of hydrogen
was obtained by electrolyzing water, with potassium hydro xid as the
electrolyte. All the precautions of rigidity of connections, insulation of
apparatus, time (which often varied) for estabhshment of equihbrium
between soil and solution, coating electrodes, and prevention of loss of
CO2 were strictly observed.
It might be said in passing that no difficulties were encountered due
to the reduction of nitrates to ammonia, as has been suggested.
In the preparation of soil suspensions, unless otherwise stated, the
arbitrary ratio of 10 gm. of air-dried soil to 100 c. c. of as pure distilled
water as obtainable was used. In the case of field soils their content was
determined as soon after sampling as practicable, and the same ratio of
soil to water maintained. All samples except those taken directly from
the field were screened through a 2-mm. sieve.
In all cases the readings became constant in a few minutes. Dupli-
cate readings on the same sample of soil could easily be read to 0.02 volt.
However, to be certain 'that equilibrium had been established the elec-
trode was allowed to remain in contact with solution for 30 minutes.
It was almost impossible to get such closely agreeing results as 0.02
volt with different samples of the same soil. This can be accounted
for in the lack of uniformity of mixing.
The results are reported in the usual manner for such measurements,
units of gram-molecules of H ion per liter. The tables of Schmidt {11)
were used in securing the H-ion concentration from the voltmeter read-
ings.
RESULTS OBTAINED
In Table I will be found the results of measurements of the H-ion
concentrations of 68 samples of untreated soils, including subsoils, which
represent a wide range in types of five series. Included in this table
are results derived from five samples of treated soil. The samples are
taken as a fair representation of the soils common to the area of the
southeastern portion of the United States, which extends from and
including the Appalachian Mountains to the Atlantic Ocean.
Jan. 7, 1918
Studies in Soil Reaction
21
Table I. — Hydrogen-ion concentration of soil suspensions
Sample
No.
Soil type.a
Depth.
Volt-
meter
readmgs.
H-ion concen-
tration.
1565
1566
1067
1068
1501
1502
1287
1288
1397
1398
1483
1484
1344
1345
991
992
1552
1029
1030
1519
1520
1499
1500
1352
1353
1287
1288
I125
I126
1261
1262
1337
1205
1333
1322
1323
1435
1436
1121
1122
1344
1345
1328
1329
1257
1258
395
396
397
398
639
640
6652
653
Norfolk sand (sand-hill phase).
....do
....do
....do
Norfolk coarse sandy loam ....
....do
....do
....do
Norfolk sandy loam
....do
....do
....do
Norfolk line sandy loam
....do
....do
....do
Norfolk very fine sandy loam.
do
do
do
Norfolk silt loam
do
do
do
Cecil coarse sandy loam
do
do
do
Cecil fine sandy loam'.
do
do
do
Cecil clay loam
do
do
do
Cecil clay
do
do
do
Iredell fine sandy loam
do
do
do
Iredell loam
do
do
do
Porter's sand
do
do
do
Porter 's sandy loam
do
do
do
Porter's loam
Inches.
0-4
4-21
0-4
4-18
0-6
6-28
0.6
6-26
0-7
7-28
0-7
7-28
0-6
6-36
0-6
6-36
0-6
6-24
0-8
8-30
0-6
6-28
0-6
6-32
c^7
7-20
c^6
6-30
0-8
8-24
0-8
8-26
0-6
6-30
0-6
6-36
0-7
7-20
0-6
6-36
0-6
6-36
0-8
8-24
0-8
8-24
0-6
6-26
0-6
6-20
0-6
6-20
0-7
7-28
0-7
7-30
0-8
700
684
712
706
684
684
696
681
674
668
662
674
660
671
645
640
657
660
614
583
574
568
560
555
712
699
706
700
702
681
699
673
652
666
659
643
641
582
646
607
732
738
711
704
692
714
684
679
673
668
640
638
648
641
656
649
644
Gram mole-
cules per liter.
o. 4X10"^
9X10-6
3 X 10-6
3X10-6
9X10 6
9X10-8
5X1O-S
oX 10-6
I X 10-5
I X 10-5
2 X 10-5
I X 10-5
2 X 10-5
I X 10"
4x10-5
5x10-5
2x10-5
2 X 10-5
I X 10-4
5x10-4
7x10-4
9Xio~4
I X lo-a
1X10-*
3X10-8-
5X10-8
3X10-6
4X10-6
4X10-8
oXlO-8
4x10-6
I X 10-5
3X10-5-
I X 10-5
2 X 10-5
5X10-5
5X10-5
5X10-4
4X10 5
I X 10-4
I X 10-6
I X 10-8
10-8
10-6
10-6
3X1
4Xj
6X:
2 X 10-6
9X10-6
I X 10-5
I X 10-5
I X 10-5
5x10-5
5x10-5
3X10 5
5x10-5
2 X 10-5
3x10-5
4x10-5
1 The writer is indebted to Dr. W. H. Mclntire, of the Tennessee Agricultural Experiment Station, for
the samples of the treated soils.
22
Journal of Agricultural Research
Vol. XII, No. I
Table I. — Hydrogen-ion concentration of soil suspensions — Continued
Sample
No.
Soil type.
Depth.
Volt-
meter
readings.
H-ion concen-
tration.
I202
1372
1373
674
684
68s
1007
1212
1077
2526
Porter's loam
do
do
Porter's clay
do
do
do
Muck
do
do
do
Chickamauga limestone soil«.
Chickamauga limestone soil &.
Cumberland loam «
Cumberland loam &
Tillico sandy loam a
Inches.
8-30
o-l
7-24
0-7
7-24
0-6
6-20
0-18
0-24
0-36
0-36
• 639
.647
• 639
•577
.566
•594
•587
•483
. 462
•536
.427
.864
.824
•893
.856
.866
Gram rtiole-
cules per liter.
.5X10-5
.3X10-5
.5X10-5
.6X10-*
.1X10-3
.3X10-*
.4X10-*
.2X10-2
.6X10-2
.3X10-3
.2X10-1
.7X10-9
.3X10-8
.2X10-9
.9X10-9
.6X10-9
o Treated in 1912 with 16,000 pounds of calcium carbonate per acre in excess of Vietch indication.
6 Treated in 1912 with MgCOsOCaCOs at rate of 16,000 pounds per acre in excess of Vietch indication.
The results shown in Table I indicate wide variations in the H-ion
concentrations of the untreated soils under experiment. This vaijies
from very nearly neutral in some of those of the Iredell series to rather
excessive acidity in the Norfolk silt loam and mucks. This is in accord
with what would be expected. The Iredell soils are of residual origin
formed from basic eruptions, mainly diorite. The amount of basic
elements supphed this soil is greatly in excess of those commonly found
in the area included in this study. The Norfolk silt loam being a trans-
ported soil, was formed under conditions through which the basic ele-
ments have been removed. It also contains rather high amounts of
partially decomposed organic matter, and should show a high concen-
tration of H over OH ions. Field and pot tests with various crops have
shown indications of excessive acidity on some of the muck soils. Indeed
on some of the muck fields from which the samples were taken little or
none of the common agricultural crops will thrive until the land has
been limed.
No definite relationship appears to exist between the H-ion concen-
tration and types of different texture. It can be noticed that there is a
tendency for the H ion to increase as the number of fine particles increase
in a series. However, there are exceptions. The clay loams and loams
of the Iredell and Porter's series show greater H-ion concentration than
do those of coarser texture. It may be observed that in many cases the
subsoil shows a greater degree of acidity than the corresponding surface
soil. The general practice in farming these soils has been the removal
from the land of all crops. Probably there is a tendency for plant roots
to remove more bases from lower depths than from the surface.
Jan. 7, 1918 Studies in Soil Reaction 23
The samples of soil which have been treated witii excessive amounts of
calcium and magnesium carbonate show a greater concentration of OH
than H ions in solution. Dr. Mclntire informed the writer that the car-
bonates have long since disappeared in these soils. Obviously the new
compounds of calcium and magnesium, whatever they may be, give a
strong basic reaction.
These results bear out very forcibly the contentions of Gillespie (7) and
Sharp and Hoagland {12) that there is a preponderance of H over OH ions
in the liquid phase of many soil suspensions. Whether the " true acidity "
as indicated from the results of Table I is developed from organic acids
or acid silicates can not be stated. In some of the sandy soils the organic
matter is quite low, yet indications lead to the conclusion that there is a
greater concentration of H ions than found in neutral solutions. With
the muck samples the organic -matter content is very high, being 90 per
cent or more, and most marked acidity is shown.
H-ION CONCENTRATION OF SOIL FILM OR CAPILLARY WATER
The question has often been raised, "Is the water held as a liquid film
around the soil particles of different reaction from that in the free state ? "
The difficulty encountered in securing any workable amount of the film
water has militated against any direct study of this question. Recently,
Morgan (9) has devised a workable modification of the "oil displacement "
method for obtaining the soil solution in an unaltered form. A study of
this water should throw much light on some of the obscure problems of
soils. One serious difficulty yet remains with the method, which is its
inability to get back all of water held as thin films around the soil grains.
However, a study of the reaction of that portion of the film water obtain-
able should give indications of value as to the conditions of the soil
solution.
The Morgan apparatus, with a few unimportant modifications, has
been used in obtaining that portion of the capillary water displaced by
paraffin oil under high pressure. The oil used in this work was the
purest obtainable. As far as could be noticed, it gave a neutral reaction
with the hydrogen-electrode apparatus and by titrating against standard
alkali with methyl orange, methyl red, and phenolpthalein as indicators.
Some of the soil samples used were taken directly from the filed;
others had been stored in the laboratory for a number of years. The
portion which passed a 2-mm. sieve was thoroughly mixed, and the water
content determined. Sufficient distilled water was added to bring the
moisture content up to about what is considered the optimum for that
soil. Water lost through evaporation was replaced at frequent intervals,
after which the soils were thoroughly packed in the cylinder and treated
with oil under high pressure.
The results of this investigation are shown in Table II
24
Journal of Agricultural Research
Vol. XII, No. I
Table II. — Reaction of soil-film water
Film water.
Free water.
5
Soil type.
1
h
m ll> 2
'o
0
a
"3
.1
5
1
"3
1
1
0
0
a
8
"o
.1
a
'•3
«
1
H-ion concentratic
Cecil clay loam
Lbs.
lO
lO
P.d.
25
25
p.d.
7-2
6.9
4
4
C..C
5°
50
G»w.
C.c.
Daj'J.
Gram-mole-
cules per
liter.
0. 8+ 10-*
Do
S88
679
593
604
684
614
624
661
666
654
694
929
877
976
806
. 4+ IC-*
Do
5
so
4
.1 + 10-*
Porter's loam
lO
lO
25
25
6. 4
4-9
,-
.3 + IO-*
Do
4 1 50
.2 + 10-*
Do
5
50
4
.9+10-8
Norfolk fine sandy loam
lO
lO
20
20
15-3
13-3
3
3
50
SO
.i + io-<
Do
.1 + 10-*
Do :
S
50
3
. 2 + IQ-*
Norfolk sand
lO
lO
15
15
18.4
20. 6
2
2
50
50
.1 + 10-5
Do
.3 + 10-5
Do
S
50
2
. 6+ 10-5
Tellico sandy loam (^
5
20
12.8
3
25
.S + iQ-iO
Do
z-S
25
3
.4+10-9
5
25
7- 7
3
25
.8+10-11
Do
2-5
25
3
.1 + 10-9
o Sufficient soil not available for duplicate extractions.
The data presented in Table II show rather conclusively that the
soil-film water has the same reaction as the free water. The difference
is only in degree of intensity. Those soils which showed an acid reaction
in suspensions gave a greater concentration of H ions in the film water.
Conversely, those which indicated a greater OH-ion concentration than
H in suspension gave a greater intensity of OH ions in the solution.
These results are not in keeping with those of Sharp and Hoagland (12),
who found no appreciable change in H or OH ions in varying the pro-
portions of soil to water in making suspensions.
With the freezing-point method for measuring the concentration of
the soil solution Bouyoucos and McCool (j) show that as the amount of
water decreases in an arithmetic progression the lowering of the freez-
ing point (increase in concentration) increases in a geometric progression.
This apparently is due to the rendering inactive of a portion of the water
in the soils, hence this water does not take part in dissolving the solutes
of the soil. That portion left free or tmcombined becomes highly charged
with soluble salts and gives phenomenal increases in the freezing-point
lowering. The same line of reasoning may be applied to the increase in
intensity of reaction of film water when compared to that of soil sus-
pensions.
With the heavier types of soil only a small percentage of the added
water was recovered. It would be of extreme interest to note the
intensity of reaction of the thinnest moisture film which could remain
in contact with the soil grains. In other words, all of the liquid added
to the soil should be recovered and studied.
Jan. 7, 1918
Studies in Soil Reaction
25
EFFECT OF FERTILIZER MATERIALS ON SOIL REACTION
It has frequently been stated that certain fertilizer materials, more
especially ammonium sulphate and acid phosphate, increase the acidity
of soils. To obtain additional evidence, the H-ion concentration of
soil suspensions have been measured in samples taken from the ferti-
lizer plots of the North Carolina Experiment Station. Some of the
soils have received annual applications of these fertilizers for as many
as 15 years. All fields are located on well-defined soil types for each
area of the State. Three distinct types have been studied.
The samples of both soil and subsoil were collected as carefully as
possible. Borings were made at several points on the plots for both
surface and subsoil samples. These were taken to the laboratory as
quickly as possible and thoroughly mixed, after which water was deter-
mined in each composite sample. The ratio of water to soil was kept
the same as with the air-dried soils.
Table III gives a compilation of the total amount of each fertilizer
material and lime which has been applied to the plots at the different
branch stations.
Table III.
-Total quantity {in pounds per acre) of fertilizer materials applied to
experimental plots
Branch station.
Ammo-
niiun sul-
phate (20
per cent
nitrogen).
Sodium
nitrate
(14 per cent
nitrogen).
Dried blood
(14 per cent
nitrogen).
Acid
phosphate
(16 per cent
phosphorus
pentoxid).
Potassium
sulphate,
(40 per cent
potassium
oxid).
Lime
(90 per cent
calcium
carbonate).
Buncombe
4,826
1,288
"■i,'638'
1,846
3>332
3,237
3,372
I, 100
540
307
457
6, 000
Iredell
577
577
577
675
675
67s
4, 000
4, 000
4, 000
Central
Edgecombe
The data secured from the measurements of the H-ion concentration
of plots fertilized with ammonium sulphate and sodium nitrate are given
in Table IV.
Marked increases in acidity can be noticed in samples taken from plots
fertilized with ammonium sulphate. In every case pronounced increase
of H ions is evident, and extends to the subsoil in all of the fields studied.
On the Norfolk fine sandy loam more acidity is noted in the subsoil than
surface.
There does not appear to be any relationship between texture and
development of acidity by ammonium sulphate. This is in agreement
with the work of Allison and Cook (/).
The results secured from the sodium-nitrate plots are in harmony with
the accepted theory regarding its effect on soil reaction. A reduction of
acidity is apparent.
26
Journal of Agricultural Research
Vol. XII. No. r
Table IV. — Effect of ammonium sulphate and nitrate of soda on the H-ion concentration
of field-soil suspensions
Branch station.
Year of
begin-
ning
experi-
ment.
Soil type.
Depth.
Treatment.
Volt-
meter
read-
ings.
H-ion con-
centration.
Iredell
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
1907
Cecil clay loam
Inches.
0-6
6-24
0-6
6-24
0-6
6-24
0-8
8-24
0-8
8-24
0-8
8-24
0-6
6-24
0-6
6-24
0-6
6-24
Ammonium sulphate . . ,
do
0 ';42
Gram-mole'
cules per
liter.
0.2X10-'
Do
do
568
694
678
656
649
548
582
671
656
613
603
556
548
666
648
646
6?8
.9X10-*
Do
do
Sodium nitrate
.6Xio-«
Do
. ...do
do
.iXic-*
Do
do
Untreated
do
Ammonium sulphate . . .
do
.2X1C-*
Do
do
.3X1C-*
Central
Do
Durham sandy loam . . .
do
.2Xio-»
• sXic-*
Do
... do
Sodium nitrate
.1X10-*
Do
do
do
.2X10-5
Do
do
Untreated
.iXio-<
Do ■
do
do
.2XlO-«
Edgecombe ....
Do
Norfolk fine sandy loam
. . do
Ammonimn sulphate . . .
.do...
.1x10-8
.2x10-8
' Do
do
.iXio-s
Do
do
do
.3X10-5
Do
do
.4X10-S
Do
1907
do
do
. 8X lo-s
EFFECT OF POTASSIUM SULPHATE ON SOIL REACTION
Skinner and Beattie {13) and others have observed that potassium
sulphate increased the lime requirement of soils. Measurements have
been made of the H-ion concentration of soil suspensions from plots to
which have been added different amounts of this salt. These data are
given in Table V.
Table V. — Effect of potassium sulphate on the H-ion concentration of field-soil sus-
pensions
Branch sta-
tion.
Buncombe .
Do
Do
Do
Do
Do
Do
Do
Iredell
Do
Do
Do
Do
Do
Do
Do
Edgecombe
Do
Do
Do
Do
Do
Do
Do
Year
of
begin-
ning
ejcperi-
ment.
Soil type.
191 1
1911
191 1
1911
1911
1911
1911
1911
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
Porter's loam.
....do
....do
do
do
do
do
do
Cecil clay loam.
do
do
.do.
.do.
.do.
.do.
do.
Norfolk fine sandy loam
do
do
.do.
.do.
.do.
.do.
.do.
Depth.
Inches.
0-&
8-24
8-24
0-8
8-24
0-8
S-24
0-8
8-36
0-8
8-36
0-8
8-36
0-8
0-36
0-8
8-24
0-8
8-24
0-8
8-24
0-8
8-24
Treatment.
Dried blood
do
Dried blood, potassium
sulphate.
do
Potassium sulphate
do
Untreated
....do
Dried blood
....do
Dried blood, potassium
sulphate.
do
Potassium sulphate
do
Untreated
....do
Dried blood
...do
Dried blood, potassiimi
sulphate.
....do
Potassium sulphate
....do
Untreated
....do
Volt-
meter
read-
ings.
0.656
.654
.623
.623
. 609
• 596
.648
.651
.682
. 671
.630
.618
.621
.608
.676
. 672
.632
. 620
. 602
•S90
•596
. 600
•634
.628
H-ion con-
centration.
Gram-
nwlecules
per liter.
o. 2X10-5
.3X10-5
.iXio-<
.iXio-<
.1X10-*
.3Xio-<
.3X10-6
.3X10-5
. I X 10-8
.1X10-5
.8X10-5
.1X10-
.1X10-
.1X10-
.1X10-'
.iXio-
.7X10-
.iXio-
.2X10-'
.3X10-
.3X10
.2X10-
.6X10-5
.8X10-5
Jan. 7, 1918
Studies in Soil Reaction
27
Slight increase in the H-ion concentration was obtained from the
plots to which potassium sulphate had been applied. However, nothing
like as marked an effect in producing "true acidity" is found on these
plots as those to which have been added ammonium sulphate. Why
this should be is not clear, unless it has been caused by nitric acid devel-
oped by the soil organisms.
EFFECT OF ACID PHOSPHATE AND LIME ON SOIL REACTION
Table VI gives the results from the plots which have received annual
applications of acid phosphate and lime.
Table VI. — Effect of acid phosphate and lime on the H-ion concentration of field-soil
suspensions
Branch sta-
tion.
Buncombe .
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Iredell
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Edgecombe
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Year
of
begin
ning_
experi-
ment.
191 1
19 1 1
1911
1911
1911
1911
1911
1911
1911
1911
1911
1911
1911
19H
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
1903
Soil type.
Porter's loam.
...do
....do
....do
....do
.do.
.do.
do.
do.
do
do
do
do
...do
Cecil clay loam.
...do
....do
do
....do
.do.
.do.
.do.
-do.
do.
.do.
.do.
.do.
do.
Norfolk fine sandy loam.
do
do
do
do
do.
.do.
.do.
.do.
do.
.do.
.do.
do.
.do.
Depth.
Inches.
0-8
8-30
0-8
8-30
0-8
8-30
0-8
8-30
0-8
8-30
0-8
8-30
0-8
8-30
0-6
6-30
0-6
6-30
0-6
6-30
0-6
6-30
0-6
6-30
0-6
6-30
0-6
&-30
0-6
6-24
0-6
6-24
0-6
6-24
0-6
6-24
0-6
6-24
0-6
6-24
0-6
6-24
Treatment.
Dried blood
.....do
Acid phosphate
.....do
Dried blood, add phos-
phate.
.....do
Dried blood, acid phos-
phate, potassimn sul-
phate.
do
Dried blood, acid phos-
phate, potassium sul-
phate, lime.
do
Lime
do
Untreated
.....do
Dried blood
.....do
Acid phosphate
.....do
Dried blood, acid phos-
phate.
do
Dried blood, acid phos-
phate, potassium sul-
phate.
do
Dried blood, acid phos-
phate, potassium sul-
phate, lime.
do
Lime
do
Untreated
...do
Dried blood
...do
Acid phosphate
...do
Dried blood, acid phos-
phate.
...do
Dried blood, acid phos-
phate, potassium sul-
phate.
...do
Dried blood, acid phos-
phate, potassium sul-
phate, lime.
....do
Lime
....do
Untreated
....do
Volt-
meter
H-ion con-
read-
centration.
mgs.
Gram-
mclecuUs
per liter.
0.656
0.2X10-5
•654
• 3X10-5
.662
. 2X10-^
•6SS
.2X10-*
.652
.3X10-5
.660
.2X10-^
.612
.1X10-*
. 610
.1X10-6
.702
.4Xio-«
.716
.2X10-*
.728
.1X10-8
•736
.iXio-«
.648
.3X10-6
■651
.3X10-6
.682
i.oXio-6
.671
.1X10-6
.674
.1X10-6
.679
.1X10-6
.684
.9X10-6
.666
.1X10-6
.613
.1X10-*
.609
.iXio-<
.712
.3X10-6
•732
.1X10-6
.746
. 7X10-'
•754
.5X10-'
.666
.1X10-6
.6si
.3X10-6
.632
.7X10-^
.620
.iXio-<
. 624
.1X10-*
.624
.iXio-*
. 614
.iXio-«
.609
.iXio-<
.600
.2Xio-<
•S92
• 3Xio-«
.712
• 3X10-6
.716
.2X10-6
.726
.iXio-«
•744
.8X10-'
.638
.SXio-6
.628
.8X10-6
28 Journal of Agricultural Research voi.xii, no. i
The results from the plots to which acid phosphate has been added
do not show any greater H-ion concentration than the ones used as
controls. These plots have received rather heavy annual applications
of this fertilizer for the past 15 years, the total amount applied being
over 3,000 pounds per acre. The fine sandy loam may be some excep-
tion. In this case the readings are so nearly the same as from those
plots which have received no fertilizer that the differences are within
the range of experimental error. Indeed there is as much evidence in
indicating an increased basicity from the use of acid phosphate in the
clay loam and loam as from an increase in acidity in the sandy loam.
This is in agreement with the more recent work of Conner (5), Brooks (4),
and Bear and Salter (2).
Additions of lime alone or lime in combinations with the fertilizer
materials have materially reduced the acidity in all plots. This is often
more marked in the subsoil than in the surface. With the exception of
the Cecil clay loam, lime has not been used in sufficient amounts to pro-
duce basicity.
EFFECT OF AMMONIUM SULPHATE AND MONOCALCIUM PHOSPHATE
ON H-ION CONCENTRATION OF SOIL-FILM WATER
The results heretofore reported in this paper with ammonium sulphate
and acid phosphate have been derived from soil suspensions. The ques-
tion arises. Is the film water of the soil affected in the same or different
manner from the free water? To secure data on this question three
soils were treated with the ammonium sulphate and monocalcium phos-
phate at optimum moisture conditions and extractions made with the
Morgan apparatus. The materials were applied in solution as a fine
spray over the soils in order to get as good distribution as possible.
The soils were well worked after the additions to get a uniform mass.
Monocalcium phosphate was substituted for acid phosphate on account
of its complete solubility. Table VII gives the data derived from
treatment with ammonium sulphate.
The film water is shown to be more strongly acid from the treatment
with ammonium sulphate than that developed when the same amount
of salt is applied in suspension. The indications from this are that
methods for estimating soil reaction or lime requirement based on treat-
ing the soil with a neutral solution do not give the total acidity in the
filtered extract.
The mechanism of this reaction has been the subject of much conten-
tion. The explanation which has been offered that the basic radicle has
been absorbed by celloidal material; and the acidity developed from
the combination of SO4 with 2H of the slightly ionized HjO leaves an
unbalanced equation. Parker {10) contends that the fine soil particles
catalyze the reaction (NHj2S04 + 2HOH = 2NH,OH-}-H2S04 with
the removal of the entire base from solution by selective adsorption
Jan. 7, 1918
Studies in Soil Reaction
29
phenomena. The contention that the base has been removed by com-
bination with the difficultly soluble acids more nearly agrees with the
results obtained. The measurements showing the effect of monocal-
cium phosphate on the H-ion concentration of soil-film water are given
in Table VIII.
Table VII. — Effect of ammonium sulphate on the H-ion concentration of soil-film
water
Film water
Free water.
■3
c2
3 2
>
B
•6
a!
4-»
1
.3
3
3
CD
3
Soil type.
^^t
O'm
u
— ij
0
"o
oit
8
>.
.^fi
0 a
.1
-1
SI
8.i
.^i
>.
J -a
fe 1
It
^
n
3 2
"0
'0
"3
1
3
II
3 2
3
a
3
as
1
§
a
OH
;^
s
>
>
s
a
Oti
a
>
a
Gram-mole-
Grantrmole-
ciUes per
cules per
Lbs.
Gm.
p. a.
p:ct.
C.c.
liter.
Gwi.
Gm.
c.c.
c.c.
liter.
Cecil day loam ....
10
10
25
6.0
so
0.523
0.5X10-3
5
0. II
5°
so
0. 644
0. 4X io-«
Do
10
20
25
4.8
SO
.518
.7X10-''
5
. 22
50
50
. 612
.iXio-<
Do
so
• 496
.1X10-2
5
• 33
50
50
• 591
■ 3Xio-«
Do
S
5°
50
• 674
.iXio-s
Porter's loam
10
10
25
13.6
SO
■532
.4X10-3
S
. 11
SO
50
. 602
.2X10-*
Do
10
20
25
14. 1
SO
•S03
.1X10-2
S
. 22
50
so
.588
.4Xio-«
Do
10
30
25
II. 2
SO
. 461
.6X10-2
5
•33
SO
so
•572
.8Xio-<
Do
s
SO
50
.686
.9Xio-«
Norfolk fine sandy
loam
10
10
20
10. 6
SO
• 540
.2X10-3
s
. II
SO
5°
•S84
• SXio-<
Do
10
20
30
20
II. 9
13-2
SO
.528
.4X10-3
i.oXio-3
s
5
. 22
• 1?
50
50
SO
S°
•578
.566
.6Xio-<
Do.
.1X10-3
Do
S
SO
SO
.670
.1X10-5
Table VIII. — Effect of monocalcium phosphate on the H-ion concentration of soil-
film luater
Soil type.
Quan-
tity of
soil.
Quan-
tity of
mono-
calcium
phos-
phate
Mois-
ture
con-
tent,
dry
basis.
Mois-
ture
recov-
ered.
Vol-
ume of
film-
water.
Volt-
meter
read-
ings.
Per
cent.
C.c.
7-4
50 0
(^•3
50
6.1
50
II. 8
50
10. 9
50
14.4
50
12. 2
50
II. I
50
II. 9
50
H-ion concen-
tration.
Cecil clay loam
Do
Do
Porter 's loam
Do
Do
Norfolk fine sandy loam
Do
Do
Pounds.
10
10
ID
10
10
Cm.
10
20
30
10
20
30
10
20
30
Per
cent.
25
25
25
25
25
25
20
20
20
562
598
602
562
602
591
Gramrmolecules
per liter.
0.4X10-4
.3X10-*
■3X10-^
.2X10-*
. 2 X IO-*
.1X10-3
.2X10-*
•3X10-"
.4X10-*
By comparing the data of Table II with those derived from this experi-
ment it is apparent that only excessive applications of monocalcium
phosphate have increased the H-ion concentration. The 20-gm. appli-
cations of the salt do not show any increase in "true acidity" with any
soil used. The clay loam and loam give a higher H-ion concentration
30 Journal of Agricultural Research voi. xii, No. x
with the 30-gm. addition. The iine sandy loam shows more "true
acidity" with a 20-gm. application than when 10 gm. are added, and
still more when 30 gm. have been applied.
The fixation or removal from solution of phosphates is supposed to
be done by the bases, such as iron, aluminum, calcium, etc., in the soil.
The clay loam and loam soils are well supplied with very fine particles
of iron and aluminum compounds. They therefore have the. capacity of
fixing more soluble phosphate than the fine sandy loam, which has a
relatively low content of bases. These data are in accord with those
obtained by Conner on soils of Indiana.
SUMMARY
The hydrogen electrode has been used for indicating soil reaction on
a number of untreated soils in suspension. The soils experimented with
represent a wide range in texture of those common to the area of the
southeastern portion of the United States, extending from and including
the Appalachian Mountains to the Atlantic Ocean. The H-ion concen-
tration varies from almost "true neutrality" to rather excessive "true
acidity" in the soils.
With the Morgan apparatus for extracting film water from soils, it
is shown that its reaction is the same as the free water, differing only in
intensity.
The effects of certain fertilizers on the H-ioh concentration of long-
time-treated plots of three soils have been measured, with the following
results: (i) Ammonia sulphate has materially increased the H-ion
concentration of all plots which have received applications of this material.
The acidity thus developed extends often to the subsoil. (2) Sodium
nitrate has slightly reduced the acidity of the plots to which it has been
applied. (3) Potassium sulphate increases the "true acidity" when
applied to soils, though not as greatly as ammonium sulphate. (4) Acid
phosphate does not appear to have affected in either direction the H-ion
concentrations of field soils. (5) Lime materially increases the OH-ion
concentration of field plots to which it has been added.
The acidity developed from ammonium sulphate is more intense in the
film than in the free water of three soils.
Monocalcium phosphate does not change in any way the soil-film water
until excessive amounts are added.
LITERATURE CITED
(i) AxLisoN, F. E., and Cook, R. C.
191 7. THE EFFECT OF AMMONIUM SULPHATE ON SOIL ACIDITY. In Soil vScicnCC,
V. 3, no. 6, p. 507-512, I fig.
(2) Bear, F. E., and Salter, R. M.
1916. THE residual effects OF FERTILIZERS. W. Va. Agr. Exp. Sta. Bui.
160, 26 p., 2 diagr.
Jan. 7. i9i8 Studies in Soil Reaction 31
(3) BouYOUCOS, G. J., and McCooL, M. M.
1916. FURTHER STUDIES ON THE FREEZING POINT LOWERING OF SOILS. Mich.
Agr. Exp. Sta. Tech. Bui. 31, 51 p., i fig.
(4) Brooks, W. P.
1915. PHOSPHATES IN MASSACHUSETTS AGRICULTURE; IMPORTANCE, SELECTION
AND USE. Mass. Agr. Exp. Sta. Bui. 162, p. 131-167, 2 pi.
(5) Conner, S. D.
^ 1916. acid soils and the effect of acid phosphate and other fertilizers
UPON THEM. In Jour. Indus, and Engin. Chem., v. 8, no. i, p. 35-40,
2 fig.
(6) FiNDLAY, Alexander.
1906. PRACTICAL PHYSICAL CHEMISTRY. 282 p., 92 fig. London, New York
and Bombay.
(7) GaLESPiE, L. J.
1916. THE REACTION OF SOIL AND MEASUREMENTS OF HYDROGEN-ION CONCEN-
TRATION. In Joiu-. Wash. Acad. Sci., v. 6, no. i, p. 7-16, 2 fig.
(8) HiLDEBRAND, J. H.
I913. SOME APPLICATIONS OF THE HYDROGEN ELECTRODE IN ANALYSIS, RE-
SEARCH AND TE.^CHiNG. In Joiu. Amer. Chem. Soc, v. 35, no. 7,
p. 847-871, 15 fig.
(9) Morgan, J. F.
191 7. THE soil SOLUTION OBTAINED BY THE OIL PRESSURE METHOD. In Soil
Science, v. 3, no. 6, p. 531-546, i pi. Literature cited, p. 544-545.
(10) Parker, E. G.
1913. SELECTIVE ABSORPTION BY SOILS. In Jour. Agr. Research, v. i, no. 3,
p. 179-188, 2 fig.
(11) Schmidt, C. L. A.
1909. TABLE OF H -f- AND OH — CONCENTRATIONS CORRESPONDING TO ELECTRO-
MOTIVE FORCES DETERMINED IN GAS-CHAIN MEASUREMENTS. In Univ.
Cal. Pub. Phys., v. 3, no. 15, p. 101-113.
(12) Sharp, L. T., and Hoagland, D. R.
1916. ACIDITY AND ADSORPTION IN SOILS AS MEASURED BY THE HYDROGEN
ELECTRODE. In Jour. Agr. Research, v. 7. no. 3, p. 123-145, i fig.
Literature cited, p. 143-145.
(13) Skinner, J. J., and BeattiE, J. H.
1917. INFLUENCE OF FERTILIZERS AND SOIL AMENDMENTS ON SOIL ACIDITY.
In Jour. Amer. Soc. Agron., v. 9, no. i, p. 25-35.
ADDITIONAL COPIES
OP THIS PXTBI.ICATION MAY BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT
10 CENTS PER COPY
Subscription Price, S3.00 Per Year
A
CONXKNXS
Pure Ctiltxires of Wood-Rotting Fucgi on Artificial Media
W. H. lONG and R. M. HAJRSCH
( Contribution from Bureau ol Plant lurtustr;.' '
Gossypoi, tbe Toxic Substance in Cottonseed
W. A. WITHERS aad FRAKK E. CAIIRUTH
( ; .-•ntribaiion Ii'cta North Carolina AKricotWral 'KxDerhjrient Station
Frait-FIj Parasitism La Hawaii Dimng 19 Hj - - -
C. E. PEMBERTON and H, F. WILLARD
( CoatdbuUan from Bureau ot Entamolosy)
S3
103,
PDBUSHEft BY AUTHORm OF THE SECRETARY OF AGRICULTURE.
WITH THE COOPERATION OF THE ASSOCIATION OF AAIERICAN
AGRirriTTT'RAf COIJ.VCVP, XMW FTPFRIMET^T STATIONS
WASMINGT01>i, r>. C
w/iMNOTOM '. aOVSR^'Mew
EDITORIAL COMMITTEE OF THE
ONITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
yOR THK OBPAJRriVlBM I
5CARL F, KRLLERMAN, Chaikman
Pkysiologisl and Associate Chief, Buretm
of Plant Industry
KDWINW. ALLEN
Chief, OfJicf. of Exi>eriment Statumi,
CHARLES L. MARLATT
BntomolQgisl and Asiixiani Chief,: B«utti>M
of Rntomotoov
fOR THK ASSOCIATION
RAYMOND PEARL*
Biologitf, Afaiue Aoricuihual Ktpfrtmm*.'
Sla^ion
H. P, ARAISBV
Director, Imlitute of AntftitJ Ntttriitm. Tkti
Pennsylvania Slate ColU'i'
8. M. FREEMAN
Bfftnnist. plant Paihotoyift and Attutant
Oean, Afriatlturet Experiment Stmi^m oi
ibeUntversiiyofMinntsota
All correspondence regarding articles from the Dep^rtmeatof Agricultimi should txe
addressed to Karl F, Kellerman, Jourtxal of Agricultural Research, Washington, D. C.
* Dr. Pearljias undertaken special work in connection with the war emei^ncy;
therefore, until further notice •all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Arniaby, Institute of Animal Nutrition .
State College. Pa,
JOraALOFAGRKlTIIALlSEARCe
Vol. XII
Washington, D. C, January 14, 191 8
No.
PURE CULTURES OF WOOD-ROTTING FUNGI ON
ARTIFICIAL MEDIA
By W. H. Long, Forest Pathologist, and R. M. Harsch, Assistant in Forest Pathology,
Bureau of Plant Industry, United States Department of Agriculture '
INTRODUCTION
The study of wood-rotting fungi by means of cultures on artificial media
has been very meager in the past compared to the almost universal use of
cultural methods by bacteriologists and workers with strictly parasitic
fungi. A critical study of the existing literature on cultures of wood-
rotting fungi develops the fact that much of this work was either not done
under proper control conditions where the purity of the organism under
investigation was guaranteed or the media used by many of the workers
consisted of pieces of wood, bread, dung decoctions, etc., and not artificial
media of such a character that others could reproduce the media, growth
conditions, etc., and thus repeat and verify the experiments.
Brefeld (i-sY, Falck (5-7), Humphrey and Fleming (8), Lyman (11),
Rumbold (12), and Zeller (75, 14) are some of the workers who have made
cultural studies of a number of hymenomycetous fungi on a rather exten-
sive scale. However, the line of investigation followed by most of them
has been more along the lines of polymorphism in spore forms, enzymic
action, the rot caused by each fungus, and the prevention or control of
these fungi in the rotting of structural timber rather than a critical study
of their cultural characters on artificial media. Probably the most seri-
ous drawback to investigators in working with wood-rotting fungi, espe-
cially the Polyporaceae, has been the fact that it was not possible under
conditions used by them to obtain with any degree of certainty the sporo-
phores of the various fungi on artificial media.
This paper deals with two lines of investigation of fungous activity
when grown in pure cultures : (i) A method by which various wood- rotting
fungi can be differentiated from each other by their cultural characters
alone when grown upon artificial media; and (2) a method by which the
1 The writers are under obligations to Dr. E. A. Burt for assistance in identifying the various species
of the Thelephoraceae, and to Dr. W. A. Murrill for identifying the more di£5cult species of the Polyporaceae
discussed in this paper.
2 Reference is made by number (italic) to " Literature cited," pp. 81-82.
Journal of Agriculture Research,
Washington, D. C.
Ip
(33)
Vol. XII. No. 2
Jan. 14, 1918
Key No. G — 132
34 Journal of Agricultural Research voi. xii. no. 2
fruiting bodies or sporophores of wood-rotting fungi can be produced from
pure cultures on artificial media. In this paper the writers have only
given in a general way the results of somewhat extended investigations
on many species of wood-rotting fungi, leaving for a later article a large
amount of detail and the discussion of special methods and culture media
which they have found very valuable in working wdth this group of
organisms.
GENERAL METHODS OF EXPERIMENTATION
ORIGIN OF CULTURES
The initial cultures of all of the wood-rotting fungi under investigation
by the writers have been obtained from the three following sources:
(i) Small pieces of diseased wood, (2) small pieces of sporophores, and
(3) spores. Pieces of inoculum 4 or 5 mm. in diameter have been found
to be better than smaller pieces. The old idea that the smaller the piece
the freer it is from contamination is good reasoning theoretically, but in
actual practice pieces of the size mentioned above have been found more
viable than small bits and as free from contamination. The larger the
piece the greater are the chances of viable mycelium being present. The
pieces should be inserted endwise into the middle of the agar slant until
about one-half of the wood is buried in the agar. Care should be taken
to avoid, as far as possible, burying the wood in the agar, since the cover-
ing of agar excludes the air and either retards or prevents entirely the
fungus from starting on the culture medium. A pair of long-handled
scissors or forceps are especial!}' suitable for this work.
In initial cultures the writers have found it very convenient to use a
series of 10 tubes, including 2 tubes each of carrot, malt, commeal,
prune, and parsnip agars. These agars have been found to give a fairly
good growth of mycelium, and at the same time indications of the fruiting,
cultural characters, etc. of the organism may be obtained even from
these initial cultures.
The writers have made approximately 10,000 cultures of wood-rotting
fungi in their preliminary studies here reported. All of the inoculations,
both initial and subcultures, have been made in an open room without
the use of any special inoculating chamber. The percentage of pure
subcultures obtained when the original tube was uncontaminated has
been very high. For instance, out of 1,000 transfers recently made only
7 contaminated tubes were found.
METHODS USED IN MAKING SUBCULTURES
The writers desire to describe here a method which they have found
very useful in making transfers of fungus cultures when 10 or more
transfers are to be made from the same tube. The instruments used in
these transfers are a pair of long-handled scissors made by lengthening
the handles of a pair of dissecting scissors, a small square glass jar with a
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 35
triangular section cut out of the aluminum screw top large enough to
hold the plug from the mother tube, and a salt-mouth bottle (holding 300
or 400 c. c.) filled about two-thirds full with 95 per cent alcohol. The
blades of the scissors used in making the transfers are kept in this 95
per cent alcohol when not in actual use. The glass jar and top are cleaned
by washing in hot water and then dried before using. The opening in the
top is thoroughly flamed over an alcohol lamp or Bunsen burner and the
jar is placed on its side with the triangular opening toward the operator.
During the actual inoculation the cotton plug from the mother tube is
placed in the triangular opening with the lower end of the plug inside the
jar in such a manner that only the sharp edges of the top of the jar come
in contact with the cotton plug. In this position the plug is protected
from outside contamination and at the same time the hands of the
operator are left free to handle the scissors, two culture tubes, and the
cotton plug from the tube to which the transfer is being made.
In making the transfers of certain standardized series it was necessary
to obtain small inocula as near the same size for each transfer as possible.
The ordinary inoculating needles and loops made either of platinum or of
iridio-platinum are too soft and in other ways unsuited for making
transfers of fungus mycelium. The writers therefore adopted the use of
the scissors for such work, since by using them the mycelial layer on the
surface of the agar in the culture tubes can be readily cut and any desired
size of inoculum transferred without loss of time and with a minimum of
outside contamination.
VEGETATIVE CULTURAL CHARACTERS ON ARTIFICIAL MEDIA
MEDIA USED
In studying the cultural characters of the various fungi as outlined
under No. i of the introduction, the following general system was adopted :
A series of 10 different culture media in agar was used for each fungus.
These 10 media were (i) 1.5 and 2 per cent carrot agar, +3.5 to +5.0;
(2) 1.5 and 2 per cent malt agar, +7.0; (3) 1.5 per cent beet agar, +2.5
and +3.0; (4) 1.5 per cent celery agar, +9.5 to +15.5; (5) i-5 per cent
bean agar, + i.o to +1.5; (6) 1.5 and 2 per cent corn-meal agar, +0.25;
(7) 1.5 and 2 per cent prune agar, +1.0 and +1.5; (8) 1.5 and 2 per
cent alfalfa agar, +13.5 to +15.5; (9) i-5 per cent parsnip agar, 4-9.0
to + 13.5; and (10) 1.5 and 2 per cent potato agar, +2.0 and +3-5- The
acidity of the media here given is based on Fuller's scale and is the actual
acidity of the media after tubing and as used in the cultures.
In any series of a given fungus each corresponding agar for each strain
had the same percentage and the same acidity. For instance, there were
nine strains of Trametes pint compared. The carrot agar used for each of
these nine sets was 2 per cent and had an acidity of +3.5.
The writers selected the 10 media for the study of the cultural characters
of the different fungi not with a view to obtaining vigorous growth but
36 Journal of Agricultural Research voi.xii, no.
to get media on which the growth on each would be different for the same
organism. In other words, the media used were not intended to develop
general characters but specific ones which might differentiate the fungus
under investigation from other closely related species.
NUMBER OF TUBES OF EACH MEDIUM INOCUL.^TED
One tube of each of these 10 media was used in the series for any given
fungus. Better results would probably have been obtained by using
three or more tubes of each medium rather than one, but the writers could
not do this in their preliminary work for lack of sufficient equipment.
However, in a great majority of cases it is believed that accurate results
were obtained with these series of 10, since many of them have been
repeated to the fourth and fifth subcultures with different batches of
media, and the resulting characters when grown under the conditions
described below were practically identical for each subculture of the
fungus for each medium.
POSITION OF CULTURE TUBES IN REFERENCE TO GRAVITY
After inoculating the series of 10 tubes, they were placed in a horizon-
tal position, side by side, in shallow boxes with the surface of the agar
slant uppermost. The boxes were from 2 to 4 cm. deep and about 14
to 14.5 cm. wide. The culture tubes (150 mm. long or longer) had their
tops resting on the upper edges of the boxes and were therefore tilted at
a slight angle. These boxes were then placed on shelves in front of
windows with a western exposure where they received all of the diffused
light which came through and during the afternoon received the direct
rays of the sun from one to four hours daily. Under these conditions
the agar in the tubes gradually dried and the upper portion of it separated
from the glass, leaving a space of varying depth between the agar and the
glass, on which the aerial mycelium could grow even to the bottom of
the tube.
AMOUNT OF DIRECT SUNLIGHT CULTURES RECEIVED
The earlier cultures of the writers received only one or two hours of
direct sunlight. As the season advanced, the quantity of direct sunlight
received became greater, until finally the amount received was judged
to be too great and the intensity of the direct sunlight was decreased,
first, by a single screen of cheesecloth tacked over the front of the frames
holding the culture tubes. Later, a second piece of cheesecloth was
tacked over the first one. The general effect of the sunlight on the cul-
tures in the tubes thus exposed was to check the growth of the fungi,
compared to similar tubes when placed in very weak, diffused light or
absolute darkness. The sunlight also seems to intensify the colors of
the aerial mycelium when it is normally other than white.
Jan. 14, 191S Cultures of Wood-Rotting Fungi on Artificial Media 37
TEMPERATURE RECORDS
During the entire time the cultures were under observation, two
thermographs were run continuously. One was placed on the shelf
with the fungi exposed to sunlight. A soil thermograph was used to
record the temperature of the tubes kept in the dark.
DEFINITION OF TERMS USED
It was foimd early in the study of the cultural characters of the fungi
under consideration that a set of descriptive terms especially adapted
to the cultures of fungi grown under the conditions here described would
have to be used. The terms employed in the tables in this paper and
in the body of the text are those usually found in ordinary botanical litera-
ture, but they have been modified somewhat to fit the conditions ob-
taining for fungus growth. The fungus growth on artificial media is
divided by the writers into two general classes, aerial and submerged.
The aerial mycelium consists of that which is on or above the surface of
the agar; the submerged mycelium includes all that is beneath the surface
of the agar.
Great difficulty was found in obtaining appropriate terms which
would express the character of growth of the aerial mycelium. In de-
scribing this aerial growth terms which are usually used in describing
the pubescence of leaf surfaces have been employed. In other words,
the surface of the agar is considered as the surface of a leaf and the
character of the mycelium growing on this surface is discussed in terms
of leaf pubescence with some minor modifications made necessary by
the character of the organism under discussion. The following terms as
used by the writers require special definition, since they depart in some
instances from the usually accepted definitions of these terms:
Appressed: Mycelium which is prostrate on ttie surface of the agar. This with
•many fungi is the first stage in the aerial growth of the mycelium. Later this appressed
mycelium may give place to other forms.
Cobwebby: Long, weak, intertangled hairs which are not thick enough to be either
woolly or felty and are not short enough to be considered as downy.
Cottony: Erect, rather long (3 to 5 mm.) mycelium spreading in all directions.
Downy: Short, fine hairs, loosely scattered over the surface of the mycelium, giving
it a downy appearance.
Felty: Matted with intertwined hairs, resembling felt.
Floccose : Scattered patches of short mycelium.
Plumose: Tufts of mycelium with a central axis from which short hyphae radiate.
Silky: Long parallel threads of mycelium, more or less prostrate, like combed silk.
Sodden: Mycelium having a water-soaked appearance; usually such myceliiun is
appressed.
Subfelty: A thin layer of mycelium consisting of short intertwined hairs.
Velvety: Layer of mycelium with distinct, dense, straight, short hairs like pile of
velvet.
Woolly: A dense mass of mycelium consisting of long, tortuous, matted hairs.
Cottony and woolly may both later become felty by the long hairs becoming matted
and prostrate.
38 Journal of Agricultural Research voi. xii, no. 2
Hyphenated compound words, like "appressed-downy," "felty-
woolly," indicate a condition intermediate between the two names, while
"downy to appressed" means that the older portions are downy, while
the yomiger portions are appressed.
In all of the tubes the growth of the fungus is both lateral and longi-
tudinal. Of course, the lateral extension is very limited, since the inside
diameter of the tubes is only about 20 mm., while the length varies with
the length of the agar in the tube. The first record for growth shown in
all the tables is always that of the lateral growth. For instance, a record
showing 20 by 30 mm. means that the lateral growth was 20 mm. and the
longitudinal 30 mm. When the same set of figures are repeated for two
intervals of time, like 60 days, 20 by 80 mm., and 80 days, 20 by 80 mm.,
this indicates that the growth of the fungus had reached the bottom of
the tube at the first record given and would therefore be the same for the
second interval of time.
IMPORTANT DIFFERENTIAL CRITERIA
The following criteria have been found of value in the dififerentiation
of the various species: (i) Macroscopic characters, including rapidity
of growth, color of aerial and submerged mycelium, character of aerial
mycelium such as to texture, etc., staining of the agar, decoloration of
the agar, the comparative rate of growth between the aerial and sub-
merged mycelium, especially when the submerged mycelium is colored
and markedly in advance of the aerial; (2) microscopic characters, such
as septation, branching, size and color of hyphae, clamp connections,
polymorphism in spore formation, etc. A few of the species of Polypora-
ceae examined by the writers have in addition to the usual basidiospores
other spore forms variously known as conidia, oidia, chlamydospores, etc.
These various nonbasidiosporic forms may be divided into two general
groups, spores which are borne on the aerial hyphae and the so-called
spores which are borne on the submerged hyphae, often referred to as
chlamydospores. The latter have been found by the writers to be more
widely distributed in the Polyporaceae than the aerial spores and their
presence and characters as well as those of the aerial should always be
noted, since they are of great diagnostic value.
It will be noted from the tables that certain fungi have colorless or
white aerial mycelium throughout on certain agars, while others have
colored depending upon the agar used. Such color differences are very
important, since they are usually constant for a given species on a given
agar.
In some species of the fungi examined the submerged mycelium in
certain media is constantly colored, while in other species, whether the
submerged mycelium is colored or colorless, seems to depend upon certain
environmental factors, such as the amount of moisture present in the
medium or the acidity or alkalinity of the medium.
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on A riificial Media 39
Some of the most important criteria for distinguishing different but
closely related fungi are found in the first 10 or 15 days of the growth of
the subcultures, such as rapidity of growth, color changes in the myce-
lium, staining of the agar, decoloration of media, etc. Important char-
acters which are sharply defined at one stage of growth often disappear
or are obscured by the later mycelial development; and for this reason
the cultural data in the tables have been given for several periods of
time in the growth of the cultures, say at 10, 20, and 30 day intervals.
INFlyUENCE OF SUNUGHT ON CULTURAI, CHARACTERS
One of the special benefits which seems to be derived from exposing
cultures to the sunlight is the accentuating of the color characteristics
and toning down of the mycelial growth of the fungus, thereby making
it more characteristic and uniform for a given species than when placed
under similar conditions in the darkness.
The differentiation of the characters of the mycelium produced, both
as to texture and color of the aerial mycelium, is very much more marked
when the cultures are grown in the presence of light at ordinary room
temperatures than when grown in incubators at the optimum and con-
stant temperature for the mycelial growth of the fungus under considera-
tion. This probably explains why no one up to the present time has
seriously attempted to differentiate the various species of wood-rotting
fungi by means of cultural characteristics alone.
Furthermore, the cultures when grown in darkness and at a more
or less constant and high temperature overrun very rapidly the surface
of the agar in the tube, thus obscuring the real growth of the fungus as
observed in the cultures subject to daylight conditions.
GROWTH OF WOOD-ROTTING FUNGI ON AGARS
Texture. — In the growth of wood-rotting fungi on agars the fungus
as it spreads from the inoculum on to the surface of the slant proper
assumes certain well-defined stages in its growth, which may be roughly
divided into two general divisions: (i) Fungi whose advancing young
mycelial zone is appressed and (2) fungi whose advancing zone is downy,
felty, woolly, etc. There is but little real difference between these
two methods of growth, since as a rule the character of the mycelium
first to appear is appressed. If the true aerial mycelium, in contra-
distinction to that which is strictly prostrate on the surface of the agar
keeps pace in its growth with the appressed mycelium, the zone of growth
will be downy, felty, woolly, etc. If, on the other hand, the growth of
the strictly aerial mycelium is much retarded, the appressed mycelium
will present a well-defined zone from one to several millimeters across.
The appressed mycelium is usually either colorless or both colorless
and sodden, and from this the true aerial mycelium usually develops.
The cottony mycelium as a rule does not persist in this condition for
40 Journal of Agricultural Research voi. xii, no. 2
any great length of time. The long, divergent, aerial strands usually
become more or less compact and finally felty. The usual steps in the
growth of the mycelium of a fungus are first appressed, then downy,
then felty, woolly, etc. Cottony mycelium usually develops the cottony
stage immediately from the appressed condition. Many of the fungi
pass so rapidly from the downy to the felty or woolly stage that it is
unnecessary in the description to indicate that there is an intermediate
downy stage.
Colors. — The colors of the fungus as a rule follow certain definite
changes. Excluding the color of the mycelium on the inoculum, the
first color which usually appears in the early stages of the fungus will
be either colorless or white, depending to a considerable extent on
whether the young mycelium is appressed or downy. The next step in
the color changes will be for the older whitish areas to become light
buff \ warm buff, antimony yellow, etc., if the fungus happens to belong
to some of the brown polyporaceae. As the culture ages, the color of
the mycelium on the older areas will assume a deeper and deeper tone
until finally a color is reached beyond which no appreciable change is
observed. In the large majority of cases the color of the older mycelium
constitutes a rather extended area compared to that of the 5^ounger
zone. In practically every instance the cultures obtained from any
given fungus on at least several of the culture media will approach very
closely the color of the sporophores as they appear in nature. For
instance, if one is attempting to grow cultures of Polyporus dryophilus ,
Fomes texanus, or other brown fungi, one would expect to have at least
several of the culture tubes with brown mycelium similar to that of the
fungus.
TABLES SHOWING CULTURAL CHARACTERS
In this preliminary report only a few tables are given out of a large
number which the writers have complied on the cultural characters
of certain species of wood-rotting fungi. The following species having
brown sporophores are given to illustrate the close resemblance in colors,
texture, etc., of the same fungus on different hosts: Four strains of
Fomes texanus (Tables I-IV) and two strains of Polyporus farlowii
(Tables V-VI).
' The colors used in this paper are according to the following standards:
RiDGWAY, Robert, color standards and color nomenclature. 43 p., 53 col. pi. Washington,
D.C. 1912.
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media ' 41
^ :-
^ a
O in
•^
"^
«
a
s
a
t.1
s
u
1— 1
3
w
0
n
J
3
n
r/;
<
' — '
H
>r, 0 "1 N i/v
g ^ i~ ^ •* to
0 m 0 -^ 0 ,
00 r^ -^ to
d
d
>.
^XXXXXXXXXX
;§o 000000000
I
•0
0
d
5
. -3 2 -S :
2
Si
5
20inuii/l>o000"0
•o
d
0 CJ2 -T
.w^ 0
?: «*^i/^r-ofow ■<r'^t^'o^
0
2
Q
U^
0 -S 12 -a
•3 Q
ixxxxxxxxxx
^0000000000>o
■o
_C3
^^ -2
s
•0
5
S
oS pq
5
CD
00
s
S3
1 2
"o
a
g" <H K-.CO « rj 0 <o U1 1/100
Igxxxxxxxxxx
^oovr>./ioooooio>n
1
C4
0
a
1
0
: a :
S i2
•Q 3
" 1
pq u
•0
Xi
J3
II 1 ti
>>
E
>
2
■j3
5
d
ca
•0
jj : vO ■ - M to rO t • «
^ -.x : ;xxxj^ ;x
/Si
a
5
c
a
d :
3
1
& ■
0
1
"o
"3
0
c)
•0
1
3
a
V
u
a
<
1
3
a :
>. :
a d
g -a
^ :
2 £
■■^5 5
§ 1 ':
•0
s8
; ; I'd ;T3 : : : :
: : :^ :^ : : : :
: : -.Z -a : : ■ ■■
;;;§:§;;;:
. . . c3 • a .
: : :ot30 : : : :
>> : : >.a >,>, : : :
r
r
0 • ■ « & (U CI • ■ •
a
■a
w
> : :><>> :
ui
tn • cX
"
0
d
•0
3 ;
1
1
"o
■35 "3
"o
0
i
"0
00
; : ••d
: : : 0.
■ : ■ 0.
'■d • ■ • •
: (u : : : ;
. i, : : . .
• 0. ■ ■ • •
• & : : : :
l5
0
1 i
•0
0 '
d 0
0
2?
did
^3
0 C
a
^ :
0
tn-Oii
q *J 0
■w 31.,
a •
^ 1
e
5
c
X
1
a
. . . C3 • 0! . . . .
: : :°-do : : : :
~° 1 3 3 OJ4 „
E*
'^.c
15 « 0 » ja*"
3 -s «
■3 ■ ■'S
> : :>«
2.i;'C • • •
<>> : : :
-'i
m
0
<) u< u -
I) 0 t/
0
0^
OJ.
■4J 0)
5
1
3 3 III ^^
a
3
ni
00
0 "^
0 >.
3
S .
"o
0
0
-U -4.
3 :
q . c
1 •
1
01
•0
... 4,
. . . u
... a
: : : c
■•d : : : :
:g : : : :
■ a : : : :
a
51
<
ll§
^-6
•Si
a
. . . ta
• • ■5"'
> : :>■
• a ... .
g2 ; ; ; ;
^ HJ V • • '
<>> ; : :
a
>.
~F
1
0 p
0 g-S
• 9 '
3 a
i 0
"o
M
ijj
"3
t-
a .Ss"
." ^^
^ "^ t
0 00:
a "S a (
a in 0
2a
■3
0
3
J3
8
a
n
- .e-s
a O-Soj
•f >'3-S c
- 3^32
3 0*11
■•d ; : : :
0
<
c
) W OM 0
55 rt c
3
•0
. i_ . . . .
. a . • • •
• a : : : :
: ^
d
:_o
^
•«....
^
•"3
. 0
0
J2
^
:>•:'.
?'2 : ; ; ;
s >>>> • ■ >>
•0
3 ■
•: I
•0 >§
a
0
J5
: ^
a
3
2
■ V ■ ■
:_> :
^
ta
4-*
3
;
3
:> : i
<5>> : ;>
^ '
^
; Qpi ps
m
H
4
tit
bt
w
a
3
5
i
n
0
P.
e
q
5 f
1<
i ■
< p
I
M
t
IS,
:«::::
pquP^ <PuPi
42
Journal of Agricultural Research
Vol. XII. No. 2
1^ «
fv.
'~*
oo
H
•d
<N)
a.
•a
- i
§ S
s 3
^ 3
g ^
I"!
s is
S bo
S5 nl
^ .=
pq
5h
Wo
S -a
3-0
m Pi 5
3.^
o H
pq:s
!SS
O oj --* O
rt o
■SI ill
j2 a
iH "" O t^
< o<
a o
^M
a o
Is*-
a 3
B) " & 0
< u
^ a . a
C t- i*
ni V- >>
*J.D a
on
it
^s^
2^o
Sogg-s^
^5
(JO
^ S O 1-1
•5 s
aS
o mo
■a 2
_. o w in N o mo o O m
■Ixxxxxxxxxx
BjOOOOOOOOON
; O O O "o O "1 O "1 O M
JXXXXXXXXXX
'loOOOOOOOOOO
-icOWWOOOOOONO
;2xxxxxxxxxx
~< U-, o >noO O 00 O O 00 O
xxxx
•g o o o aj tl t) ° o o
•^ • • • o.'oJ'qJ
> : : ■■<»
ti ° ° ° Jitit! « wjj
*S ■ • ■ Q.'oJ'S - - -
> ■■ ■ ■<»
•do
■SooofJ-ScJooo
<J»
•y O O 1> o \„ ^
0Sffl0K0fl,.^|l<fL,
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 43
s ^
II
S 3
o — .
K 8
i-t ^
<
Q "I
n S
5 a .J. 0 >.
go §^£§
■^ P 3 ^ S fe «
^ 3 o-g o.^ = &
Pi (3
« > S
V a SiJ
O O
o<J
5S&
So
■w o b
^
u< o w
< m
<1 UhI o w
« o
^ s
2 J -J
^
On-t
*i a
E -^
a
I si
■a
ran Qj Olol-''^ MO
;XXXXXXXXXX
= 0000000000
CO loO « lAOoO O i-^.O
. PO^N TON •*rO'^'*)M
SXXXXXXXXXX
^jCOOOO'^OOOOOO
XXXXXXX
tloooii-sjooo-tj
<5>
,01-*-'
■C^.2 o i; ai t; ° o
•a o
a""
. '■ 6. ,
i:S-a
S a it's (u Q CSa c« o
44
Journal of Agricultural Research
Vol. XII. No. a
-^ 2
v8
S
H
«5
a.
A
1
k
a
g
w
en
s
■*i-
m
•> -a
1
■**
^
s
a
e
*»^
^ 1
Si o
Y
> -3
n
<
^ J' =■
• 3 ■* JS
S3
o a
eg
•aM
•2^
r
a^ a
-*^ • Sj= 9 2
(u g a S 43 »j
H P en a; .^ .-^
Pw og M.S C SS 1=^ en S.«„
3^>.j
CJ
^ i!) 3 3.fl-Q 3
3 OJ3— '.a-~ 3X1
Si
s
o
O
3
n
a 2
feja
2gta
O ^ -
oja o „ _
g|E2
5 P o
p-a
B "^ 'S -^ _, y
fc -s
» S rt
2 n omo fw <:p4
i^5RvS5
oooooooooo
SOOO'Nio>/iO»^"Ooo
^gxxxxxxxxxx
^oooo>oooooo
grtu^OOOOOnOv)
:2xxxxxxxxxx
'S'S
o. a (u
«>
•SjoddSSJ-gd p_^
"S • • • o, a'oJ ■ ■ -
> : •• ■■«>
rt « S'S 8 p Sag 5 o
Jan. 14. 1918
Cultures of Wood-Rotting Fungi on Artificial Media 45
s
•«i.
^ ^ i
S . . <"£ Sn .
V 6 o ax) u o
•3 *jt« o
O WO
6 6 6
PQO
• n w ■ •
5 : : «5 K :
i o o 0"^^ o
> ■ -w o •
■5 a
^ .2
e 3
.ir>0>nOQOOOOv>
iOO OvCO 00000 Q\ O^ 0\ t^
ixxxxxxxxxx
•0000000000
: 00 acO 0> t^OO CC CT» Ov t^
:xxxxxxxxxx
"ooooaooooo
~ioOOOOOOO>"0
^xxxxxxxxxx
^0000000000
pS«
:^n
c c c
So £
J3 OX!
c & c
It
:-° 9
ic
O iJ 4J i-
aj2 1-
;X :xx
• * • fi ^ ■ >> ^ *
: ". : & c : ^ c :
- • ■ o <y -^2 <u ■
; *. : TD "O 'WO
'o'o'o'o ^''o^'o'oS
OOOO^SO-jO^O-g
• >, ■
:xi :
•X3 •
. c* •
; & :
•J3 ■
■ O >.
'■ 0*3
a o
Tl '^
>.8^
^^c^S ti o S lu S
•o-o2o&>-a.2£S
^^u^O :oOo
■ <o
s 0^2-52
46
Journal of Agricultural Research
Vol. XII. No. a
*<.
U
< -
4' i_
^ a
<>
2 -2
> 5 -3
i
vn 0 Q m'r,
t^ 00 00 t^ t^
3 w m >n 0 1
J, NO I^ 00 t^
SX X XXX X XX X X 1
g
:§s
0
> 0 0
3 0 0 0 0
3
31A10 0 mio 10 0 1
CS
>>
^X X XXX X XX X X
*o
•a
•0
CI
•§ 8
D
DOO 0 00 0 0
w
m
. 0 00
^00 0 >
0 :,. 0 .0
>>
g N M
-< • N « H
«
4;X X XXX X :x X X
•a
^8
>
3 11 0
n . VI 10 1^1
■*
^"" » :« M « 1
J
; i'O
: : >>
Ji
S
. . lU
>,
"O
• ■ s .
0
•0
•3 ^ •o-aara-3 >>>• >> "o
0 ^ ■ ■ * ■
. . U. ^ 0 ^^ T* •"
n
.3
> ^
;
• • ^ tn ^y 0) a» oj
(U
^Si-a S
: >>
a
•c
1
•T3
a
'■ c
+■
3 c ij c
; :2
>.
rt
^
NO
f ^
3 0 a ^ c
^^ ^ 1
^
r'
:^<5 P
. fefe p:, ^
3
1
0
• i-o
• 0
H
en
>.
: ■ 01
: >»
^ s
•> '. V ■ ^
, :« •>;
•0
° >>t
i
>-§ag|
:2>>a 0 1
Tf
z^l
c
::::a2 -q 1
O'tn 4
c
' :^"^
■ o'«-g
0 >
C
:<i) c
;U 0
a
>.
S
rt
V3
•C
0 c
c
00 c
00 0 0
V,
0
"o •=
•c
"o/q t:
»oxi -a -a 1
^
tS
a
"9
ui
t/i
o
t-.
>>
01
■a
0
"S c
c
00 0
00 0
c
1
NO
•3 •=
•c
•o-q "c
•a -a -c -u 1
a
3
in
O
tn
S
b
>.
*— <
_o
•a
0 0
c
00 0
000 0 1
"c
a "
•c
•W "C
•o-o "0 -o 1
(J
^
a
: : fl
: : fl
u
. . f-i
. . u
0
: ■ 0
. : ; ,^ .^'
>.
J3 .
• : J
rl
+j c3
• jn ,j_,
c : : -^ S43
T3
^& 0
0
d.^ ^
^06 •« S e
o-o.y <-< ^ S
vg
OgT3
•a
•Og 0
3.a
' '3 "^
.D -''^ ai^ci
g
m
:t3 «
:^ pq ?:
.2
"v
a
■ • «
• • a
a
>>
. . ;-<
. . 1-
>.
0
• • 0
'■ '■ M
. : ; - .?■
"rt
CO
-4-1 Q '
6M -^
a : • -^^ C.C
'O
^ & d
0
e 0 (u -^ ? B
«
NO
^s-o
•0
•a. s "
0 73 ,+^ 0 C g
■3
3X! •
pq
:c3 «
5 -t? ^.C^
,
• • J. ■
U
"3
• • "3
: : >.
CO
•0
1 «
g 1
1 ^
1
e-S a
^ 0 g
?^3 ^
: : >> d
• : g 3
On
1 ll^
B
_3
^
: ! "a
•5
1
•4-)
«
CS3 CO 0
U
s
n
u« 0
PL,< fL,
(k
1
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 47
DISCUSSION OF CUIvTURAI^ TABIvES
Fames texanus. The cultural characters for all of the four strains
show remarkable uniformity for each of the corresponding agars in the
series of ten. This fungus seems to be more susceptible to differences
in the amount of sunlight it received than any species thus far inves-
tigated. When the intensity of the sunlight was diminished by the
cheesecloth screens previously mentioned, this species on several of the
agars immediately responded to the decrease in Ught by making a more
vigorous growth and turning a lighter shade of brown.
Polyporus jarlowii. This fungus is also very uniform in growth on
each of the 10 agars as the tables show. When grown in the dark, the
strain from Acer negundo developed a submerged mycelium on prune
and corn-meal agars which was Mars brown in place of colorless. Whether
this change in colors of the mycelium was due to the darkness is doubtful
since the other strain from Populus iialica when grown in darkness still
retained its colorless mycelium the same as when grown in the light.
Series were grown of nine strains of Trametes pini obtained from
material collected in four States and growing on seven hosts — ^viz, Pinus
echinata, P. fiexilis, P. ponderosa, Picea engelmannii, Pseudotstiga taxi-
folia, Abies arizonica, and A. lasiocarpa. This series represented sub-
cultures ranging from i to 7. While there was some slight variation,
especially in the colors of the mycelium of the various strains, the dif-
ferences were not so marked as to constitute real specific characters.
There was practically no difference between the cultural characters
obtained from the different subcultures. This would indicate that the
general fundamental characters of the fungus are not materially changed
through successive subcultures, at least in this instance to the seventh
subculture.
As will be noted from the temperature record, there was rather a wide
variation in temperature during the time the various series of cultures
were growing. Nevertheless the general cultural characters as shown in
tables are practically identical. This identity of characters for a given
organism on a given culture medium is still more marked when a series
of from three to six tubes of the same agar for the same strain is made
at the same time and then compared as the growth progresses.
CULTURAL CHARACTERS FOR DIFFERENT STRAINS OF SAME FUNGUS
The sources of the initial cultures from the nine strains of Trametes
pini were five from sporophore tissue and four from infected wood, rep-
resenting seven hosts, while the cultures from Fomes texanus all were
from tissue but from two hosts. The Polyporus farlowii cultures were
both from tissue but from different hosts. A careful comparison of the
cultural characters of the various strains of each of these fungi shows
no appreciable differences between cultures of a given fungus whether
obtained from infected wood or from sporophores; neither do the hosts
48 Journal of Agricultural Research voi. xii. no. 2
of the fungus seem to make any marked changes in the fundamental
cultural characters, as is clearly shown in the various tables when strains
from different hosts are compared. There may be minor differences
due to the host from which the strain came, but nothing more.
The comparison of the cultural characters of many species of parasitic
fungi has long been recognized as a reliable index to the identity of the
fungus under investigation, and there is no reason why the cultural
characters of wood-rotting fungi which are just as uniform and depend-
able should not be used for identification purposes.
The writers have purposely avoided going into a discussion of the
results of the use of synthetic agars and of other special media, as these
will be taken up in a later article. They have presented here the results
obtained from agars easily made and apparently of a uniform enough
composition for similar cultural characters to appear on different batches
of the same agar even when the acidity, alkalinity, and water content
vary considerably. Just how great a variation in these factors must
occur to produce a decided change in the cultural characters is a problem
for future investigation.
EXAMPLES OF THE DIAGNOSTIC VALUE OF CULTURAL CHARACTERS
IN SPECIES DETERMINATION
In comparing the cultural characters of closely related but really dis-
tinct species marked and constant differences in the character of the
mycelium will be found on certain corresponding agars in the series of
cultures representing the two species, while if the two fungi are really
the same species no constant differences of specific rank will be found.
The following fungi will illustrate the diagnostic value of the cultural
characters in determining the real position of the species.
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 49
t-.
00 C8
C\J g
^ "5
s
^ a
u a
Is
ils 8.2 sis
^ ai^ ai^ a
*T3.2 iJf-d.S "-0.2
^"sl^a-i^sa-
at ^
u
s 6
0*^
To
u ^
•la
a
:-a E ° "s a
3 1 C 3 I "2
^ 2. ^1j 2 bs i*
. o ■* >> o " "^ 'J
acn-°a«-^a^
SI'S n c« o » "o
y o o
i-i O S U
•? a- -rfl.S-r
U <
X j2 "> a
^ n a
•o 'd 'O -d
o • o !5 O.J,
.2 J3
8 S
fl-H H.S
Sg eg
ca O 3
'3&
>. o
u
a o
a °
o .^
o • *- u > -, d
5^.cSJ2.Q>?a
IS
.2f a
rt O 3
u n
as
cc u a
fi a
J. O . _!,
o '"Ma
rtXl aJ3i3.Sf.2
0 < >-i^
1 1
-I
i4
O CBi'm>--^j3
« . O C >■ (U O r; J3 .
>>"'o o-yo o o in !;
rt03-wc«03-S.?0
U « ** "^ "^
;^ «
■•r ^
.2 0
O ."o •
"8
1
e 8
1-30
■>-' j5 ^ Bii
do- .2f 3 -J * o
o : a pq ^
(3 ;^ n
•a 2
(1< <! PM P^
27805°— 18-
Jxxxxxxxxxx
.^0000000000
.OOw. mOOOO*^ iy>vO
.:;xxxxxxxxxx
»e^OOOOOOOOOO
.00 O O vnti^^oo « »^to
ixxxxxxxxxx
;0000000000
•a ;-0 ;
<u M <u S!
i-o
ao aS
Sg-rt a
: «
13
.2" "'.2' nl
• c ^
S?
•i|'-32&«ga
O g O (U-C ^ cs o
ir 1-1 5« •'^
* nl £• OJ $• : 2irrH ^
■XI o a oj'o^ « o
Q C. O. t -. u
>.C ^■d oS g'g^
fl-s d"S ♦J o ^ «i a
n o 0. ■*-' ^ c3 o
•i! >>5 >.'>,S'S ° >>
•31; o^ o 03^1; o
•§•§£■§§1 :?^|
t*:: <uii !» " 3.iS b ■«
OSWy«U(l<<fr<P<
50
Journal of Agricultural Research
Vol. XII, No. a
I §
CI, M
o
a .9
s «
•S ^
^ 1
< e
.1 3
">^
m
<
o a o
r
U Q U
as
9 . in '^3 g . w
">— !? ?? -^ <" -^ f^
So So
c; >, -^ (U >, a; >. .
! I- Is Is
"3 d
" o o S « >- o o
rt 0-3 a a ca o-S
So So
_o -* a S " **
o <
.£f a
— T-S M _ a
Isssia
p « p
;-a is iJ
- 2*^ as ^"5
o < n IS
3 -^ i
fi 'S
-■9
.^ ^ 4-» a»
- "o " •S.'C
5 « hJ
■s
fiii *§
>.a'3 o
^- ^'
3l J
.2*9
i4
^ « S pq a
tt^
ca
a^ o
2 btTn
btTo ■ a
s Sin '^
a=3ig
t; o >. S
a-w «ja
O O
o S
^ o a
a -!.
oS 3
3 -"^.Sf o
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 51
10 I-
r> r»
0
invO
«
0 « -^ 1
ij i>* r^ fo (>■ m 10 r>-vO r* fo
.fxxxxxx xxxx
p^oooooo 0000
>,
CD
•0
0
-r
. 1/1 1000 000 "1 0 t^ n
gr^ot^r^i^io o^or*^
sgxxxxxx xxxx
i^oooooo 0000
i
T3
u
0
a
to
"S
.00 1^. ftO"lM >ni/iioO
N
S
a
i XXX XXX XXXX
^000000 0000
•0
0
'nmSSmm SmHS
ixxxxxx xxxx
,^OODOi/-.00 0«i/li«
&
•c
0
i'S^i'S >" ':
is^sa ;
•0
U ?) V,—
u . .
: aSi Gen
iti . ;
• a a c. a ^- •
• c3 a M «^ •
Qj ; •
(0
:>.«>.-« :
a : •
>>
tnaiul
ainly
mainl
ty to
ppres
3wny
toap
felty.
•0
0
•t
ytoi
tom
ytoi
y-fel
ny a
tod<
felty
y to
own
elty
own
own
dow
elty
Iky-
elty
own
fc
IPtoOQ PhW^Q 1
•■0 ••^ >,
. <u-a (J ^
■i^^a
•0 ;
■ t-l y^ ^4'-^
0
: a u an
\ -J.
:&g^a^
l--\
in
:>.«>.-,«
a • w
^
>.'a >.'5 S 0 >. -r >v 5 1
a .2 — .9 "rc^ea
s
•0
0
to dowi
y to ma
to main
y toma
y-felty
■ny app
to dowi
felty to
to dowi
.y to ap
3
1
a
"a
1
fefaCfeOQ toiofefi
•d'St
*Vi
0 s «
0
" S^
•d : •
v
u . :
^
l-o&g.S
?! :-d
1
i
ny.. .
pressc
ily ap
inly a
to ma
owny
appn
•essed
presse
Oj
•a
0
to dowi
y toap
to mair
y to ma
y-felty
scd.
Ity to d
felty to
to appi
,y to ap
elty,
elty
own
city
own
own
pres
Jbfe!
Iky-
elty
own
p:,{i,G'^aQ wjjfaQ
L
a
>!>•'■.'■.■
1
•S 1 : ■••a'
1
1
>.
i l^;i
M
&
2 2:^ :&
•0
0
elty
elty
to si
[elty
oap
>» >> y) >. .^
•v-2-v>.T*'
4-»+J{/i^-»>l?^W^♦.•,^?^
ou im a 53 ^ go ^ g
k.
to
M
i
•3
3
5
*j 4-
M (
'■■I
a ,
s
0
2s3 « 0
SP
il
>p:
.S
p
Lc<pHfl4
52
Journal of Agricultural Research
Vol. XII, No. 2
k
">^
"*i.
^
^ 3
S .2
§ '3
a I
HI
2 °i .5?i .
n m <u ^ a/ u
t-cT! a; ■ H V ■ n v
^.2'S B'rt" B'a'S
S Ri -gi
2cj &8 SI
8 I |-S 8
<2 .Oo<n'2:30 ."'
8sa-oS§8£3!3.
Sua S u
o P fl
e2a
i-Sla
o '§ o
1- ■ n v> •
a.a ^S 2.2
ss 'Ss ss .
5o .S-g tS-3 6
"" -a^ ss
3 «
43 T)
" O
^ g !
2 5°
o -a .3
^j2 a ^.sfo
•as
n
^ ^'
:2 s ^8.2
3 «
+3 o
o o
O J,
*^ o
u
•^^^^
g2|s
I'll
Is 5
2 MO •
■5 o ^-S
flii rtXI
< u
2, *§
(3.0
1^ u 13
J3
0
0
0
^;s^
>,
■S >>3
cn
u
0
2 -w J
53 * <i>
(J ;s «
^ o
^ a ^
(!< «<
Jan. 14. i9i8 Cultures of Wood-Rotting Fungi on Artificial Media 53
;xxxxxxxxxx
JOOOOOOOOOO
•xxxxxxxxxx
;oooooooooo
.r«g«OOt/iOO>n<n
jXXXXXXXXXX
;0000000000
Jxxxxxxxxxx
1^ o o "oo >o<» 0000
rtS^-a -3!
>>S 92pS
. 3 aHii o a
' d cQ o <i> "^ cQ
>>'0 '
o rt a H *i a
"O d cJ o m rt
■•a >.;
>''a ?! !; o 0. M p >?
oHa^'iioao"
ot" o .-."i ^o -^f
: O-
: a
"a o Qa 00 -30
„ . . ^ • • oj ^ '5 2
E-aJfirt E I « t5
54
Journal of Agricultural Research
Vol. XII, No. 2
^ T^
2 -
"^ s.
<
s z.
S Z
i
-e
ri
a
S
N
^
a
n
U
11
^ 1
■"tj _; ot3 • >n-a
•5 Si o
iS » i; « S w t/ >. o
">- a". 2" c.S"^ B.2 '>.•« 4)
u u o u o
00 oj . a
J, n
sea w
•2,.sa|g
a!2
CJ
o 0
"n3foaf'ci"<C^
01 =i D i^' (u !J ° o o
«^ gmXl ?»!£ g'o '^ "
4! .5 s .5 S3 .£ u^2
o n « o H CS o R CJ.2 Ho
(J O U U O
S^a
|og|
.2 5^ a .2
o o
'S
m-o a o-a a "^-a
7agiag7c
M>,o. i-'>>a)i-">.
T fl.a'n a.S~ a.
-ia^^.i.ai'— <a
CJ o u
V 01
o o
u u
1^
a-s
o
m"^ S
-0.2.2
^ *i ^ o
O J, o
-S 2
3S 3
r5^.s-a
d^H^ —
^5
O O -O ™
o u . o o
8 • -S5 o
«^«l32
o o m
ll
S2
o a o rt o o
go toS m w
gja 30 s . 3
n ^^ 2 M 2 M-3 M^
... o — -
o
8
o o
O B
i.nj3.- a
►4 h-t
bo's 60 O
.-< o — to
h4
S u u • t-*J3 ia
o o ^ li! o 3
0 o m
'•3 ^
j^ai^s
3 3
■4J-S *J 1
jao ja;;
Mo M ;
&
^ S5
J- ti^.-
^ 3
« s n =» a
U h4
^-ti
J3 5 *.
^4" '3
3 .2
a;j= o "in
O CU <J
Jan. 14, i9i8 Cultures of Wood-Rotting Fungi on Artificial Media 55
^
aXXX XXXXXXX
sooo 0000000
CS
"0
0
^««<, n«c««c«
*
000 Oln'nOu-.Oul
i
. r^O ^ «^po»^*^»^t^(*l
SXXX xxxxxx
i
•0
0
itMOoo 0000000
"3
..
>n 0 f
0 0 w 0 « 0 0
1
i
,0»ot*5 loro-*On^<*5
gXXX XXXXXXX
•a
0
v^ooo 0000000
wo C
000 >/^CC ■/! 0
S5
«5 Ci « M « « N
^
SXXX XXXXXXX 1
•0
0
M
^88«
0 "noo 0000
0.
i
c
1
>
1
■* 1-
*
a
0 4| a a
0
c
a
0 a 0 "
1
c
owny
owny to
Ho
c
■a
nbfelty t
elty to a
elty to d
owny to
m
C
Q
codtP^P
6.
a
•? •=
•c
_c
1
>.T3
^ 1
i
>.£
> cn pt ki
eg
c
0.
0 ji a a
•0
& ,
a
•0 0.& 0,
a
0
0 =
en
0 a 0 *
3
>.
a
fO
>
"33
elty to d
owny 1
pressed,
owny to
do. ...
c
abfelty t
elty to a
elty to d
owny to
fetoC
P
c/^bfaP
6.
cd
0
01
Z
>"
3
•a
•B 1
s
7
01
>.T3 >.S
in
*?
s
C S C M
H
>.
>.£
£
& s s g
a
a
a
0 J* 0 a.
"d
& ,
a
■a Q-o a
0
o.£
a
0 a 0 "
•d
0 ..
•dS
*J >> Ul >. _
*j *5 *j :>.
•53^ 0
0
•0
Ilbfel
elty
Ilbfel
own
fefeS
P
wfewP
•2'
en
&i
•d
•d
•0
>>s
^
V
>>
=S rt
0
S
>
S
•^ 0
k«
^
•3
2 >.
a
a
a
a
a
a
0
+J
a
B
0
et
•^
.&~7
0
0
S-0 0
*i-ii Oi
tr>.
^ >.r
>>>.>._
&& 0
. ftmu 1
3 0 0
li 000
•3 a 0
W<!P
wQO
ppp
3
:«
*
*»
•
I
0
fc
a
■4-1
Si
2
t
^
2
?3
M 0
0
S
n
c
n
U
(^
<
Oi
!k
56
Journal of Agricultural Research
Vol. XII, No. 3
g
">-^
\ I
^ 1
<
O I-.
.10! i
I M . O
a^ "3
O U (J u So
o -^ja
"3 >■!-
u a <u
t.^2^
^° o
2 8 "
.•la l2s.-s|
H-t n i-t
S-3
•a 3
a
Sta
o XI ^ o;S
u </; S ot^r
o ^ eq
I O o
S ri
S ?: « o
g s
tn-S
P p
s-a
0.3
Si S
^ S o sS
>>S u o o o
O PQ «
o O
5o
D < i-f w '-'
is ""
O O o
u W
^ 2 3
.2 M
>>3 q
0:0.2
aa^so
*3 o >■ 5
.iS'-" O — '
0
d -i
aa^
5 & a 4;
" Q b b
-a 'a
a s
•g 11 .2
U Ph
(11 (X«
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 57
tA 0 l^ VI
0 VO I^ »nOO fO
i
gXXX XX XXXXX 1
sooo 00 00000
(S
•a
0
g««« «<
M « « « N «
•*
i
SXXX XX XXXXX
s
•0
gooo 00 00000
(^
0
CO
^
0 u^ 0 OT I
n n i/i N 0 0
v
. t^vo 'n 10 r
in
gXXX XX XXXXX 1
■o
Sooo 00 00000
0
S M « « « C
0 >o u> 00 >/> N lAoo 00 0
a
. •* «) w « c
gXXX XX XXXXX
•a
0
Soom 00 oo>nooo
^
•« -2
t -s'l i
T
'
»
m
s?
a 0.8 0 ^ ^ >." 1
iS
a
■w 0 « C B.
T3
'S>..
^■7 >i 0 0 '^
J>.cn-H>>i?iS^-i-'<-'-t-'>> 1
'3
SS29£5ii>.>.H
0 0.-^ 0 O.o-q'Su 0
\ii
« wC
Cc/3pt,feQ
6.
6. ^•
0
g
•a p g>
>•□
"E
CO
M « n
^ av
» 1
C3
^ 1^
5ii&&
a
•0
0
2 S =
^lll^
-.3
vX.
^•V >,o o-^
?<
•>,ffi-w>>in>-w+j-w>. 1
^■a 9 22 g
£ 52 >-->.g
a
*C/
0 C-b c
0 tflC
*rt
fe
pMt.feQ
1
i
"O
•z
CI
u
3
•a -a
•a
'4-1
a 4)
a >>'0
en
t/i
« V!
en a <"
»
H
>.
>.a s
0 E: S >.K
Id
c a
■S 0 « c a
•a
'3 %
■0 -; 0?
>.'a a^ g-
0
il 0 ao ">
^ -w COT) 0
": Td +j 4j
0 . cu ^
**•* ^ 0 0 t^
« >.S >. ir
>,*j -w -i; >.
±i±i i> t: i> c
"S'S 0 a 0 c
1, 0 ^ (u 0 0
%U^P> P<
•s
•s
(A
tfl
ss
s
s. -g
g-
"S
a w
a
u
ui
« S
<A
>.
S
>>
0 C
0
M
t
B
■>-' Q.
a
T)
>> a
i
^ >.d.
0
^ M
oil«
^
s o-a
"o
iijj 0
^1 •*-'
lu V a
a^ 0
>
"a
2| i§
"1°
. 0
p^wQ P<^
fi
w>Q
3
a
'^
I
^
1
4J
PC
>
"a
0
S
pq
1
a
1
J
a
0
S
58
Journal of Agricultural Research
Vol. XII. No. a
I -5
s
►J
PQ
<
< 3
i ;
<,
o o [H o 1- o
0*3 n cs M o'
•3 I
' oa ' 0;:= ' oa
iJ>>o o>>o S>>0
"5 S'h'5 o'Cb h'C b
ogsoaaiOHiio
£ . ' °l^
S rn O - 4> fl
8*0 K _• M "^
CD OT* "" —
O
ot3 d """d
•""d
©a ' oa ' o-A
^ <u o .^ 11 o K^ ai
< ^ >. l» >. 4) ><
; B'Cb u'cn a-j:"5 s^'cB'Cm
) o o o u u S
2"
00
>. >. So 3
Eg S 2
5 a .^a
^3.
M y rt « .
^ O P
O o m
O ^xi.g
|o.ra
'I ^s
o « o
rtX! 3
g 8 XI " 5
^;
^" 3
1^ X
:§ia
-.2 ^
XI a t -a f
|a :1a Js^ Js
™ 8jj ^8
POP
Is-- a
5 o
t« J3 3 <
•°i ^X -"i
|8 :1a |8
p i3 P
^^Isim 3 8 -a a .3 8 : .«!
2 2
W hi
S So .
S
2
1
as :S
2 2
o « o*j
_-3j3qJ3
CO o M <> bfl
^ a 3
S.2*-'.S
as-as
•J o ^ o
■5 hJ
Jan. 14, i9i8 Cultuves of Wood-RoUiug Fungi on Artificial Media 59
6o
Journal of Agricultural Research
Vol. XII, No. 2
^ <u
S "I
1& ca
<3
^
X s
l-t «
•o u;5
u 0
6 6
t23
>> 1- lU >>
a o-r; a
&'o o &
01 u o («
H OH
a ot; a
.Si S'.s o
o .tiS'
o o
•05
:3
■^.a .g.
5 5
o
in K
3 2
.CJ3
.ti n
^ 2
.ti fl
& p
6 >■
-O rt g 3 >,3
o o-j; a t;
& o o fc ^
cil u o =« o
H U
55
•^ .ti
a
a a
» 9 -S
^xxxxxxxxxx
Soooooooooo
O >^ i^.^O <r) 1/1 o o o 00
. 00 r^ wi >Avo r^oo ^ 00 «
SXXXXXXXXXX
s.,0000000000
JXXXXXXXXXX
(OOOOOOOOOO
SXXXXXXXXXX
VmOOcCOOOOOOO
J o g o G£oi::£a
5-0 0-0 0-3 -o-o •35
s o
^a-o
CI o
a o a a.i'.i'.r^a
*j : 4-> 4J o o o ■»->
o ; o o>>> o
_>.
>,>,>,>, ^ .^ _^
a a a a^^^a
000 o-^TJ-rr q
a; o a o o o^.^.S?a
o o o*i
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 61
r^ 10 u^ 0
0 i^ r. Co
" °
S
?:R
>.
-■xxxx
XX
X
X
XX
ca
s 0 0 0 0
00 0
0
0
0 0
ui
•0
0
;^«...
A "
0 «
^
0
"& 6 6
odd 6
606
ui
0 u^ « ui
. «0 0 CI 't
00 w
0
-o'
■^
^S
^
1 C C
0
u
oaa 0
POO
>>
C3
SXXXX
XX
X
X
XX
ca
■a
0
*2 0 0 0 0
^ C< « C4 C<
00 0
C3 "
0
0
0 0
0 0 « r.
CO CO
0
w
t 0
1
SXXXX
XX
X
X
XX
a
_3
>.
•a
0
^ 0 0 "1 0
00 0
0
0
0 0
Ti
•a
tn' • •
<^
13
0
.^ d d
6 6 6 0
d d d
«i
0 CO CI C
«8
0
0
00 >n
a
S "^ ■^.
•o-c-a TS
73 •« -o
>>
ca
SXXXX
XX
X
X
XX
"T;
•g
3 :
•0
0
^ 0 0 00 c
CO "•
e "
0
0 0
^
Ov
t:£
1
H
11
0
a
a
U3
v
(U
bl
01
a
3
1 •
(A
s
E
1
8
a
a,
S
"o
0
"5
tn
5
M :
<3
0
0
3
•0
0
.2 0
0
•0
0
z
d 0
•C T3
0
•a
0
•a
0
•a
ca
>
at 6
2 a
a »
>
a
0
• >
Z a
•a &
°2
a
■3 ;
2
1
0
■0
%
2 Mg.^
n
.>;2
01 a
■0
0
11
3-0
•'°'2
<>>
°w ^5
ii 0 >
1^
3^
5 >;>.
u
^Hii.
b)
"S.^
s
Q (D tu 0
*J 0 a*
■w 0
■w ly D
■■i-
cjfcfefe
uto
U
fe
;^(:4
0
•a
S
:
ft,
1
0
.200
000 c
000
s '«. ■«
•q-CT) t:
73 T) -q
>
a
"3
E
0
s
0
0
0
.V
1
1
1
3
>.
■0
0
^ 1 'S
0
0
0
0 c
■2 g .a
g . ca .
^ « u
1
>.
ca
•a
0
0%
II
0
0
a a
0 0
0
>
V >
C
>
>
0
d
3
'3
•*
a
0 0
0 C
0
0
.s
rtj2 3S.2
1-0 rt.Sf « «
C8
00
OCJ
0
0
if
u PQ M
Oh}(J c
•g
—
— |-
S
a
OS
-M a ^
6
■u a
0 K -M
ca
s
ca
•a 1 ^
0 J3 3 c
0 ..g wX
ia
.2= X
- 2 %
"0
^
>
a
a •
0 :
0 •
^ ;
b ■
0
§
i
0
0
2 \
^3 0 2
«5o 0 .
1
1
>
CJ
0
•0
0
0
3
y
rtJ3 3-W.2
0 W 1-
8^ o(ci <:
a-3 s oiQ
-a
5 8
>
>.
0 ■
-0
iS
>>
a
>,fi >. 3 >
^ P ca asi
^00
0
4<
^-"
0 ;
"i
s
^
0 CO M ca^a CI
t Oh4U C
5 "
0>
^ >. >
K
s
III!
11
J
c
!•§
s
•H
■" a
li
■!-> 4
0 «i -y
?i
•a I
= 2d
0 X 3
.3 ■*■
■u -y j:
•^ *■
4-
3
0
•is
t
3
3
C9
a
IS
. 1 1 f
0 0 0 c
OUOO
0 c
00
c
a
S i
1
0 .i: c
>
X
0
0
•a
0
0 o-y
0 W P
> • 0 ^
5 ap 8«
s
a
^cj.M tax 0
• Onto C
J ^ f a asi
> ^ U U
i
•0
3
a
<LI
S
E
<
c
E
<
ja
■t-j
■ 0 4.
: 5 5
H
is
; 1
:^ -i
' ."^
0
0
"c
1.
1
-
•0
0
0
0
8d4
j 1
: 5
^1
••a . <
'di
-52
: 2 •§
0
i '^w a^s
l-S
2
>•>
a c
2i
OS
H
0 e
■ -i
c
c
d
•q
06 ta)"^ '^ c
3 fS D ts
1-
t
C!
E
1
1
'"rt
i J
i 1
3 p
; i
5 C
J i
;P
;a ,
5(3 P
.( 0
c
i
5 <
<
1
P
<
2
c
) P
E
1
<
c
■E
P
' d
la
0
P<
S : a
62 Journal of Agricultural Research voi. xii, no. »
Tables VII to XII show the cultural characters on the lo agars for
Fomes rimosus (two strains) from Acacia roemeriana (catclaw), F.
rimosus from Siderocarpos flexicaulis (Texas ebony), F. rimosus from
Prosopis juli flora (mesquite), F. robiniae from Robinia neomexicana and
a species of Fomes from Juglans rupestris (?). A study of these six
tables fails to show any marked and constant differences in cultural
characters for any of the lo agars for the six strains given, although four
of the six strains represent Fomes rimosus, one F. robiniae and one a
Fomes on Juglans rupestris ( ?) which the writers have been referring to
F. everhartii. Some of the cultures on carrot, malt, and parsnip show
slight color differences, but none great enough to be of any specific value.
On parsnip and malt, the submerged mycelium for all of these strains is
colored, while carrot has colored to colorless mycelium. The dominant
characteristics of Fomes rimosus is seen on beet, celery, bean, and alfalfa
agars. On each of these agars there is but little really aerial mycelium
which is downy to mainly appressed, while the submerged mycelium
extends far beyond the aerial mycelium, giving a peculiar and very
characteristic glassy appearance to the surface of the agar. This peculi-
arity of growth of the submerged mycelium on these four agars is seen
in all of the six strains and indicates that they are one and the same
species — viz, that Fomes robiniae is only a form of F. rimosus, as many
scientists have always believed, and that the specimen of Fomes sp.
supposed to have been collected on Juglans rupestris also belongs to this
species. These six examples show how valuable the cultural characters
can become in determining the specific position of closely related or
identical species. Only one viable sporophore of the Fomes supposed to
have been collected on /. rupestris was available for culture work, and
since this specimen may have been wrongly labeled as to host, the
writers do not wish to place the Fomes so common on /. rupestris through-
out the southwestern United States as belonging positively to Fomes
rimosus until further cultures with specimens absolutely known to have
grown on this host have been made.
The cultural characters of Fomes everhartii, F. arctostaphyli (lo), F.
igniarius from Populus tremuloides, and a species of Fomes from Alnus
sp., sent to the writers as F. igniarius, when compared showed that the
Foines from Alnus is more closely related to F. arstostaphyli than to
F. igniarius from P. tremuloides, but that it is apparently neither of
these two species; nor did it have any of the cultural characters of F.
everhartii. In this instance the sporophore from Alnus sp. was very
similar in all its characters to the sporophore of F. igniarius from P.
tremuloides, yet the cultural characters instantly showed the two were
not the same species.
Polyporus farlowii, P. dryophilus, and P. texanus are three closely
related species which differ but little in their general sporophore charac-
ters. Tables V, VI, XIII, and XIV show the cultural characters of each
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 63
of these three fungi. The cultural differences between P. texanus and
P. dryophilus are much more marked than those between P. farlowii
and P. dryophilus. A study of the tables giving the characters of these
three fungi at once shows marked and constant differences between each
. of the three species. It will be seen that general color resemblances are
not as distinctive in differentiating specific characters as are certain
other factors. Many of these species have a buckthorn brown color on
several of the agars, which indicates their general relationship while
specific characters must be sought in the differences of growth on certain
agars.
The difference in the fruiting of related species on the various agars is
also of much value in differentiating species. For instance, P. farlowii
fruits vigorously on several of the ten culture media producing perfect
and typical pores, P. dryophilus very rarely fruits on the culture media
here given unless the inoculum is fresh sporophore tissue, while P. texanus
has so far fruited on only two of them.
IDENTIFICATION OF UNKNOWN ROTS BY CULTURAL CHARACTERS
The cultural character method here given can be used to determine
what fungus produces a given rot. In a large majority of cases rots,
both heart and saprophytic, will be found without any sporophores
being present to indicate what fungus produced the rot. If careful
inoculations are made from such infected wood, it is a comparatively
easy matter, as a rule, to obtain pure cultures of the causative organisms
and later grow them again on the 10 media given, thus determining their
cultural characters and from them the fungi producing the rots.
The number of species of fungi producing heartrots in living trees is
not very great. A few of them produce rots which can usually be
identified by the character of the rot alone, such as Trametes pini, but
the great majority of them can not be certainly determined by the rot.
For instance, the rot produced in conifers by Polyporus schweiniizii,
P. stUphureus, and Fonies laricis are so similar that no one can be certain
by examining the rot alone which of these three fungi was the cause of
the rot in question.
The rots of structural timbers are more numerous than the true
heartrots and the causative organism producing each rot is more difficult
to determine from the rot alone, since the majority of the structural-
timber rots are very similar in general appearance. Most of these rots
are of the carbonizing type, where a brownish, brittle rot is produced
which on dessication breaks up into little cubical blocks of varying sizes.
Pure cultures from these structural timber rots will differentiate them
as to their causative organisms at once. The value of such means of
determining the causative organism from the rot in the absence of any
sporophore is very evident to anyone who has had to deal with such rots.
64 Journal of Agricultural Research voi. xii. No. a
EXAMPLES OF IDENTIFICATION OF UNKNOWN ROTS BY CULTURAI,
CHARACTERS
The following examples illustrate how the cultural characters here
described may be used to determine the causative organism of unknown
rots:
(a) There are two common heartrots found in conifers in the western
United States. One is the redrot caused by Trameies pint and the other
is what the senior author has previously called western redrot (9) . In
their early stages of growth these two rots resemble each other very
much. Pure cultures from each of them differentiate the two immedi-
ately, since the pure cultures of the fungus which causes western redrot,
called previously Polyporus ellisianus, produces entirely white cultures
on all of the 10 media, while the cultures of Trameies pini are varying
shades of brown.
(h) On a recent field trip in eastern Texas, when the senior writer was
studying the rots of bridge timbers and railroad ties, he often found a
certain rot in driven bridge piling made of creosoted longleaf pine (Pinus
palustris). At that time in the year there were no fruiting bodies present
on the rotting piling and the question at once arose as to what fungus
was the cause of this serious rot. From the peculiar odor of the freshly
opened wood as well as the character of the rot it was believed that it
was caused by Leniinus lepideus, but nothing could be determined
definitely from an examination of the rot alone. Specimens of the
rotted piling were forwarded to the laboratory and cultures made from
them. Pure cultures were obtained showing all of the characters of
L. lepideus when grown on the 10 cultural media. Later many of these
cultures developed the typical sporophores of L. lepideus. Cultures of
this fungus can usually be recognized by the presence of abortive sporo-
phores which develop on the surface of the agar and also by the presence
of large, thick-walled, obovate to subglobose spores in the submerged
mycelium.
(c) While studying the rots of cypress (Taxodium disiichum) ties in
wet locations in eastern Texas two unknov^Ti rots were found in this
wood. Pieces of the rotting ties were sent to the laboratory and pure
cultures of two different fungi were obtained. One set of these cultures
showed all the cultural characters of Leniinus lepideus, while the other
series produced a sporophore of an unknown species of Poria.
{d) A rot in the heartwood of Quercus gamhelii, collected in New
Mexico, was determined in the field as caused by Polyporus dryophilus.
Cultures of the diseased wood gave all of the cultural characters of
Fames everhartii and none of those of P. dryophilus.
(e) A specimen of rot in Pinus ponderosa, which was supposed to be
caused by Polyporus sulphureus, was received from Oregon. Cultures
from the wood showed that the rot was unquestionably caused by
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 65
Lentinus lepideus. Since the rot was sent in as a specimen of heartrot
found in the western yellow pine in Oregon, the writers are wondering
whether L. lepideus produces a real heartrot of living pine in that State
or whether this rot came from a dead area on a living or from a dead tree
which had been attacked by this saprophytic fungus.
Many other instances could be cited of the determination of the causa-
tive organism of a rot by use of the cultural methods here outlined.
Usually it is not even necessary that the sporophore stage should be
developed, since the vegative cultural characters on the 10 special media
will usually determine the identity of the fungus. The practical impor-
tance of such a method of determination is of great value and is easily
recognized by anyone who has worked for any length of time with organ-
isms of this character. In fact, one of the worst stumbling blocks to a
successful study of the various rots of wood, both saprophytic and
heart rots, has been the lack of methods by which the organisms producing
these rots could be grown and identified in pure cultures on artificial
media.
SPOROPHORE PRODUCTION
The fact has long been known that the production of sporophores in
nature in many of the Hymenomycetes was more or less dependent
upon light. This fact has also been demonstrated for a few species mainly
by gross cultures on dung decoctions, pieces of wood, bread, etc.
Buller {4) has shown among other things how the light influences sporo-
phore production for a few species, mainly Agaricaceae, but none of his
experiments were made with pure cultures on artificial media. It has
been taken for granted that light was essential to the formation of
sporophores of the wood-rotting fungi, including the Polyporaceae, but
such had never been proved with pure cultures under control conditions on
artificial media with a sufficiently large number of species to determine
the actual influence of light as well as other factors on sporophore pro-
duction.
The studies here made indicate that there are many Polyporaceae
which fruit in diffused light of varying degrees of intensity and others
apparently require the direct rays of the sun to produce perfect spore-
bearing sporophores, while some can form sporophores in absolute
darkness. Since this is only a preliminary report, no attempt is made to
determine a large number of factors which should be ascertained in a
complete study of this phase of fungus life. For instance, there must
be a minimum, optimum, and maximum condition as to light, heat, and
moisture under which a given fungus will produce sporophores.
It will be seen by consulting Table XVI that only three species,
Polyporus farlowii, Trametes serialis, and P. cinnabarinus , were able to
develop sporophores in absolute darkness. Only the first two of these
fungi produced both sporophores and spores. P. cinnabarinus pro-
duced fairly typical pores, but no spores were found.
27805°— 18 3
66 Journal of Agricultural Research voi.xii.No.a
METHODS OF EXPERIMENTATION
In the earlier sporophore study the tubes containing the cultures
were kept both horizontally and vertically. If the species under investi-
gation produced sporophores at all, they were able to produce them in
either position. However, the practice was soon abandoned of placing
the tubes in a vertical position, since it was found very difficult to obtain
uniformity in the proper Ughting of the cultures and to get spore prints
from the sporophores produced in such a position.
The general method followed in the sporophore studies was to take a
series of tubes on different agars and place them in the same general posi-
tion that was described in the study of the cultural characters. It was
found very important early in the study that the slant side of the tube
should be kept uppermost and that the relative position of the tube
in reference to gravity and sunlight should be the same throughout the
experiment. The tubes were so placed that the cotton plugs were
away from the simUght and the bottom of the tubes faced the light.
As soon as there was any indication of a hymenium forming, the tubes
were placed with the slanting surface downward. After this was done,
the sporophores usually continued to develop normally, and in due season
spores would be formed and discharged against the side of the tube op-
posite the hymenium.
The first sets of fungi which were kept in darkness were placed in
pasteboard boxes in a horizontal position and these boxes inclosed in
other pasteboard boxes. It was found, however, that some diffused
light reached the tubes in spite of the double-box arrangement. Inside
one of these boxes the recording tube of the soil thermograph was
placed, while the registering portion of the instrument was kept on a
shelf outside the boxes. These boxes were kept on the shelf beside the
other bcJxes containing the tubes which were exposed to the direct rays
of the sun. This was done in order to obtain as near as possible the
same environment for the tubes kept in the sunlight and those kept in
the darkness, except for the single factor of Ught.
In a later series of experiments, the ones which are recorded in detail
in this paper, the cultures were kept in pasteboard boxes in a horizontal
position and these boxes were placed in a photographic dark room from
which all light was excluded by means of sheets of cardboard being placed
over the ruby lights. By this arrangement absolute darkness was ob-
tained for the cultures. The soil thermograph was arranged in the same
manner in the dark room as when on the outside. Table XV shows the
highest maximum and minimum, the lowest maximum and minimum,
the highest and the lowest mean, and the average of the daily maximum,
minimum, and mean temperatures for each month as recorded by the
two thermographs.
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 67
Table XV. — Temperature records (°F.) for cultures grown in daylight and in darkness
Date and degree recorded.
In light.
In dark.
Maximum.
Minimum.
Mean.
Maximum.
Minimum.
Mean.
December 18-31, 1916:
Highest
Lowest
80
72
66
55
73
65
75
66
65
57
69
64
Average
76
61
68
71
61
66
January, 1917:
Highest
Lowest
89
65
70
52
77
58
78
63
73
58
74
60
Average
79
60
69
71
68
69
February, 191 7:
Highest
97
74
67
56
81
66
86
68
72
59
78
65
Lowest
Average
87
60
73
77
65
71
March, 1917:
Highest .
96
71
72
50
80
62
88
73
80
62
83
68
Lowest
Average
86.5
59
72.7
79
68
73
April, 1917:
Highest
96
69
66
46
77
65
86
68
80
65
80
Lowest
67
Average
82
59
70-5
78.7
70.8
74-7
May, 1917:
Highest
100
61
67
49
83
56
77
60
74
57
75
59
Lowest
Average
So
60
70
71
68
69-5
June, 191 7:
Highest
no
82
78
50
93
68
92
70
88
69
90
70
Lowest
Average
96
■ 70
83
84
80
82
EXPLANATION OP TERMS USED IN SPOROPHORE TABLE XVI.
In Table XVI most of the headings are self-explanatory; however, there are a few
that need special explanation. The heading for the sixth column is " Inoculum and
date of inoculation." Under this heading the character of the inoculum used when
sphorophores were produced is given, as well as the date or dates when each tube
which produced sphorophores was inoculated. "Wood" means the inoculum used
was infected wood; "tissue" means pieces of sphorophore tissue, while "malt,"
"potato," etc., signifies that mycelium, was used as an inoculum and that it was
taken from a cultiu-e on malt agar, potato agar, etc. " Inoculum as used means
the piece of wood, tissue, etc., used to inoculate the media. " Development period of
sporophores" means the number of days from the time the inoculation was made to
the first evidence of the formation of a hymenium.
68 Journal of Agricultural Research voi. xii, no. 2
Table XVI shows that the authors have obtained under the cultural
methods described 629 sporophores in the light and 11 in darkness, 640
in all, representing 4 families (Agaricaceae, Polyporaceae, Thelepho-
raceae, and Tremellaceae), 16 genera (Coprinus, Daedalea, Exidea, Fomes,
Ganoderma, Irpex, Lentinus, Lenzites, MeruUus, Panus (?) Pleurotus,
Polyporus, Polystictus, Poria, Stereum, and Trametes), 42 species, and
97 strains from 65 host species collected in 11 States. If the genera
given in North American Flora are used, there would be 24 genera, since
the following 8 genera would be added : Coriollelus, Elfvingia, Inonotus,
Laetiporus, Pycnoporus, Pyropolyporus, Spongipeilis, and Tyromyces.
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 69
M O H
ti u" B u
+^" s, S a S
o a_.-
Ph «JJ
, CT. 'J Ov i^ Cv ij O I
i M M '^ w
K 'i !iS <'' S> ">
4j - " *; . " .
o " -
» 3 "
'^cB
0.T)
o> >> >, >■
iSSS
^" - ? ^ K"
>.>.>■
>'.P
jQ
8 a? E-S-
so"or?o"oS-"oc"SSHSoS~"SS3oSJ5dSdSaoi5dogK*'S
fi 3. 3; a 4; 8^3- vj a.^" «j"j3'^. 34
•- a
S^
f-;S
^H
^H
o o o— o
•0T3T) g-O
8
• Si
ji a
.-^ 0 c a ~ -^ 2 -^ ,
o
s S
!°-S.9-$.8
aS-S-al'^ ssis "so
*.2-§ ° O o.^:-^
2.2.*.
5<^
o S S S o >~-Si^ « •? <j ;;ra ;j o o si-S:
a,'>'>.oa,'<a,'§>-ja,'<a,i/osai 0,0,0,
a :'^
acfo, Q,
I
:!■§•
a &
O o
Ocni^i^mo'da^ 6 6 6 6 6 6 §!S d d
:S:Sgg : :•£! ::::::«; g : :
■fHS^H : ■.<iz : : : : : :o^ • •
3 § S
.s.S-
:*.t:-~>
s 2 S i- 00^ S'
ooa'HSSooS -5
It, f*.
^ ^ s*
►2 ::§
8-3
0,0,
8 S
00 00 >Soo oo o i^t^oo^ooof'^ooo wiowTj-wrroooo^oo ^o oor^t^t^"^ *ooo r^ jj-oo
70
Journal of Agricultural Research
Vol. XII, No. 2
o
*i C
§
.9 3
o
"o g
S
41 O
o
t->
1
1
*«
SJ
1-.'^
p
t^
n a
f,<
•3ii'
n
S
u
«^
o?
U
< < ^
Ph W O W Ph Pk O
0>M
< aa
^"2 2
■u* aT o • iT
t^ t^ p,
Oi o> ..
o> oi 2' 2' o\ I
tCi-1 M *"> o! ti « N M n" N '^ IT'S " '^ § 9 '^ f^ a
^ ;?
■©,
I 1
fell
a s is
a, 0,^0,
h a a
a, a
>
<
3 5 =3
fe <J :2;
;2;3
S d g
:q, q.
0\ Oi r* 00
O O C* CO
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 71
72
Journal of Agricultural Research
Vol. XII, No. a
o 3 te
5 5
J2
a
o
.^•i
H s O
PQPhM PiW cw
o o u
> l; S
O- .000000-- •■ ■ • o
00 r^ »^00 ro t^ « 00 CO
SO fO«c« f^Ooo O
3 O J5
o ■ -o-ooooo---- o
S "3 o K E
SB^B
rt 1 t:
a"!-
o o.
u o
set: . „
C c« p m
S fl <-> o J5
0. b _- . .-• n,
g « a rt
Sa
^•50 J
ii c H a S gifi" n d 3 p g ^J-a dpSi{|HS«|dSag5iiSg o - j
3 •- S >-'h~ rt
TrSje "S « s 2
III
s SS
28
S^:
O O Qf^ o o
.2 2 ■ wis I* K ■ -.2 ^ •
6 6 6 6 6 ui'^
'O'O'd'd'O ca t»
H^H
■2.8
11
!•-
.§■§
o d 6 *^ ^ 6
qtij
2<
a S
^JJ^Kj
f^ O t^ in foco M 00 't w 00 M ro
00 O t^ r^OO "^OO O t^O OD w >0
HMrtMHMMWMMlHOlO>
ovO O On t/lvO
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 73
2 a!
Bo
g
.S 6
^ 5
M 3
£it:i «^
a 0)
ngUsh
dine re
renadiii
scarlet.
Do.
So
w 0
P^
5«
50.5
A -i
i. i
iij 1^ so.
rt ffl O »
>^s^^«
oH V O
\A 3
"S -^^'^-St!" 8 S p §
•^■5t5-^5ca^-">;n
*-' u - O ±^ rt J=^
•- O 9 S <
5^
13 a ;3 „-
o -o S S o u
-oi
aa
s a . a
aa oS
ggiioSaBgS§S2s^--g
ga«aafeSaj5 5SS8a«„-
U S OU OP4UCJ Sfu
<.> o
a a
a a a.
So
<5i
1
1
.2 ^
3 o
t, <
o o ^ja
2-
ci.a,
00 00 f*l*0 fO N
74
Journal of Agricultural Research
Vol. XII, No. 2
s|.^ .
h R o _: o
0
0 0
Q
PQ
•3 « .Ef^
AhU h4
-0
S S J'
> S g
h2
'^1-6
•o 0..2
►g; a a
as
Sh«d
sy
^13
O o
a CO
5 '«'3 "m
16 „
^ad
S " ao a a
0 o
8 H.
^■- E
i'^'d!
c. 3
S a
- . 5 1) O H >v
i'>--wr<-»!3<j-i-rrr!-w'''?<-i-'t.!3C^'-22i-i
3 aaggj
-2 "3 •
■2 -5
■S S
•o-a ^
gS
a;
^; <'A
o <u
3T3
!2; H
§2
0,0, :o.
£5 20
O O Oi 00
t^ O O O 00
00 O^ OtOO
Jan. 14. 1918 Cultures of Wood-Rotting Fungi on Artificial Media 75
i «±; 01 o
"0 t- 'o +j -*j
J3 CI rt « a
o o d.3 oj-H 4) ca g a 2 ^ •» '
« 0
ca rt 3 —
Au
o caja o
OOP
2
.hi
t>. o»
00
t^
000
i.
in
k
J,"
1^00 ^a
i a
ca o cij
t! S d dg fc .0.
•ast:at5S
■H b^fl hrS"
oH
flS« § art O-a « n Qg rt &iH H w^ Ubj H o art
G a~
(-.a
a, a,
-idS
^3i '
i
xo do t^ o o\
76 Journal of Agricultural Research Voi. xii. No. »
Sporophores were produced when various kinds of inocula were used,
such as wood, tissue, spores, or the myceUum from other cultures and on
a great variety of artificial culture media. The wide range of genera
and species covered by this investigation, together with the large number
of sporophores produced, clearly proves that the sporophore production
here reported is not an accident, but is a constant and permanent per-
formance determined in the great majority of cases by the presence of
light and to a very limited extent by the character of the substratum.
Aeration and humidity are two other factors which also enter into sporo-
phores production on artificial media.
INFLUENCE OP SUBSTRATUM ON CHARACTER OF HYMENIUM
The influence of the host or substratum on the character of the spore-
bearing surface is well illustrated in the several strains of Polysticius
hirsutus when grown in artificial cultures. In all of these strains the best
developed and well-defined pore surfaces are produced on carrot, malt,
and parsnip agars. On these the pore surface is fairly typical of that
produced in nature, both as to size and color of pores. In the prune agar
the pore surface is usually reduced in the majority of the tubes to a few
scattering irpiciform spines. On some of the prune-agar tubes the spines
are not flattened like those of an Irpex, but are round like those of a
Hydnum. In the corn-meal tubes the hymenium in a large majority of
cases is very similar to that produced in various species of the Thele-
phoraceae, being reduced to a smooth or slightly granular surface in which
there are no definite pores. In all of the tubes (carrot, malt, prune, corn-
meal, and parsnip agars) there is an enormous production of spores irre-
spective of the character of the hymenium. In fact, in some of the tubes
the first evidence of any sporophore production is the deposit of spores on
the opposite side of the culture tube when even a careful examination
with a hand lens fails to show any signs of pores or spines.
The size, shape, and coloring of the pores and tubes produced in artifi-
cial cultures on many of the agars are practically identical with those
found in nature for a given species. However, it often happens that in
abortive sporophores on certain agars the coloring is not as pronounced
as on agars where the sporophores reach their full development. In
Polyporus cinnabarimis, for instance, the sporophores produced on
potato agar are nearly white, while the same strain will produce the
typical grenadine- red to flame-scarlet pores on malt.
TRUE PILEI IN ARTIFICIAI, CUI.TURES
One of the most interesting facts brought out in this investigation was
that in all of the thousands of cultures made with the hundreds of sporo-
phores produced not a single one had a typical pileus, unless the fungus
was a gill-bearing form ( Agaricaceae) , when the culture tubes were so
placed that the slant faced the light in such a manner that its rays were
Jan. 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 77
more or less at right angles to the agar surface where the hymenium was
being developed.
At first it was believed that the absence of pilei was probably due to
the fact that the tubes were left in a horizontal position and therefore
pilei had no chance to develop. The same results were obtained when
the tubes were placed vertically with the slant side facing the light as
when placed horizontally. Very recently, however, the writers have
devised a method by which small but otherwise typical pilei have been
grown on artificial media in the test tubes. The method was as follows :
In place of arranging the culture tubes so that the light fell directly on
the slanted surface the tubes were placed in one of two positions (i)
vertical in opaque boxes in such a manner that the rays of light would
fall on the tops of and parallel to the tubes and none on the sides or
bottom, and (2) nearly horizontal, with slanted surface of the agar
turned downward but ^^dth the light again falling only on the tops and
parallel with the tubes. Typical sporophores were produced by this
means for Polyporus dryophilus , P. hirsutus, and Fomes rimosus, the
only species tried so far. Whether this method will produce pilei with
all species is not known. There are objections to both methods where
one desires to obtain spores for plating in order to obtain individual spore
colonies — viz, when the culture tubes are kept vertically the discharged
spores fall on the agar rather than on the inner surface of the culture
tubes and when the culture tubes are kept nearly horizontal with the
slanted surface downward from the first the mycelium grows around the
edges of the agar onto the glass, thereby covering the surface of the tubes
where the spores will fall. There is also this further objection to both
methods — viz, that the agar in drying separates from the glass tubes in
a very irregular manner in place of from only the top surface as it does
when the tubes are kept horizontal with the slanted surface uppermost.
INFLUENCE OF LIGHT ON THE FORMATION OF PILEI
The pilei of the Polyporaceae always developed in such a manner that
their tops were directly toward the sunlight. Also the pilei of the
Agaricaceae when grown in cultures were strongly proheliotropic from
the very beginning of their formation. This positive heliotropism was
especially marked when the sporophores of Lentinus lepideus and Pleuro-
ius ostreatus were developing. The writers tried P. osireatus in three
different positions: (i) Culture tubes placed vertically but with the
lower part so shaded that the light entered at the top of the tube; (2)
tubes placed horizontally and covered with black paper so that the light
entered only at the top of the tube and this placed toward the light;
(3) the third experiment was made in a flask in which the medium was
slanted on the side of the flask. The flask was placed upright and the
development of the sporophores from this more or less slanting surface
was observed. In every instance from the earliest development of the
78 Journal of Agricultural Research voi. xii, No. 2
sporophore to the complete expansion of the pileus, the sporophores
always pointed directly toward the light. In the case of the vertical
tubes it was to be expected that normally these sporophores would grow
vertically, since it was presumed that gravity as well as sunlight might
be a factor in the upward growth of the sporophore. When the tube
was placed horizontally, a totally dififerent condition existed. In this
case the sporophore developed directly toward the sunlight but at right
angles to the force of gravity. In the experiment with the flask the
upper two-thirds of the flask was covered with dark paper. Ten or
fifteen sporophores developed on the slanting surface of the agar in the
flask, all of them without exception pointing downward toward the source
of light.
Lentinus lepideus was also grown on a more or less vertical slant with
the face of the slant turned towards the light. Perfect sporophores
which developed under these conditions always turned directly toward
the sunUght.
INFIyUENCE OF GRAVITY ON THE FORMATION OF PORES
In all of the experiments conducted with the Polyporaceae the formation
of the pores was always parallel to the action of gravity. If the tubes
were left in their original horizontal position with the slant side upward,
pores developed on the slant with their mouths pointing upward, thus mak-
ing them parallel to gravity. If, on the other hand, the culture tubes were
placed in a vertical position, the pores were formed on the sides of the
slant in such a manner that their mouths pointed more or less downward,
or both upward and downward in some cases and parallel to the force of
gravity. In fact, the position of the pores of a fungus which produces a
vigorous sporophore in artificial cultures could be governed at will by
simply changing the position of the tube in reference to gravity. It seems
therefore that while light is usually the main factor governing the initia-
tion of a hymenium, gravity is the dominant force which determines in
what direction the pores will point, irrespective of the incident light or
whether the spores when discharged will fall onto the walls of the porec
or not.
INFLUENCE OF SUBSTRATUM ON PORE FORMATION
Although it is a well-known fact that cultures of strictly parasitic
fungi can fruit on a wide range of media irrespective of the special host
on which the parasite usually thrives, the idea seems to have been gener-
ally accepted that some special culture media would have to be used in
order to obtain sporophores of the Polyporaceae on artificial media in test
tubes, flasks, etc. The experiments here given show that the same gen-
eral rule as to food material appUes to the wood-rotting fungi as a whole —
viz, that no special decoction is necessary for each species in order to
obtain sporophores; in fact, the wide range of artificial media upon which
many of the Polyporaceae have fruited is rather remarkable when the
nature of these fungi is taken into consideration.
Jan, 14, 1918 Cultures of Wood-Rotting Fungi on Artificial Media 79
Table XVI shows that there were 56 strains of wood-rotting fungi
which produced sporophores on malt agar, 52 on corn meal, 48 on prune,
44 on carrot, 27 on parsnip, 14 on potato, 11 on celery, 11 on beet, 10 on
bean, and only i on alfalfa agar, while there are four of these agars pre-
eminently suitable for sporophore production — viz, malt, com meal,
prune, and carrot, in the order given.
POSITION OF SPOROPHORES ON MEDIA
One most interesting fact in reference to sporophore production on
artificial media is the fact that a very large percentage, probably more
than 95 per cent, of the sporophores were developed on the upper half of
the slant. No perfect sporophores have ever been observed by the writers
other than on the slant, although the agar in the tubes often dries out
sufficiently to leave ample room for the sporophores to develop from the
slant to the bottom of the tube over a distance of from 60 to 80 mm.
Whether this means that the formation of the sporophore was dependent
upon a small amount of moisture or whether it needed the greater aeration
which the upper end of the tube afforded is not known. In many in-
stances, especially with the Stereums, the fruiting surface was formed in
a narrow zone at the extreme upper limit of the agar slant. The studies
so far made indicate that an important factor in this case is probably
the drier condition of the agar at the upper end of the slant. It would
seem that if the drying of the agar was the only requisite for the forma-
tion of sporophores that as the agar dries in the lower portion of the
tube a condition would be reached which would normally produce sporo-
phores; yet such is not the case. The by-products produced by the
fungus may be more or less deleterious to the formation of sporophores;
and, since such by-products would be more abundant in the lower portion
of the tube than under the slant, it follows that few, if any, sporophores
would develop there.
DENSITY OP MYCELIUM AND SPOROPHORE PRODUCTION
Another very peculiar fact develops that the sporophore production
usually occurs on that portion of the agar slant where the aerial growth
of mycehum is the least. If a dense mass of mycelium forms over the
entire agar slant, the chance for the formation of pores on such a surface
is materially lessened. This may explain to some extent why sunlight
plays an important part in the production of sporophores, since cultures
kept in the dark usually develop a denser mass of mycelium on the surface
of the slant than corresponding cultures in the light. There are many
species of fungi which have been tried in both light and darkness in
which the aerial growth is very limited, even when grown in the dark;
and still no sporophores were produced when in darkness, but were pro-
duced in the light.
8o Journal of Agricultural Research voi. xii, no. 2
INFLUENCE OF INOCUI.UM ON SPOROPHORE DEVELOPMENT
In a few of the species of fungi the interesting fact developed that
when tissue from sporophores was used for the inocula the presence of
this tissue materially shortened the development period for sporophores
on the agar. For instance, in Polyporus alhidus from Pinus ponderosa
(FP 21875), when tissue was used as an inoculum on corn-meal agar,
the development period of the sporophore was 5 days; when mycelium
from potato agar was used the development period was 1 1 days ; for
P. anceps from Thuja plicata (FP 21 801), when tissue was used on malt
the development period for the sporophore was 5 days; when mycelium
from potato agar was used for the inoculum on malt agar the development
period was 12 days. This rapid formation of the sporophore when tissue
was used as an inoculum was characteristic of many of the strains of
this group of fungi.
This shortening of the development period of the sporophores is also
found occasionally in other species of fungi than this group. For instance,
it was rather marked in Polyporus dryophilus when fresh sporophore
tissue was used, in various strains of Trametes peckii, and to a slight
extent in Fomes roseus. In several cases not only was the development
period for sporophores shortened by the presence of pieces of sporophore
used as the inoculum, but cultures made from infected wood, mycelium
or spores produced sporophores only on one or two agars, while the same
species would produce sporophores on several media when the inocula
were pieces of sporophores. In those cases where the sporophore tissue
shortened the development period, the pores usually but not always
start directly on the tissue inoculum and then spread rapidly often over
the entire agar slant. In no case was the development of the pores
limited to the pieces of inoculum, while in many instances the pores
would start on areas not immediately adjacent to the inoculum.
SUMMARY
(i) The following criteria were found of value in the differentiation of
the various species: (a) Macroscopic characters, including rapidity of
growth, color of aerial and submerged mycelium, character of the aerial
mycelium as to texture, etc., staining of the agar, decoloration of the
agar, the comparative rate of growth between the aerial and submerged
mycelium, etc. (b) Microscopic characters, such as septation, branching,
size and color of hyphge, clamp connections, polymorphism in spore for-
mation,' etc.
(2) The sunlight was found to accentuate the colors and tone down the
mycelial growth of the fungus, thereby making it more characteristic and
uniform for a given species than when placed under similar conditions in
the darkness or in weak diffused light.
(3) The cultural characters of vegetative development of the various
strains of a given species of fungus show no appreciable difference between
cultures of this fungus whether obtained from infected wood or from
Jan. 14. 19^8 Cultures of Wood-Rotting Fungi on Artificial Media 8i
sporophores; neither do the hosts of the fungus seem to make any
marked changes in the fundamental cultural characters when strains
from different hosts are compared. There may be minor differences due
to the host from which the strain came but nothing more.
(4) When the cultural characters of closely related but really distinct
species are compared, marked and constant differences in the character
of the mycelium will be found on certain corresponding agars in the
series of cultures representing the two species, while if the two fungi are
really the same species, no constant differences of specific rank will
occur. Unknown rots can also be identified by making pure cultures of
the causative organism from the diseased wood and determining from the
cultural characters of the fungus thus isolated its identity.
(5) The presence of light is essential to the production of sporophores
when grown on artifical media in the great majority of fungi here inves-
tigated, while the character of the substratum plays only a very minor
roll in sporophore initiation.
(6) The medium on which the fungus is grown often governs to some
extent at least the form of the hymenium which develops.
(7) The size, shape, and color of the pores and tubes produced in
artificial cultures on many of the agars are practically identical with
those found in nature for a given species.
(8) The pilei of both the Poplyporaceae and the Agaricaceae when
grown in pure cultures on artificial media are from the very beginning
of their formation strongly proheliotropic, while the formation of the
pore tubes in the Polyporaceae is always such that they are placed
parallel to the action of gravity.
(9) In a few species of fungi the presence of tissue as the inoculum
shortened the period of sporophore development from one to several days,
(10) Workers with wood- rotting fungi now have the following means
for determining the identity of a given fungus or the causative organism
of a given rot : (a) The sporophore characters as usually found in nature,
(b) the characters of the rot produced, (c) the vegetative characters
developed when grown in pure cultures on artificial media when exposed
to light, and (d) the characters of the sporophores and various spore
forms when produced on artificial media.
LITERATURE CITED
(i) BrEFELd, Oscar.
1877. BOTANISCHE UNTERSUCHUNGEN USER SCHIMMELPILZE. III. BASIDIO-
MYCETEN I. 226 p., II pi. Leipzig.
(2)
1888. UNTERSUCHUNGEN AUS DEM GESAMMTGEBIETE DER MYKOLOGIE.
VII. BaSIDIOMYCETEN. II. PROTOBASIDIOMYCETEN. 178 p., II pi.
Leipzig.
(3)
1889. UNTERSUCHUNGEN AUS DEM GESAMMTGEBIETE DER MYKOLOGIE.
VIII. BASIDIOMYCETEN. III. AUTOBASIDIOMYCETEN. 305 p., 12 pi.
Leipzig.
27805°— 18 4
82 Journal of Agricultural Research voi. xii.no. a
(4) BuLLER, A. H. R.
1909. RESEARCHES ON PUNGi. 287 p., 83 fig., 5 pi. London, New York, [etc.]
(5) Falck, Richard.
1902. DIE CULTUR DER OIDIEN UND IHRE RUCKFUHRUNG IN DIE HOHERE
FRUCHTFORM BEi DEN BASiDiOMYCETEN. In BeitT. Biol. Pflanz.,
Bd. 8, Heft 3, p. 307-346, pi. 12-17.
(6)
(7)
1909. DIE LENZITES-FAULE DES CONIFERENHOLZES. In MoUer, A. Haus-
schwamm-Forschungen. Heft 3, 234 p., 24 fig., 7 pi. Jena.
1912. DIE MERULius-FAULE DES BAUHOLZEs. In MoUer, A. Hausschwamm-
Forschungen. Heft 6, 405 p., 73 fig., 17 pi. Jena.
(8) Humphrey, C. J., and Fleming, Ruth M.
1915. THE TOXICITY TO FUNGI OF VARIOUS OILS AND SALTS, PARTICULARLY
THOSE USED IN WOOD PRESERVATION. U. S. Dept. AgT. Bul. 227
38 p., 4 pi. Bibliography, p. 37-38.
(9) Long, W. H.
191 7. a preliminary report on the occurrence op western red-rot
IN PINUS PONDEROSA. U. S. Dept. AgT, Bul. 490, 8 p.
(10)
1917. THREE UNDESCRiBED SPECIES OP POLYPORES. /» PapcTS New Mexico
Chapter Phi Kappa Phi, v. i, no. i, p. 1-3.
(11) Lyman, G. R.
1907. culture studies on polymorphism of hymenomycetes. in proc.
Boston Soc. Nat. Hist., v. 33, no. 4, p. 125-209, pi. 18-26. Literature,
p. 203-209.
(12) RUMBOLD, C.
1908. BEITRAGE ZUR KENNTNIS DER BIOLOGIE HOLZZERSTOrENDER PILZE.
In Naturvv. Zt«chr. Forst. u. Landw., Jahrg. 6, Heft 2, p. 81-140,
6 fig., I pi. Literatiu-, p. 139-140.
(13) Zeller, S. M.
1916. studies in the physiology op the pungi. ii. lenzites saepiaria
FRIES, WITH SPECIAL REFERENCE TO ENZYME ACTIVITY. In Ann.
Mo. Bot. Gard., v. 3, no. 4, p. 439-512, pi. 8-9. Bibliography, p. 504-
509-
(14)
1917. STUDIES IN THE PHYSIOLOGY OP THE FUNGI. III. PHYSICAL PROPER-
TIES OP WOOD IN RELATION TO DECAY INDUCED BY LENZITES
SAEPIARIA FRIES. In Ann. Mo. Bot. Gard., v. 4, no. 2, p. 93-164,
pi. 9-13, II charts. Bibliography, p. 154-155-
GOSSYPOL. THE TOXIC SUBSTANCE IN COTTONSEED
By W. A. Withers and Frank E. Carruth,
Chemical Division, North Carolina Agricultural Experiment Station
REVIEW OF PREVIOUS WORK
Since our previous publication (77)^ on this subject, several articles
have appeared in which other explanations of cottonseed-meal poison-
ing have been offered.
Thus, Rommel and Vedder (14) have suggested that poisoning by
cottonseed meal is similar to beriberi, and is caused by deficient diets.
This view was based on the similarity of post-mortem symptoms noted
in pigs fed on rice and tankage.
Wells and Ewing (15) have concluded that cottonseed-meal injury is
due in large part to incomplete diets.
Richardson and Green (11) fed white rats and concluded that cotton-
seed meal and flour are not actively toxic, but contain insufficient
minerals and possibly inadequate amounts of the fat-soluble growth-
promoting substance.
Osborne and Mendel (10) have secured results similar to those of the
last-named authors with cottonseed meal and flour, but on subsequently
feeding raw cottonseed kernels supplied by us, they have corroborated
the results which we had obtained with the kernels. They admit the
presence of a deleterious substance in raw cottonseed, but apparently
still hold the view that cottonseed meal, the product resulting from
cooking the kernels and pressing out the oil, is nontoxic, at least for rats
and chickens (9).
Inasmuch as no comparative experiments with an isolated and purified
substance have been reported, we present the results of additional
experiments with various animals to supplement those given in our
previous experiments, in which rabbits and fowls were used.
The toxic efltect of an ether extract of raw cottonseed has been well
shown in the rat-feeding experiments described by McCollum and co-
workers (6) and by Osborne and Mendel (10). This extract contains
about 2 per cent of gossypol, which is equivalent to about 0.6 per cent
of the weight of the kernels from which the extract is obtained. Our
rat diet, containing 20 per cent of this extracted oil, caused prompt
decline in grown rats. Osborne and Mendel (10) used as little as i per
1 Reference is made by ntunber (italic) to "I,iterature cited," pp. loo-ioi.
Journal of Agricultural Research, Vol. XII, No. 2
Washington, D. C. Jan. 14. 1918
Im Key No. N. C— 8
(83)
84 Journal of Agricultural Research voi.xii. No. a
cent of the extract (equivalent to about 0.02 per cent of gossypol in the
diet) and found that the growth of the rats was greatly retarded. Our
experiments have led to the conclusion that raw cottonseed kernels are
highly toxic to rats, but that cooked cottonseed is only slightly toxic.
Whether cottonseed meal made from cottonseed sufficiently cooked
with moist heat is toxic to rats seems to depend on the diet in which it
is fed. In a short feeding experiment a diet such as was used by Rich-
ardson and Green (ii) (45 per cent of cottonseed meal, 17 per cent of
whole milk powder, 10 per cent of starch, and 28 per cent of lard) has
shown no definite toxic effect ^ on our rats even when it contains a
short-cooked (28 minutes) meal. When the meal is the sole source of
vitamines, protein, and minerals, we have rarely had such favorable
growth as is reported by Richardson and Green (//) and by Osborne
and Mendel (/o).
If compared with ether-extracted raw cottonseed or with soybean
meal, the rate of growth has been very small. The explanation of this,
according to Osborne and Mendel, might be that the diet was unpalatable
and that consequently less food was ingested. Unpublished experi-
ments indicate to us that there still remains something toxic in long-
cooked cottonseed meals which in restricted diets is objectionable to
rats, causing a lower food intake, but the effect of which is overcome in
supplemented diets.
This seems to be the same phenomenon discussed by McCollum (7),
who finds that in diets very well supplemented the toxic effect of the
fat of wheat embryo and other slightly toxic substances is overcome.
The evidence of a toxic factor of moderate intensity for rabbits and
pigs is also furnished in our rabbit and pig experiments.
TOXICITY OF RAW COTTONSEED KERNELS
In a large number of experiments with rats we have found that the
effect of cottonseed products on rats can be predicted accurately when
chemical tests indicating the presence or absence of gossypol have been
made (Table I). In order to appreciate the significance of some of the
experiments we may briefly describe the properties of gossypol. A
yellow plant pigment having the apparent formula C30 Hjg O9 (molecu-
lar weight 532), not soluble in and not extracted by petroleum ether,
readily soluble in acetone and ether, moderately soluble in alcohol,
benzene, and chloroform, dissolves readily in sodium hydroxid (NaOH)
and sodium carbonate (NajCOg) and is slowly soluble in sodium bicar-
bonate (NaHCOg). It may be titrated as a dibasic acid with aqueous
alkalies. It crystallizes well from a mixture of ether and acetic acid as
a sparingly soluble substance containing 10. i per cent (i molecule) of
'Data ia article not yet published.
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed
85
acetic acid. When dissolved in ether or oil and treated with anilin, it
forms a bright-yellow insoluble substance which is apparently a di-
anilin salt of gossypol (i molecule of gossypol to 2 molecules of anilin)
CgoHjgOg, 2C6H5NH2. This substance is very insoluble in most solvents,
even aqueous sodium hydroxid.
Table I. — Results of feeding raw cottonseed kernels in milk diets to rats
Rat
No.
Weight.
Period.
Remarks.
Initial.
Final.
Change.
Diet 366 (59 per cent of ker-
nels, 17 per cent of whole-
milk powder, 10 per cent of
starch , 1 4 per cent of lard ) .
Do
43
44
45
46
71
72
74
75
58
59
60
60
47
48
45
46
Gm.
^33
217
156
139
154
186
130
117
160
170
131
105
112
184
128
104
Gm.
no
179
128
109
100
112
87
80
107
112
103
122
185
188
177
159
Percent.
-17
-17
-18
-25
-32
-40
-33
-31
-33
-34
— 21
+ 16
+ 65
+ 2
+ 38
+ .S3
Days.
7
7
7
7
18
21
23
19
61
61
59
20
85
85
14
Died nth day.
Discontinued.
Do
Recovered on ether-
Do
extracted kernels.
Do.
Diet 377 (30 per cent of ker-
nels, 38 per cent of milk-
powder, II per cent of
starch, 14 per cent of lard, 7
per cent of butter).
Do
Died.
Do.
Diet 378 (28 per cent of dry-
heated (no" C.) kernels,
equivalent to 30 per cent of
raw, 40 per cent of milk-
powder; otherwise like diet
377)-
Do
Do.
Do.
Diet 379 (10 per cent of ker-
nels, 50 per cent of milk-
powder, 12 per cent of
starch, 28 per cent of lard).
Do
Do.
Do.
Do
Discontinued.
Control diet 364 (60 per cent of
milk-powder, 12 per cent of
starch, 28 per cent of lard).
Diet 367 (39 per cent of ether-
extracted kernels, 17 per
cent of milk-powder, 10 per
cent of starch, 34 per cent of
lard).
Do
Do.
Subsequently de-
clined from dis-
ease.
Do.
Do
Were on unextracted
Do
kernels for i week
previous.
Do.
86
Journal of Agricultural Research
Vol. xn, No. 2
TabIvE I- — Results of feeding raw cottonseed kernels in milk diets to rats — Continued
Rat
No.
Weight.
Period.
Initial.
Final.
Change.
Gm.
Gm.
112
Per cent.
-19
Days.
6
"'I
200
lOI
185
118
-19
— 10
— 12
8
8
8
140
107
-23
8
113
162
133
94
144
106
-16
— II
— 20
4
4
7
141
102
-28
112
80
81
71
70
— II
~^3
108
112
102
138
+36
21
Remarks.
Diet 368 (20 per cent of ether-
extract(equivalent to about
60 per centof raw kernels),
replacing 20 per cent of
lard in control diet 364).
Do
Do
Do
Diet 369 (0.4 per cent gossypol
(equivalent to about 20 per
cent of ether extract), re-
placing 0.4 per cent of
starch in control diet) .
Do
Do
Do
Diet 373 (o.i per cent of gos-
sypol added to control diet).
Do .
Do
The survivor, rat 58, was then
put on the same diet minus
gossypol.
51
30
31
32
58
59
60
Died.
Discontinued.
Do.
Do.
Do.
Do.
Do.
Died.
Discontinued.
Died.
Do.
Discontinued.
The sparingly soluble compounds of gossypol w^ith acetic acid and
with anilin have been used to estimate the amount of gossypol present
in cottonseed kernels. Both methods have given results which show
that gossypol exists in cottonseed kernels to the extent of approxi-
mately 0.6 per cent.
In order to explain the change in toxicity in cottonseed after being
cooked in the mill, we offer the following hypothesis: Under the action
of moist heat the gossypol streams from the glands and is spread over the
seed tissue. Part is oxidized to a less toxic substance which we may for
convenience call " D-gossypol; " part is left in combination with the bases
or protein as a salt of gossypol; and part is expressed in the oil. The
degree to which these changes take place is dependent on the method of
cooking and the condition of the seed. In dry heating to 100° C. there
is practically no decomposition of the gossypol. In very dry seeds the
gossypol may not spread over the seed tissue and be changed unless much
moisture is added and the cooking prolonged.
Some quantitative data on the amount of gossypol left in the seed
have been obtained from samples of kernels cooked various lengths of
time — 5, 10, 20, and 28 minutes. The percentage of gossypol extracted
by ether in these cases was, respectively, 0.62, 0.24, o.io, and 0.07.
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed
87
METHOD OF REMOVING THE TOXIC SUBSTANCE FROM THE ETHER
EXTRACT
By treatment of the ether extract of raw kernels with an excess of
anUin the gossypol is practically quantitatively precipitated. The
dianilin salt produced is extremely insoluble in most solvents except
hot anilin and alcoholic potassium hydroxid (KOH). The substance
itself is not toxic because of its insolubility. It passes through the
alimentary canal unchanged, as can be seen by a glance at the feces.
One-half gm. doses of this anilin compound were fed for seven con-
secutive days to a rabbit without result, and it was also given to rats
in a milk diet (0.3 per cent). The food intake of the rats was not dimin-
ished; nor were the rats affected perceptibly
Gossypol "acetate" was then prepared from this compound as follows:
The substance was decomposed by means of an alcoholic alkali. The
anilin was steamed off, and the gossypol was extracted with ether and
crystallized as the "acetate" by the addition of acetic acid. This was
fed in amounts (0.25 per cent) equivalent to the anilin compound (0.3
per cent) fed previously. The rats which had not been affected by the
anilin compound were promptly affected and consumed but little food
(see Table II).
By passing steam through the extract from which the anilin com-
pound had separated the excess anilin was removed, and the resultant
oil did not prove toxic to rats. The result of this experiment has led
us to believe that gossypol is the only substance in raw cottonseed
possessing marked toxic properties. This conclusion was indicated
in our previous experiments {17), wherein we found that the gossypol
extract freed from gossypol was not toxic.
Table II. — Results of feeding gossypol "acetate" to rats
Rat
No.
Weight.
Period.
Remarks.
Initial.
Final.
Change.
Diet 436 (gossypol "acetate" (0.25
per cent) prepared from decompo-
sition of the insoluble nontoxic
gossypol-anilin compound, added
to the control diet).
Do
201
202
203
160
171
175
94
96
90
Grams.
166
122
128
117
155
95
124
122
Grams.
119
91
79
190
190
118
167
137
Per cent.
-28
-25
-26
— 32
+ 27
+ 25
+ 24
+35
+ 12
Days.
12
12
14
14
39
39
39
39
39
Died.
Do.
Do
Do.
Do
Do.
Diet 448 (12.5 per cent ether extract
freed from gossypol by treatment
with anilin, replacing 12.5 per
cent of lard in control diet).
Do
Alive.
Do.
Do
Do.
Do
Do.
Do
Do.
88
Journal of Agricultural Research
Vol. XII. No. 3
TOXICITY OF GOSSYPOL TO RABBITS
In our previous paper (17) most of our feeding experiments with
gossypol were with the "acetate," a crystalline substance containing
acetic acid in its composition. This product had the same toxic action
as the product precipitated by petroleum ether; therefore we inferred
there was no change wrought by crystallization. We have recently
fed gossypol in amounts equivalent to a toxic weight of cottonseed
kernels and have found it to produce serious results in every case. Where
gossypol itself is added to a diet in appreciable amounts, the toxic
effect is marked.
Gossypol was mixed with the feed in four forms: (i) Precipitated (by
petroleum ether), (2) recrystaUized "acetate" (lo.i per cent of acetic
acid), (3) "free" gossypol, a very pure product, and (4) as the sodium
salt of gossypol (gossypol "acetate" neutralized with three molecules of
sodium hydroxid, 10.5 c. c. of N/2 alkaU to i gm. of substance). The
results are summarized in Table III.
Table III. — Results of feeding gossypol to rabbits
Rabbit
No.
Weight.
Quan-
tity of
gossy-
pol
eaten.
Equi-
valent
in ex-
tracted
ker-
nels.
Feed-
ing
period.
Diet.
Initial.
Final.
Gain
or
loss.
Result.
Gossypol precipitated from
ether solution by petroleum
ether.
Do
990
996
998
994
1,001
21
22
17
18
19
Grams.
1,490
2,440
i»o7S
2,330
975
2,000
1,500
950
850
700
Grams.
1,205
2,10s
850
2, 180
820
1,900
1,440
800
750
590
Grams.
-28s
-335
-225
-150
-155
— 100
— 60
-150
— 100
— no
Grams.
1. 81
2.61
.68
.87
•47
0.4
•33
•433
•333
-153
Grams.
200
290
75
97
53
Days.
24
26
14
14
8
4
4
Died.
Do.
Do
Do.
RecrystaUized gossypol "ace-
tate."
Do
Do.
Do.
Gossypol free from acetic acid
(o.i gm. daily per animal,
equivalent to about 17 gm.
of-raw cottonseed kernels).
Do
Died isth day.
Died.
Do.
Do
Died nth day.
Sodium salt of gossypol
Died 13th day.
A peculiar feature about the effect of gossypol and oftentimes of
cottonseed kernels is that the animals may eat these substances for
several days without being affected, then they may suddenly cease eating,
waste away, and finally die. This was the case with rabbits 21 and 22
in this experiment.
In our pre\dous paper (77) we described the nontoxic product obtained
by oxidation of gossypol by action of air on its alkaUne solution. This
oxidation product may also be formed to some extent in the cooking of
cottonseed, but there is no evidence of it. The meal still contains con-
Jan. 14, 19x8 Gossypol, the Toxic Substance in Cottonseed 89
siderable amou nts (about i per cent) of a substance which we have called
"D-gossypol."
"D-gossypol" is very slightly soluble in ether. For rabbits ether
extraction does not render the meal nontoxic. But where we find in the
meal after 6 hours' extraction with ether considerable amounts of a
substance giving color reactions for gossypol or "D-gossypol," it would
seem that the substance is bound in some way. To explain this we have
assumed that it may be combined with the protein or some other con-
stituent. There is some evidence of this in the properties of these sub-
stances. We have mentioned that gossypol combines with anilin and
with acetic acid to form less-soluble compounds. Possibly similar com-
bination may take place with free amino and free carboxyl groups in the
protein molecule. Marchlewski (<?) mentioned the fact that gossypol
behaves like tannin toward basic dyes. Tannin also precipitates pro-
teins as insoluble compounds. A similar combination of gossypol with
protein may occur in the cooking of cottonseed.
Confirmatory evidence that these substances may be the cause of
cottonseed-meal poisoning is given in the previous publications of this
Station. Thus, when cottonseed meal is treated with an alcoholic
alkali (jp), the meal is rendered nontoxic to rabbits. When the meal
is fed with iron salts to pigs {18) and rabbits {16) , the toxic effect is greatly
diminished. Pigs fed on cottonseed meal and corn meal (1:3) with
ferrous sulphate (copperas) did not die in 180 days, whereas without
ferrous sulphate all the animals died. Rabbits were fed 106 days with
ferric ammonium citrate without harmful results. We have explained
this by assuming that the alkali treatment promoted oxidation of the
gossypol and by assuming that the iron salt formed an insoluble pre-
cipitate with the gossypol, or possibly assisted the organism to oxidize it.
The experiments referred to in this article support our previous view
that gossypol is toxic and that it is the only toxic substance in the raw
kernels.^ Extensive experiments with various meals with rats, rabbits,
fowls, and swine show that there still remains, even in thoroughly cooked
meals, an injurious factor. Such thoroughly cooked meals are harmful
to rabbits and swine, but seem to have little effect on rats and fowls when
fed on adequate diets.
PRELIMINARY EXPERIMENTS WITH PIGS
In order further to test the correctness of our view that cottonseed-meal
injury is due to a toxic substance rather than to dietary deficiencies, we
have conducted a few preliminary experiments with small pigs.
It seemed desirable, in view of the extreme position taken by Rommel
and Vedder (14) to ascertain (i) whether gossypol is toxic to pigs; (2)
' The results of other experiments showing that cooking exerts a profound influence on the toxicity will
be published at an early date.
90
Journal of Agricultural Research
Vol. XII, No. 2
whether extraction of gossypol from cottonseed by a solvent renders the
residue nontoxic; (3) whether by the addition of vitamine-containing
feeds, cottonseed-meal poisoning can be averted.
Four small Duroc-Jersey pigs were confined in small pens about 3.5 by
8 feet. The pens had a concrete floor which was bedded with pine
shavings. The water used was secured from the city mains. The pigs
were fed the diets given in Table IV.
Table IV. — Percentage composition of diets for pigs
Pig I.
Pig 2.
Pig3-
Pig4.
Feed.
Period i
(ist-28th
day).
Period 2
(29th-38th
day).
Period 3
(39th-45th
day).
Cottonseed meal
25
75
21. 7
30
45-5
45-5
Com meal
49- 75
0. 25
50
75
Gossypol
Wheat bran
25
65
60
Ether-extracted cottonseed kernels
Milk (solids)
13
(a)
ID
(a)
A
Green feeds
a About one-half pound daily.
Pig 2 received at the start 1.22 gm. of gossypol daily. This figure is
based on yields (about i per cent) of crude crystalline gossypol acetate,
obtained from "oil-free" cottonseed kernels. Gossypol "acetate"
equivalent to the required amount of gossypol was dissolved in ether; the
acetic acid present was removed from the ether solution by agitation with
water. The ether solution of gossypol was then spread over a part of
the corn meal and the ether evaporated. This was no doubt an unneces-
sary procedure, as we have found no difference in the action of the
"acetate" and the "free" gossypol.
Pig 3 was fed on cottonseed kernels from which practically all the
gossypol had been removed by percolation with ether.
Pig 4 was fed with a view to supplying any deficiency of vitamines in
the cottonseed meal by wheat bran, whole milk, and some green food,
chiefly leguminous. It is not possible to give the exact composition of
the ration of this pig. The green feed (about J/z pound daily) was not
always consumed. By disregarding the green feed eaten and by assuming
that the milk contained 12 per cent of solids, the composition of the diet
was approximately as given above.
Figures i and 2 and Tables V to VII summarize the important data of
the experiment.
Jan. 14. i9i8 Gossypoly the Toxic Substance in Cottonseed
91
Fig. I. — Graphs of the growth of pigs i, 2, and 3. "C. S. K."=^ cottonseed kernels.
•^^
Fig. 2.— Graphs of the gains per week of pigs i, 2, and 3. VC. S. M."= cottonseed meal.
92
Journal of Agricultural Research
Vol. XII, No. 3
Table V. — Results of feeding various diets to pigs
Pig
No.
Feed.
Cottonseed meal
(jossypol
Ether-extracted cottonseed kernels
(Cottonseed meal, milk, etc
Weight.
Initial.
Pounds.
29-5
28. 25
20.7s
19-5
Final.
Pounds.
31-5
27
« 44- 75
35
Pounds.
2
-1.25
24
15-5
Result.
Died 50th day.
Died 48th day.
Lived.
Died 45th day.
» Weight of pig 3 on the 50th day. This pig weighed 128 pounds when the experiment was discontinued
on the isstb day, a daily gain of 0.69 pound.
Table VI. — Weight of cottonseed feed consumed, by weeks «
Week.
Pig I
(cotton-
seed
meal).
Pig 2 (gossypol).&
(extracted
cotton-
seed ker-
nels).
Pig 4
(cotton-
seed
meal).
1
Pounds.
1.68
1-7
1.63
1.58
1.36
1.27
I. 04
Gm.
8.37(1-82)
9. 76 (2. 15)
8.04(1.77)
6. 76 (i. 50)
4. 06 ( . 90)
5-56(1.22)
3-5o( -77)
Pounds.
1-3
1-5
1-75
2. 18
2. 18
2.63
2-53
Pounds.
1-3
I. 52
1-43
I. 62
c
2. 14
6
2.8
7
I. I
Total cottonseed feed eaten
Total feed eaten
10. 26
41. 0
49. 95 (lo- 13)
40. 5
14.07
56-3
II. 91
46. 0
o Maximum estimates are given for pigs i, 2, and 4.
6 Figures in parentheses give the weight (in pounds) of oil-free kernels, which correspond to the gossypol
eaten.
Table VII. — Comparison of post-mortem notes on pigs i, 2, and 4<^
Organ, etc.
Pig I.
Pig 2.
Pig4-
Lungs
Congested, edema-
tous.
Thrombus
Congested, edema-
tous.
Extremely edema-
Heart
tous, with some
congestion.
Thrombus.
Chest cavity ....
Abdominal cav-
ity
Small intestines .
Penis
2 to 3 ounces of serous
fluid.
Slight excess of fluid .
Considerable injec-
tion of blood ves-
sels.
Sheath swollen
Poor
4 ounces of fluid
Slight excess of fluid .
Deeply injected
Sheath swollen, or-
gan paralyzed, and
protruding.
Very poor
About 16 ounces of
fluid.
Slight excess of fluid.
Inflamed areas.
Sheath swollen.
Nutrition. . .
Ck)od.
o These pigs were examined by Dr. G. A. Roberts, Veterinarian of this Station.
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed 93
GENERAL DISCUSSION
For the first few days of the experiment all the pigs ate well, and all
gained in weight. Pigs i and 2 occasionally left part of their feed. In
two weeks' time all except pig 3 began to show loss of appetite and reg-
ularly left a portion one-half to one-fourth of their feed. On the
twenty-fifth day the pig 2 (fed gossypol) was quite sick and not able to
walk well. At this time pig 3 was the thriftiest of the four, while the
rations of No. 2 and 3 were reduced on account of refusal to eat. On
the twenty-ninth day the ration of pig 4 was changed to i part of cotton-
seed meal and 2 parts of bran. On the thirty-second day the feed of
pig 2 was changed. The wheat bran was replaced by middlings, for which
the pig had a better appetite. On the thirty-ninth day the wheat bran
in the ration of pig 4 was replaced by corn meal, the pig getting equal
parts of cottonseed meal with corn meal. At that time this pig was
leaving one-half to two-thirds of the wheat-bran mixture. For three or
four days he ate the new mixture with much better appetite, but then
refused a large part and died on the forty-fifth day. Pig 3 maintained
perfect appetite up to the forty-ninth day, when she did not clean up
the last trace of feed as usual. When removed to the yard to be photo-
graphed, it was noted that she had an abnormal gait in walking, the
forelegs showing a tendency to double under her. This animal seemed
to have a great desire to eat dirt, manure, etc. However, on being
allowed the freedom of a large lot, the animal soon recovered. She
was given a little ferrous-sulphate solution, chalk, and milk on the
fiftieth day. Her normal appetite returned, and in three or four days
she was able to trot. At no time did this pig show the rough coat and
lack of appetite that characterized the others. On the fifty-second day
the feed of cottonseed was increased slightly, and the pig received about
one-half pint of milk daily for the following nine days. Whether this
pig was suffering from deficiency of some sort in the ration, from lack of
exercise, or from the daily intake of a small amount of gossypol in the
kernels, we are not able to say. Even granting that this pig had a slight
attack of beriberi, we can reasonably conclude from the experiment
that the deficiency factor is one quite secondary to the toxicity factor.
Plate I shows the condition of these pigs at various periods of the ex-
periment.
This preliminary experiment was originally planned to run for a short
period, but as the pig on the extracted kernels seemed to be in a path-
ological condition, in that the forelegs tended to double under her when
she attempted to run, it seemed desirable to continue the animal on this
diet. At this time, when pig i, which had been given the diet most
closely resembling that of pig 3 in chemical composition, had died in a
rather emaciated condition, pig 3 was a plump, very-well nourished
animal (PI. i, B). It is quite possible that the above-mentioned condi-
94
Journal of Agricultural Research
Vol. XII. No. a
tion was due both to a slight toxicity of the extracted kernels and to the
restricted diet. This condition was remedied, as previously described.
The animal was continued on the outdoor turf lot for 35 days and then
removed to a small indoor pen, where for 70 days longer she gained
steadily.
Throughout the experiment the daily feed was maintained at i per
cent (of body weight) of extracted kernels plus 3 per cent of corn supple-
ment.
ysc?
A/O
\
i. <so
\
'^O
^o
:
•'
' y
/v
/
y^
y
y'
lO
, "S
s-...
^
^
"X"
.''
^ £>
'ifi?
4
y
/
[1
e
/'
A
^
p
«/■
^'
/
•c
<<
•
y
i^-
/
4-
r
4
)
.ft
>
,fi'
A
y
f r
,.5-
i-/
''
.(?
U
7
u^
?'-'
■ih
^
/
c
\0 , •■•
V''
t^
V
V
r
■~/
ix
r
..•o
^
^
^ ^ e* 3 /o
Fig. 3. — Graphs of the growth of pigs 3, s,6,7,and8. " C.S. M."=cottonseedmeal: "EE.C.S.K."=
ether-extracted cottonseed kernels.
Two Other small pigs, No. 7 and 8, were also fed on this diet of ex-
tracted kernels and corn meal on the same basis as in the previous case.
On a few days it was necessary to use whole or cracked corn in place of
corn meal, owing to a lack of meal. The results in general were the same.
There was fair growth, but after six or eight weeks the pigs lost their
keen appetites and developed a tendency to squat on their hindquarters
and to walk stiffly. The growth curves in figure 3 show the general
results. The gain was fair, 0.581 pound per day for the female and
0.556 pound for the male pig. The experiment was discontinued on
the eighty-eighth day.
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed
95
The female pig was observed for a few weeks after the experiment.
The animal continued to increase in weight, but still retained the squat
ting tendency. Addition of small amounts of milk and outdoor exer-
cise did not eliminate this condition. The nutrition of the animal was
excellent throughout.
In view of the slight toxicity of the extracted kernels, as shown in our
rabbit experiments, it is possible that this slightly pathological condition
may be due to a toxic factor, although with such very young pigs it may
be due to the limitations of this food mixture.
No doubt if these pigs had been continued on this diet confined to
pens, they would have ultimately failed, as do swine fed on restricted
diet of cereal grains (see Hart and McCollum, 2). This phenomenon,
however, should not be confused with what is commonly understood as
cottonseed-meal "injury" or poisoning.
We do not claim, however, that this diet is an adequate one, and
it is quite possible that the condition described was due in large part
to the inadequacy of certain dietary factors. Just what factors are in-
sufficient in this particular diet is not at present apparent. In consid-
ering this question we have taken the view, tentatively, that the supply
of vitamines in cottonseed is similar to that of other seeds and that
the mineral content is very much better.
The fact that cottonseed is cooked, and subsequently pressed, raises
the question. Are the vitamines thus rendered partially inactive or re-
moved by the crude oil ? In answer to this, we may point to the excel-
lent growth of rats reported by Richardson and Green (//, 12, 13,) and
Osborne and Mendel {10) as evidence that even the cooked meal is as
well supplied with vitamines as any similar vegetable food.
With the dietary factors more favorable, as in the cottonseed-meal,
experiment to be described, probably the pigs would not have manifested
this stiflFness of gait and squatting tendency. It is also of interest to
compare the mineral content of a diet of cottonseed meal and com meaJ
(i :3) with the ash content of a diet found successful for growth in rat?
(vSee Table VIII.)
Table VIII. — Average mineral content of a diet containing I part of cottotiseed meal ard
J parts of corn meal; also the analysis of m,inerals of other materials
Feed.
Cottonseed meal
GDm meal
Average mixttire 1:3
Pig and rat diet A (2)
Diet BO)
Dry skim milk (/)....
Pro-
tein.
36
87
15-5
Ash.
Sodi-
um.
O. 26
113
14
022
029
Potas-
sium.
1.66
• 19
• 55
•335
. 076
I. 27
Calci-
lltQ.
o. 27
.015
.08
.266
.080
1-34
Magne-
sium.
0-55
. 122
. 22
•275
. oog
. 146
Chlorin
04
07
055
041
057
935
Phos-
phorus.
51
35
264
248
186
979
S'lV
phur.
o. 49
. 20
. 20
.089
. 141
•357
96 Journal of Agricultural Research voi.xii.No.a
Rat diet A represents the mineral content of a diet which was found
successful for growth of pigs on artificial rations by Hart and McCollum
(2).
Rat diet B represents the mineral content of a diet which did not pro-
duce pronounced stunting in rats (4). The diet was satisfactory in other
factors.
The recent work of Hogan (5) and of McCollum, Simonds, and Pitz (5)
indicates that the deficiencies of corn lie in the poor ash content, in poor
proteins and low amounts of protein, and inadequate amounts of fat
soluble A.
Since our pigs on ether-extracted kernels plus corn meal made much
better gains than could be expected on whole com alone under these
conditions, our opinion is confirmed that the addition of this cottonseed
feed to a corn diet furnishes a greatly improved protein and a mineral
basis for nutrition. It is then evident that, where death ensues or poor
growth is manifested, this is a result of an injurious substance rather
than of dietary deficiencies.
FURTHER EXPERIMENTS WITH GOSSYPOL
Gossypol was fed to two other pigs of approximately 50 pounds' weight.
In one case a rather large amount was given in the ration. Pig 5 ate a
slop made of i pound of a mixture of corn meal and soybean meal (1:3)
containing 4.5 gm. of gossypol. The pig showed a poor appetite for the
same amount given the next morning, but ate it slowly after a pint of
milk was poured into it. Next, a half dose (containing 2.3 gm.) was
offered, one-half of which was refused. For the next five days the pig
was offered smaller doses, but refused it all or in part, even when tempted
to eat by using bran and milk in the feed. The pig showed a good appe-
tite for other feed that did not contain the gossypol mixture. On the
fifth, sixth, and seventh days the pig showed no appetite at feed time
but seemed sluggish and showed a desire to lie down. The pig lost in
weight. The experiment was then stopped. A control pig, No. 6, fed
on a similar ration without gossypol showed a good appetite and devel-
oped no such symptoms.
After a rest of six days, the previous control pig. No. 6, was used
for a gossypol experiment, and No. 5 served as the control. Gossypol
was fed in amounts equivalent to that in 0.5 to 0.6 pound of oil-free
cottonseed kernels, approximately 2.2 gm. daily. An ether solution of
gossypol was dried on corn meal and this was mixed with more corn
meal and wheat bran. The other pig received the same ration without
gossypol. The gossypol pig ate practically all its feed for a week, and
then began to show a poor appetite for it and refused part. On the
eighth day one-half the feed was left. On the ninth and tenth days the
pig ate scarcely any feed. From the eleventh to the fifteenth day the
animal was given dry recrystallized gossypol acetate (10 per cent acetic
Jan. 14, i9i8 Gossypol, the Toxic Substance in Cottonseed 97
acid) as a finely crystalline powder mixed v/ith the feed in place of gossypol
evaporated on corn meal. The pig ate it for one day, and then refused it
as before. By using com meal only the animal ate one more feed, but
refused it when repeated. During the last few days of the experiment
the goss3^1 animal seemed to be growing weaker and very preceptibly
thinner. In the last week the animal lost 3 pounds^ while the control
gained 2 pounds. This experiment was discontinued on the fifteenth
day because of the refusal of the animal to eat.
In both of these cases of feeding gossypol it was very evident that the
animals were physiologically afifected at an early date. All told not
over 15 gm. of gossypol were eaten in the first case and not over 22 gm.
in the second case. These amounts of gossypol are equivalent to about
4 and 6 pounds of oil-free cottonseed kernels, respectively.
Besides the direct proof of the existence of a toxic substance in cot-
tonseed meal, we have a strong argument against the deficiency theory
in the results of feeding rabbits on cottonseed meal treated with boiling
alcoholic alkali. This treatment, which would be expected to destroy
the natural vitamines present, so changes the meal that it becomes non-
toxic to rabbits. This change in toxicity we have shown is explained
by the ease with which gossypol undergoes oxidation in alkaline solution.
This was confirmed by feeding to rabbits the products formed by oxida-
tion of gossypol in alkaline solution by air.
CAN COTTONSEED-MEAL POISONING BE OVERCOME IN A FAVORABLE
DIET UNDER FARM CONDITIONS?
The two largest and oldest pigs (No. 5 and 6) were fed in a turf lot
about 50 feet square. Grass was abundant, and a good part of the time
there was water in the lot from frequent rains which also kept the turf
soft. These pigs were fed on a mixture of equal parts of cottonseed
meal, corn meal or com, and wheat bran, with a pint of milk apiece each
day. The cottonseed meal fed each day was about 1.33 per cent of body
weight, rather higher than has been the practice at this Station. It was
thought that for a while, about the fiftieth day, the pigs acted somewhat
suspiciously. One showed a lack of appetite for the mixture. Its eyes
seemed partly closed and somewhat watery. The pigs also seemed rather
short-winded. They retained, however, perfect control of their limbs
and were able to mn very well at all times. Finally, after 160 days of
high feeding of cottonseed meal, these pigs were put under cover in a
small pen. The pigs were soon eating sparingly and losing weight. The
pig which had acted somewhat queerly at times in the experiment
became sick and died on the one hundred and ninety-eighth day, show-
ing typical symptoms, although there was also a pneumonic appearance
of the lungs. The experiment with the other pig was then discontinued.
On changing this pig's feed to corn, he began to regain weight. This
animal was later fed corn, wheat bran, meat scraps, etc. (Table IX.)
27805°— 18 5
98
Journal of Agricultural Research voi. xii, no. «
Table IX. — Results of feeding cottonseed meal to pigs 5 and 6 under favorable conditions.
Weight.
Time of weighing.
Pig 6.
Pigs.
Gain per day.
Pig 6.
Pigs-
Day,
Pounds.
55-5
129
158
150
142
Pounds.
50-5
112. 5
153-5
145
145
(^)
158
C177
Pound.
Pound.
0.8
.64
0.67
9^
160
.65
'■n
^■^b •'o"
a Dead.
6 Discontinued feeding cottonseed meal.
c Slaughtered.
As long as these pigs were kept in the outdoor lot, no marked symptoms
of cottonseed-meal poisoning were noted. It was noted that when kept
itJkdoors they lost both appetite and weight.
It is evident from our indoor and outdoor experiments that the effect
of cottonseed meal is more severe on pigs kept in pens, a fact that has
long been known ; however, past records show that typical sudden deaths
from "acute cottonseed-meal poisoning" may also occur among pigs
receiving cottonseed meal when on pasture. Such deaths may follow
excellent gains and may be without previous sickness, often occurring
when the animals are exercised violently. Consequently, a conclusion
that the meal was without effect during the outdoor experiment can not
be drawn. Certainly the subsequent loss in weight after removal from
the turf is suspicious. It is plausible to suppose that the outdoor con-
ditions stimulate metabolism so that the animal is enabled to overcome
or resist the injurious factor. Possibly the difference in effect of cotton-
seed meal on rats and pigs may be in part explained by the more vigorous
metabolic activity of the smaller animal.
In a series of three recent articles, Richardson and Green {11, 12, 13)
have well shown the high nutritive efl&ciency of cottonseed meal and flour
for rats, indicating the economic value of this substance. In some points,
however, we believe that they have misinterpreted facts. Thus, they
speak of the flour as a "highly milled" or "refined" product and account
for the apparent slight nutritive superiority of the unbolted meal over
the flour by stating that —
This suggests a greater amotmt of the growth-promoting substance associated with
certain fats in the less highly milled product.
It seems hardly possible in the case of cottonseed meal to effect such a
change in nutritive value by mechanical means. We would suggest
that the difference was accountable on the basis of different conditions
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed
99
in the cooking of the products, as we have found these to be the greatest
cause of variation in toxicity of meals.
Richardson and Green (u, p. 316), state in conclusion:
Our results indicate that cottonseed meal does not contain sufficient minerals for
growth, is not actively toxic, contains efficient protein and perhaps fat-soluble, growth-
promoting substances, similar to those of butter fat, but in less adequate quantities.
Our own extensive unpublished experiments on the toxicity of cotton-
seed products indicate that the toxicity of cottonseed meals varies with
the conditions of cooking the raw seed. While we find that the flour
and thoroughly cooked meals have no apparent toxicity for rats when
fed in diets supplemented by milk powder, these same products fed in
unsupplemented diets are inferior to ether-extracted cottonseed kernels.
Even thoroughly cooked cottonseed meals are definitely injurious to
rats and pigs.
The ash analysis of cottonseed flour given by Richardson and Green
(ii, 12, ij) in each of these three articles differs radically in some respects
from that given by Forbes (i) (Table X).
Table X. — Ash analyses of cottonseed flour
Constituent.
Inorganic salts
Silicic oxid (Si02)
Chlorin
Sulphur trioxid (SO3)
Phosphorus pentoxid (P2O5)
Potassium oxid (KgO)
Calcium oxid (CaO)
Magnesium oxid (MgO)
Sodium oxid (NajO)
Analysis
of cotton-
seed flour
according
to Rich-
ardson
and
Green.
Per cent.
5-5°
o. 14
None.
.06
2-57
2. 01
.26
•25
None.
Constituent.
Ash
Chlorin. . . .
Sulphur. . . .
Phosphorus
Potassium . .
Calcium
Magnesium .
Sodium . . . .
Analysis
of cotton-
seed meal
according
to Forbes.
Per cent.
7. 629
. 042
•536
1.479
I. 81
. 291
•599
.283
Richardson and Green's data are from an analysis of material left after
ignition, which, as is well known, causes loss of elements, such as sulphur
and chlorin. While the elements chlorin and sodium are not necessary
for plant growth and the amounts present may vary, it seems hardly
possible that they are entirely lacking in cottonseed flour. They are
certainly essential to animals. It may be noted that the rats of Richard-
son and Green "have grown and maintained body weight for 135 days"
on a diet containing cottonseed flour as the sole source of minerals.
In their second article Richardson and Green (12) have attempted to
repeat some of our work with extracts of cottonseed. Instead of using
unheated cottonseed kernels, as we did, they used kernels heated to
120° C. Thus, they fail to find toxic the ether extract of petroleum-
loo Journal of Agricultural Research voi. xii, no. 2
ether-extracted kernels. Their results may be due to the influence of
previously heating the kernels to 120° C, thus possibly decomposing
some of the gossypol, and to incomplete extraction so that the remain-
der was left in the three fractions, the extracts and the residue. They
also assume that the ethyl-ether extract of petroleum-extracted kernels
is always 2 per cent of the weight of the kernels. This is the case only
with long-continued extractions.
SUMMARY
Raw cottonseed kernels contain about 0.6 per cent of gossypol and are
highly toxic to rats. Ether extraction renders the material nontoxic
and gives a highly toxic extract containing about 2 per cent of gossypol.
Gossypol fed in milk diets in amounts equivalent to those contained in
the raw cottonseed diets has proved as toxic as raw cottonseed. Gossy-
pol may be quantitatively removed from the ether extract by precipita-
tion as its insoluble anilin compound. The extract is thus rendered non-
toxic to rats. The insoluble anilin compound of gossypol is not toxic
because of its insolubility. Gossypol prepared from this compound
possesses its original toxic properties.
Cottonseed meal is much less toxic than raw cottonseed, owing mainly
to the oxidation of gossypol during cooking.
Cottonseed meal, ether-extracted cottonseed, and gossypol have been
fed to small pigs in pens under comparable conditions. Cottonseed
meal has been found definitely injurious, while the ether-extracted raw
seed does not appear to cause cottonseed-meal poisoning. Gossypol has
been found toxic to pigs.
If the presence of an injurious substance in the meal is disregarded, a
diet of cottonseed meal and corn meal has nutritive limitations which
may, under restricted conditions of living, lead to failure of pigs to thrive.
Such failure is a phenomenon distinct from cottonseed-meal poisoning.
Outdoor exercise, access to forage and soil, and improved diets tend
to postpone or avert cottonseed-meal poisoning of swine. The defi-
ciency hypothesis that cottonseed-meal poisoning of swine is similar to
beriberi is untenable.
LITERATURE CITED
(i) Forbes, E. B., Beegle, F. M., and Mensching, J. E.
1913. MINERAL AND ORGANIC ANALYSES OF POODS. Ohio Agr. Exp. Sta. Bul.
255, p. 2II-23I.
(2) Hart, E. B., and McCollum, E. V.
1914. INFLUENCE ON GROWTH OF RATIONS RESTRICTED TO THE CORN OR
WHEAT GRAIN. In Jour. Biol. Chem., v. 19, no. 3, p. 373-395, 11
charts, i pi.
(3) HOGAN, A. G.
1916. THE NUTRITIVE PROPERTIES OF CORN. In Jour. Biol. Chem., v. 27,
no. I, p. 193-208. Bibliography, p. 208.
Jan. 14, 1918 Gossypol, the Toxic Substance in Cottonseed loi
(4) McCoLLUM, E. v., and Davis, Marguerite.
191 5. THE INFLUENCE OF THE COMPOSITION AND AMOUNT OP THE MINERAL
CONTENT OP THE RATION ON GROWTH AND REPRODUCTION. In Jour.
Biol. Chem., v. 21, no. 3, p. 615-643, 11 charts.
(5) SiMMONDS, Nina, and PiTz, Walter.
1916. THE DIETARY DEFICIENCIES OP THE MAIZE KERNEL. In Jour. Biol.
Chem., V. 28, no. i, p. 153-165, 10 charts.
(6)
I916. THE DISTRIBUTION IN PLANTS OF THE FAT SOLUBLE A, THE DIETARY
ESSENTIAL OP BUTTER FAT. In Amer. Jour. Physiol., v. 41, no. 3,
p. 361-375, 2 fig., II charts.
(7)
1916. THE NATURE OP THE DIETARY DEFICIENCIES OF THE WHEAT EMBRYO. In
Jour. Biol. Chem., v. 25, no. i, p. 105-131, 19 charts.
(8) MarchlEwski, L.
1899. gossypol, Ein bestandteil der baumwollsamEn. In Jour. Prakt.
Chem. N. F., Bd. 60, Heft 1/2, p. 84-90.
(9) Osborne, T. B., and Mendel, L. B.
1916. THE effect of THE AMINO-ACID CONTENT OP THE DIET ON THE GROWTH
OF CHICKENS. In Jour. Biol. Chem., v. 26, no. 2, p. 293-300, i pi.
(10)
1917. THE USE OF COTTON SEED AS FOOD. In Jour. Biol. Chem., v. 29, no. 2,
p. 289-317, 5 charts.
(11) Richardson, Anna E., and GrEEn, Helen S.
1916. nutrition INVESTIGATIONS UPON COTTONSEED MEAL. I. In Jour. Biol.
Chem., V. 25, no. 2, p. 307-318, 5 charts.
(12)
1917. NUTRITION INVESTIGATIONS UPON COTTONSEED MEAL. II. In JOUT.
Biol. Chem., v. 3c, no. 2, p. 243-258, 13 charts.
(13)
I917. NUTRITION INVESTIGATIONS UPON COTTONSEED MEAL. III. COTTON-
SEED PLOUR. the NATURE OF ITS GROWTH-PROMOTING SUBSTANCE
AND A STUDY IN PROTEIN MINIMUM. In Jour. Biol. Chem., V. 31,
no. 2, p. 379-388, 4 charts.
(14) Rommel, G. M., and Vedder, E. B.
1915. beriberi AND cottonseed POISONING IN PIGS. (PRELIMINARY NOTE.)
In Jour. Agr. Research, v. 5, no. 11, p. 489-493.
(15) Wells, C. A., and Ewing, P. V.
1916. COTTONSEED MEAL AS AN INCOMPLETE FOOD. In Jour. Biol. Chem., V.
27, no. I, p. 15-25. References, p. 24-25.
(16) Withers, W. A., and Brewster, J. F.
I913. studies ON COTTONSEED MEAL TOXICITY. II. IRON AS AN ANTIDOTE.
In Jour. Biol. Chem., v. 15, no. i, p. 161-166.
(17) and Carruth, F. E.
191 5. GOSSYPOL, the TOXIC SUBSTANCE IN COTTONSEED MEAL. In Jour. AgT.
Research, v. 5, no. 7, p. 261-288. Literature cited, p. 287-288.
(18)
1917. IRON AS AN ANTIDOTE TO COTTONSEED MEAL INJURY. In Jour. Biol.
chem., V. 32, no. 2, p. 245-257, 4 charts.
(19) and Ray, B. J.
1912. A method FOR the REMOVAL OF THE TOXIC PROPERTIES PROM COTTON-
SEED MEAL. A PRELIMINARY REPORT. In Scicncc, n. s., V. 36, no.
914, p. 31-32.
PLATE I
Effect of feeding cottonseed feeds to pigs:
A. — Pig 3, showing condition on the ninety-fourth day on a feed containing ether-
extracted cottonseed kernels.
B. — Pig 3, showing condition on the fiftieth day.
C. — Pig 4, showing condition on the twenty-seventh day on a feed containing
vitamines.
D. — Pig I, showing condition on the fiftieth day on a feed containing cottonseed
meal. See figiire G.
E. — Pig 2, showing condition on the twenty -seventh day on a feed containing
gossypol.
F. — Pig 3, showing condition on the twenty-seventh day on a feed containing
ether-extracted cottonseed kernels.
G. — Pig I, showing condition on the fiftieth day on a feed containing cottonseed
meal.
(I02)
Gossypol, the Toxic Substance in Cottonseed
Plate 1
Journal of Aericultural Research
Vol. XM. No. 2
FRUIT-FLY PARASITISM IN HAWAII DURING 1916
By C. E. Pemberton, Assistant Entomologist, and H. F. Willard, Fruit-Fly Qtiar-
antine Inspector, Mediterranean Fruit-Fly Investigations, Bureau of Entomology,
United States Department of Agriculture
Since the introduction of parasites of the Mediterranean fruit fly
{Ceratitis capitata Wiedemann) into the Territory of Hawaii in 191 3, by
the Territorial Board of Agriculture and Forestry, more or less continuous
notes have been kept, from month to month, indicating the extent of
parasitism exerted upon the larvae of the fruit fly by these parasites and
by other species subsequently brought in. Papers presenting sum-
maries of these records, separately for the years 1914^ and 1915,' have
already been published. Opportunity for special investigations of fruit-
fly parasites in Hawaii in 191 6 has made possible the accumulation of
much careful data on fruit-fly parasitism during this year, of much the
same nature as that given in the publications just cited, and it is the pur-
pose of this paper to give, possibly more in detail, some results of the work
in 1916.
It is felt that a separate record of the conditions of parasitism as existing
in Hawaii in 191 6, three years after the first and probably the most
important of these introductions, will be of no little vaTue and interest to
entomologists, by way of comparison with the parasitism in 1914 and
1 91 5, for interesting developments in connection with the question of
general parasite introductions and as a necessary contribution to the history
of fruit-fly parasitism in Hawaii.
The tabulation during most of the year of the exact degree of infesta-
tion of large quantities of host fruits of the fruit fly, from many localities,
has been an important part of this work (Tables I-III). All fruits col-
lected for parasitism records on their contained maggots have been
accurately counted and placed in separate boxes over sand. The fruit
is then kept in the boxes until practically all the contained fruit-fly
larvae have developed, emerged, and entered the sand below for pupation.
A record of the total number of larvae thus developing and pupating is
secured. The larvae quickly pupate after leaving the fruit. The pupae
are placed in vials and later carefully counted in determining the degree
to which they have been parasitized. The filuig of exact data of this
nature from year to year is necessarily the most reliable and positive
method of ascertaining the actual degree to which the parasites now
' Back. E. A., and Pemberton, C. E. parasitism among the larv.E of the mediterranean fruit
FLY (c. capitata) IN HAWAU IN 1914. In Bien. Rpt. Bd. Comrs. Agr. and Forestry Hawaii. 1913/14. P-
ISJ-161. 1915.
S parasitism among the I^ARViB op the mediterranean fruit fly (c. capitata) in HAWAU
IN 1915. /n Jour. Econ. Ent.. V. 9. no. 3, p. 306-311. 1916.
Journal of Agricultural Research, Vol. XII, No. a
Washington, D. C. J*°- ■■*. 1918
Is Key No. K— 59
(103)
I04 Journal of Agricultural Research voi. xn. no. a
established are contributing toward a control of the fruit fly. Such data
not only indicate the extent of fruit infestation from various localities,
but also the amount of parasitism among the larvae from month to month
and the seasonal efficiency of each parasite.
Seasonal diflferences in the value and prolificness of certain species of
the introduced parasites have been most striking. This is suggested by
an examination of any of the parasite notes from almost any locality
and by a comparison of emergences of different species for each month
of the year. But most convincing proof that seasonal differences exist
is obtained by the inspection of records from fruit collected from the same
localities month by month. Some species of trees in Hawaii bear fruit,
normally a host of the fruit fly, almost continuously throughout the year.
The systematic collection of fruit from such trees and the filing of exact
data bearing on the extent of parasitism of fruit-fly larvae secured from
such individual trees throughout the year have thrown most light upon
the seasonal values of the different parasites.
Mention of possible fluctuations in the abundance of different species
has already been made.^ The work of 191 6, wherein it has been possible
to concentrate parasites in fruit collections from individual trees,, has
most impressively shown the rise of the parasite Diachasma tryoni
Cameron in the summer and fall of the year and its certain decline during
the winter and particularly the spring months. Changes in temperature,
of no great magnitude, alone seem responsible for this. The parasite
Opius humilis Silvestri, more hardy and prolific than any of the other
introduced species, has been overshadowed by the other species, partic-
ularly by D. tryoni, and has had its seasonal rise and fall directly the
reverse and entirely dependent upon the rise and fall of this species of
Diachasrha. The slight seasonal changes have little visible effect upon
the activities of Opius humilis, however, for in the winter and spring,
with the decrease in abundance of D. tryoni, it rapidly ascends and be-
comes the most effective check upon the fruit fly (Table III). These
interrelations are treated elsewhere by the writers.
The problem of control of the fruit fly in Hawaii through parasites is
only partially solved. The four species already established are accom-
plishing a certain control, particularly in the coffee districts, but a casual
survey of the extent of infestation of most host fruits as shown in Table I
will convince one of the continued destructiveness of this pest in Hawaii.
An average parasitism of 40 per cent of all of the larvae developing is,
numerically considered, of much importance; but from the standpoint
of the practical needs of the horticulturist it brings little relief.
> Back, E. A., and Pemberton, C. E. parasitism among the larv.b of the mediterranean fruit
Fi,y (c. capitata) in Hawaii in 1915. /« Jour. Econ. Ent., v. 9, no. 2, p. 306-311. 1916.
Jan. 14. 1918 Fruit-Fly Parasitism in Hawaii during 19 16
105
Table I. — Extent of infestation of host fruits by larvcB of Ceratitis capatata in Hawaii
during igi6
Host fruit.
Number of
fruits col-
lected.
Number of
larvsE of C.
capitala
emerging.
Average
number of
larvpp per
•fruit.
Kamani ( Terminalia catappa)
Mango (Mangifera indica)
Coffee {Coffea arabica)
Strawberry guava {Psidium cattleianum)
Black myrobalan {Terminalia chebula). .
Peach (Prunus persica)
Rose-apple (Eugenia jambos)
Chrysophyllum monopyrenum
Brazilian plum (Eugenia braziliensis) . ..
French cherry (Eugenia uniflora)
Mimusops elengi
Ochrosia elliptica
Kamani (Calophyllum inophyllum)
BestiU (Thevetia neriifolia)
Averrhoa carambola
Chinese orange (Citrus japvnica)
Noronhia emarginata
Guava (Psidiu^n guajava)
15.723
1.317
41, 605
13.825
6,615
669
1,258
1.956
4,398
7, 627
11,883
77
218
1.532
159
1.588
5.296
I. 791
149.415
2,291
21, 224
22, 017
46, 639
13, 738
7,001
4.034
3,808
6, 617
63,017
240
737
5,540
214
4, 843
9.032
12, 248
Table I shows few fruits that have a yearly average infestation of less
than two larvae per fruit. Considering the large quantity of fruits col-
lected during the year, from which the records have been made, an aver-
age of two larvae per fruit is high. It means that great numbers of fruits
from all localities are nearly always heavily infested. The fruits col-
lected for such a record are not selected with the purpose of obtaining only
heavily infested fruits or only sound fruits. All available fruits that have
matured are gathered and brought in whether infested or not. In this
manner the exact average condition of fruit-fly abundance, injury, and
parasitism is obtained.
To refer again to the tables particular attention should be called to the
mango (Mangifera indica), guava (Psidium gtiajava), Mimusops elengi,
Noronhia emarginata, and the Chinese orange (Citrus japonica). Great
numbers of fruit-fly larvae develop in these fruits and are but slightly
parasitized, as shown in the total column for these fruits in Table 11.
Certain characters of these fruits prevent the parasites from reaching the
larvae within. This in part accounts for the constant presence of this
pest, in spite of the establishment of parasites well adapted to the condi-
tions of the country and of great prolificness.
io6
Journal of Agricultural Research voi. xir, no. a
Table II. — Percentage of larval parasitism of Ceratitis capitata in Hawaii, 1916'^
Host fruit.
Month of col-
lect iou.
Num-
ber of
larvae
emerg-
ing
during
first
2 to 6
days.
Percentage of parasitism.
Opius
humi-
lis.
Dia-
chasma
tryoni.
Dia-
chasma
fulla-
■wayi.
Teiras-
iichus
giffar-
dianus.
Kamani (Terminalia catappa)
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Mango
Do
Do
Do
Do
Coffee b
Do
Do
Do
Do
Do
Do
Do
Do
Strawberry guava
Do
Do
Do
Do
Do
Black myrobalan
Do
Do
Peach
Do :
Rose-apple
Do
Do
Chrysophyllum mo nopyremmt
Do
Do
Do
Do
Brazilian plum
Do
Do
French cherry
Do
Do
Do
January. .. .
March
April
May
June
July
August . . . .
September.
October. .. .
November.
December.
June
July
August ....
September .
October. . . .
February. .
March
April
May
June
August . . . .
October . . .
November. .
December. .
April
May
June
July
October . . .
November. .
October . . .
November..
December. .
March
April
May
June
July
January. . . .
February.. .
March
April
May
June
November..
December. .
January
March
April
May
115
2,792
9.558
I. 391
3.094
3.569
4,017
3.526
3.403
2,299
1,408
283
299
47
53
39
390
62
I, 621
105
131
308
288
1,192
643
2,705
48
979
1,013
189
34
3,081
2,518
1,319
2,3"
951
170
1,089
14
996
702
378
78
634
1,306
78
5Z
41
95
25
Ii4>
22. 6
30.5
46. I
24. 7
6.8
2.4
8.4
11. 8
IS- 5
8.3
9-
9-
9-7
6.3
I.
5-
60. 7
4
60.3
57-1
85-4
17. 2
59- o
22.3
12. 2
38.1
27. I
10. 2
10.3
3- I
27.
3-
15-
4-
6-5
74-3
57- o
19-5
7.6
6.0
34- ij
8.4
64.0!
10.5!
0.6
I. I
16.5
7-3
27.7
53-9
58. 5
51-5
49
3
12. o
4'
5-6
12.8
6.4
•07
•4
•9
6.7
2,-Z
29.7
5-7
66. I
4-3
•9
52.9
10. 4
II- 3
14. 6
9-4
12. I
27.9
17. I
2-7
6.7
46. o
14.8
41. 1
3-8
8.6
6.0
3-5
-5
37-6
62. 4
57-1
I. I
5-8
16.6
16. 7
3-0
1-5
-5
I. o
12.8
-9
22.3
29.4
21. 2
14. 6
32. 4j 9. 6
0-5
-03
1-7
2-3
1. 2
-3
1-3
2. I
•4
-5
•3
I. I
12.8
a Most of the fruits represented in this table were collected about Honolulu at low elevations; the cofTee,
however, was collected on the island of Hawaii, in addition to localities in Honolulu, and much of it came
from points 1,000 to 2,000 feet above sea level.
b The June collection of coffee came from the Waianae Mountains, where only Opius humilis was
established.
Jan. 14. 1918 Fruit-Fly Parasitism in Hawaii during igi6
107
Table II. — Percentage of larval parasitism of Ceratitis capitata in Hawaii, IQI'^ — Con.
Month of col-
lection.
Num-
ber of
larvse
emerg-
ing
during
first
2 to 6
days.
Percentage of parasitism.
Host fruit.
Optus
humi-
lis.
Dia-
chasma
tryoni.
Dia-
chasma
fulla-
wayi.
Teiras-
tichus
Qiffar-
dianus.
Total.
French cherry
June. . . .
479
60
862
181
314
976
10, 514
2,535
103
18
819
210
167
116
18
22
85
173
274
392
84
159
32
84
143
319
39
258
76
1,767
853
403
781
248
1,670
496
4.8
1.6
3-4
4.4
.6
3-5
8.4
10. 0
IS- 5
16.6
.8
1.9
4-7
.8
5-5
6.2
1.6
10.7
II. 4
48.3
8.4
•5
Do
Tulv .
22. 4
51.5
Do
December. . . .
January
February
March
April
Mimusops elengi
22. 5
Do
4. 9
6
Do
3-5
8 A1
Do
•3
36.8
5-5
. 01
•3
23-3
22. 2
. 2
.02
. I
. 2
Do
May
"• 43
10.7
75-6
44-3
I. 2
1.9
4.7
8
Ochrosia elliptica
May
Do
June
Kamani {Calophyllum ino-
phyllum)
January
March
April
Do
Do
Do
December. .. .
March
May
Bestill
16.6
Do
Do
June
7.0
10. 9
3-6
3-3
9.4
3-4
2-5
3-3
9.4
2-3
22. 6
19. I
39-2
16. 9
^•7
S8
9.6
7-1
13^8
25.8
18.3
34-5
35-3
46.3
31-9
40. 6
13-0
4.8
. 2
Do
August
September . . .
October
November. . . .
December
September . . .
October
January
March
Tulv
Do
Do
Do
Do
I. 2
40. 6
13.0
4. I
Averrhoa carambola
Do
•7
. 2
5-1
2.7
Do
Do
"3-'8"
23.6
. 2
•4
3-4
2. 0
2-5
5-2
4.6
7-7
1-5
12 8
Do
August
September . . .
June
8 0
Do
23.6
1.7?
Noronhia emarginata
1-5
1.6
•4
3-6
2. I
•OS
Do
July
2. 0
Guava
Mav
3-8
2. 2
Do
June
Do
Tulv
•4
•OS
6^5
7-35
4.6
Do
September . . .
October
Do
The data in Tables I and II covering the guava are of unusual interest.
This shrub grows wild and luxuriantly over most of the uncultivated
portions of the islands up to an elevation of 1,500 to 2,000 feet, and fruits
throughout the year. It is not generally considered by the layman of
Hawaii as a favored host of the fruit fly, though of the 18 host fruits
given in Table I, it stands fourth in degree of infestation, showing, from
1,791 fruits collected during the year, an average infestation of 6.8 larvae
per fruit. Infestation of this fruit is not easily detected until it has
decayed. The larvae are small and nearly all inconspicuous at the time
the fruits are picked and eaten or converted into preserves or jelly.
This fruit, though heavily infested in most localities, protects the larvae
io8
Journal of Agricultural Research
Vol. XII. No. 2
from parasite attack and thus constantly liberates great numbers of
flies throughout the year and serves but in a small measure toward the
building up of favorable quantities of parasites.
Table III. — Total parasitism by month of all larvcB of Ceratitis capitata collectedin
Hawaii during igi6
Month.
Number of
larvse.
Percentage of parasitism.
Opius
humilis.
Diachasvta
iryoni.
Diackasvia
fuUawayi.
Teirasti-
chus gif-
fardianus.
Total.
January....
February..
March
April
May
June
July
August. .. .
September
October. . .
November.
December.
2,295
1, 406
7,161
21, 619
5,525
10, 013
6,134
4,803
5,631
7,972
6, 205
4,540
0.4
1-7
2
6
15
13
27
34
27
25.2
20. 9
0.08
. 2
.04
.09
. 009
.02
•9
•4
1.4
1.4
4.6
6.98
19- S
14.7
37-64
26. 69
27. 809
18.52
37-5
45-2
44-3
44-3
44.1
A comparison of fruit-fly parasitism data secured during the years
1914, 1915, and 1 91 6 would indicate that the parasites now present in
the Territory have reached their maximum degree of development and
can hardly be expected to attain a greater control of the fruit fly than
that evidenced in 191 6. There has been some variation during the past
three years in the activities of the different species introduced, as already
noted in regard to the fluctuations in abundance of Diachasma iryoni and
Opius humilis; but the check upon the work of this pest by the present
parasites can hardly exceed its present limits. Some hope, however, is
yet felt for the parasite Tetrastichus gifjardianus Silvestri. It has grad-
ually increased in numbers about Honolulu since its establishment late
in 1 914. Certain valuable points in its favor may enable it, after further
acclimatization and general adaptation to new environment, to exceed
the work of the braconids and thus increase the total average parasitism.
ADDITIONAL COPIES
OP THIS PTTBLICATION MAY BE PROCURED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASHINGTON, D. C.
AT
20 CENTS PER COPY
Subscription Price, $3.00 Pee Year
V
Vol. XH JANUARY 21, 1918 No. 3
JOURNAL OF
AGRICULTURAL
RESEARCH
CONXKNTS
Page
Irrigation Experiments on Apple-Spot Diseases - - - 109
CHARLES BROOKS and D. F. FISHER
( Contributkin from Bureau of Plant Industry )
Relation of Carbon Dioxid to Soil Reaction as Measured
by the Hydrogen Electrode - - - - - - 139
D. R. EOAGLAND and L. T. SHARP
( Contribution from California Agricultural Experiment Station)
PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE.
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASWINOXOM, E). C.
WASHmOTON ■■ OOveRNMtHT piflNTlNO OKFIC"
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT QF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Physiologist and Associate Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Siatiorts
CHARLES L. MARLATT
Entomologist and Assistant Chief, Bureau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricultural Experitneni
Station
H. P. ARMSBY
Director, Institute of Animal Nulrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pallwlogist and Assistant
Dean, Agricultural Experiment Station of
the Universily 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.
*Dr. Pearl has imdertaken special work in connection with the war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Annsby, institute of Animal Nutrition,
State College, Pa.
JOIML OF ACMIMIRAL RESEARCH
Vol. XII Washington, D. C, January 21, 191 8 No. 3
IRRIGATION EXPERIMENTS ON APPLE-SPOT DISEASES^
By Charles Brooks, Pathologist, and D. F. Fisher, Assistant Pathologist, Bureau
of Plant Industry, United States Department of Agriculture
THE PROBLEM
The present paper deals with the effects of soil-water supply upon
bitter-pit, Jonathan-spot, and certain other nonparasitic spot diseases
of the apple (Mains sylvestris). It also includes notes upon the relation
of the time of picking to the development of apple-spots in storage.
BITTER-PIT
HISTORICAL REVIEW
Bitter-pit was first described by Wortmann {22) ^ under the name
"Siippen." It has been frequently discussed in the publications of the
State Experiment Stations under the name "Bald win -spot" and was
referred to in the Nineteenth and Twentieth Reports of the New Hamp-
shire Experiment Station as "fruitpit." Lewis (8) and Allen (7) appar-
ently used the term "fruitpit" to refer to the troubles discussed later
in this paper under the name "cork," and McAlpine (9-12) apparently
included cork and also drouthspot under the name "bitter-pit."
Various explanations have been offered as to the cause and nature of
bitter-pit. Wortmann (22) reported that the disease was due to abnor-
mal transpiration conditions and that varieties of apples in which the
water was conducted most readily from the deeply seated cells to replace
that lost by transpiration were least susceptible.
Sorauer (19, p. 80) thought that the pits were produced by rupturing
of the cells during the process of swelling. In a later publication {20, p.
116-169) he stated that the disease was worst on porous dry soils and
suggested that the pits were produced by an overrapid maturing of
certain cell groups resulting from the checking effect of drouth upon the
accumulation of organic material.
Evans (5) reported that the disease was due to a bursting of cells in
the apple tissue that resulted from the sudden checking of transpiration
at night while the root action of the tree remained vigorous.
' Studies on Fruit Rots and Spots: III.
2 Reference is made by number (italic) to "Literature cited," p. 136-137.
Journal of Agricultural Research, Vol. XII, No. 3
Washington, D. C. Jan. 21, 1918
Iv Key No. G-133
(109)
no Journal of Agricultural Research voi. xii. no. 3
Ewart (6, 7) concluded that the disease was the result of local poison-
ing and mentioned spray materials and the toxic salts of the soil as
possible causes.
White {21) considered that the disease was the result of the poisoning
effects of arsenical compounds and other spray materials.
McAlpine (9-12) thought that the disease was produced by a shortage
of water in the affected tissue and that the condition might be brought
about either by transpiration exceeding the water supply or by the growth
of the pulp tissue being too rapid to allow time for the formation of the
new vascular tips needed to supply it with water. He found that there
was slightly less of the disease on trees receiving two irrigations than on
those receiving one.
DESCRIPTION OF BiTTER-PiT
Bitter-pit makes its first appearance as water-soaked bruiselike
spots on the surface of the apple. The epidermal tissue is at first entirely
normal, the spotted effect being due to the breaking down of cells in the
subepidermal region. The spots soon become depressed into rather
definite pits, 2 to 6 mm. in diameter, hemispherical in shape, and fairly
regular in outline. They develop a higher color than the surrounding
surface of the apple, becoming a deeper red than the adjacent tissue
when occurring on the colored portion of the fruit and a darker green
when on the lighter parts (PI. 2, A). As the disease advances farther,
the spots take on a brownish color owing to the dead pulp cells beneath
the epidermal layers, and in late stages of the trouble the affected area
may entirely lose its normal color, becoming a deep brown (PI. 3, A).
The diseased tissue is dry and spongy, the cells are collapsed but still
full of starch, and the cell walls show no sign of thickening or disinte-
gration. The affected tissue often has rather a bitter taste, and this
together with the sunken nature of the spots has given rise to the term
'' bitter-pit."
The pits are usually associated with the terminal branches of the
vascular bundles, and the surface spotting is often accompanied by a
browning of the vascular tissue deeper in the fruit, giving the appear-
ance of numerous brown spots in the flesh when the apple is cut (PI. 2, B).
This internal browning is especially common in the tissue within a centi-
meter of the surface of the apple. While the internal browning and sur-
face pitting are commonly associated, either may occur without the other.
Bitter-pit is often confined to the calyx half of the apple. Baldwin,
Northern Spy, Grimes, Jonathan, and Yellow Bellflower are especially
susceptible to the disease; and Rome Beauty and Winesap are fairly
resistant; but almost all varieties are sometimes affected.
Bitter-pit is very similar in appearante to rosy-aphis stigmonose, but
the latter disease is not accompanied by a browning of the vasculars
and the subepidermal tissue has a firmer texture and a darker color than
is the case with bitter-pit. Stigmonose is found only on limbs that
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases iii
were infested with aphids earlier in the year, and the spots usually
appear several weeks before picking time, while bitter-pit is rather
evenly distributed over the tree and is found only on mature or nearly
mature fruit. Bitter-pit can be distinguished from fruitspot by the
fact that with the latter disease there is an almost entire absence of
subepidermal browning, and the spots have an irregular outline and a
flecked or speckled appearance.
EXPERIMENTAI, WORK
The writers were convinced by earlier investigations that bitter-pit
was not due to fungi or bacteria. They had frequently seen unsprayed
fruit that was seriously affected with the disease, thus making the
theory that spray materials were responsible for the trouble seem en-
tirely untenable. Drouth had frequently been mentioned as a cause of
bitter-pit, and several writers had suggested an excessive or uneven
water supply as a possible cause, but little experimental data had ever
been furnished in support of any of these theories. The question of the
influence of soil-water supply seemed to the writers to be an extremely
important one, and a series of experiments were started to determine
the effect of irrigation upon the disease.
The work has been located at Wenatchee, Wash. The climate of the
section is arid, but little precipitation occurring from April till October,
thus making the trees almost entirely dependent upon irrigation for their
soil-water supply during the growing season. Except where otherwise
mentioned, the water was applied by the furrow method (PI. 4, A).
The contrasts in the amount of water on the different plats were secured
by varying the frequency and duration of the irrigations and, in some
cases, by differences in the number of furrows supplying water to the
row and by variations in the head of water at the flume.
The amount of water in the soil was determined by means of samples
taken with a soil auger. In the beginning of the work samples were
taken at depths of 6, 18, 30, 42, and 54 inches — that is, from the middle
of each of the first 5 successive feet of soil — but in the final experiments,
as reported later, samples were taken only from the one or two depths
that seemed most important in determining the condition under which
the tree roots were working in the particular orchard.
Some difficulty was found in securing samples that represented the
average moisture conditions of the tree row. The lateral movement of
soil water is very slow, resulting in considerable contrast between the
amount of moisture beneath the irrigation furrow and a few feet from it,
especially in the upper layers of the soil. Samples were usually taken
at a distance from the furrow equal to one-fourth the space between
the furrows, thus securing soil from a point midway between the wettest
and dryest areas. The plan of sampling was always the same for the
different plots of a particular orchard. Samples were taken at intervals
of 7 to 10 days, and usually just before and i or 2 days after an irriga-
tion, thus obtaining a record of the extremes in soil-water conditions.
112
Journal of Agricultural Research
Vol. XII, No. 3
As soon as obtained, the samples were transferred to tin cans with
tightly fitting lids, and the cans immediately closed. The weight of
the fresh soil sample was determined and a second weighing made after
the soil had been reduced to constant weight in a drying oven, the
difference between the two weighings being taken as the moisture con-
tent of the sample. The percentage of saturation was determined by
comparing the moisture content of the sample with the total water-
holding capacity of the soil. In the experiments of 1914 and 191 5 the
latter was secured by taking the average water-absorbing capacity of
a large number of samples, but in 191 6 saturation tests were made on
each soil sample.
Notes were taken on the amount of bitter-pit at picking time, and
later notes were taken to determine the increase in storage. The apples
were cut open at the time of the last note-taking and a record made of
the amount of internal browning. An apple was counted as affected
with bitter-pit if it had either internal or external evidence of the dis-
ease, but very few apples showed internal browning that did not also
have the external pitting.
EXPERIMENTS ON GANG APPLES IN 1914
The irrigation experiments were begun in 191 3. The results of the
first season were of little value, since the main trouble in the experi-
mental orchards was found to be stigmonose instead of bitter-pit. In
1 91 4, the data from the most promising orchard were lost on account
of mistakes of the picking crew, but some interesting contrasts were
obtained in an orchard of Gano apples. The trees in the latter orchard
were 11 years old and thrifty; the soil was a volcanic ash, uniform in
texture to a depth of 6 feet. The orchard had been under clean culti-
vation but at the time of the experiments was sown to vetch. There
were four trees in each plat. The soil-moisture condition for the season
is shown in figure i. All of the plats became quite dry the middle of
August on account of trouble with the irrigation canals.
Table I. — Percentage of bitter-pit on Gano apples in igi4
Irrigation treatment.
Total
num-
ber
of ap-
ples.
Percentage of apples of following sizes:
Per-
Plat
No.
3K
to 4
inches.
3J4
t0 3K
inches.
3K
to 3'^
inches.
3 to
inches.
2H
■ ^V
inches.
ageof
bitter-
pit-
I
2
3
4
Heavy throughout season
Medium throughout season
Light throughout season
129
152
176
144
2.7
.8
. 0
.6
25. 2
18.8
7.0
21. 9
27. 0
30. 6
30. 6
23-9
36.6
41.9
39-9
41. 0
8-5
7-9
22. s
12. 6
7.0
2.6
2-3
. 0
Medium till Aug. i, then light. . .
The crop was quite heavy, averaging about 25 bushels per tree. There
was no bitter-pit on the fruit at picking time. Five boxes of apples
from each plat were placed in cold storage and held for three months.
Tab e I gives the results of notes taken at the end of this storage period.
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases
113
The results make it evident that heavy irrigation favored bitter-pit,
and also increased the size of the apples. It can be seen that there is
60^
GO^o
r?
■^ Co
I
^O "^ *0 ^ "^
N.
lis
I.
xis
^S
^ ^ ^ ^ '^ '^ <Q
^0 O ^ 'v Itn "^ "O
N. N^ < »0 < N) n;
I I I I I » V
N. <0 "v <6 V. (o "^
I I
•O >x »0 >* ^
< K> < fO >
^ Co N. <o ^
^ ^> $ § ^ ^ J*J
< no N. fo
I I I I
N. CO "^ <0
^^^^
^
K
^
^
10 u
(3: q; ^ ^
Fig. I. — Diagram showing the soil-moisture conditions in irrigated plats of Gano apples in 1914. The
results show the average percentage of saturation for each half month based on the average of soil samples
at depths of 24 and 36 inches.
a very close relation between the size of the apple and the amount of
the disease, but there is hardly sufficient parallelism to justify the con-
clusion that the increase in bitter-pit is entirely due to increase in size.
114
Journal of Agricultural Research
Vol. XII, No. 3
EXPERIMENTS ON GRIMES APPLES IN I915
Similar irrigation experiments were carried out in 191 5 on Grimes
apples. The experimental plats were located on bottom land along the
Wenatchee River. The soil was of an alluvial nature composed of
so^
60%
*^) >v Wv >j U
^ ^ s^' to q:
^ i> "^ > "O
k
Fig. 2. — Diagram showing the soil-moisture conditions in irrigated plats of Grimes apples in 1915. The
vertical bars show the average percentage of saturation for each half month based on the average of soil
samples at depths of 24 and 36 inches.
medium heavy sandy loam with considerable clay and was uniform to
a depth of 3 feet. It was kept under cultivation in the tree row. The
trees were 5 years old and making a vigorous growth. They were quite
uniform in size and vigor and satisfactory in every respect for compara-
tive experiments. Three trees were used in each plat. The fruit was
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases
115
kept practically free from stigmonose by means of a late dormant spray
of lime-sulphur and nicotin sulphate.
The irrigation of the orchard was not satisfactory because of a short-
age of water resulting from trouble with the canals. A cloudburst on
July 26 gave an indiscriminate watering to all the plats. Irrigations
were made according to plan on August 24 and September 12. The soil
moisture conditions for the latter part of the season are shown in figure 2.
It will be noted that in spite of the unfavorable conditions a decided
contrast in soil moisture was secured on the different plats. Plat i
was given heavy irrigation throughout the season; plat 2, medium;
plat 3, light; and plat 4, medium, followed by heavy. The percentages
of soil saturation given do not indicate any decided contrast between
plats 2 and 3, but the condition of the trees in the two plats made it
very evident that a distinct contrast in soil-water conditions had been
secured.
The yield in the orchard was light, being about a bushel to the tree.
The fruit was gathered on September 22, about 10 days later than the
average commercial picking of Grimes apples in that section. It was
placed immediately in cellar storage at a temperature of about 50° F.
Notes were taken on bitter-pit seven days later. The results are shown
in Table II.
Table II. — Percentage of bitter-pit on Grimes apples. September 2g, igi§
Plat
No.
Irrigation treatment.
Heavy
Medium
Light
Medium till Aug. 24, then heavy
Total
num-
ber
of
apples.
299
222
156
175
Percentage of apples
affected with bitter-pit.
Wind-
falls.
90
36
77
Picked
fruit.
43
17
14
49
Total.
56
25
23
59
The contrasts are quite striking and make it evident that heavy
irrigation tended to increase the amount of bitter-pit. It is interesting
to note that plat 4, which was heavily irrigated late in the season,
showed a greater percentage of the disease on the picked fruit than
plat I, which was heavily irrigated early as well as late.
All of the fruit that was apparently free from bitter-pit was returned
to cellar storage and notes were taken again on November 9. The
results are given in Table III. All of the percentages but those in the
last column are based on the number of apples returned to storage
and not on the number in the original yield from the plats.
ii6
Journal of Agricultural Research
Vol. XII, No. 3
Table III. — Percentage of bitter-pit on Grimes apples in storage. November g, igis
Plot
No.
Irrigation treatment.
Total
num-
ber
of
apples.
Percentage of
apples of the
following sizes.
Larger
than
inches.
2Vi
inches
and
smaller.
Percentage of apples
that developed bitter-
pit from Sept. 29 to
Nov. 9.
Apples
larger
than
2^
inches.
Apples
inches
and
smaller.
Total.
Total
per-
cent-
age
of
bitter-
pit
de-
vel-
oped
by
Nov.
Heavy
Medium
Light
Medium till Aug. 24, then
heavy
132
166
106
72
92. 2
83.0
08.6
18. 2
7.8
17. o
1.4
63- 9
48.4
33-7
63-4
4.2
. o
II. o
79
59
52
62. 5
P£P CENT OF Brrrep-PJT
\ PEO CENT or B/rrEQ-Pir om s£pr. 29.
\p£»,CENr OF Birrc/f-piT dcvclopfo bcfween S£Pr.39ANO nov.s.
\P£a CE/VT OF BITTEQ-PIT DEVELOPED BFTWEEr^ N0V.3 AND UAN.^*.
Fig. 3.— Diagram showing the amount of bitter-pit on Grimes apples in 1915. The black portions of
the bars indicate the percentage of apples affected with bitter-pit one week after picking; the shaded por-
tions, the amount developed between September 29 and November 9; and the white portion, the amount
between November 9 and January 4. All of the percentages are based on the number of apples at the
beginning of the experiment. See figure 2 for soil-moisture conditions.
A study of Table III shows that nearly all of the bitter-pit occurred
on the apples that were larger than 2^ inches. The percentages in
the next to the last column show that the contrasts in bitter-pit on
the stored samples were similar to those found a week after picking,
and indicate the importance of orchard conditions in determining the
susceptibility of the fruit in storage. These percentages are estimated
on the basis of the sound apples left on September 29. If the original
number of apples were taken as a base in estimating percentages, these
contrasts would partially disappear, as is shown in figure 3; but this
would be an unfair comparison, so far as determining behavior in
storage is concerned, as a large number of the apples had already been
eliminated from the experiment. The last column in Table III shows
the total amount of bitter-pit to November 9, estimated on the basis of
the original number of apples.
The sound fruit from the above experiment was returned to cellar
storage and a third set of notes taken on January 4, the fruit being cut
open at this time to determine the amount of internal streaking or
browning. But very few specimens of bitter-pit were found, and these
gave but little contrast between the fruit from the diflferent irrigation
plats.
Jan. 21. 1918 Irrigation Experiments on Apple-Spot Diseases
117
The results obtained on Grimes apples in 191 5 are shown in graphic
manner in figure 3. The contrasts for the season are similar to those
obtained on September 29.
EXPERIMENTS ON GRIMES APPLES IN 1916
In 1 91 6 the experiments were continued in the Grimes orchard de-
scribed above. Five trees were included in each plat. It was possible
ll
I
60% ■
I
^ r> < o N »o <
t I I I I I I
< 10 V ^0 >n' (o V
I I I I I I t
^ «o •>. to "^ © >.
■>v v> ■>»
•M
K.
II I I I I. I
N. to ■«> to •»> to ■^
<o O "o -v "O V ^0
V *5 X lo V »o s,
I I I I I I I
V to ^ to ^ to ^
N >^ >.
•<! Uj ^ X , • ■ •<
"o O "0 ^ 'O X "o
. < 10 >. »0 V f^ <
I I I I I I I
X to •>» to ■>■ to V
N. N S.
So ijj i. X . k-
> s vj vi to o Q
s "0 s "0 ^ •> >:
I I I I I I I
X b V. to ■>. to ■"^
5 $ y SI
Fig. 4.— Diagram showing soil-moisture conditions in irrigated plats of Grimes apples in 1916. The results
show the average percentage of saturation for each half month based on the average of soil samples at
depths of 24 and 36 inches. Plat i received heavy irrigation throughout the season; plat 2, medium;
plat 3, light; plat 4, medium till late in July, then heavy; plat 5, medium in June, heavy in July, and
light in August and September; and plat 6, heavy throughout, with the exception of a sudden drop to
medium in July.
to carry out the irrigation schedule much more satisfactorily than in
the preceding year. The soil-moisture conditions for the season are
shown in figure 4.
The fruit on plat 3 was noticeably smaller than that on the other plats
as early as August i , and by the close of the season the effects of irriga-
tion were quite evident in the size of the fruit from the various plats. At
picking time the fruit on plat 3 was found to be somewhat riper and more
highly colored than that on the other plats. The apples were picked
ii8
Journal of Agricultural Research
Vol. XII, No. 3
on September i6. The crop was uniform and quite heavy, making it
possible to secure approximately 2 bushels from each tree for storage.
The fruit was placed in cellar storage in open packages. Hygrother-
mograph records showed that from September 19 to October 18 the
temperature of the cellar averaged 55° F., and the relative humidity
approximately 55 per cent; that from October 18 to November 9 the
average temperature was 48° F. and the average relative humidity 68
per cent, and that from November 9 to March 20 the temperature was
fairly constant at 35° F., the relative humidity averaging 80 per cent.
Notes w&re taken on September 19, when the fruit was picked, and on
October 18, November 9, and March 20. At the time of the last note
taking the apples were cut open and a record made of the internal brown-
ing as well as the bitter-pit spots. The vascular bundles of about half
the pitted apples were browned, but the apples that showed no external
Pea CENtoFBirreR-P/r
o s 10 IS \?o 2s 30 35 ^o
MIIIIIIIHA pea CENTOF BITT£fi-PlT/>T PICK/NG TIME.
^^BB •°^* CCNTOF BITT£Q-PIT DEvetOPED BETWEEN SEPT. 1641^0 OCT 10.
\ I PEO CENT OF BlTT£!i-Pir DEVEL OPED BETWEEN OCT. IB AND N0\/. 9.
I I PEQ CENT OF BITTE/I'PIT OEVELOPEO BET»V££NN0V.9ANDf1AOC/iS0.
Fig. s.— Diagram showing the amount of bitter-pit on Grimes apples in 1916. The diagonally shaded
portions of the bars indicate the percentage of apples having bitter-pit at picking time; the solid por-
tions, the percentage developed between September 19 and October 18; the horizontally shaded portions,
the amount developed between October 18 and November 9; the white portions, the amotmt developed
between November 9 and March 20. See figure 4 for soil-moisture conditions.
evidence of bitter-pit were free from internal browning. The bitter-pit
results are given in Table IV. The percentages in the first and last
columns are based on the total number of apples, those in the other col-
umns on the number of sound apples at the previous note taking.
Table IV. — Percentage of bitter-pit on Grimes apples
Percentage of apples affected with bitter-pit.
Plat No.
At picking
time, Sept. 19.
Developed be-
tween Sept. 19
and Oct. 18.
Developed be-
tween Oct. 18
and Nov. 9.
Developed be-
tween Nov . 9
and Mar. 20.
Total.
I
I- 5
I. 0
•7
1-3
• 4
4. I
19. 8
9.8
12. I
28. 7
5-3
35-2
0.8
1. 0
2. 0
1-5
1.6
5-5
2>-Z
2. 0
4.1
2. 0
.6
3-2
23.0
12. 7
17. 6
31- I
7.6
40.8
2
-J
4
c
6
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases 119
A study of the table shows that nearly all of the disease developed
during the first month of storage. The contrast between the plats,
however, makes it evident that the development of the disease was
largely determined by orchard conditions. The apples from the heavily
irrigated plats were in all cases more susceptible to bitter-pit than those
from the lightly irrigated ones. The amount of disease was much greater
on plats 4 and 6, which were irrigated heavily only late in the season,
than on plat i, which was heavily irrigated throughout the season. It
was less on plat 5, which had heavy irrigation followed by light, than it
was on plat 3, which received light irrigation throughout the season, or
k
StZ£OF/lPPL£S J PEP CENT OF B/TTEQ'PIT
/O 20 JO ^O
6\
4\
2 3/4 /M AND /
SMALLER Z
2
6
6
/
2% TO 3/M J
2
Jl
6
4^
Z/N.AND /
LARGER J
2
J
J
Fig. 6. — Diagram showing the relation of the amount of bitter-pit to the size of apples. The bars show
the amount of disease on the different plats and are grouped according to size of apples. It will be noted
that heavy irrigation increased the disease as much on the small fruit as on the large. For the irrigation
of the different plats see figure 4.
on plat 2, which received medium irrigation throughout the season.
The results indicate that the character of the irrigation during the last
weeks in which the apples are on the trees largely determines the amount
of bitter-pit developed in storage.
The total amount of bitter-pit for the season is shown in graphic
manner in figure 5. All of the percentages are based on the original
number of apples.
In the note taking of October 18 the apples were graded according to
size, and the record on bitter-pit made accordingly. The results are
given in Table V and figure 6.
I20
Journal of Agricultural Research
Vol. XII, No. 3
Table V. — Percentage, according to size, of Grimes apples affected with bitter-pit.
October i8, IQ16
Total
num-
ber of
ap- .,
pies.
Percentage of apples of various sizes.
Percentage of apples of various sizes af-
fected with bitter-pit.
Total
per-
Plat
No.
3'A
to
inches.
3 '4 to
3^
inches.
3%
inches.
2?4
to 3
inches.
2K
inches.
and
smaller.
sJ^to
3K
inches.
3% to
3^^
inches.
3 to
inches.
2K
to 3
inches.
2K
inches
and
smaller.
cent-
age of
bitter-
pit.
I
2
3
4
5
6
776
775
879
560
690
715
0.4
. I
•4
10. 8
6.4
10.3
12.5
6.7
20. 7
30.8
26. 5
28.3
38.6
24.8
31.2
37-9
50. 2
48.8
39- I
45-1
40. 8
20. I
16. 9
12.5
9.8
23-4
6.9
100. 0
100. 0
100. 0
52-4
36. 0
14-5
54-3
7.0
31-9
20. 9
13.6
26. 7
31.0
15.2
50.0
13-6
6.9
9.6
23-3
5- 1
31-5
10. 9
2-3
3-6
9.1
I. 2
24-5
19-8
9-8
12. 1
28.7
5-3
35-2
The large apples were much more susceptible to bitter-pit than the
small ones, but evidently size can not be taken as a measure of sus-
ceptibility, since the small apples on the heavily irrigated plats often
developed more disease than the large ones on the lightly irrigated ones.
(Table V; fig. 6.) A study of the table shows, however, that the same
soil conditions that favored bitter-pit also tended to increase the size
of the fruit, the plats standing in practically the same order as to
percentage of apples larger than 3X inches as they do in percentage of
bitter-pit.
EXPERIMENTS ON JONATHAN APPLES IN 1915
Irrigation experiments were made on Jonathan apples similar to those
already reported on Grimes. The work was carried out in an orchard
at Wenatchee, Wash. The soil was a rich gravelly loam, with a con-
siderable percentage of clay, underlain at a depth of i6 inches with
a layer of medium fine gravel. For several years previous to the
beginning of the experiments the orchard had been heavily manured with
slaughterhouse refuse, and during the time of the experiments it was
kept in alfalfa. The trees were 6 years old, and there were 5 trees in
each plat. The experiments were begun in 191 5. Breaks in the irri-
gation canals at various times and a rainstorm on July 26 made it impos-
sible to secure much contrast in the different plats before the first of
August. All the trees but those of plat i were extremely dry the latter
part of June and the first half of July. A further report of this condition
is given later in this paper under the head "Drouthspot." There was
a shortage of water several times in August, plat 5 suffering severely
from drouth at this time and finally losing more than 75 per cent of its
foliage and considerable of its fruit (PI. 4, B). Plat 3 suffered from
drouth the latter part of August, but no defoliation occurred. Plat 2
was practically as wet as plat i during the latter part of July and first of
August, but became quite dry about the middle of August. The moisture
conditions for the season are given in figure 7.
Jan. 21, 191S Irrigation Experiments on Apple-Spot Diseases
121
The first picking was made on September 3, when the apples were
rather green, a second on September 17, when they were right for
commercial picking; and a third, October i when the fruit was dead
ripe. In most cases a bushel of apples was secured from each tree at
each picking. There was practically no bitter-pit on the fruit at picking
time. The apples were placed in cellar storage at an average tempera-
ture of about 47° F., and notes were taken November 10. The results
are given in Table VI.
All apples more than 2% inches in diameter were counted as large,
and the others as small. There was little contrast as to size in the fruit
• 40^
40^'
il
.1
^
,
1
B
S ^ J^ "^ ■> '^
^ ? ^ > V >
?>?^^^7^
^i5J55iQ?5lQ
^ Us >> <^ >v Irs
•^ n: "^ X n vi
?!Q^5^^5!Q
<§<50 < !fc<
^ < i$ «!. (§ <
S$ < S§ < J§<
<$< SQ< i$ -1.
io < !§< (5 <
lls^^l"
l^llll
IsIlP'
|S3^^^
^ ^ ^ ^ i$ ^*
5 5 5 ^ ^ "Si
'^ ^ '^ ^ "^ to
^ "5 ^ ^ "^ ^
^ ''j *> "^ '^ <0
^ ^ ^ ^^ to
*^^ ^ ^^ «0
>»
\
^
^^
«0
K
%
k
k
K
5
5
*i.
^
*^
Q.
^
Flo. 7. — Diagram shor/ing the soil-moisture conditious iu irrigated plats of Jonathan apples in 1915. The
average percentage of saturation is given for each half month and is based on soil samples taken at
a depth of 16 inches. Plat i was to receive heavy irrigation throughout the season; plat 2, medium;
plat 3, light; plat 4, mediiun in August, then heavy; and plat 5, heavy till August i, then light. The
schedule was followed as closely as the water supply ^v■ould allow.
of the different pickings, and all three were combined to obtain the data
given on size.
The large apples again have much more bitter-pit than the small ones.
The apples of the first picking had more than twice as much bitter-pit as
those of the second and those of the second several times more than those
of the third. It might be suggested that a part of this contrast should be
attributed to the fact that the earlier pickings had been in storage longer,
but the later development of the disease in storage gives no support for
this hypothesis. The more mature fruit was apparently much less sus-
122
Journal of Agricultural Research
Vol. XII, No. 3
ceptible to the disease. A study of the total bitter-pit as given in the
last column of Table VI shows effects from irrigation similar to those
obtained on Grimes apples. The fruit from the trees receiving medium
irrigation followed by heavy irrigation late in the season had the most
bitter-pit, and that from the trees irrigated heavy both early and late
the next in amount. As has already been mentioned, the contrast
between plats i and 2 in the amount of irrigation was not as great as
intended; the latter, however, received less water and had less pit than
the former. Plats 2 and 5 had but little bitter-pit, even on the large
apples. The fruit from plat 2, however, was of an inferior quality on
account of the sunscald that resulted from the defoliation of the trees.
Table VI. — Percentage of Jonathan apples affected with bitter-pit. November 10, IQI5
Plat
No.
Irrigation treatment.
Percentage
of apples of
following
sizes.
Larger
than
inches
inches
small-
er.
Percentage of apples affected with bitter-pit.
First
picking.
Ap-
ples
larger
than
. iVi
inches
Ajj-
ples
inches
or
small-
er.
Second
picking.
Ap-
ples
larger
than
2H
inches
Ap-
ples
inches
or
small-
er.
Third
picking.
Ap-
ples
larger
than
2H
inches
Ap-
ples
inches
or
small-
er.
Total.
Ap-
ples
larger
than
2H
inches
Ap-
ples
inches
or
small-
Large
and
small
Heavy
Medium
Light
Medium, followed by heavy
Heavy, followed by severe
drouth
91.9
82.4
66.9
92.9
49.8
8.1
17.6
33-1
7-x
so. 2
32-9
32- S
13-7
44.4
4-S
22.7
S-9
8.2
16.7
4.4
15- o
XI- 7
S-4
23.0
16.7
2.0
1.4
8.3
4.0
1-3
18.5
16.1
5-9
25.8
3-4
22.0
2-3
3-8
9.1
3-3
18.6
13-6
S-7
24-6
3-4
The above fruit was held in cellar storage and a second examination
made on February 7. At this time the apples were cut open, and any
that had either browning of the vascular tissue or surface pitting were
counted as affected with bitter-pit. The results are given in Table VII,
the percentages being computed on the number of apples that were
free from bitter-pit at the time of the last note-taking. There was
little contrast in the amounts of disease on the different pickings, and
the three are considered together.
Table VII. — Percentage of bitter-pit on Jonathan apples. February 7, igi6
Plat
No.
Irrigation treatment.
Heavy
Medium
Light
Medium, then heavy. .
Heavy, then very light
Percentage of apples
that de-
veloped bitter-pit in storage
from November lo to
February 7.
Large
Small
L,arge and
Apples.
Apples.
apples.
7.6
5-5
7-4
2. I
.6
1-3
3-6
1.9
3-3
1.9
. 0
1.6
•9
I. 2
I. 0
Total
percent-
age for
season.
24. 6
14.7
8.8
25.8
4-3
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases
123
But little bitter-pit had developed on any of the apples during the
three months of cellar storage. This may have been because the sus-
ceptible apples had already been eliminated, or may have been due
to the fact that the apples were in an open package and finally became
slightly shriveled. The relative amounts of disease on the apples from
the various irrigation plats is little different from that given in Table I.
The results for the season are shown in the last column of Table VII
and also in figure 8.
EXPERIMENTS ON JONATHAN APPLES IN I916
The bitter-pit experiments were continued in 191 6 in the Jonathan
orchard already described. The irrigation conditions were much more
yIPPLSS PICk£0 ^SPTEMBEff \S
APPLES PICKED OCTOBER I.
mi
■ PEO CENT OF BITTER-Pir DEVELOPED BY NOV. /O.
3 PEP: C^ENT OF BiTTEP-PtT DEVELOPED BETWEE^t NOV.IOANO FEB-T
c
Fig. 8.— Diagram showing the amount of bitter-pit on Jonathan apples in 1915. The solid portions of the
bars indicate the percentage of apples affected with bitter-pit on November 10. the white portions the
percentage developed between November 10 and February 7. All of the percentages are based on the
number of apples at the beginning of the experiment. See figure 7 for soil-moisture conditions.
satisfactory than in 191 5. The percentages of soil saturation main-
tained on the different plats are shown in figure 9.
All of the trees were in vigorous condition except those of plat 3,
which were apparently suffering from the effects of the drouth of 191 5.
The apples of this plat were very highly colored, while those of plats
5 and 7 were rather low in color. The first picking was made on Septem-
ber 22 and a second on October 2. The apples of the first picking were
undercolored and immature, while those of the second were well colored
and suited for commercial picking. Approximately 3 bushels of apples
were saved from each plat in the first picking, and approximately 2
bushels from each in the second, and placed in cellar storage. There
was no bitter-pit on the apples at picking time and none had developed
by October 24. The results obtained from notes taken on November 14
27806°— 18 2
124
Journal of Agricultural Research
Vol. XII, No. 3
and on March i8 are given in Table VIII. From the time of the storage
of the fruit till November 14 the average temperature of the cellar was
approximately 50° F., and the average relative humidity about 61 per
cent. From November 14 to March 18 the temperature averaged 38° F.,
and the relative humidity 80 per cent.
Table VIII. — Percentage of Jonathan apples affected with hitter-pit in igi6
Plat
No.
Irrigation treatment.
Heavy
Medium
Light
Medium, followed by heavy
Heavy, followed by light
Alternating, heavy, medium, heavy.
Alternating, heavy, medium, heavy,
medium
Percentage of apples affected with bitter-pit.
Nov. 14.
First
pick-
ing.
2.9
Second
pick-
ing.
D. O
•7
. o
. o
•4
•4
Total.
1.6
Mar. 18.
First
pick-
ing.
4.9
2.9
2.6
II. I
1.8
3-0
S-8
Second
pick-
ing.
2-3
1-7
. o
2. 2
•3
5-2
Total
for
year.
3-5
2-3
2. 2
7-5
1.4
3-7
4.2
60;t
I t I I I I I I I I I t I I
V <o ^ "o V <o >. N <o ^ <o V to ^
^ •O ^ "0 ^ *0 ^
I I 1 I I I I
"^. to ^ to •'^ <o ^
V *\ >. to > »0 ^
I r I I I I I
V to ^ to "^ <<> ^
iy !y ^ >> ..; L< J^
3(
I r
< I I
■ to V to ^ tos
60^
Fig. 9. — Diagram showing the soil-moisture conditions on plats of Jonathan apples in 1916. The average
percentage of saturation is given for each half month and is based on soil samples taken at a depth of 16
inches. Plat i received heavy irrigation throughout the season; plat 2, medium; plat 3, Ught; plat 4,
medium followed by heavy; plat s, heavy followed by light; plat 6, heavy in June, medium in July, and
heavy in August and September; plat 7, heavy till the middle of July, medium till August, heavy the
first half of August, and medium the remainder of the season.
Jan. 21. 191S Irrigation Experiments on Apple-Spot Diseases
125
The relative susceptibility to bitter-pit of the apples from the dif-
ferent plats was the same as in previous experiments, the fruit from
the trees receiving heavy irrigation late in the season having the largest
amount of disease, that from those heavily irrigated throughout the
season the next, and that from those receiving heavy irrigation followed
by light having the least (fig. 10). As was found in the experiments
of 1 91 5, the apples from the early picking showed much greater sus-
ceptibility to bitter-pit than those of the late picking.
The size of the apples from the various plats and the relative suscep-
tibility of the different sizes to bitter-pit is shown in Table IX.
P£Q CENT OF B/Tr5P'PfT
3 4 S
77t
P£P CENT OF BiTTER-PtT D£V£LOPeD BY NOV. I-*.
c
J pea CENT OF BITTER-PIT DEVELOPED BETWEEN NOV 14. AND MARCH IB.
Fig. 10. — Diagram showing the amount of bitter-pit on Jonathan apples in 1916. The black portions of
the bars indicate the percentage of apples affected with bitter-pit on November 14 and the white portions
the percentage developed between November 14 and March 18. See figure 9 for scil-moisture conditions.
Table IX. — Percentage, according to size, of Jonathan apples affected with bitter-pit.
March 18, igi6
Total
num-
ber
of
ap-
ples.
Percentage of apples of various sizes.
Percentage of apples of various sizes
afiected with bitter-pit.
Total
per-
cent-
age
of
bitter-
pit.
Plat.
3'Ato
.f3K
mches.
3Kt0
inches.
3 to
3K
inches.
2j<tO
3
inches.
inches
and
smaller.
i'Ato
.3K
mches.
3Kto
mches.
mches.
23^ to
3
mches.
inches
and
smaller.
I
2
3
4
S
6
7
579
593
445
440
946
566
542
0-5
•4
5-7
.8
•7
17.7
1-3
9.6
4. I
3P-9
22. 5
1. 1
42.8
16.3
43-3
32-4
59-2
65- 9
40.5
34.5
68.4
43-3
59-6
4.2
10.8
57-7
4-5
14. 0
3-4
3-9
50.0
. 0
18.2
. 0
. 0
23. I
8-3
20. 0
9.1
5-°
3-8
20. 0
6.4
3-9
2.9
6.3
1-3
2. 2
4.4
1-3
•9
I. 2
3- I
0. 0
1-5
• 4
. 0
. 0
. 0
. 0
3-5
2-3
2. 2
75
1-4
3-7
4. 2
The plats receiving heavy irrigation late in the season had more
large apples than the others. The amount of bitter-pit on the fruit of
126 Journal of Agricultural Research voi. xii, no. 3
a particular size was hardly sufficient to form a basis for conclusions,
but it is evident that the disease was worse on the large apples than on
the small, and that with the exception of one or two cases where there
were but few apples heavy irrigation increased the amount of disease
on the medium-sized as well as on the large fruit.
DISCUSSION OF RESULTS OF BiTTER-PiT EXPERIMENTS
The results of the various experiments have been uniformly consistent
in showing that heavy irrigation favors the development of bitter-pit.
Heavy irrigation throughout the season has given less of the disease
than medium irrigation followed by heavy, and light irrigation through-
out the season has resulted in more bitter-pit than heavy irrigation
followed by light. Heavy irrigation the first half of the season caused
the trees to develop a more luxuriant foliage and probably produced a
lower concentration of cell sap in the apples, both of which facts would
tend to make the fruit less susceptible to the forcing efifects of late irriga-
tion. The amount of irrigation in August and September has apparently
largely determined the amount of disease.
Sudden changes in the amount of soil water do not appear to have
had any effect upon the amount of disease. No evidence has been
found that bitter-pit is brought about by a rupture or bursting of the
cells.
Large apples have been more susceptible to bitter-pit than small
ones, but the increase in the disease from heavy irrigation has been
almost as great on the small and medium sized fruit as on the large.
This fact is brought out in Tables I, V, VI, and IX, and in a particularly
striking manner in figure 6. Apparently apples are not susceptible to
bitter-pit merely because they are large, but rather because of condi-
tions that may sometimes accompany an increased growth.
The results as a whole point to the harmful effects of heavy late irri-
gation regardless of the size of the fruit. In looking for the final cause
of the disease not only the direct growth -forcing effects of the water
should be considered but also the effects of the excess water upon the
soil flora and soil solutes. This subject will be more fully discussed in a
later publication upon the effects of fertilizers.
JONATHAN-SPOT
HISTORICAL REVIEW
Jonathan-spot was first reported by Scott (17). He suggested the
possibility that the trouble might be due to the effects of arsenate of
lead. Later Scott and Roberts {18) gave a fuller report on the disease,
showing that it could not be due to the effects of spraying and that
while fungi were sometimes present in the spots they could not be taken
Jan. 21. 1918 Irrigation Experiments on Apple-Spot Diseases 127
as the C3,usal agency. They considered the disease of a physiological
nature and found that it could be partially prevented by early picking,
prompt cold storage, and early consumption after removal from storage.
Norton (15) reported that spots practically identical in appearance
with the Jonathan-spot could be produced by the gases of ammonia and
folmaldehyde.
Cook and Martin (5) considered Jonathan-spot to be a form of rot
caused by a species of Alternaria. In a later report (4) they made a
distinction between the small, nearly black, typical Jonathan-spots that
were more commonly confined to the dark area of the skin, and the
larger light-brown "Alternaria" spots that were more common on the
lightest area of the skin. They reported that they were able to reduce
the amount of the disease by keeping the apples covered with glassine
bags during the latter part of the summer, and considered that this
fact furnished further evidence that the spots were of fungus origin.
DESCRIPTION OF JONATHAN-SPOT
"Jonathan-spot" is the term applied to superficial black or brown
spots that are especially common on Jonathan apples. The trouble is
also found on Esopus, Yellow Newtown, Stayman Winesap, and other
varieties. In the early stages of the disease only the surface color-bear-
ing cells are involved and the spots are seldom more than 2 mm. in diame-
ter, but later the spots may enlarge to a diameter of 3 to 5 mm., become
slightly sunken and spread down into the tissue of the apple to a con-
siderable depth. In this later stage of the disease rot fungi are often
present, Alternaria being particularly common.
EXPERIMENTAL WORK
The Jonathan-spot experiments were carried out in the same Jonathan
orchard and on the same apples as the bitter-pit experiments, and the
details in regard to soil, irrigation, time of picking, and condition of stor-
age have already been given.
In 1 91 5, plat 5 suffered severely from drouth the latter part of the
season, the trees finally losing more than three-fourths of their foliage
and the fruit becoming badly bronzed by the sun. Plats 2 and 3 also
became very dry in August but there was no defoliation. The soil moist-
ure conditions for the season are given in figure 7. The first picking was
made September 3. The fruit at this time lacked fully 10 days of being
at its best stage of maturity for picking. A second picking was made
on September 1 7 and a third picking on October i . The fruit of the last
picking was highly colored and dead ripe.
There was no Jonathan-spot at picking time. The results of notes
taken on November 10 and February i are given in Table X.
128
Journal of Agricultural Research
Vol. XII. No. 3
Table X. — Percentage of Jonathan apples affected with Jonathan-spot in iqi^
Percentage of apples afff ected with Jonathan-
Percentage of
spot.
apples of follow-
ing sizes.
Plat
No.
Total
November
10.
February
I.
Date of picking.
ber of
apples.
Larger
2H
Ap-
ples
Apples
inches
and
smaller.
Ap-
ples
Apples
inches
and
smaller.
than
inches
and
larger
than
Total.
larger
than
Total.
inches.
smaller.
2H
inches.
2H
inches.
I
281
84
16
24
II
22
52
50
51
2
237
86
34
18
0
IS
45
32
41
Sept. X
3
4
338
250
62
38
7
5
13
3
6
4
12
54
38
72
28
62
'^^r^ • O
93
38
I S
179
49
51
2
3
3
22
18
20
I
260
98
2
34
0
33
67
17
65
2
291
81
19
18
4
15
48
45
47
Sept. 17
3
4
354
195
79
94
2 I
23
17
16
2 1
77
93
65
67
76
92
•^'^r *" */
6
8
16
I S
56
52
48
3
0
2
38
15
27
I
183
95
5
17
22
18
79
53
77
2
198
78
22
3
I
2
72
30
63
Oct. I
3
267
60
40
11
4
8
79
35
61
4
I S
I
173
92
8
4
0
2
62
79
64
724
92
8
26
12
25
64
47
62
Total for all
2
726
82
18
14
2
12
52
37
49
pickings
3
959
67
ZZ
14
6
II
71
58
67
4
618
93
7
12
5
II
62
54
62
I 5
235
50
50
3
3
3
26
18
22
There was more Jonathan-spot on the large apples than on the small
ones, and at the time of the first note-taking there was more on the fruit
of the first and second pickings than on that of the third. Irrigation
apparently had but little effect upon the disease. The apples from plat
5 had the least Jonathan-spot; but, as already mentioned, these were
badly sunburned and therefore not suitable for use in comparison with
those of the other plats.
In 1 91 6 the experiments were continued in the same orchard. All the
trees were in a healthy, vigorous condition except those of plat 3. These
were the same as used in plat 5 the preceding season and showed the
effects of the previous year's drouth in their thin foliage and short twig
growth. The soil-moisture conditions for the various plats are given in
figure 9. Pickings were made on September 22 and October 2. The
fruit from plats 5 and 7 was rather poorly colored, while that from plat
3 was very highly colored. The conditions of storage have already been
given in connection with the notes on bitter-pit. The results for the
season are shown in Tables XI and XII.
Jan. 21. 1918 Irrigation Experiments on Apple-Spot Diseases
129
Table XI. — Percentage of Jonathan apples affected with Jonathan-spot in igi6
Plat
No.
Irrigation treatment.
Percentage of apples affected with Jonathan-spot.
November 14.
First
pick-
ing.
Second
pick-
ing.
Total.
March 18.
First
pick-
ing.
Second
pick-
ing.
Total
for
year.
Heavy
Medium
Light
Medium, followed by heavy
Heavy, followed by light
Alternating, heavy, medium, heavy
Alternating, heavy, medium, heavy,
medium
1-3
3-8
0-3
o. o
0.0
.6
o. o
•4
0.8
2. I
II. 8
3-0
•3
2. I
55-8
51-2
72. 2
66.3
18. 2
53-3
82.8
16. 2
33-9
65-4
54- 2
66.5
69. I
24. o
64.7
23.8
Table XII. — Percentage, according to size, of Jonathan apples affected with Jonathan-
spot in igi6
Total
nimi-
ber of
apples.
Percentage of apples of various sizes.
Percentage of apples of various sizes affected
with Jonathan-spot.
Plat
No.
3Hto
3K
mches.
3KtO
mches.
3 to
mches.
inches.
inches
and
smaller.
3Mto
3K
mches.
3J<to
mches.
3 to
. 3K
mches.
2j^tO
3
mches.
. 2H
mches
and
smaller.
Total.
I
2
3
4
5
6
7
579
593
445
440
946
566
542
0-5
•4
5-7
.8
•7
17.7
1-3
9.6
4.1
30-9
22. 5
I. I
42.8
16.3
43-3
32-4
59-2
65- 9
40-5
34-5
68.4
43-3
59-6
4.2
10.8
57-7
4-5
14. 0
3-4
3-9
72.8
60. 0
61.5
53-4
60. 0
68.7
18.8
65-3
22. 7
66.8
52- 9
58.3
77-7
24.7
69.8
25- I
66.6
62.5
73-2
70. 0
26.6
63.2
23-4
65-4
54-2
66.5
69. I
24. 0
64.7
23.8
100. 0
55- 0
33- i
38.2
13.6
On November 14 the Jonathan-spot was worse on the apples of the
first picking than on those of the second, but by March 18 this condi-
tion had in most cases been reversed. There was little contrast be-
tween the amount of disease on the fruit of different sizes. The con-
trasts between the irrigation plats were not very consistent, but in
general indicated that heavy irrigation favored the disease. Plats 5
and 7, on which the fruit was lowest in color, had least of the disease.
DISCUSSION OF RE;SULTS on JONATHAN-SPOT
The experiments on Jonathan-spot have furnished little in the way
of consistent positive results. In both 191 5 and 1916 the apples of
the early picking had more of the disease than those of the late picking.
In 1 91 5 the large apples developed more Jonathan-spot than the small
ones, but this did not hold in 19 16. The results of both years gave
some evidence that heavy irrigation was more favorable to the disease
than light irrigation, but there was nothing to indicate that the amount
of soil moisture was an important factor in determining the amount
of Jonathan-spot.
130 Journal of Agricultural Research voi. xii. no. 3
OTHER PHYSIOLOGICAL SPOT DISEASES OF THE APPLE
DROUTHSPOT
The term "drouthspot" (2) has been applied to certain fairly large
areas of dead brown tissue usually occurring near the surface of the apple,
but sometimes found deeper in the flesh. The disease may appear
at almost any stage in the development of the apple, but the fruit
appears to be more susceptible after it is one-third grown. The spots
are usually located on the blossom half of the fruit. In typical cases
the trouble first appears as large, irregular, water-soaked spots that
often have a reddish margin and are usually covered with drops of a
yellowish, sticky ooze that is sweetish to the taste, and later hardens
into a brittle, crystalline-like deposit (PI. 3, D). At this stage the spots
resemble fireblight infection (caused by Bacillus amylovorus) and are
sometimes mistaken for it. Upon cutting the apple open a very shal-
low layer of dead brown tissue is found in the region of the vascular
network just beneath the skin. Occasionally brown streaks follow the
vascular bundles deeper into the apple pulp. The afifected tissue is
very bitter to the taste. The skin of the apple over the diseased area
finally regains its normal appearance; but growth is arrested at this
point, and the enlargement of the surrounding tissue soon gives rise
to a much misshapen apple (PI. 3, E, F). On account of its manner
of development, the disease has sometimes been referred to as "spot-
necrosis" {13). Mix {14) has given a full discussion of the characters
of the disease as it occurred in the Champlain Valley of New York.
The above description applies particularly to the trouble as it has
been observed on Winesap and Stayman Winesap apples in the irrigated
sections of the West. It was first produced experimentally at Wenatchee,
Wash., in 191 3, by subjecting Winesap trees to a sudden and severe
drouth when the fruit was about i inch in diameter. At about the
same time it was observed at Peshastin, Wash., on Ben Davis trees that
had suffered from a similar drouth. It occurred again at Wenatchee in
1 91 4 and in 191 5, always on trees that had been subjected to a sudden
and severe drouth and that had been making a normal or vigorous
growth earlier in the season. The drouth periods resulted from trouble
with the irrigation canals. The affected trees were usually located on
shallow soils or on soils underlain with coarse gravel at a slight depth,
thus making them peculiarly susceptible to drouth.
In 1 91 5 a series of drouth periods occurred, the first and most severe
coming the latter part of June and the first of July, the second the latter
part of July, and the third about the middle of August. At the time of
the first drouth even the trees on deep soil began to suffer, and those on
shallow soil lost a large percentage of both their foliage and fruit. The
fruit that remained on the trees was much shrunken in size, sometimes
being reduced to two-thirds its normal diameter. White Pearmain ap-
Jan. 21 191 8 Irrigation Experiments on Apple-Spot Diseases 131
pies became very badly shriveled and wrinkled (PI. 5, C), and Jonathan
and Delicious apples showed slightly less serious effects; but with the
return of irrigation water all of these regained their turgor without spot-
ting. The Winesap and Stayman Winesap apples did not become as
badly shriveled as the White Pearmain, but they developed typical
drouthspots before they became shriveled. It was also observed that
the oozing of the fruit sap, as well as the spotting of the fruit, preceded
the renewal of irrigation. The apples subjected to the early drouth were
also involved in the later ones, and the result was a series of spots on the
same apples that could be distinguished as to time of formation by the
color of the skin and the depth of the pitting.
On September 3 samples of fruit were obtained from the Jonathan
trees that had suffered most severely from drouth, and on October 13
similar samples were secured from the Winesap, Stayman Winesap, and
White Pearmain trees. All of the apples were placed in cellar storage
until January 13 and were then cut open and examined. The Jonathan
and White Pearmain apples had developed no spots, but their flavor was
decidedly poor. With the Winesap and Stayman Winesap apples the
spots had not enlarged, and there was but little brown tissue beneath
the skin (PI. 3, F). The flavor of the affected tissue was bitter and
acrid, but that of the rest of the apple was normal.
The above trees that had suffered from drouth appeared to recover
largely before the close of the season and their leaves came out normally
the following spring; but a number of them died a few months later, and
the remainder showed a lack of vigor throughout the summer. Their
foliage was thin and they appeared to suffer from drouth even with a
slight decrease in the percentage of soil moisture. The usual number of
irrigations were made, and there were no real drouth periods; yet more
than half of the apples on some of the Winesap and Stayman Winesap
trees developed typical drouthspots. The weak condition of these trees
and the death of others earlier in the summer probably resulted from the
destruction of some of the smaller roots during the drouth of the preced-
ing season.
CORK
The disease or group of diseases called "cork" may be similar to
drouthspot in cause, but is distinctly different in many of its gross char-
acteristics. Instead of being subepidermal, the spots are located in the
pulp of the apple, often quite deeply seated and often closely associated
with the larger vascular bundles (PI. 5, B). The patches of dead, brown
tissue are usually much larger than in the case of bitter-pit and much
deeper than in drouthspot. They resemble the internal browning of the
former disease, but are firmer in texture, more corky, and less spongy.
Affected apples are often slightly less firm than others, and usually have
a cheesy consistency when cut. When the spotting occurs near the core
132 Journal of Agricultural Research voi. xii, No. 3
only, there is usually no external marking to indicate the disease; but
when the outer pulp tissue is affected, depressions occur over the dead
spot, and the apple becomes more or less roughened or corrugated (PI.
5, A, B), The development of the disease in the case of these corrugated
apples is similar to that of drouthspot in many respects. It appears first
as reddish stains on the surface of the apples, and these stained areas
may gradually become water-soaked and covered with a sticky yellow
ooze. Later the skin regains its normal color, but large areas of dead,
brown tissue are left in the pulp.
Apples affected with cork are sometimes also affected with a condition
known locally as "apple-blister." The trouble first appears as slightly
raised brown or reddish spots on the skin of the apples (PI. 5, E). The
center of the raised portions is very hard and corky, but only the outer
epidermal layers are involved. As the apple develops, the blisters crack
and scale off, exposing a rough corky layer that has formed beneath.
The later stages of blister have usually been found on apples that were
also affected with cork, but blister appears 'early in the spring, very
often becoming evident as soon as the petals have fallen.
Troubles identical with cork, or very similar to it, are quite widely
distributed. They have been observed by the writers in the Wenatchee,
Entiat, Spokane, Okanogan, and White Salmon districts of Washington,
in the Willamette and Hood River Valleys of Oregon, in the Okanogan
district of British Columbia, in the Champlain Valley of New York, and
in various apple sections of Virginia and West Virginia. It is evident
from McAlpine's reports {9-12) that the disease is of considerable
importance in Australia.
McAlpine's {9-12) photographs indicate that he included the disease
under the name "bitter-pit." Lewis {8) included " corerot " and " dry rot "
as forms of fruitpit or bitter-pit. Allen (j) referred to the disease as
"fruitpit." Mix {14) has very carefully distinguished between cork and
bitter-pit. In British Columbia the disease is known as " malformation "
and in Washington as "dryrot." A trouble known in Virginia as "York-
spot," or "punky disease" {16), and in California as "hollow-apple"
are apparently very closely related to cork.
The losses from the disease are usually local, but sometimes severe.
At Entiat, Wash., in 1916, two carloads of apples from one 20-acre
orchard were rendered worthless on account of cork. On the lower flats
of the Okanogan Valley in British Columbia it is regarded as the most
serious of all apple troubles, and in certain sections of the Hood River
Valley, Oreg., it was the cause of considerable annual loss prior to the
introduction of systematic irrigation.
The cause of cork is not known. Allen (j) has reported that fruitpit
is worse on trees in a dry soil or in a soil lacking in organic matter.
The disease is apparently not produced by fungi or insects. The
writers have made repeated attempts to isolate an organism from the
Jan. ax, is»i8 Irrigation Experiments ou AppleSpot Diseascs 133
affected tissue, but with negative results. Close observations have been
made on the work of insects in orchards where the disease was serious,
but no evidence has been secured to indicate the association of any insect
with the production of the disease. Orchards affected with rosette are
sometimes also affected with cork, but the latter disease occurs in orchards
that are free from the former. In nearly every case where the disease has
been observed either in the East or West, its occurrence in the orchard
has been closely correlated with certain peculiar soil conditions; some-
times an excess of alkali or an outcropping of slate, but more often a
shallowness or openness of the soil. In most sections cork has been most
serious when there was a shortage in soil-water supply, either resulting
from light rainfall or a lack of irrigation.
An orchard at Entiat, Wash., that has been seriously affected with
cork has been under close observation for the past three years. The
orchard is located on a low bench near the Columbia River, and has had
a permanent cover crop of alfalfa. Soil samples from the orchard showed
that in the sections where spotting had been most prevalent the surface
soil was only about 3 inches deep and was composed of a coarse sand
with only a small percentage of humus. The subsoil, which was more
than 6 feet deep, differed from the surface soil only in the absence of the
humus and was underlain with coarse gravel. In sections of the orchard
where spotting had been less prevalent, the soil was found to be a much
finer sand, and in sections where no spotting had occurred it was a typical
volcanic ash, very fine in texture, closely compacted when wet, and very
retentive of moisture. Soil-moisture determinations made soon after
the spring rains showed that while the surface soils in the different orchard
sections retained their moisture fairly well, the subsoil in the first section
dried out quickly and that in the last section was very retentive of its
moisture. It will be seen that the occurrence of the disease varied with
the character of the soil, particularly with the water-holding capacity of
the subsoil.
The irrigation of the orchard was inadequate. The furrows were 5
feet from the tree rows, and alfalfa growing near the trees and in the tree
rows was yellow, frequently wilted, and very evidently suffering from
lack of water. The trees suffered from drouth, especially in the spring,
before the irrigations were begun. In 191 6 the first irrigation was made
several weeks later than usual and the trees became very dry. Later
the apples developed an unusually high percentage of cork, the disease
first appearing in blister form soon after the petals had fallen. The con-
ditions in the orchard indicated that the soil-water supply was at least
one important factor in determining the amount of disease.
The circumstances under which cork and drouthspot have occurred
in the Champlain Valley have been quite fully described by Mix {14).
A special form of cork known in certain sections as "Yorkspot" and
in others as "hollow-apple" has been found particularly common on
134 Journal of Agricultural Research voi. xii, no. 3
York Imperial apples and has also occurred on the Gano and the Esopus
varieties. The disease has been under close observation for several years
at Wenatchee, Wash., and in the summer of 191 6 a careful study was
made of it in orchards at Staunton, Va. In the latter case the disease
v\ras found only on York Imperial apples. It could not be correlated
with any peculiar soil conditions, but was found decidedly worse on trees
that were lightly loaded than on those with a medium or heavy crop. It
was much worse on the south side of the tree than on the north side and
slightly worse on the east side than on the west. It occurred almost
exclusively on apples well exposed to sunlight, always on the blush side
of the fruit, and always on fruit surfaces that would receive the oblique
rather than the direct rays of light. The spots were similar in appear-
ance to cork, but, instead of being scattered over the apple, were often
located in a crescent-shaped line at the edge of the blush surface of the
fruit. In some cases there was a definite ring almost entirely surrounding
the point which received the most direct sunlight (PI. 5, F, G). The
skin of the apple was alwa3^s normal, and the corky tissue beneath was
usually indicated by surface depressions. While it seems probable that
Yorkspot is in part an effect of drouth, its occurrence is undoubtedly
greatly influenced by sunlight and possibly by soil conditions and other
agencies.
The observations reported above seem to indicate that cork is a form
of drouth injury; yet the disease appears to differ from typical drouth-
spot, both in characteristics and conditions of occurrences. With certain
varieties of apples drouthspot can apparently be produced on any soil
under conditions of sudden and extreme drouth. Cork seems to be the
result of a less severe but more chronic drouth on trees located on certain
peculiar soils, especially on soils that are lacking in humus and are not
retentive of moisture. Blister is closely associated with cork and is
probably produced by the same agencies.
It should be noted in this connection that the harmful effects of
drouth are not always in proportion to the degree of desiccation. Other
factors must be considered in a study of drouth troubles, and among
these are the percentage of harmful substances in the soil water and the
general growth condition of the plant.
SUMMARY
(i) Bitter-pit and Jonathan-spot are distinguished from rosy-aphis
stigmonose, drouthspot, cork, and blister. Bitter-pit usually appears
first as spots of dead, brown tissue in the subepidermal portion of the
apple. These spots are associated with the terminal branches of the
vascular bundles and in later stages of the disease the browning often
follows the vasculars deep into the flesh of the apple. Rosy-aphis stig-
monose is characterized by similar brown spots in the subepidermal region
jaii.21, i9i8 Irrigation Experiments on Apple-Spot Diseases 135
but the affected tissue is firmer than in the case of bitter-pit and there
is no association with the vascular bundles. The early stages of Jona-
than-spot are confined to the color-bearing cells of the skin of the apple.
Drouthspot is characterized by the checking of the growth at certain
points on the apple without the production of any large quantity of corky
tissue. Cork differs from the drouthspot in the presence of compara-
tively large spots of brown corky tissue and in the fact that these are
usually rather deeply seated in the flesh of the apple. Blister is a super-
ficial lesion associated with cork and characterized by its blister-like
appearance.
(2) Drouthspot has been produced by sudden and extreme drouth.
It has occurred on trees that were favorably located as well as on those
that were growing under rather unfavorable soil conditions. Cork is
apparently also a drouth effect, but it differs from drouthspot in the
fact that its occurrence is usually associated with certain peculiar soil
types.
(3) Experiments have shown that there is a close relationship between
the soil-water supply of the orchard and the development of bitter-pit
in storage. Heavy irrigation has greatly increased the disease, but not
so much as medium irrigation followed by heavy irrigation. Light irri-
gation has greatly reduced it, but heavy irrigation followed by light has
resulted in the lowest percentage of the disease. Sudden changes in the
amount of soil water have apparently not increased the disease.
(4) Heavy irrigation may have been slightly favorable to the develop-
ment of Jonathan-spot, but the contrasts have been too slight to justify
definite conclusions.
(5) Large apples have shown greater susceptibility to bitter-pit than
small ones, but with Jonathan apples heavy irrigation increased the
disease on the medium-sized fruit as well as on the large, and with
Grimes the percentage of increase from heavy irrigation has been even
greater on small apples than on large ones. Apparently, large apples
are not susceptible to bitter-pit merely because they are large, but
rather because of certain conditions under which they become large.
(6) In 1 91 5 there was more Jonathan-spot on the large apples than on
the small ones, but in 191 6 there seemed to be no correlation between
size of fruit and severity of disease.
(7) During the first weeks of cellar storage there was always more
Jonathan-spot developed on apples that were picked early than on apples
that were picked late, but with longer periods of storage these contrasts
seemed to largely disappear. The results indicate, however, a greater
susceptibility in the early-picked fruit.
(8) Bitter-pit was worse on the Jonathan apples that were picked early
than on those that were picked late.
136 Journal of Agricultural Research voi. xii, N0.3
LITERATURE CITED
(i) Allen, R. W.
191 5. CONDITION OF ROOT SYSTEM OF APPLE TREES IN THE HOOD RIVER DISTRICT.
In Oreg. Agr. Exp. Sta. Rpt. Hood River Branch, 1914-15, p. 20-24,
fig. 7-8.
(2) Brooks, Charles, and Fisher, D. F.
1916. spot diseases of the apple. In Proc. i2t±i Ann. Meeting Wash. State
Hort. Assoc, 1915, p. 46-51, i fig.
(3) Cook, M. T., and Martin, G. W.
1913, THE JONATHAN SPOT ROT. In Phytopathology, v. 3, no. 2, p. 119-120.
(4)
1914. THE JONATHAN SPOT ROT. In Phytopathology, v. 4, no. 2, p. 102-105.
(5) Evans, I. B. P.
1909. BITTER-PIT OF THE APPLE. Transvaal Dept. Agr. Tech. Bui. i, 18 p.,
5 pi. Bibliography, p. 16.
(6) EwART, A. J.
1913. ON BITTER PIT AND THE SENSITIVITY OF APPLES TO POISON. (2nd. Paper)
In Proc. Roy. Soc. Victoria, n.s., v. 26, pt. i, p. 12-44, pl- 3-5-
(7)
1914. ON BITTER PIT AND SENSITIVITY TO POISONS. (3rd Paper) In Proc. Roy.
Soc. Victoria, n.s., v. 26, pt. 2, p. 228-242, pi. 23.
(8) LEWIS, C. I.
1915. FRUIT-PIT STUDIES IN THE WILLAMETTE VALLEY. In Ore. AgT. Exp. Sta.
2nd Bienn. Crop Pest and Hort. Rpt., 1913-14, p. 35-37, fig. 8.
(9) McAlpinE, D.
i911-12. bitter pit investigation. the past history and present posi-
tion of the bitter pit question. first progress report. 197
p., 34 pi. Melbotune. Literature, p. 111-117.
(10)
(II)
(12)-
1912-13. BITTER PIT INVESTIGATION. THE CAUSE OF BITTER PIT: ITS CONTRIBU-
UTING FACTORS, TOGETHER WITH AN INVESTIGATION OF SUSCEP-
TIBILITY AND IMMUNITY IN APPLE VARIETIES. SECOND PROGRESS
REPORT. 224 p., 61 pi., map. Melbourne. Literature, p. 96.
I913-14. BITTER PIT INVESTIGATION. THE CONTROL OF BITTER PIT IN THE GROW-
ING FRUIT. THIRD PROGRESS REPORT. 1 76 p., 38 pi., 5 maps.
Melboiune. Literature, p. 96.
1914-15. BITTER PIT INVESTIGATION. THE EXPERIMENTAL RESULTS IN THEIR
RELATION TO BITTER PIT, AND A GENERAL SUMMARY OF THE INVES-
TIGATION. FOURTH REPORT. 178 p., 41 pi. Melbourne. Litera-
ture, p. 84.
(13) MELANDER, a. L., AND Heald, F. D.
1916. THE CONTROL OF FRUIT PESTS AND DISEASES. In Wash. Agr. Exp. Sta.
Pop. Bui. 100, 61 p.
(14) Mix, a. J.
1916. CORK, DROUTH SPOT AND RELATED DISEASES OF THE APPLE. N. Y.
Geneva Agr. Exp. Sta. Bui. 426, p. 473-522, 12 pi. Literature cited,
p. 520.
(15) Norton, J. B. S.
1913. JONATHAN FRUIT SPOT. In Phythopathology, v. 3, no. 2, p. 99-100.
(16) Reed, H. S., and Crabill, C. H.
1915. NOTES ON plant DISEASES IN VIRGINIA OBSERVED IN 1913 AND 1914.
Va. Agr. Exp. Sta. Tech. Bui. 2, p. 37-58, 17 fig.
Jan. 21, 1918 Irrigation Experiments on Apple-Spot Diseases 137
(17) Scott, W. M.
191 1. A NEW FRUIT SPOT OP APPLE. In Phjrtopathology, v. i, no. i, p. 32-34.
(18) AND Roberts, J. W.
1913. THE JONATHAN FRUIT-SPOT. In U. S, Dept. Agr. Bur, Plant Indus.
Circ. 112, p. 11-16, 2 fig.
(19) SORAUER, Paul.
1900. scHUTz DER OBSTBAUME gegEn krankheitEN. 238 p., iio fig. Stutt-
gart.
(20)
1909. handbuch DER pflanzenkrankheiten. Aufl. 3, Bd. I. Berlin.
(21) White, Jean.
1911. bitter pit in apples. In Proc. Roy. Soc. Victoria, n. s., v. 24, pt. i,
p. 1-19, 9 pi.
(22) Wortmann, Julius.
1892. UEBER die sogenannte "stippen" der aepfel. In Landw. Jahrb.,
Bd. 21, p. 663-675.
PLATE 2
A. — Early stage of bitter-pit on Northern Spy apple from Westminister, Vt., No.
vember i6, 1916.
B. — Cross section of the apple shown in A. Brown spots are evident just beneath
the skin, and a few others can be seen deeper in the flesh of the apple.
(138)
Irrigation Experiments on Apple-Spot Diseases
Plate 2
/
Journal of Agricultural Research
Irrigation Experiments on Apple-Spot Diseases
Plate 3
*v .
#■
¥
' i
Journal of Agricultural Research
Vol. XII, No. 3
PLATE 3
A. — Late stage of bitter-pit on Rhode Island Greening apple.
B. — Internal browning accompanying bitter-pit.
C. — Jonathan-spot on Jonathan apple.
D. — Early stage of drouthspots on a Winesap apple from Wenatchee, Wash. The
drops of exudate can be seen on the surface of the apple.
E. — Late stage of drouthspots on a Winesap apple. Note the deep depressions
scattered over the surface of the apple .
F. — Cross section of the apple shown in E. Note the almost entire absence of brown
corky tissue.
27806°— 18 3
PLATE 4
A. — An apple orchard showing the furrow system of irrigation employed in the ex-
perimental work at Wenatchee, Wash.
B. — ^Jonathan apple tree showing the effects of drouth, Wenatchee, Wash. Photo-
graphed on September i, 191 5.
Irrigation Experiments on Apple-Spot Diseases
Plate 4
Journal of Agricultural Research
Vol. XII, No. 3
Irrigation Experiments on Apple-Spot Diseases
Plate 5
Journal ot Agricultural Research
PLATE 5
«
A. — Cork on Yellow Newtown apple from Hood River, Oreg. Note the roughene,
appearance.
B. — Cross section of the apple shown in A. Note the area of brown corky tissue.
C. — "White Pearmain apple showing the severity of the 1915 drouth at Wenatchee,
Wash. No drouthspots were developed on such apples.
D. — Cork, or " dryrot ", on a King apple. Note the brown corky tissue near the core.
In surface view such an apple appears normal.
E. — Blister on an Esopus apple from Entiat, Wash.
F. — ^An extreme case of Yorkspot on a York Imperial apple. Note the circular
nature of the injury.
G. — Cross section of the apple shown in F. Note the pockets and the brown corky
tissue beneath the surface depression.
RELATION OF CARBON DIOXID TO SOIL REACTION
AS MEASURED BY THE HYDROGEN ELECTRODE ^
By D. R. HoAGtAND and L. T. Sharp,'
Assistant Chemists, Agricultural Experiment Station of the University of California
INTRODUCTION
In a previous article (ii) ' the authors have presented data con-
cerning the question of soil reaction as determined by the hydrogen
electrode. Since this work did not include direct measurements of the
effect of carbon dioxid on the reaction of soils, it was thought desirable
to carry out further experiments on this point. Before discussing the
data obtained in these additional experiments it will be well to emphasize
again the fundamental principles upon which the conclusions of our
first paper were based.
The present tendency to advance involved explanations of the nature
of soil acidity seems to be unnecessary, for the simple conception of the
relations of H- and OH-ion concentrations are in accord with the facts
so far ascertained and are warranted by the accepted teachings of chemis-
try. The lack of agreement in the literature appears to be due to the
attempt to use interchangeably the terms "lime requirement" and
"soil acidity." In the methods of determining the lime requirement it
is proposed to measure the amount of lime required to bring the soil to an
end point dependent upon arbitrarily selected conditions. These methods
are in themselves so varied and the final measurement of reaction so dif-
ficult, that the discordant results which appear in the literature are wholly
to be expected.
On the other hand, the term "soil acidity" has a definite and precise
meaning — ^namely, that condition of the soil in which its aqueous solu-
tion contains H ions in excess of OH ions. In our opinion it would be
preferable to refer to soil acidity, soil neutrality, and soil alkalinity as
those phases of soil reaction in which the H-ion concentration is respec-
tively greater than, equal to, or less than the OH-ion concentration.
These H- ion concentrations may be definitely determined by measure-
ments with the hydrogen electrode.
The lime requirement, in so far as it is related to soil acidity, would
consist of that amount of lime necessary to bring an acid soil to the neutral
point as ascertained by the above-mentioned procedure. Such a lime
requirement implies that the dissolved and total undissolved soil acids
have been neutralized. To put this procedure into practice may involve
1 From the Divisions of Agricultural Chemistry and Soil Chemistry and Bacteriology in equal cooper-
ation.
2 Reference is made by number (italic) to "Literature cited," p. 147-148.
Journal of Agricultural Research, Vol. XII, No. 3
Washington, D. C. Jan. 21, 1918
Iq Key No. Cal.— 13
(139)
140 Journal of Agricultural Research voi. xii, N0.3
certain inherent difficulties. The reaction is so prolonged either by the
slow rate of solution of the soil acids or their slow diffusion through the
soil particles that the point of neutrality may not be easy to establish
and maintain permanently. Thus an apparent state of equilibrium at
the neutral point may be attained with a subsequent slow return to an
acid condition, owing to the solution and diffusion of the soil acids.
But once the total soil acids have been neutralized, a further return to
an acid condition can come about only through leaching processes or,
possibly, in a few cases through decomposition of organic matter.
As pointed out, some soils whose solutions are neutral or alkaline
remove considerable quantities of calcium hydroxid from the solution
without materially increasing the OH-ion concentration of the soil
suspension. In certain cases this reaction might be erroneously attrib-
uted to the soil acids. Although the term "lime requirement" is in
common usage in agricultural literature, it has been variously interpreted
by different investigators. At present the term is devoid of scientific
significance. In distinction thereto soil reaction whether acid, neutral,
or alkaline, is capable of precise definition and determination.
EFFECT OF CARBON DIOXID ON SOIL REACTION
In considering the effect of carbon dioxid on soil reaction Maclntire
(8) states that many acid soils when extracted with water saturated
with carbon dioxid yield alkaline extracts. He also makes the following
conclusion :
Since we admit that the soil solution is the medium through which a plant absorbs
its mineral supply, we are compelled to conclude that a plant's sotu-ce of nutrition
is almost always alkaline, but of varying degrees of alkalinity.
From the nature of the chemical equilibria involved {2, 9, 4), we have
been unable to reach the conclusion that a solution existing in contact
with an acid soil can ever become alkaline owing to any change in the
partial pressure of carbon dioxid. An increase in carbon-dioxid tension
would either be without effect upon the H-ion concentration, or else
would increase it, depending upon the relative dissociation constants of
carbonic acid and the soil acids. It is conceivable that certain acid soils
when extracted with water saturated with carbon dioxid might yield
filtrates which would give an alkaline reaction after their carbon-dioxid
content had come into equilibrium with the partial pressure of atmos-
pheric carbon dioxid, but the equilibria governing the reaction of the soil
solution in contact with the soil are by no means identical with those
regulating the reaction of the filtrate obtained from such a soil.
In order to obtain direct evidence on the influence of carbon dioxid on
soil reaction, the experiments reported in this paper were undertaken.
The method of procedure was similar to that previously described by the
authors, with the addition of an arrangement for controlling the partial
pressure of carbon dioxid in the atmosphere above the soil. By adopting
Jan. 21, i9i8 Relation of Carbon Dioxid to Soil Reaction 141
a simplified form of the apparatus used by McClendon and Magoon (7)
and McClendon (6) for investigating the H-ion concentration of sea water,
we have been able to obtain the desired data.
DESCRIPTION OF APPARATUS AND RESULTS OBTAINED
The hydrogen-electrode apparatus was the same as that previovisly
described in this journal. To provide a chamber for mixing the hydrogen
and carbon dioxid a graduated i ,000-c. c. cylinder, the base of which had
been cut off, was immersed in a larger cylinder filled with mercury. The
upper end of the inner cylinder was tightly stoppered and contained two
capillary stopcocks for admission and outlet of the gases. A definite
quantity of purified hydrogen, electrolytically generated, was admitted to
this cylinder through one stopcock. Through the other stopcock there
was admitted from a gas burette a known quantity of pure carbon dioxid.
Both gases were measured at atmospheric pressure. A sufiicient time
„was then allowed for the thorough diffusion of the gases, which was aided
by raising and lowering the inner cylinder. The reservoir of mixed gases
was then connected to the hydrogen-electrode chamber which contained
the soil suspension. Forty to seventy c. c. of the gas mixture were forced
into the space above the soil suspension, adjusted to atmospheric pressure,
and the hydrogen-electrode cell was then closed. Equilibrium was
hastened by the shaking method, and the voltmeter readings were noted.
This procedure was repeated with new portions of the gas mixture until
the voltmeter readings were constant to within 0.005 volt. The experi-
mental details and results are recorded in Table I.
The term "With previous car bona tion" signifies that carbon dioxid
has been passed through the soil suspension for a period of >^ to 2 hours
previous to the determination of the H-ion concentration. The purpose
of this step was to ascertain the effect of thoroughly saturating the soil
with carbon dioxid upon its subsequent reaction. This also proved to
be necessary in the case of some of the alkaline soils to insure complete
saturation of the carbonates present, thus making it possible to attain
the final equilibrium when using the smaller percentages of carbon
dioxid, without the preparation of excessive quantities of the gas mix-
ture. By satisfying in this manner the capacity of the soil to combine
with carbon dioxid, it is possible to reach a partial pressure of carbon
dioxid above the suspension in the hydrogen-electrode cell comparable
to that in the mixing cylinder.
As expressed in Table I the o per cent of carbon dioxid means that
several portions of pure hydrogen were used to obtain equihbrium.
With such a technic the loss of carbon dioxid from the soil is minimized .
Thus, the atmosphere above the suspension will undoubtedly contain a
small percentage of carbon dioxid.
142
Journal of Agricultural Research
Vol. XII, No. 3
Table I. — Effect of carbon dioxid on the reaction of soil suspensions
Description of soils.
Quan-
tity
of
soil.
Quan-
tity
of
water.
Per-
cent-
age of
carbon
dioxid
in gas
mix-
ture by
volume.
Reaction without
previous carbonation.
Reaction with
previous carbonation.
Soil
No.
Read-
ings on
volt-
meter."
H-ion con-
centration.
Read-
ings on
volt-
meter."
H-ion con-
centration.
I
Fine sandy loam (California)
do
Gm.
10
C.c.
so
0. OO
•42
1.90
9. 00
.00
. 22
.42
•49
1.90
4. 80
9. 00
.00
.42
1.90
4.80
9. 00
. 00
. 22
1.90
9. 00
.00
4.80
. 00
1.90
.00
1.90
. 00
1.90
.00
. 20
2.00
.00
. 20
.42
1.90
4.90
.00
•50
1. 90
4.90
0.589
•590
Gram-mole-
cules per liter .
0. 40 X 10-^
.4oXio-<
0.592
Gratn-mole-
cules per liter.
0.37X10-'
I. . . .
do
• S89
do
• SSo
•783
.S9Xio-<
.17X10-'
2
3
Silty clay loam (California)..
do
lO
SO
• 763
.742
.38X10-'
.92X10-'
do
•747
.73X10-'
do
•744
• 726
.SsXio-'
. 18X 10"*
do
. 726
.709
.688
.896
.18X10-6
.34X10-'
.8oXio-«
. 20X lO"*
do
do
3
Clay loam (California)
do
10
5°
3
•773
.768
. 26X lO"'
do
• 33 Xlo-»
do
•7S3
. 762
. 700
.687
.59X10-'
.42X10-'
.49Xio-«
.84Xl0-«
do
•7S6
.712
•53X10^
.3iXio-«
4
4. . . .
Clay loam (Pennsylvania)...
do
lO
SO
do
.685
.9oXro-«
do
• 66s
.667
• 644
.658
.20X10-*
.18X10-*
.46X10-5
.26X10-6
s....
Clay loam ( Louisiana)
do
10
50
.66s
•647
.663
•653
.685
.664
.602
. 20X10-*
. 40X 10-*
.21X10-*
.32X10-6
. 90X io-«
. 2lXlO-*
6....
6....
Silty clay loam (Wisconsin).
do
lO
SO
7
7
Clay loam (Louisiana)
do
10
SO
.684
.94Xl0-«
8....
8....
Silty loam (California)
do
10
SO
. 604
• 603
.792
.23X10-^
.24X10-^
.13X10-'
.25X10-*
do
10
SO
.789
.770
.726
.14X10-'
. 30X I0-'
. I8XI0-*
do
9. . . .
do
10
Silty clay loam (California)..
lO
SO
•7^0
.30X10-'
.764
•7S8
.724
. 700
.38X10-'
. 49X IO-'
.i9Xio-«
.49X10-8
do
10. . . .
do
do
do
lO
so
.782
.19X10-'
II
do
•75S
•735
• 718
.55X10-'
. 12X10-*
do
do
.24X10-0
o Mercury cell, with mercuric chlorid and potassium chlorid in Nlio concentration
Since the measurement of the H-ion concentration is based on hydrogen
at atmospheric pressure, any diminution in this pressure caused by the
admixture of carbon dioxid would result in a certain lowering of the
electromotive force. Loomis and Acree (5) have determined the changes
in electromotive force resulting from the diminution of the partial pres-
sure of hydrogen within certain ranges. Their data indicate that these
changes are so slight as to be without significance in the present investi-
gation, hence no corrections for this factor have been made. Most of the
determinations reported above are the averages of duplicates. The
agreement of these duplicates was in nearly all cases within 0.005 volt.
The data in the above table have to do with the efifect of carbon dioxid
on three general types of soil reaction. The acid type of reaction is
Jan. at. 1918 Relation of Carbon Dioxid to Soil Reaction 143
represented by soils 1,4, 5, 6, 7, and 8, the neutral or slightly alkaline
type by soils 2, 9, 10, and 11, and the strongly alkaline reaction by soil 3.
Considering first the case of the acid soils, we note that the increase in
H-ion concentration of the soil suspensions, in contact with even the
higher percentages of carbon dioxid, is scarcely greater than the errors
of observation. In soils i and 8, whose suspensions give H-ion concen-
trations of the magnitude io~*, there has been no perceptible change in
reaction due to the presence of carbon dioxid. The other acid soils,
which have a smaller concentration of H ion, show some slight increase
in acidity when in contact with the mixture of gases containing carbon
dioxid. This might be expected from a consideration of the relation of
the dissociation of carbonic acid to that of the soil acids. Those soils
whose acids dissociate comparably to carbonic acid are not measureably
affected by the partial pressures of carbon dioxid used in these experi-
ments, while soils containing less dissociated acids have their H-ion con-
centration increased to a sHght extent by the carbon dioxid.
The H-ion concentration of suspensions of the slightly alkaline soils is
appreciably increased by increasing the partial pressure of the carbon
dioxid. The degree of increase in acidity seems to be dependent upon
the proportion of carbon dioxid in the gas mixture. In fact, the higher
percentages of carbon dioxid brought about a slightly acid reaction in
the suspensions of these soils.
In case of soil 3, a so-called "alkali" soil, which had a very low H-ion
concentration, a large increase was caused by the introduction of carbon
dioxid into the hydrogen-electrode cell.
DISCUSSION OF RESULTS
The results recorded in this paper show that the effect of carbon dioxid
on the H-ion concentration of soil suspensions is not an insurmountable
difiSculty in obtaining the reaction of soils under various conditions by
means of the hydrogen electrode. The technic followed in the former
investigation in which the loss of carbon dioxid is minimized, although
not entirely avoided, evidently gives results of the same order of magni-
tude as would be obtained if there were no loss of carbon dioxid. The
adoption of such a view is warranted by the fact that the maintenance
of a small partial pressure of carbon dioxid above the soil suspension in
the hydrogen-electrode cell, only altered the H-ion concentration by
less than a magnitude. It should be remarked, however, that in soils
containing alkali carbonates, and having a high OH-ion concentration,
the partial pressure of the carbon dioxid exercises a very pronotmced
effect upon the reaction, as instanced by soil 3. For a clear exposition
of equilibria governing such systems the reader is referred to Johnston {4) .
So far we have not considered the question of H-ion concentration in
soils under field conditions. By determining the carbon-dioxid content
of the soil atmosphere under field conditions and then duplicating the
144 Journal of Agricultural Research Voi.xii. No. 3
partial pressure of carbon dioxid in the manner suggested in this paper
it should be entirely practicable to obtain a measurement of the H-ion
concentration identical with that of the soil in the field. Russell and
Appleyard {10) have found that the carbon-dioxid content of the soil
atmosphere under different conditions varied between 0.02 and 2 per cent,
the general mean of arable soils being 0.25 per cent. If these percentages
of carbon dioxid are to be regarded as typical for field soils, then in view
of the present experiments the changes in carbon-dioxid content during
sampling and laboratory manipulation would not invalidate our inferences
with regard to the reaction of soils under natural conditions.
The a priori considerations already presented in the first portion of this
paper with respect to the effect of carbon dioxid on the reaction of soils
are entirely substantiated by the experimental data, which lead to con-
clusions at variance with those of Maclntire {8) on this point. Even
saturating the soil suspension with carbon dioxid previous to measuring
the reaction did not decrease the H-ion concentration. Therefore it
follows that an acid soil would never present to the plant a soil solution
of alkaline reaction, notwithstanding any increase in the partial pressure
of carbon dioxid. It should be recalled that the criteria heretofore
used for judging the reaction of soils do not always permit of an accurate
distinction between soils of different reactions. Some soils may be
judged as acid from the standpoint of certain lime requirement methods,
when in reality their reaction may be alkaline. For this reason it may
be doubted whether some of the soils reported as yielding alkaline
extracts were in fact acid. Although in certain instances the application
of lime may be followed by an increased crop yield, this result may not
be dependent upon any change in the reaction of the soil due to the addi-
tion of lime. The more accurate interpretation of liming experiments
demands that an attempt be made to differentiate lime as a neutralizer
of acidity and the other directly or indirectly beneficial effects of lime
on the soil or the plant.
The statement frequently met in the literature that the extracts from
soils considered acid have a neutral or alkaline reaction has led to the
conclusion that water-soluble acids are not found in acid soils. We
desire to emphasize again the point that extracts of acid soils, especially
those prepared with carbonated water, might become neutral or alkaline
after the loss of carbon dioxid from the extract. Moreover the reported
results on the extracts are likely to be misinterpreted, for they do not
take into account the H-ion concentration but are based upon titrations,
using indicators whose end points may be removed from neutrality by
several magnitudes. Indeed Gillespie {3) has found that soils determined
to be acid by the hydrogen electrode, yield extracts whose reactions,
colorimetrically estimated, were in close agreement with the hydrogen-
electrode measurements.
Jan. 21. 1918 Relation of Carbon Dioxid to Soil Reaction 145
One of Bouyoucos's (i) conclusions from his valuable and ingenious
application of the freezing-point method to soil investigations is as fol-
lows:
Since no mineral soil out of a great number tested gave an acid curve but only an
absorption curve, and inasmucb as the free acid, and acid salt produced in these soils
when they were treated with neutral salts, or acid and acid salts, were carried away by
washing and the soils then gave an absorption ctu^ve, the conclusion seems to be that
the presence of soluble acids, or acid salts, in the mineral soils under favorable natural
conditions is only temporary, if ever present, and never permanent. The acidity
or lime requirement of soils, therefore, seems to be due mainly to the insoluble acids
of the soil, the silicic acid, silica, acid alumino-silicates, and perhaps to the insoluble
organic matter. There appears to be then practically no active acidity in the mineral
soils, but only negative. Exceptions to these general statements are probably very
few.
Contrary to the above conclusion, the data presented in this and other
papers {3, 11), show that the solution in equilibrium with the soil par-
ticles of certain soils contains H ions in excess of OH ions. Such soils
are therefore necessarily acid and they include various types, many of
which would be called mineral soils. Furthermore, it is well to bear in
mind that the hydrogen electrode is capable of measuring specifically the
H-ion concentration, while the freezing-point method is unsuited to this
purpose. This is especially true in dealing with such heterogeneous
systems as those of the soil mass. These statements are not to be con-
strued as denying the possible value of the freezing-point method in
estimating the total "lime requirement."
The soil acids are frequently referred to as insoluble, but such insolu-
bility is, as a matter of fact, only relative, for complete insolubility is
practically unknown and the soil acids must therefore have a definite
solubility although it may be slight. The important consideration is that
the solution of an acid soil must be continuously acid in just the same
way that a solution in contact with silicic acid or other slightly soluble
acids is always acid. It is quite true that when the soil is treated with a
base in order to bring about a condition of alkalinity, by far the greater
part of the base is used in combining with acids which at any given
moment were not in the solution. But this is in no way opposed to the
conclusion that the solution in contact with an acid soil is an acid solution
and would accordingly offer an acid medium for plant growth. The
effect of such a medium would be related to its H-ion concentration and
not to the total quantity of base required to neutralize all the soil acids
present. In other words, the reaction of a soil is concerned with the
dissolved fraction, which is in equilibrium with the undissolved soil mass.
If any added substance — for example, calcium carbonate — disturbs this
equilibrium, then it is clear that the undissolved portion of the soil will
enter into the reaction in accordance with the laws of mass action.
146
Journal of Agricultural Research
Vol. XII, No. 3
EFFECT OF POTASSIUM CHLORID ON THE H-ION CONCENTRATION
OF SOILS
As previously shown, neutral salts added to suspensions of certain
soils considerably increased their H-ion concentration. This fact has
brought up the question as to the effect of the diffusion of potassium
chlorid from the agar connecting tube into the soil suspension. In
order to determine the magnitude of the possible changes induced in the
reaction of soil suspensions by the escape of potassium chlorid into them,
several experiments were undertaken. For this purpose it was necessary
to eliminate as far as possible the diffusion of potassium chlorid into the
soil suspension. As shown in Table II, one procedure consisted in
bringing the soil suspension into equilibrium with hydrogen without
any possible chance for contamination with potassium chlorid, then the
electromotive force was read just as the agar tube touched the suspension,
thus reducing the diffusion of the potassium chlorid to a minimum.
This reading was then compared with subsequent readings made in the
manner heretofore described. In addition, some agar tubes prepared
with soil extracts were substituted for the potassium chlorid tubes.
The sensitivity of our galvanometer did not allow of a greater accuracy
than 0.02 volt when the soil-extract tubes were used.
Table II. — Effect of potassium chlorid
on the reaction
7/ soil sus
pensions
Laboratory
number of
With first contact of
potassium chlorid tube.
After many contacts of
potassium chlorid tube.
Soil-extract tube.
soil.
Readings
on volt-
meter.
H-ion concentra-
tion.
Readings
on volt-
meter.
H-ion concentra-
tion.
Readings
on volt-
meter.
H-ion concentra-
tion.
7
0.681
Gram-molectUes
per liter.
0. IlXlO"5
0. 664
.764
•589
•675
.651
•675
.699
.608
•738
.660
•653
Gram^nwleciUes
per liter.
0. 21 X 10-5
.38X10-^
.40X10-"
.13X10-5
•35X10-5
.13X10-5
.5iXio-«
. 19X10-"
. IlXlO-8
.24x10-5
.33x10-5
0. 661
.771
Gram-molecules
per liter.
0. 23X10-5
. 29X10 ^
2C
I
•595
.677
•653
.678
. 702
. 610
■739
.660
•653
•33X10-"
.12X10-5
.32X10-5
.12X10-5
.46Xio-»
. 18X10-"
I. oXio-7
.24X10-5
•33X10-5
26
27
28
2Q
^I
7.2
■i-x
?4
It is evident from the data in Table II that the slight diffusion of
potassium chlorid from the agar tube has a tendency to increase the
H-ion concentration of the soil suspension. In almost all cases this
increase corresponds to less than 0.005 volt. For most agricultural
purposes this difference has no significance. By bringing the entire
system into equilibrium with hydrogen before immersing the agar tube
and by keeping the tube out of the suspension except momentarily
at the time of reading the electromotive force, it is believed that the
error will be entirely negligible.
Jan. 21. 1918 Relation of Carbon Dioxid to Soil Reaction 147
SUMMARY
(i) The H-ion concentrations of soil suspensions have been measured
under various partial pressures of carbon dioxid.
(2) The H-ion concentration of suspensions of acid soils is not
markedly affected by increasing the content of carbon dioxid up to
10 per cent. The H-ion concentration of slightly alkaline soils is slightly
increased by such treatments. A notable increase in H-ion concen-
tration is observed when soils containing ^Ikali carbonates are similarly
treated.
(3) It has not been found that any treatment with carbon dioxid can
produce an alkaline reaction in the suspension of an acid soil.
(4) When the original conditions are restored, no permanent change
in soil reaction could be attributed to the carbon dioxid.
(5) Further experiments with the hydrogen electrode have confirmed
the point of view that solutions in equilibrium with acid soils contain
H ion in excess of OH ion.
LITERATURE CITED
(i) BouYOUcos, G. J.
I916. THE FREEZING POINT METHOD AS A NEW MEANS OF DETERMINING THE
NATURE OF ACIDITY AND LIME REQUIREMENT OF SOILS. Mich. Agt.
Exp. Sta. Tech. Bui. 27, 56 p., 18 fig.
(2) Cameron, F. K., and Bell, J. M.
1907. THE ACTION OF WATER AND AQUEOUS SOLUTIONS UPON SOIL CARBONATES.
U. S. Dept. Agr. Bur. Soils Bui. 49, 64, p., 5 fig.
(3) Gillespie, L. J.
I916. THE REACTION OF SOIL AND MEASUREMENTS OF HYDROGEN-ION CON-
CENTRATION. In Jour. Wash. Acad. Sci., v. 6, no. i, p. 7-16, 2 fig.
(4) Johnston, John.
I916. THE DETERMINATION OF CARBONIC ACID, COMBINED AND FREE, IN SOLU-
TION, PARTICULARLY IN NATURAL WATERS. In Jotir. Amcf. Chem.
Soc, V. 38, no. 5, p. 947-975-
(5) LooMis, N. E., and Agree, S. F.
I916. THE EFFECT OF PRESSURE UPON THE POTENTIAL OF THE HYDROGEN
ELECTRODE. 7« JouT. Amcf. Chem. Soc, v. 38, no. 11, p. 2391-2396.
(6) McClendon, J. F.
I916. THE COMPOSITION, ESPECIALLY THE HYDROGEN ION CONCENTRATION
OF SEA WATER IN RELATION TO MARINE ORGANISMS. In JotU". Biol.
Chem., V. 28, no. i, p. 135-152, 2 fig. Bibliography, p. 152.
(7) and M.'VGOON, C. A.
I916. AN IMPROVED HASSELBALCH HYDROGEN ELECTRODE AND A COMBINED
TONOMETER AND HYDROGEN ELECTRODE, TOGETHER WITH RAPID
METHODS OF DETERMINING THE BUFFER VALUE OF BLOOD. In JoUT.
Biol, Chem., v. 25, no. 3, p. 669-681, 3 fig. References, p. 681.
(8) MacIntire, W. H.
i916. factors influencing the lime and magnesia requirements of soils.
Tenn. Agr. Exp. Sta. Bui. 115, 48 p.
(9) MiCHAELis, Leonor.
1914. DIE W.VSSERSTOFFIONEN-KONZENTRATION. 2IO p., 41 fig. Berlin.
Literatiirverzeichnis, p. 196-207.
148 Journal of Agricultural Research voi. xii, N0.3
(10) RusSEL, E. J., and ApplEyard, Alfred.
I915. THE ATMOSPHERE OF THE SOIL: ITS COMPOSITION AND THE CAUSES OP
VARIATION. In Jour. Agr. Sci., v. 7, pt. i, p. 1-48, 17 fig.
11) Sharp, L. T., and Hoagland, D. R.
1916. ACIDITY AND ADSORPTION IN SOILS AS MEASURED BY THE HYDROGEN
ELECTRODE. In Jour. Agr. Research, v. 7, no, 3, p. 123-145, i fig.
Literature cited, p. 143-145.
ADDITIONAL COPIES
OF THIS PUBLICATION MAY BE PEOCtJEED FEOM
THE SUPEEINTENDENT OF DOCUMENTS
GOVEENMENT FEINTING OFFICE
WASHINGTON, D. C.
AT
15 CENTS PER COPY
Subscription Peice, S3.00 Per Year
A
Vol. XII JANUARY 2S, 1918 No. 4
JOURNAL OP
AGRICULTURAL
RESEARCH
COISTXE^NTS
Pag*
A Study of the Plow Bottom and Its Action Upon the
Furrow Slice ----- _ _ - 149
E. A. WHITE
( Contribution from Cornell University Agricultural Experiment Station )
Influence of Nitrates on Nitrogen-Assimilating Bacteria - 183
T. L. HXLLS
(Contribution bom Wisconsin Agricultural Experiment Station)
PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE.
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINOXON, r>. C.
WASHINQTON : GOVERNMENT PRINTINQ OFFICE : Itit
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
ICARL F. KELLERMAN, Chairman
Physiologist and Assodai-e Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Exherimeni Sfatians
CHARLEvS L. MARLATT
Entomologist and Assistant Chief, Bureau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricullural Experiment
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
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.
*Dr. Pearl has undertaken special work in connection with the war emergency;
therefore, imtil further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
JOINALOFAGRKMTIALISEMCH
Vol. XII Washington, D. C, January 28, 1918 No. 4
A STUDY OF THE PLOW BOTTOM AND ITS ACTION
UPON THE FURROW SLICE ^
By E. A. White, ^
Assistant Professor of Farm Mechanics
College of Agriculture of the University of Illinois
INTRODUCTION
The most ancient records show that from a very remote period man
has used the plow, in one form or another, to assist him in stimulating
the earth to bring forth a more bountiful harvest. As has been the
case in many other lines of endeavor, theory has trailed far behind
observation and experience in developing this implement. In fact, as
far as can be ascertained, it was not until the last half of the eighteenth
century that any serious attempt was made to develop a plow bottom
from a theoretical standpoint, and even then the productions of Jefifer-
son, Lambruschini, Small, Rham, and others can not be considered as
thoroughly grounded upon well-developed theories; rather their works
should be looked upon as hypotheses (fig. i). Experience in the field
generally proved that the machines designed by these men were not all
that could be desired — ^for example, it is reported ^ that when Lam-
bruschini's helicoidal moldboard was taken into the field for trial the
driver of the draft animals immediately observed that the force required
to move this plow was too great for the results obtained. To be sure,
geometrically exact moldboards furnished the basis in many instances
for more perfect developments, but the results obtained by empirical
plow designers who worked in the field were so far superior to the results
obtained by the men who worked in the laboratory that the theorists
were soon completely outstripped and even held up to ridicule by the
men who developed their machines in the hard school of experience,
until at the present time we find special types of plow bottoms designed
1 Approved for publication in the Journal of Agricultural Research by the Director, Cornell University
Agricultural Experiment Station.
* The experimental work for this paper was done under the direction of Prof. H. W. Riley, of the De-
partment of Rural Engineering, Cornell University, and the mathematioal developments were prepared
under the supervision of Prof. F. R. Sharpe, of the Department of Mathematics. In addition to the above,
grateful acknowledgments are given to the following: To Prols. James AIcMahon and Virgil Snyder, of
the Department of Mathematics, for their most timely and helpful suggestions; to Mr. J. E. Reyna,
Instructor in Drawing, College of Agriculture, Cornell University; and to Mr. L. S. Baldwin, Instructor
in General Engineering Drawing, University of IlUnois, for making the drawings.
3 Lambruschini, R. d'ijn nxjovo orbcchio da coltri. In Gior. Agr. Toscano, v. 6, p. 37-80. 1832.
Journal of Agricultural Research, Vol. XII, No. 4
Washington, D. C. Jan. :8, 1918
It Key No. N. Y. (Cornell)— 3
(149)
I50
Journal of Agricultural Research
Vol. XII, No. 4
to meet certain field conditions; but no well-developed theory is avail-
able to serve as a guide in this work.
This paper is an attempt to begin a fundamental analysis of the plow
bottom and its work, in the hope that some light may be thrown upon the
theory of this humble but perplexing machine, and other attempts stimu-
lated to delve further into the secrets which are still to be revealed
regarding the theory of this important implement. Empirical methods
have given the world plow bottoms which work well. It is still to be
hoped that scientific investigation can refine and further perfect, supple-
ment as it were, the productions of experience.
The work undertaken by the writer can be naturally divided into three
parts: (i) A study of the forms of plow bottoms; (2) an attempt to
Date
Name
Generatrix
Directrices
Eiquation ofSurfaa
Small
Stnjiight line
Sfmigiit line E, Catena/y
Stephens
5tra/gbt b'ne
Straight line and
arc ofCirc/e
^ = tan[fMJ
1768
JefTerson
5tra/ghtb'ne
Straight lines
l^//-^--
1616
Da^/s
/\nc ofCin:Ie
/ires of Circle
1632
Lamb/vschini
5tra/gbt Line
Straight Line and
tierir
^ = fanfaz)
1639
MtherviY
andP/erce
Arc ofCyc/o/c/
Arcs ofCycloiaf
1640
Rham
Straight line
Straight l/nes
f-^=2nz
f340
Rham
Stnsight line
Curves
/852
Knox
Stroig/it Line
/Ires of Circles
f?o/ed sunbce of
e/ghth order
1854
Gibbs
Straight line
Arcs of Circles
r.-f,-^-o
/863
l\deoa/
Straight Line
Arcs of Circles
/867
tio/b/vo/r
Stroightline
Straight l/nes
1634
Jacobs
/t port/on from each of 2 Sur/iices; each surface
haWng S sets ofsfraighf //ne generators.
Pig. I. — Diagram giving the generatrices, directrices, and equations of surfaces of historical plow bottoms .
analyze the motion of the soil particles as they pass over the surface,
and (3) a mathematical analysis of the surfaces of the most important
historical plow bottoms which were designed to be geometrically exact.
It was, and still is, hoped that a knowledge of just what the plow bottom
is and how it performs its work will be of material assistance in developing
a theory which will furnish a very definite basis for the proper design of
this fundamental implement of tillage.
FORMS OF THE PLOW BOTTOM
A study of modern American-manufactured plow bottoms reveals
the fact that a large number of these are so constructed that their surfaces
contain sets of straight lines, each set consisting of an infinite number of
straight lines, so related that an equation or equations satisfied by the
coordinates of points on the surface can be found.
Jan. 25. 1918 Study of Plow Bottoms 151
Plate 6, A, represents a bottom with two sets of straight lines. The
few lines shown in the illustration indicate that through every point of
the surface two straight lines can be drawn which lie wholly on the surface
until they pass off the edges of the bottom. These straight Unes furnish
the basis for the proof that such a surface is a portion of an hyperboloid
of one sheet (for the form of this surface see fig. 3 to 7) whose equation
can be developed and studied with mathematical exactness. The
method of developing this equation will be given later, but at present we
are mainly interested in the fact that there is a classs of plow bottoms on
whose surfaces lie sets of straight lines, and, further, that one equation can
be developed which Vv'ill approximately represent the working surface of
such a bottom.
Further study shows that the surfaces of other plow bottoms contain
sets of straight lines, but that one equation will not completely describe
such a surface. In Plate 6, B, a bottom is shown whose surface is com-
posed of a portion of each of two surfaces. Plate 6, C, shows a similar
bottom, but in this case the two surfaces merge into each other farther
back upon the moldboard.
In Plate 6, D, a class of bottoms is represented whose entire surfaces
do not contain an inlinite set of straight lines. It is true that the share
and back end of the moldboard exhibit the same characteristics that the
first two classes have shown, but the lines do not continue to the fore part
of the moldboard.
Plate 7, A, shows a plow bottom with a convex surface which has two
sets of straight lines.
The American-manufactured plow bottoms studied can thus be
divided into three general classes: (i) A portion of one quadric surface;
(2) a portion of each of tv/o quadric surfaces, and (3) nonquadric sur-
faces. Nearly all forged bottoms belong to classes i and 2 with the
majority falling into class 2, while most of the cast bottoms belong to
class 3. It should be noted, however, that some recently designed cast
bottoms depart from the general characteristics of class 3 and show
clearly the two quadric surfaces of class 2. The lines running in the
general direction, front to rear, marked "/," (PI. 6, A) will be called
longitudinal lines, and those running in the general direction, top to
bottom, marked "t" (PI. 6, A) will be called transverse lines.
For the purpose of studying the forms of the various surfaces under
consideration, a machine, illustrated in Plate 7, B, was designed and built
for measuring the space coordinates of any desired point. ^ By means
of slots and a system of pulleys attached to the drafting board the cross-
bar can be kept horizontal and be moved both laterally and vertically,
while the drafting board is attached to a frame which can be moved
' Similar machines are described in the following publications: Gould, J. S., et al. report on trials
OP PLOWS. /» Trans. N. Y. state Agr. Soc, V. 27, pt. I, 1867, p. 426. 1868.
Giordano, Fedetigo. le ricerche sperimentali di meccanica agrarl\. p. no. Milanu, 1906.
152
Journal of Agricultural Research
Vol. XII, No. 4
backward and forward upon guides so marked that the board in all posi-
tions will be squarely across the guides. When a plow bottom is properly
placed upon the platform the x, y, and z coordinates of any point upon
the surface can thus be recorded upon coor-
dinate paper fastened upon the drafting
board.
/
/
/■>^ey6'26y
a'^zVs^e
Y
y
y
DBVEI<OPMENT OF THE EQUATION
.yCxy^'
Q
Fig.
From a mathematical standpoint the sur-
face shown in Plate 6, A, presents the
fj problem of finding the equation of a surface,
x,yA given two sets of straight-line generators.
This can be done if the equations of any
three lines in the same set are known.
Select three Unes ah, cd, and ef (fig. 2).
Let %!, yi, Zi, and X2, y^y z^ be the coordinates
of two points upon line ab; x^, y^, z^, and x^, y^, z^ of two points upon line
cd; and x^, y^, z^, and Xq, y^, z^ of two points upon line ef.
The equations of the lines ab, cd, and ef, are
(I)
(2)
(3)
and
X-
-X,
^2-
-yi
-yi
z-
-2l
*2-
-X,
-2l
X.-
z3.
y-
Ji-
2 -
-Z3
*4-
-*3
y^-
-ys
24-
-23
X-
-^5
-Iz
-ys
z -
3
■^5 ^6-^5 2^5-
From (2) the following equation for a plane perpendicular to the XY-
plane and containing the line cd is obtained :
W4=(«- X3) O4 - ys) -(y-ys) (x^ - X3) = O.
(4)
Similarly from (2) the equation of a plane perpendicular to the YZ-
plane and containing the line cd is
^5=(y-y3) (24- 23) -(^-23) iyi-y3)=o.
(5)
From (3), the equation of a plane perpendicular to the XV-plane and
containing the line ef is
i(x-x,)(y^-y^)-(y-y,)(x,-x,) = 0.
(6)
Similarly, from (3) the equation of a plane perpendicular to the VZ-plane
and containing the line ef is
Uj^(y-y^){z^-z,)- (z-z,)(y^-y^) = 0.
(7)
Jan. 28, i9i8 Study of Plow Bottoms 153
Consider
Ui = Au^. (8)
where A is a constant. This is the equation of a plane which contains
the intersection of planes (4) and (5) ; hence it contains the line cd.
Similarly
u^ = Bu^ (9)
where 5 is a constant, is the equation of a plane which contains the line ej.
If A and B have such values that the point %', y' , z' is on (8), (9), and
(i), the line of intersection of (8) and (9) meets (i) and is a generator
(see fig. 2). Hence,
_ {x'-x^) {y- y^) - {y'-y^){x- x^) .
{y'-y,){z-^,)-{z'-^^)iy.-yz)' ^ ^
_ {x ' - Xs) (ye - ^5) - (:^ ' - yo) K - x^ .
iy'-y.)i?.-^.)-i?'-z,)iy,-y,)' ^"^
and
x'-x^ y'-yi z'-Zi
= K; (12)
«2-«i y2-yi 22-2i
where 7v is a constant.
From equations (12)
x' = K(x2-Xi)+Xi (13)
y' = K(y^-y,)+yi (14)
z'=K(z2-Zi)+Zi (15)
From equations (10), (13), (14), and (15)
i[K{x.2-Xi)+x^-X3](y^-y3))-([K{yr.-yj)+y^- y3](x^-Xs))^ ,^.
ilK(y2-yr)+yy-y3]{Zi-h))-i[K{^2-2i)+^i-h]iy*-y3))' ^ '
and from equation (8)
J^JKy■-^■^{y^-y^-{y-y■^{'^^-'>^.
(y- J'3)(24 -%) - (2-23)(j'4-3'3)
From equations (11), (13), (14), and (15)
(17)
(\K{Xi-x^-\-x^-x^{y^-y^y)-{\K{y.2-y;)-\-y^-y^{:x^-x^)^ , ^.
^~ ^\K(y.,-y,)\y,-y^{z^-z-^)-{\K(z^-z,nz,-z^\y^-y^)r ^'""^
and from equation (9)
jj_^(^-^5)(y6-y5)-(y-?'5)(^6-^5) t \
{y-yC){2t-z^)-iz-zi){y^-y^) ^'
Eliminating A^ B, and K from (16), (17), (18), and (19), we have the
equation of a surface through the lines ab, cd, and r/. The equations are
left in this form because numerical substitutions are more easily made
154 Journal of Agricultural Research voi. xii, No. 4
at this point than would be the case if the indicated operations were first
performed with the symbols/ The general form of the equation resulting
from the previous operations is
ax' + by^ -{-cz^ + 2fyz + 2gxz + 2hxy + 2lx + 2my + 2nz+d=^ O. (20)
To reduce equation (20) to its simplest form the axes must be trans-
lated and rotated.
TRANSLATION OF AXKS"
The origin of equation (20) is translated to the center by putting
x=x'+x„, y-y'+y'o, z+z'+z^; (21)
he values of x„, yo, and Zg being obtained from the following:
ax„ + hyo+gzo+l = 0 (22)
hxo + byo+lzo + ^n = 0 (23)
gXo+fyo+cz„ + n = 0. (24)
These substitutions give, after dropping the accents from x', y' , and z' ,
an equation of the following form :
ax^ + fer^ + cz' + 2fyz + 2gxz + 2hxy + (7 = 0; (25)
where G = /.ro + wj,, + w2o + rf. (25a)
ROTATION OF axes'
Equation (25) can be further reduced by a rotation of the axes. This
is accomplished by means of a cubic equation
k:^-{a-\-b-\-c)k-' + {ab+ac + bc-p-g--h-)k-D = 0; (26)
where D =
a h g
h b /
n j c
(26a)
Let the roots of (26) be A',, k.^, and k^. The desired equation, after trans-
lating and rotating the axes is
k,x'+k,y!' + k^-^^^^^j^=0:' (27)
' A numerical problem is ilcvclopeJ by this motho<l upon pages i,i6 lo i6o.
' Snydkr, Virgil, and Sis/vm, C. H. analytic gbombtry oP spacu,. p. 77. Now York, 1914.
'Idem, p. 79.
Mdcm, p. 86.
Jan. 28, 191S
Study of Plow Bottoms
155
SKeleton. Hyperboloid of One Sheet
Fig. 3.
Section z=o. Fig 3.
Fig. 4.
Section y=o, Fig. 3.
Fio. s.
o y
Section K'O, Fig 3
Fig. 6.
Hyperboloid of One Sheet, showing Lines
upon the Sorfoce
Pio. 7.
156 Journal of Agricultural Research voi.xii.No.4
where A = DG. (27a)
The direction cosines X, m, v, of the angles which the new X-axis makes
with the original axes are obtained from the following:
(a—ki)\+hij.+gv = 0 (28)
h-\ + {b-k,)ti+jv = 0 (29)
9X+/M + (c-fei> = 0 (30)
2X+m2+v2=i. (31)
Similarly, the direction cosines of the angles which the Y- and Z-axe
make, after rotation, with the original axes are found by substituting
k^ and ^3, respectively, for k^ in equations (28), (29), (30), and (31).
When equation (27) was developed from the surface of a plow bottom
having two sets of straight-line generators, it had the following general
form:
This is the equation of an hyperboloid of one sheet, a vase-shaped figure,
the skeleton of a section of which is shown in figure 3. When 2 = 0
equation (32) becomes -^ + r2=i, and the cross section through the
plane z=0 (fig. 4) is an elUpse. When y = 0,\.\\e equation becomes
-^ — -3= I, and the section through the plane y = 0 (fig. 5) is a hyperbola.
Similarly, when % = 0, ^--3= i (fig. 6). Figure 7 indicates the two sets
of straight-line generators which lie on the surface of an hyperboloid of
one sheet .-^
APPLICATION OF THB DEVELOPMENT TO A PROBLEM
In order to develop the equation which will describe the surface of a
plow bottom, it is necessary to obtain the data called for in equations
(16), (17), (18), and (19). This application of the development will be
carried through for the bottom represented in Plate 6, A, which bottom
was placed upon the machine shown in Plate 7, B, so that the origin of
JThe constants a, b, and c of this equation do not necessarily have the same numerical values as in
previous equations.
* The method for obtaining the equations of any line on the surface is given in Snyder, Virgil, and
SiSAM, C. H. op. cit., p. 93.
Jan. 28, J918
Study of Plow Bottoms
157
coordinates came at O, figure 8. The plane y=0 contains the points O,
m, and n; and the plane x=0 contains the points O and m and is per-
pendicular to the plane y = 0. The plane 2 = 0 is perpendicular to both
the planes y=0 and x=0. The axes are considered to be positive in the
directions indicated by the arrowheads (fig. 8). Three transverse lines,
ab, cd, and ef (fig. 8), were selected and the following data obtained:
Fig. 8.
TABtE I. — Values (in inches) developed for the surface of the plow bottom shown in
Plate 6, A
Xi= 2.04
J'i= 5-7
2i = l6.0
*4= 8.54
74= 6.43
24=23.0
X2= 7.42
J2= 3-78
22=19.0
X5= 9-7
)'5= 10.88
2t = 26.0
X3= 4-42
J'3= 8.74
23 = 20.0
0:6=12.58
>'6= 7-65
2fi=28.0
When the above values are substituted in the equations already
developed,
From (16)
-2.28i<: + i3-83
(33)
158
Journal of Agricultural Research voi.xii, No. 4
From (17)
From (33) and (34)
K =
A =
— y— 1.78^ + 20
i.299)/+3- 31.35'
From (18)
From (19)
From (36) and (37)
K
- 15.73; + io.04y- 13.832+ 1 1.95
x-i.i77y- 2.282+51.45 '
1.584K + 6.335,
B =
B =
K-7.29
— %— .892}/ + i9.4
.619^+2-32.74 '
7.29^ + 2.58^-6.3352 + 65.85
x+. 0887-1.542 + 32.46
(34)
(35)
(36)
(37)
(38)
By eliminating K from equations (35) and (38) the following equation
for the surface of the plow bottom is obtained :
3.9*2 +/ + 3.4522- 7.53:^/2- 7.28x2 + 6.79%)/
+ 87.i«+i2o.75>'- 75.052 + 227.25 = (9.
(39)
Table II is compiled for purposes of checking the values computed from
equation (39) with those obtained by measuring.
Table II. — Values {in inches) for the surface of the plow bottom shown in Plate 6, A,
obtained by measurement
X computed
z
y
X
from
equation (39)
Difference.
10
2
2.9
2. 27
0.63
15
6
1-53
1.56
- -03
IS
4
3-58
3-77
- .19
15
2
6.9
6.32
•58
20
10
3-72
^•^
- .08
20
8
4.73
4.76
- -03
20
4
7-83
7-94
- -13
25
12
8.22
8.12
. I
25
9
9.07
9.2
- • 13
25
6
10.43
10. 46
- -03
30
10
14
13.86
• 14
32
9
16.5
16. I
•4
Jan. 28, 1918
Study of Plow Bottoms
159
To find the geometric center, substitute the coefficients from equation
(39) into equations (22), (23), and (24). Solving, we find
X(,= — 1.405 inches.
yo= 6.52 inches.
2o= 16.4 inches.
This translation of axes is shown in figure 9. From equation (25a)
G=— 57.3. From (25) the equation of the surface referred to parallel
axes through the center is
3.9XH/+ 3-452"- 7-5372- 7-28«2+ 6.79x^-57.3 = 0. (40)
y
Fig. 9.
To find the equation of the surface referred to the principal axes through
the center, substitute the coefficients from (39) into (26), and we have
fe'-8.35fe2_ 20.17^+2.45 = 0. (41)
On solving by Horner's method
ki= 10.27
^2= 0.128
^3= -2.05
Substituting the values just found for k^, k^, k^, D, and G in equation
(27), we find
or
10.27*^+. 128^^—2.052-= 57. 3
^ y^
I.
(42)
(2.36)2 (21.2)2 (5.29)^
The direction cosines of the angles which the axes make after rotation
with the original axes are obtained by making the proper substitutions in
equations (28), (29), (30), and (31).
i6o
Journal of Agricultural Research
Vol. XII, No. 4
For the X-axis
For the V-axis
For the Z-axis
7= T0.6136
H= =Fo.48
v= ±0.627.
7= ±0.7515
M==Fo.i437
u= ±0.6445.
T0.1415
±0.828
±0.5425-
Figure 10 shows the axes after translation and rotation and the por-
tion of the hyperboloid of one sheet which is a close approximation
to the surface of this plow
i^ bottom .
SURFACES ONE PORTION FROM
EACH OF TWO QUADRIC
SURFACES
Fig, 10.
were obtained from the share
board.
By the use of the method
which has just been em-
ployed to develop the equation
of the surface of the plow
bottom shown in Plate 6,
A, two equations can be
developed which will approx-
imately represent the surface
of the bottom shown in
Plate 6, B. By taking the
origin as at O, figure 8, the
data of Tables III and IV
and the front portion of the mold-
Table III. — Values {in inches) developed for the surface of the share and front portion
of the moldboard of the plow bottom shown in Plate 6, B
Xi= 3.92
yi=
2i =
^4=
3^4 =
.8
8.0
= 6.78
= 1-75
= 16.0
X2= 7-4
y2= -75
£2=12.0
.1:5= 2.36
r5= 4-05
25=15-0
••^3= 1-73
)'3= 2.67
23=12.0
•1^6= 5-87
}•&= 2.7
0.25%' + 2.34^^ + 0.462^— 3.2572— o.77a!;2: + 2. 66:^^
+ 6.88a; + 32.3>/-
5.812-4.4 = 0 (43)
Jan. 28, igiS
Study of Plow Bottoms
161
Table IV. — Values (in inches) for the surface of the share and front part of the moldboard
of the plow bottom shown in Plate 6, B, obtained by measuring
X computed
2
y
X
from equa-
tion (43)
Difference
10
I
4-75
4-75
0. 00
10
2
1-75
1.54
. 21
15
I
8.37
9. 00
- .63
15
2
5-47
5-64
- .17
15
3
3-77
3.82
- -05
15
4
I. I
1-3
— . 2
From the remaining surface of the moldboard the following data of
Tables V and VI were obtained :
Table V. — Values {in inches) of rest of surf ace of moldboard shown in Plate 6, B
x^= 8.67
yi= 4-95
21=24.0
x^= 9.08
J'4= 9.0
24=27.0
X2= 4.96
y2= 8.64
22=22.0
3:5=13.62
^5= 6.23
25 = 33-0
^3=11-73
yz= 4-8i
23=29.0
3tg=I2.24
j'6=ii-89
26=3 1 -O
i.o73c^— Loyy^+z^— S-ggyz— 1.53:2-1-16.37x7
-f6o.55x-f 1 25. ay- 48.52+ 109.5 = 0 (44)
Table VI. — Values (in inches) for the rest of the Tnoldboard surface shown in Plate 6, B,
obtained by measurement
2
y
X
I computed
from
equation (44)
Difference.
20
2
8.85
8.68
0.17
20
4
6.67
6.78
— . II
20
6
4.9
4-95
- -05
20
8
3-4
3-5
— . I
25
3
10. 6
10. 5
. I
25
5
9-3
9.2
. I
25
7
8.23
8. 12
. II
25
25
9
II
7-4
6.82
7-34
6.77
.06
•05
30
5
12. 2
12. I
. I
30
7
II. 7
11.63
.07
30
9
11.38
"•35
•03
30
II
"•3
II. 24
.06
30
13
II. 4
"•3
. I
35
35
5
7
14.65
14.72
U-53
14.66
. 12
.06
35
9
15
14-93
.07
35
35
II
13
15-45
16. I
15-32
15-85
•13
•25
40
8
17-57
17. 62
- -05
40
10
18.5
18. 52
— . 02
i62 Journal of Agricultural Research vo1.xii,no.4
From a study of Tables II, IV, and VI it is evident that the share can
not be as accurately described by mathematical equations as can the
moldboard. However, the differences even upon the share are not very
great. It must be remembered that these surfaces have been developed
empirically; experience and an extensive knowledge of the conditions to
be met have been the chief guides. Yet this implement produced in the
school of experience has a surface approximately mathematically exact
in form. Further, the surfaces of cast bottoms, which, because of the
difficulty of manufacture, are not changed unless necessity demands,
consist in some cases approximately of a portion from each of two quadric
surfaces. It will be shown later in discussing the history of the plow that
the surfaces of the Holbrook bottoms were designed to be portions of
hyperboloids of one sheet. In the Utica (N. Y.) plow trials these
machines received many first awards and much commendation from the
judges for the excellence of their work. In addition to this, Mr. J. J.
Washburn, of Batavia, N. Y., who knew Mr. Holbrook and was
present at the Utica plow trials, stated that the Holbrook plows did as
good work as any that it has ever been his pleasure to witness. Thus,
there is considerable evidence, based upon field experience, which indi-
cates that a portion of a hyperboloid of one sheet is the proper form for
the surface of a plow bottom. So far as is known, this hypothesis awaits
definite proof.
MOTION OF THE SOIL PARTICLES IN PLOWING
For the purpose of studying the motion of the soil particles in plowing,
the work was limited to sod ground available in the vicinity of Ithaca,
N. Y. From observations on a sod plow at work in the field (PI. 7, C),
the following general facts regarding the furrow slice were noted :
The lower outside ^ edge of the furrow slice did not appear to be either
stretched or compressed.
The upper outside edge of the furrow slice appeared to be compressed.
The inside of the furrow slice was stretched, the lower edge more than
the upper edge.
As the furrow slice passed over the moldboard the cracks, which had
formed on the inside in traveling over the share and the front portion
of the moldboard, closed up as the soil passed over the rear of the plow
bottom, indicating a point of maximum stretching.
The above considerations made it evident that a more detailed study
of the behavior of the furrow slice was desirable. For this purpose rows
of pins were set in the unplowed ground, the pins being driven in the
ground to the estimated depth of plowing, as shown in Plate 4, A. The
longitudinal rows are parallel to the line of motion of the plow, which is
also parallel to the Z-axis (fig. 8) and the transverse rows perpendicular
»The portion of the furrow slice immediately adjacent to the furrow is called the "outside."
Jan. 28, 1918 Study of Plow Bottoms 1 63
to this same line of motion. The longitudinal rows are numbered from
II to VI (Row I was omitted because the colter upset the pins), and the
pins in each row numbered from i to 10, as shown in figure 11. When
the part of the furrow slice in which the pins were set was upon the mold-
board, it took the form shown in Plate 8, B. In order to obtain the
X, y, and z coordinates of points in the furrow slice upon the moldboard,
the apparatus shown in Plate 9, A, was used. In this apparatus the
axes have the same relation to the plow bottom as those shown in figure
8. This more detailed study of the furrow slice upon the moldboard
revealed the following:
The length of Row II, pins i to 10, on top of the furrow slice was
greater than the length before the soil had passed upon the moldboard,
indicating that this por-
tion of the furrow slice yt
had been stretched.
The length of Row
II, pins I to 10, was
greater upon the bot-
tom of the furrow slice
than its length before
the soil passed upon
the moldboard.
The length of Row
VI, pins I to 10, on top
of the furrow slice was
less than its length be-
fore the soil passed upon the moldboard, indicating that this portion
of the furrow slice had been compressed.
The length of Row VI, pins i to 10, on the bottom of the furrow slice
was greater than its length before the soil passed upon the moldboard.
The lengths of Rows IV and V, pins i to 10, on top of the furrow slice
were approximately the same as their lengths before the soil passed upon
the plow bottom, indicating neither compression nor stretching.
The lengths of Rows IV and V, pins i to 10, on the bottom of the
furrow slice was greater than their lengths before the soil had passed
upon the plow bottom.
The z distances of pin 10 on top of the furrow slice were approximately
the same for each row, but less than the distance which the plow had
moved forward.
The z distances of pin 10 on the bottom of the furrow slice were
approximately the same for each row and equal to the distance which
the plow had moved forward. (The coordinates of the pins at the
bottom of the furrow slice were measured by cutting away a portion
of the soil but leaving the pins in place.)
Fig.
164
Journal of Agricultural Research
Vol. XII, No. 4
Fig. h.
These observations reveal, first, that when a cross section of the furrow
slice is considered (fig. 12) the portion marked "A" is compressed in
plowing and the portion marked "B" is stretched, while the soil in the
position of line /; is neither compressed nor stretched; and, second, that
there is a definite relation between the z coordinate of a soil particle and
the distance the plow has moved for-
ward. This relation is developed on
pages 164 to 167.
The next step was to analyze in detail
the motion of the soil particles. This
study was limited to the soil particles
upon the bottom of the furrow slice, but
the methods developed are applicable to
other portions. The paths of the soil particles upon the bottom of the
furrow slice can be very accurately traced from the scratches which
they make upon the moldboard. Plate 9, B, shows the paths of five soil
particles. By taking the axes as shown in figure 8, a projection of these
paths upon the plane z-0 showed a very uniform set of curves. Each
of these curves (fig. 13) can be very accurately described by equations
of the general form
aoc^ + b'f + lx + 'my + d = 0. (45)
When these same paths are projected upon the plane y = 0,Q. set of curves
resulted (fig. 14), each of which could be very accurately described by
equations having the following general form :
ax'^-\-hz'' + lxz + inx + nz + d = 0, (46)
From equation (45)
-~- and -jZ the veloc-
dt df
ity and acceleration,
respectively, of a soil
particle in the y direc-
tion can be found if
dx d^x
-jr and -j^ are known.
dt df
dx
The values of —r,
dt
(Px
df
and
- I I I I I r+y t-'r
/S I* 13 12 // 10
can be found from
dz
equation (46) if -7- and
d^z
-Ta are known. Thus,
to analyse the velocity and acceleration of any soil particle whose path
upon the surface of the plow bottom is known, an equation must be
found between z and time (/).
Fig.
-Projectiou of tlie paths shown in Plate 9, A, upon plane
2 = 0.
Jan. 28, 1918
Study of Plow Bottoms
165
This was accomplished by comparing the z coordinates of the bottom
ends of the pins with the distance which the plow had moved forward.
The distance which the plow moved forward is designated by s, so that
s = vi, (47)
where v = velocity of the plow, and i = time.
By the use of the apparatus illustrated in Plate 9, A, the data given in
Table VII were obtained for the soil particles upon the bottom of the
furrow slice whose paths are shown in Plate 9, B. These data are typical
of 12 sets of observations.
Table VII. — Values (in inches) of points in the furrow slice
Row II.
Row III.
Row IV.
Row V.
z
J
z—s
z
s
z—s
z
s
z—s
1
z 1 s
\
z—s
i6i
is!
\
16
iSf
\
15^
iSf
*
i5i
I Si
0
2C4
iqI
i
20|
iQi
^
19*
i9i
A
2oi
iqi
i
24
2,a
X
4
23^
23^
*
24i
23!
i
23i
23i
0
2^\
27f
0
27^
27i
-*
27-1
27i
-i
27i
27i
-i
32i
.^I*
h
32*
3ii
i
3%
3ii
i
31*
3ii
*
35I
35i
-*
3 si
3 Si
0
35i
35f
*
35i
3Si
-i
39I
Z9l
-I
Z9h
39i
-i
39f
39i
-i
39i
39i
-i
Unfortunately the soil available in the vicinity of Ithaca was not well
adapted for taking observations of the kind reported in Table VII. This
soil is not uniform in texture, contains many stones, cracks much more
readily than it stretches, and the surface is not as level as could be desired
for this work. At times it was difficult to drive the pins straight into
the ground. The data of Table VII show, however, a distinct tendency
for the difference between z and s to reach a maximum value and then
decrease again to zero; and also a slight tendency for this maximum
difference to decrease from Row I to Row V. When the work was begun,
it was hoped that sufficiently accurate data could be obtained from which
a law between z and s could be developed, but on account of the difficul-
ties already explained this was impossible. Consequently, in order to
develop a method for future work, a set of conditions were assumed which
agreed qualitatively with the observed facts. It should always be kept
in mind that this was done simply as an hypothesis whose exactness should
be thoroughly tested upon a soil better adapted to this work. The
conditions assumed for the relations between z and s are as follows :
(A) That, for each path, when 2 = 40, 5 = 40.
(B) That there was no stretching or compression in the outside bottom
edge of the furrow slice up to the point z = 40.
(C) That the maximum difference, z—s, for Path I was 1.05 inches.
27807°— 18 2
i66
Journal of Agricultural Research
Vol. XII. No. 4
\
•
1
^.
^
1
V
vv
\
V^
^>
v\
^
\
\
^
\\
\\
\
\
\N
X\
\
vC
\
\^
A^
\\
\
\\
y
\
\
\
\\
\
I \
\^
\\
\
\\
V
\
\\
\
\
\
\
i \
\
■■i
\
\ \
A
\
\
\ \
\
\
\ \
I
\
1
\ ^
\
\
\
\
\
\
\
'
\
\
\
c
\
\
\
1
Fig. 14. — Projection of the paths shown in Plate 9.
the plane j'=0.
A, upon
(D) That the maxi-
mum difference, z—s,
for each path decreased
uniformly across the
furrow slice. Thus, for
Row I, ji(;= 0.85 inch, the
maximum z—s=i.o5
inches, and when
3^=13.6 inches, the
width of the furrow-
slice, the maximum
z—s=0; so when
x=7.$ inches, the max-
imum z—s for Row V
is 0.45 inch.
(E) That the stretch-
ing in each row took
place uniformly up to
the maximum point
and then decreased
uniformly until it was
zero when z=s = 40.
(P) That the maxi-
mum stretching
occurred midway be-
tween the point where
the soil particle passed
upon the plow bottom
and the point ^ = 40.
Thus, for Path I where
the soil particle passed
upon the moldboard at
the point ^ = 0.6:
40 —0.6 = 39.4 inches.
39.4 -^ 2 = 19.7 inches.
19.7 + 0.6 = 20.3 inches,
4 For Path I the point
of maximum stretch-
2 ing was at ^• = 20.3
inches.
^ The computations
below show that for
Path V, where the soil
4iP
36
34
32
30
28
26
24
22
ZO
16
10
Jan 28,1918 Study of Plow Bottoms 167
particle passed upon the share at the point i^=ii.6, the point of
maximum stretching occurs at 5^ = 25.8 inches.
40 — 11.6 = 28.4 inches.
28.4^2 =14.2 inches.
14.2+ 11.6 = 25.8 inches.
The following is the simplest form of a function which meets the
requirements imposed by the above conditions and, when the constants
are determined, will describe the relations between 2 and ^ for a soil
particle on the bottom of the furrow slice as it passes over the surface of
the plow bottom:
z — s = a{r + hs+cf (48)
From equations (47) and (48)
z-vt=a[{vtf + hvt+cY; (49)
dz d^z
From (49) -j7 and -^ , the velocity and acceleration, respectively, of a
soil particle in the z direction can be obtained.
From equation (46) by differentiation we have
dx dz
{2ax+lz+fn)-^+ {2bz+lx+n)-j- = 0; (50)
and
+ (26^+/x+«)y,+(.6^ + /^j^-0. (51)
Similarly from equation (45) we find
and
(2a^ + 0j+(26:v+m)^=O; (52)
(2a. + //^+ -(!>+ (^^y+-)g^+ ^<|>= O. (53)
. dx dy
From equations (50), (51), (52), and (53) the velocities -j., ^-, and the
d X d/^
accelerations -yw, -r^- of a soil particle on the bottom of the furrow slice
dz d/Z
can be obtained when -n and -j^ are known.
at dr
In this problem, however, we are interested in the accelerations in the
directions of the normal to the surface, designated by "A^," the tangent
to the soil path " T," and the perpendicular to the plane formed by the
normal and the tangent "7?."
1 68 Journal of Agricultural Research voi. xii. no. 4
We can find Xi, /Xi, Wj, the direction cosines of the angles which N makes
with the X-, Y-, and Z-axis in either of the following ways :
If (20) (the equation of the surface of the plow bottom) is known, we
have by differentiation
Xi ^^1
ax^ + byo + gzo + l hxg + byo + fzo + m
= ^1 _ (rA)
gxo+fyo+czo+^i
I
v
l(aXf, + by^ + gz^ + lf+ {hx^ + by^ + fz^ + mf'
or if the paths of the soil particles are known but the equation of the
surface is unknown the angle A^^- can be measured by means of a pro-
tractor and plumb bob, as shown in Plate 9, C. The direction cosines
Xi and Ui can then be computed from the following :
(Xi)'+(mi)^+ (1^1^=1 (55)
^ dx ^ dy ^ dz ^ . ^,
where the values for ^, J-, and -y. can be obtained from (49), (50), and
(52).
, dx dy , dz
The direction cosines of T (X^, yUj, 1^2) are proportional to ^, ^-, and ^'
Hence
dx dy dz //^Y , /^V + /^— V (57)
dt dt dt '\\dt)'^\dt)^\dt)
The direction cosines of T (X3, ^3, v^ can be computed from the follow-
ing:^
(X3)'+(M3)'+(i'3)'=I- (58)
^' ^ ^^— =±i. (59)
M3l^2 ~ l^3M2 /^3«^1 — l^sMl Ml 1^2 — l'lM2
The components in the directions N, T, and R of the forces acting on a
soil element of mass M, moving with the component accelerations
d^x d^y J d^z
W df ^^^ dt^ ^'^
d^x d^v d^z
Fn = M(\^^+^,J + v,-i^,) (60)
FT-MCX.g^+^f+v.lf) (6.)
F. = M(X^ + ,,^, + v,^). (6a)
' Snyder, Virgil, and Sisam, C. H. Op. cit., p. 40.
Jan. 28, 1918 Study of Plow Bottoms 1 69
EVALUATING THE CONSTANTS IN EQUATIONS (48), (46), AND (45)
The methods of evaluating the constants in equations (48), (46), and
(45) for a given soil path will now be considered. For this purpose
Path V (PI. 9, A) will be taken. The general form of equation (48) is
z-s = a(s^ + bs + cy. (48)
From the assumptions that have already been made (p. 164 to 168) the
following data for this curve are obtained :
5-
2
II.6
11.6
25-8
26.25
40.0
40.0
On substituting the above values for s and z in equation (48), three
equations are obtained from which it is found that
a = o.ooooiii4
&=-5i-6
0 = 464
giving
z— s = 0.00001 II 4{s^— ^i. 6s +4.64)^ (63)
To determine the values of the constants in
ax^ + bz^ + lxz + mx+nz + d^O, (46)
the origin is moved to ^=7.65, 2=11.6. For this point as origin an
equation of the following form describes the curve ;
a(x'y + b(zy + l^x'z' + m,x'^0. (64)
Taking a=i, only three constants, b, l^, and m^, remain to be evaluated.
From the trace of Path V on the surface of the plow bottom the following
data were obtained :
x'
I
3
6
z'
13-55
20.05
27-15
Substituting these values for x' and z' in equation (64) gives
tions from which
three
equa-
b-
= —0.019
(x')^-
0. oig(z'y — (
= -0.453
= 8.63
D.453x'z' + 8.63x'=6>.
(65J
Translating the axes
back to the
x' =
z' =
original origin,
X- 7.65
2— II. 6
gives
rt^- 0.0192^-0.453x2- I. 4501+3. 912-49. 92 = 0. (66)
lyo Journal of Agricultural Research voi. xit. No. 4
To determine the values of the constants in
ax^ + by'^ + lx + my+d=0, (45)
the origin is moved to, x = 7. 65, y = o. 2. This changes the form of the
equation to
a{x'f + h{y'f^l^x'-\-my^O. (67)
Taking a=i, three constants remain to be evaluated. From the trace
of Path V upon the surface of the plow bottom,
x' y'
I 3.1
4 5.45
7 6.68
Substituting these values of x' and y' in equation (67) gives
a= I
6= 4.29
/i=-3o.85
Wi=- 3.67
(x')' + 4.29(/)'-3o.85%'-3.67/ = 0. (68)
The axes are translated back to the original origin by substituting
x = x'-7.6s
y = y'—o. 2
in equation (68), which gives
ac2 + 4. 29^-46. 15a;- 5. 397+ 295. 45 = 0. (69)
Numerical Example
The surface of a plow bottom is represented by the equation
o.54x:2— 1.527^+ 1. 123^—3.6972;— 1.62x2 i- 2. o^xy
+ 53.63X+ ii^-goy- 46.42+ 49.4 = 0.
The motion of a soil particle which passes upon this bottom at the point
x=6.g, y=o.2, 2 = 9.5 is described by the following equations:
2 = 0.00001622(^2 — 45. 55'+ 342)2 + 5' (70)
— 0.II922— 1.126^2+ 20.78:^+ 10.032;— 201.63 = 0 (71)
x^ + 1. Sy^ - 42. 4XX- 1. 5y+ 245.25 = 0 (72)
s = vt. (47)
Jan. 38, 191S
Study of Plow Bottoms
171
From equations (70), (71), (72), and (47) the following are obtained;
Table VIII. — Values (in inches) for —
s
c
z
y
18
27
36
18.4
27.4
36.0
7-55
II- 5
19-5
3-6
8.25
II. 0
dz
■^=o.oooo2244[{vH^-4S.5vt+s42){2vH-4S-5'v)]+'o
d^z
^=o.oooos244[{vH^-4S.5vt+s42){2V')+(2vH-45.svy]
^^ (.2382+1. i26x-io.o3)^
dt 2x — 1.1262+20.78
(,.,38,+..,.6,-,o.03)g-.(g)'+o..38(g)-+.../(g)(g)
2a:— 1.1262+20.78
dy (-^^+4^-41)^
dt
(73)
(74)
(75)
(76)
(77)
(78)
(79)
3-6>'-i.5
^^ (-2X+42.4x)g-2(g)^-3.6(g)'
dt^ S.6y-i.s
The plow moved forward with a velocity of 36 inches per second, giving
.^=36^ (80)
From equations (74), (75), (76), {77), (78), (79), and (80) the values
listed in Table IX are computed.
Table IX. — Values for —
di
rf»i
dy
£^'y
d2
d^z
di
dt^
dt
dfi'
dt
dt^
Sec.
18
%
7.09
53-6
16. 9
28.4
37-7
- 9-07
27
'A
25-15
47-75
17-32
— 50. 0
34-44
— 10. 21
3&
I
38-4
41. 6
3-44
-74.8
36
29.52
By making the proper substitutions from (80), (74), (76), and (78) in
equations (54), (57), (58), and (59) the values of the direction cosines
for the normals A^ the tangents to the path T, and the perpendiculars
to the planes formed by the normals and tangents R for three points
are computed and listed in Table X.
172
Journal of Agricultural Research
Vol. xn. No. 4
Table X. — Values of the direction cosines for normals, tangents to the path, and per-
pendiculars io the planes
«=7-55 5^3-6 2= 18.4
COS N^=
cos iYy=
COS iVj=
0.549
.716
- .429
cos rx= 0. 169
COS Ty= . 4025
cos Ta= . 9
cos i?x=
cos Ry =
cos R^=
0.817
- .564
.0977
1=11.5 y=S.25 2=27-4
COS N.^=
cos A^y=
COS N^=
0.728
. 229
- .646
COS T^=o. 546
cos Ty= . 376
COS Tj= . 749
cos R^=
cos Ry=
cos i?8=
0. 4145
- .897 •
.149
X=l9-5 ^=11 2=36
COS A^x=
COS A^y =
COS N^=^
0.698
- .215
- .683
cos Tx=o. 728
COS T'y= . 065
cos T,= . 683
cos R^=
cos i?y =
cos /?2=
0. 102
- -975
. 2
For the purpose of computing the forces a block of soil 2 inches wide,
I inch long, and )4 inch thick is taken. The mass of this soil is
M=
(2. 1. 5)62. 5p 0.0362P
(81)
1728.32.2. 12 32.2.12
p= density.
By the proper substitutions from Tables IX and X into equations (60),
(61), and (62) the forces necessary to produce the accelerations are com-
puted and listed in Table XI.
Table XI. — Forces necessary io produce acceleration in soil particles
x= 7-55
Fn= .00S03P
x=ii.S
Fii= .00281P
x=ig.s
Fn= .00234P
y= 3-6
Ft= .001 i6p
y= 8.25
Ft= .000248P
y=ii
Ft= .00428P
2=18.4
Fe= .002 52P
2=27.9
^E= •OO592P
2=36
Fb= .00778P
A soil particle in passing over the surface of the plow bottom will be
acted upon by the following :
(a) A force from the surface of the bottom acting in the direction of
the normal.
(b) Gravity.
(c) Pressure from the weight of the soil above the particle.
(d) Friction between the particle and the surface.
Jan. 28, 1918 Study of Plow Bottoms 1 73
ie) Shearing, stretching, or compression on each of the remaining five
sides of the particle, due to its contact with other soil particles.
The force which produces the movement of a soil particle in any direc-
tion will be the resultant of the components of the above-listed forces
which act in the direction of the movement.
The preceding analysis of the motion which certain soil particles
have in the operation of plowing has not been developed from- as refined
methods nor as uniform data in all cases as could be desired, but the re-
sults obtained furnish abundant evidence that the problem here at-
tempted is by no means hopeless. The study should be continued upon
a tough sod, which would stretch more uniformly, and some apparatus
which would remove the necessity of certain soil particles remaining iu
line with each other should be substituted for the pins.
HISTORY OF THE DEVELOPMENT OF PLOW BOTTOMS
The Annual Report of the New York State Agricultural Society for
1867 contains an ex-
cellent treatise giving
the geometrical con-
struction of the sur-
faces of many histori-
cal plow bottoms, but
no attempt has been
made in that report
to classify these sur-
faces upon the basis
of their mathematical
forms. Using the above-mentioned work as a basis, the author has
attempted to work out the mathematical forms of the most important
of these historical surfaces with a view to making fundamental compari-
sons with present-day plow bottoms.
JEFFERSON'S PLOW BOTTOM
In 1788 Thomas Jefferson, while making a tour in Germany, devel-
oped what appears to be one of the first methods recorded for making the
surface of the moldboard geometrically exact in form.* He argued that
the offices of the moldboard were to receive the soil from the share and
invert it with the least possible resistance. In order to do this, Jeffer-
son developed a surface which he considered best adapted for the work
of plowing, but attention should be called to the fact that no evidence"
is offered to prove the assertion. Figure 15 shows the framework for
generating the Jefferson moldboard, in which lines em and oh are the
directrices. To generate the surface a straightedge is laid upon eo and
1 GouiD, J. S., et al. Op. cit., p. 403.
from Report N Y. Stole /Igric.Soc. 1867
Fig. is.
I y4 Journal of Agricultural Research voi. xii. No. 4
moved backward, the straightedge remaining parallel to the plane
z=0. By taking the point o as the origin, the equation of the surface is
T,hyz—2dxz—2hly + 2hdz=0'^ (82)
h = breadth of furrow
c^= depth of furrow
/ = length of moldboard.
On rotating the XV-axes through tan-^= 2^/36, the equation is
{()h^' + J[(P)y'z-^hdlx'-6bHy' + 2hd^Jghi'+^(Pz=0. (83)
On rotating the yZ-axes through tan-^>^V2, the equation 13(96^ +4^^)
{iy"f-{z'Y\-?>hdlx'
^-2{hd^|^WT^-2>hH^fi)\y" + z']=-0, (84)
Translating the axes to the points
y =>' +yo
z' = z" +Zo
where yo has such a value that
2{9f^ + 4(P)yo+ 2[bd^iSb^+Sd^- 3m-yf^]=0, (85)
and Zo has such a value that
-2(9b^ + 4d^)Zo+2[bd-yJi8b' + 8d^-3m^]=0, (86)
gives
(9b^+4d^)[(j"'y- (z'y]-Sbdlx'+ (>'o'-2o')(9&' + 4cP)
+ (yo+Zo) {2bd^iSb^+Sd'- 2,m4i) =0. (87)
Letting the constant terms in (87) equal C gives
(962 + 4d2)[(y'")2- {z"y]-2>bdlx' + C=0. (88)
Translating the axes to the point x' = x"-\-Xo where Xq has such a value
that
-8bdlXo-VC=0
gives
(962+4<P)[C|/'")'- {z"f] = 8bdlx". (89)
This is the equation of a hyperbolic paraboloid.^
IvAMBRUSCHINl'S PI.OW BOTTOM
Lambruschini,^ an Italian, describes a method for generating the sur-
face of a plow bottom which he considered to be more efficient than
the surface developed by the Jefferson method. Lambruschini proposed
» The method of developing the equation for this surface is given upon pages 150 to 156.
*Snyder, Virgil, and Sisam, C. H. Op. dt., p. 73.
» lyAMBRUSCHINl, R. Op. cit., p. 37-80. 1832.
Jan. 28, 1918
Study of Plow Bottoms
175
a helacoid generated as follows : Lay out a rectangle opan (fig. 1 6) twice
the desired width of the furrow and of an empirically determined length.
Take the point m midway between points 0 and p and draw the line mm
parallel to pq. A straightedge laid upon mo and moved backward along
the line mm,^ being kept
parallel to the plane
z = 0, and with an
angular rotation pro-
portional to the move-
ment toward m^, gen-
erates the surface of
the Lambruschini bot-
tom. The point of the straightedge which was at 0 will describe the
helix 00i7 (fig. 1 6) . The equation of this surface is
-=tane,
X '
where 9 has uniformly increasing values as 2 increases.
Then d=f (z), when 0 = 90°=— radians,
Fig. 16.
/ = length of line mnii
^ = 2'
2 2
n
Hence,
f=Kr>
(90)
small's plow bottom ^
About 1760, a Scotchman, James Small, established a factory in
Scotland for the manufacture of plows. The surface of Small's mold-
board is obtained by
laying a straightedge
upon op (fig. 17) and
moving it backward
parallel to the plane
2=0, with the line pvi
and the curve oh as
directrices. The equa-
tion of the curve, a half
catenary, is obtained by
drawing a line og (fig. 18) the length of line og (fig. 17). At o erect a line
ooj perpendicular to line og and equal in length to line gh (fig. 17).
' Gould, J. S., et al. Op. at., p. 415.
rrom Report of N.Y.Sfote Agric50C. I3S7
Fig. 17.
176
Journal of Agricultural Research
Vol. XII, No. 4
Through point Oj (fig. 18) draw a line oji parallel and equal to line og.
With h and 0 as points of suspension describe a catenary with its lowest
point at O. Taking the point O (fig. 18) as origin, the equation of the
catenary is
y _ _ /g21z /3ba ^ g-21z y3ba\ ^
a = Og.
Transferring the origin to the point 0 gives
yz= — (g21z /3ba ^ g-21z /3ba\ _ ^j
(91)
(92)
as the equation of the catenary oh (fig. 17). The equations of line pm
(fig. 17) are
Fig. 18.
Any plane parallel to the plane 2 = 0 is given by 2==c, and this plane cuts
the line pm at the point
z^ = c.
It also cuts the catenary oh at the point
^^~2l^
J2 = /(C)
Z2 = C.
The equation of the fine in the plane z = c which cuts the line pm and the
catenary oh (fig. 17) is
x—h _ y—0
(93)
or
3& , }{c)-0
(94)
Jan. 28, 1918
Study of Plow Bottoms
177
As this line is always parallel to the plane 2; = O, it follows that c = 2 and
f(c)=nz).
From equations (92) and (94) then,
(x - &)[ J (e^iz /3ba + g-2iz /3ba) _ ^1 _ y ?^^ -b\ = 0, (95)
which is the equation of Small's moldboard.
STEPHEN'S PLOW BOTTOM *
About the same time that Small brought out his moldboard another
Scotchman named Stephens developed a method for forming the surface
f/x>m Report ofNYSfoHAgric SoC- 1867
Fig. 19.
From Report of N Y Star-^ Aqric Soe /36T
Fig. 21.
V^A
Fig. 20.
of a moldboard the general plan of which is shown in figure 19. The
generator for this surface is a straightedge laid upon op (fig. 19) and moves
backward parallel to the plane z=0 with the line on and the curve ph as
directrices. Stephen designed his surface by taking a quarter cylinder
opmnhg and laying out p^vti (fig. 20) equal in length to pm (fig. 19).
Perpendicular to line pj^m^^ draw mjii equal to the length of arc mh (fig.
19). Through points p, h^ (fig. 20) pass a circle of radius 2nb. The plane
figure p^m^hyh^ (fig. 20) is then laid upon the quarter cylinder (fig. 19)
so that /?! falls upon p, m^ upon m, and h^ upon h. This will locate the
curve ph (fig. 19), leaving a figure as shown in figure 21. It will be
'GoutD. J. S., ct al. Op. cit., p. 431.
178
Journal of Agricultural Research
Vol. XII, No. 4
observed in figure 21 that - = tan 6 where d has gradually increased values
from O at 2=0 to 90° at z = L Further, ^ = ?. radians where 7 represents
the lengths of arcs 11', 22', etc.; then | = tan QY From figure 20 the
equation of the circle with its center at O, taking p^ as the origin is
In figure 20
F=2nb cos <l>
(y-Fy+(z+Gy = 4n^b^
<t>+ 7= 90°;
B+B'= 90°;
a + B'+ 7=180°;
<l) = a-B;
(96)
(97)
G=2nb sin ^
= zzzzz — +"
(98)
I
2n6*
(99)
Substituting the values for F from equation (98) and for G from equation
(99) gives
| = tan[/(2)], . (100)
which is the equation of the surface.
rahm's plow bottom ^
In 1846 Rev. W. L. Rham, an Englishman, brought forward the
theory that the lines of the moldboard running in the longitudinal
direction should be
T— jL..^^^^ y straight, but that the
section of the mold-
board formed by any
plane z = c (fig. 22)
should be a straight
line or a curve, ac-
cording to the phys-
ical characteristics of the soil to be worked. Mr. Rham agreed that
for medium, mellow soils the surface of the moldboard should be
' Gould, J. S., et al. Op. cit., p. 442.
^<J
From Report of N. Y. State Agric. Soc. 1S67.
Fig. 22.
Jan. 28, 1918
Study of Plow Bottoms
179
generated by laying a straightedge upon oe and moving it backward
parallel to the plane z=0 with the lines e^h and em as directrices.
This surface will be a portion of a hyperbolic paraboloid, the same
general type as the surface which Jefferson proposed. The orthogonal
projection of the generator in various positions upon the plane z=0
will look as shown in figure 23. For stiff, clay soils the lines (fig. 24)
e o
From Report of N. Y. State
Agric. Soc. 1867
Fig. 23.
e O
From Report of N. V. State
Agric. Soc. 1867
Fig. 24
e O
From Report of N. Y. State
Agric. Soc. 1867
Fig. 25.
are made concave and for loose, sandy soils (fig. 25) they are made
convex. As no exact description was given regarding the shape of
the curves (fig. 24, 25), it has not been possible to develop equations
for the surfaces. However, as it is known that these surfaces have
straight lines in one direction and can not be described by an equation
of the second order, they are of the fourth order or higher.
KNOX'S PLOW BOTTOM*
In 1852 Samuel A. Knox, of Worcester, Mass., applied for a patent upon
the surface of a plow bottom which was certainly unique. The skeleton
of this surface is shown
in figure 26. The seg-
ments of circles I,
II, and III are placed
in parallel planes 12
inches apart, so that a
series of straight lines
will cut the three cir-
cles. Circles I and III
have equal diameters
and the diameter of
circle II is one-half
that of circles I and
III. As the equation of this surface is of the eighth order, it will not be
worked out in detail, but a development will be given to show how the
equation could be obtained.
Let the equation of the three circles be ^
from ifepon of r*Y stole Agric Soc /e67
Fig. 26.
Z=0,
{x-aY+iy
z=k
H'^'
' Gouij>, J. S., et al. Op. cit., p. 49-.
* This development is the work of Virgil Snyder, Professor of Mathematics, Cornell University.
i8o Journal of Agricultural Research voi xii. No. 4
and (x-c)-+{y-dy- = R'
Z=2k.
Draw the line from a point {x^, y^, O) on the first circle to a point {x^, y^, 2k)
on the third. Its equations are
from which
x-x^_ y-yi_ z ^
^2~% y2~yi 2fe
2k(x—X,) + z(x,—c)
- = x^ — c.
Z
2k(y-y^) + z(yi-d)
■■y^-d.
Since
{x,-cy+{y,-df = R\
we have, after simplifying,
^\{x-x,f+{y-y,y]+Akz[{x-x,){x,-c)
+ {y-yd{y-d)]+z\{x-cf+{y-dy-R']=o. (loi)
This is the equation of a cone with vertex at {x^, y^, O) and passing through
the third circle.
In the same way, find the equations of the line from (x^, y^ O,) to
(«3. yst k) on the middle circle
x-x^^ y-yi_z ^
^3-^1 yz-yi k'
k(x—x^) + z(x^ — a)
z
Hy-yi)+z(yi-b).
■ ^'3 — <i>
-ys-^-
Since
(x,-ari-(y,-b)2 = (^fj'
we have, after simplifying,
k\ix-x,y+(y-y,r]+2kz[(x-x,)(x,-a)
+ {y,-b)(y-y,)]+2'^(x,-ar+(y,-by-~j=0. (102)
When equations (loi) and (102) are multiplied out, it will be seen that
x\, fi always enter in the form x\ + fj^ = R\ By substituting R- for
0^1+ y^i in each, the equations are of the forrr
Ax,+ By,==C,
A%+B'y, = C'.
Jan. 28, i9i8 Study of Plow Bottoms i8i
Solve these equations for x^, y^ and put their values in
A = [4^2 (x+ c) — 2cz^ — 2>xhr],
B = {^kzly -\-d)- adz" - 8yk%
C= [4R~k^-4k^(x^+y^) - 4kxz-4kzy- 4.kRh-hz^(c^ + d?)]
A ' = [2kz(x+d) — 2xk'^— 2az^],
B'=[2kz(y+b)-2yk^-2bz^],
a = [R'k^ + k\x^ + f)-4kz(ax+by-R^)i-z\a^ + b^ + ^R^)].
_B'C-BC'
^'~AB'-A'B'
_C'A-CA'
^' AB'-A'B'
hence {B'C-BCy+{C'A-CAy = R\AB'-A'B)\ (103)
CYLINDRICAL PLOW BOTTOMS
In 1854 an American, Joshua Gibbs/ patented a plow bottom the
surface of which is a portion of a circular cylinder. Taking a point upon
the axis of the cylinder as the origin, the equation of this surface is
x^ y"^
-2 + p-I=C> (104)
In some foreign countries, notably Germany, the hyperbolic cylinder has
been suggested as suitable for forming the surface of the moldboard. In
this cotmection it is interesting to note that any cylindrical surface can be
described by an equation of the general form.
^,±^±.=0 (105)=
mead's plow bottom ^
In 1863 a Mr. Mead, of New Haven, Conn., patented a plow bottom,
the surface of which conformed exactly to a portion of a frustrum of a
cone. The general equation of this surface is
^ ,^2 ^2
a' + p-3-« COS)
holbrook's plow bottom
The Report of the New York State Agricultural Society for 1867 con-
tains a very complete report of the plow trials held at Utica, N. Y.,
in 1867, at which trials a line of plows designed by F. F. Holbrook, of
Boston, Mass., showed general superiority to all other makes. The
»GouLD. J. S., etal. Op. cit, p. 502. » GotJiD. J. S., et al. Op. cit.. p. S05.
* Snyder. Virgil, and Sisam, C. H. Op. dt., p. 82.
27807°— 18 3
i82 Journal of Agricultural Research voi.xii, no. 4
following quotation gives a very good description of the Holbrook sur-
faces :
We 1 were interested in the most minute details of these plows by Gov. Holbrook
and the trials at Utica and subsequently at Brattleboro, Vt., showed very clearly the
influence of the warped surface which is generated by his method upon the texture
of the soil. Gov. Holbrook is as yet unprotected by a patent on his method, and we
are therefore most reluctantly compelled to withhold a description of it but we have
no hesitation in saying that it is the best system for generating the true cvirve of the
moldboard which has been brought to our knowledge. This method is applicable to
the most diversified forms of plows, to long or short, to broad or narrow, to high or
low, no matter what the form may be, this method will impress a family likeness
upon them all. There will be straight lines in each running from the front to the
rear and from the sole to the upper parts of the share and moldboard. None of these
lines will be parallel to each other, nor will any of them be radii from a common cen-
ter. The angle formed by any two of them will be tmlike the angle formed by any
other two; a change in the angle formed by any transverse lines will produce
a corresponding change in the vertical lines, and there will always, in every
form of this plow, be a reciprocal relation between the transverse and vertical ^ lines.
Plows made upon this plan may appear to the eye to be as widely different as it is
possible to make them, and yet, on the application of the straightedge and protractor,
it will be found that they agree precisely in their fundamental character. The
siuface of the moldboard is always such that the different parts of the furrow slice will
move over it with unequal velocities.
From the above description it is evident that the surfaces of the Hol-
brook plows are portions of a hyperboloid of one sheet whose general
equation is
-2+52 ^-i
MISCElvIyANEOUS PLOW BOTTOMS
In addition to the surfaces already described there remain at least
three which show unique characteristics, but data were not available for
developing the equations.
In 1 81 8 Gideon Davis,^ of Maryland, patented the surface of a plow
bottom which was obtained by using the segment of a circle as a gen-
erator and two segments of another circle as directrices. Somewhat later,
1834, James Jacobs,* another American, brought out a plow bottom the
surface of which was a combination of two mathematical surfaces, each
of which had sets of straight lines in two directions.
In 1839 Samuel Witherow, of Gettysburg, Pa., and David Pierce, of
Philadelphia, Pa., brought out a plow bottom whose surface was gen-
erated by the most ingenious use of the arc of a cycloid. A more detailed
description of this plow can be found in the Report of the New York
State Agricultural Society for 1867,^
* GoTJXD, J. S., et al. Op. dt., p. 586.
* It should be noted that the hnes here called transverse are designated as longitudinal (PI. 2, A), and the
]ines called vertical are designated as transverse.
^GoxJiD, J. S., etal. Op. dt., p. 452.
* Idem, p. 486.
*Idem, p. 491.
PLATE 6
A. — A plow bottom with two sets of straight lines.
B. — A plow bottom, the surface of which is composed of each of two surfaces.
C— A plow bottom similar to B, but with the smfaces merging into each other
farther back on the moldboard.
D. — A plow bottom, the surface of which does not contain an infinite set of straight
lines. '
study of Plow Bottoms
Plate 6
Journal of Agricultural Research
Vol. XII, No. 4
study of Plow Bottoms
Plate 7
Journal of Agricultural Research
Vol. XII, No. 4
PLATE 7
A. — A plow bottom with a convex surface whicli has two sets of straight lines.
B. — Instrument for measuring the space coordinates of any point of the plow bottom.
C. — A sod plow showing the furrow slice tiuned by it.
PLATE 8
A. — Rows of wooden pins driven into the sod for estimating the stretch of the fur-
tow slice.
B. — Furrow slice showing the position of the pins when on the moldboard.
study of Plow Bottoms
Plate 8
^Mj^mfs^^
5s.^ "•. 't'jfr yr x
Journal of Agricultural Research
Vol. XII, No. 4
study of Plow Bottoms
Plate 9
Journal of Agricultural Research
Vol. XII, No.4
PLATE 9
A. — Plow showing attachment used to obtain the x, y, and 2 coordinates of points in
the furrow slice.
B. — Moldboard showing the paths of five soil particles.
C. — Meastirement of the angle Ny by use of a protractor and a plumb bob.
INFLUENCE OF NITRATES ON NITROGEN-ASSIMILAT-
ING BACTERIA^
By T. L. Hn^ivS,-
Research Bacteriologist, Idaho Agricultural Experiment Station
INTRODUCTION
REIvATlON Olf NITRATES TO VARIOUS FORMS OF PLANT LIFE
The importance of nitrogen to plant life can not be overestimated.
It is one of several elements essential to plant growth, one, moreover,
which is apt to be deficient in arable soils. These facts are well brought
out by the almost innumerable investigations which have been made
concerning the source of nitrogen for plants.
The influence of nitrate nitrogen on various plants has been the con-
trolling idea in many of these experiments. Very little attention has
been placed on the effect of nitrates on the lower plants, especially the
bacteria. Because of the relation that exists between higher plants and
bacteria it seems advisable to consider the effect of nitrates on the soil
bacteria. Indeed, progress in the knowledge of nitrogenous fertilizers
depends on a study of the effect of the fertilizer on the soil organisms as
well as on the higher plants. The action of fertilizers on the different
groups of soil organisms, the relation of these organisms to higher plants,
and the separation of the important from the unimportant groups are
some of the factors involved in the problem of soil fertility.
REVIEW OF LITERATURE
The relation of nitrates to the germination of seeds has been studied
by De Chalmot {iiy, who found that corn germinated in solutions con-
taining nitrate was more robust than com germinated under similiar
conditions without nitrate. He also noted that if too concentrated
solutions of nitrate were used germination was retarded rather than
hastened. The presence of nitrate also increased the amount of al-
buminous material in the seed.
The direct influence of nitrate nitrogen on the growing plant is too
well known to justify any lengthy discussion here. Jost {26, p. 134)
gives the results of experiments made by Boussingault, who grew the
sunflower {Helianthus argophyllus') in sand with and without nitrate.
'Major portion ot a paper submitted in partial fulfillment of the requirements for the degree of doctor of
philosophy in bacteriology in the Graduate School of the University of Wisconsin, December, J916.
2 The \vriter wishes to acknowledge his appreciation oi the sugsiestions and criticisms obtained through-
out the progress ot this work from Prof. E. B. Fred and E. G. Hastings, of the University of Wisconsin.
' Reference is made by number (italic) to "Literature cited." pp. 227-230.
Journal of Agricultural Research, . Vol, XH, No. 4
Washington, D. C. J^" '^- ^^iS
1- Key No. Wis. — 10
(X83)
i84
Journal of Agricultural Research
Vol. XII. No. 4
During the three months' growth of the plants 1.40 gm. of potassium
nitrate were added. At the end of the period the dry weight of the
plant supplied with nitrate was nearly 60 times greater than that of the
plant where no nitrate was added. The relation between the growth of
nonleguminous plants and the amount of nitrate nitrogen supplied is
shown in a very striking manner in the following table taken from Hell-
riegel and Wilfarth {21, p. 53-54)'.
Nitrogen as Ca (N03)2 added to pots,
gm
Dry weight of oats (grain and straw) .
gin. .
None
0.3605
.4191
0.056
5.9024
5-8510
5- 2867
10. 9814
10. 9413
21. 273a
21.4409
o-33<5
But little work has been done on the direct influence of nitrates on the
development of the Eumycetes. Some investigations have been made as
to the ability of certain fungi to assimilate nitrate nitrogen directly.
Ritter {42) studied many species and found that some forms would
assimilate nitrate directly, while others reduced it to nitrite and am-
monia. He found some forms which failed to grow on media containing
nitrate. Kossowicz {28) found that various fungi utilized nitrates and
that nitrite and ammonia were produced.
Miinter {36) studied the influence of inorganic salts on the growth of
various Actinomycetes. He found that potassium and sodium nitrates
in quantities equivalent to 5 per cent permitted good growth of the
organisms but retarded spore formation. Calcium, barium, and stron-
tium nitrates in small quantities affected some species but not others.
Small quantities of these nitrates did not affect growth to any extent,
but larger quantities were detrimental to growth and spore formation.
Silver nitrate in all amounts studied almost entirely prohibited growth.
Nitrates appear to exert some influence on the yeasts. Drabble and
Scott {13) studied the effect of sodium nitrate on these organisms. They
found that the greatest reproduction took place in solutions containing
0.2 gram-molecule of the nitrate. Increasing amounts of the salt led
to a decrease in reproductive activity until with 0.7 gram-molecule
present no reproduction took place. From their results it is evident
that small quantities of nitrate stimulated reproduction, whereas larger
amounts proved detrimental. Kayser (27) studied the effect of man-
ganese nitrate on yeasts. He found that the amount which produced
the maximum increase in the alcoholic fermentation of sugar varied with
the strain of yeast employed. He likewise found that manganese nitrate
produced greater increase than did the same quantity of potassium ni-
trate. Fembach and Lanzenberg {14) concluded that nitrates hindered
the rapidity of cell multiplication of yeasts but greatly accelerated the
action of the zymase. More alcohol was formed in the presence than in
the absence of nitrate. According to Kossowicz {28), nitrates are not a
suitable source of nitrogen for yeasts.
Jan. 28, i9i8 Nitrogen- Assimilating Bacteria 185
The direct influence of nitrates on bacteria has been studied to a limited
extent. The influence of various nitrates on soil bacteria has been
studied by Greaves (19). He added sodium, potassium, calcium, mag-
nesium, manganous and ferric nitrates to soil in varying quantities.
The amount added to the soil was such that in each case equivalent
quantities of the anion (NO3) in the various forms were added. The
effect of these salts on the bacteria was determined by using ammonifi-
cation as an index of the bacterial activity. He found that sodium-
potassium, manganous and ferric nitrates in small amounts, approxi-
mately 0.97 to 5.5 mgm. of nitrate in 100 gm. of soil, slightly stimulated
ammonification. Greater concentrations of these salts proved toxic as
evidenced by a decrease in the amount of ammonia formed. Sodium
nitrate was much more beneficial to ammonification than potassium^
nitrate. From his results as a whole Greaves concludes that it is the
electronegative ion which stimulates bacterial activity. Calcium and
magnesium nitrates proved toxic in all concentrations studied.
However, a majority of the investigations have been directed toward
a determination of the effect of the bacteria on the nitrates. But little
work appears to have been done on the direct action of nitrates on
bacteria. Pfeffer (j8, p. 351) cites some experiments showing the
repellant action of potassium nitrate toward certain bacteria. Spir-
illum undula was repelled by a solution of potassium nitrate having an
osmotic concentration equivalent to 0.5 to i.o per cent. With Spirillum
voluians a much higher concentration was necessary to bring about the
same reaction. It was found that different organisms required different
quantities of the same nitrate to repel them.
It can be readily seen that by far the greatest amount of work on the
relation of nitrates to plant growth has been done in the realm of the
higher plants. Obviously further investigations should be made in
respect to the effect of nitrates on the lower forms of plant life, especially
the bacteria. In this paper an attempt is made to set forth the results
secured in a study of the influence which nitrates exert on certain groups
of soil bacteria, including not only their reproduction but also some of
their physiological properties.
EXPERIMENTAL WORK
OUTLINE OF PROBLEM
The results of much careful experimentation show that nitrate nitro-
gen is most readily assimilated by higher plants. As a rule it seems to
stimulate the plant to increased activity. In some cases this is un-
doubtedly due to increased nutrition, while in others it is a result of
nuclear stimulation with a consequent cell multiplication. No sharp
line can be drawn between these two effects. Probably one overlaps
the other, and the increased growth of the organism can be attributed
to a combination of the two actions.
1 86 Journal of Agricultural Research voi. xii, no. 4
From a practical standpoint the relation of nitrates to the nitrogen-
assimilating organisms of the soil is of importance. Hence, it was
arranged to study the effect of nitrates on soil bacteria, especially those
forms concerned with the fixation of atmospheric nitrogen. The work
naturally falls into two rather distinct lines of investigation. First, the
influence of nitrates on Azotobacter was determined. Here studies were
made of the effect of nitrates on the growth of the organism in soil and
also the effect of these salts on the nitrogen-fixing property of these
bacteria. The action of Azotobacter on nitrates in solution, the relation
of nitrates to pigment production and to the formation of volutin bodies
were studied. Second, the influence of nitrates on the growth of Bacillus
radicicola in soil was studied. The action of B. radicicola on nitrates in
solution and the possible nitrogen-assimilating properties of the legume
in the presence of nitrates were investigated. Also the influence of
nitrates on gum production was determined. The latter part of the
investigations included a study of the relation of nitrates to nodule
formation on alfalfa.
METHODS USED IN EXPERIMENTS
Nitrates were determined by the reduction method with Devarda's
alloy and also by the phenolsulphonic acid (colorimetric) method.
The total nitrogen content of all samples was determined by the
modified Kjeldahl method with sulphuric acid, salycilic acid, sodium
thiosulphate, and copper sulphate. Where nitrate nitrogen was present,
50 c. c. of concentrated sulphuric-salycilic acid (25 c. c. of concentrated
acid plus 25 c. c. of distilled water) were added to the cultures slowly
and with constant stirring. This acid was allowed to react for a few
days, after which the usual procedure was carried out. Digestion was
continued for five to six hours subsequent to the clarification of the
liquid.
The amount of ammonia was determined by dist'ilation with steam
in the presence of magnesium oxid.
Nitrites (qualitative test) were tested for with Trommsdorf 's reagent.
In all distillations NI14 acid and alkali were used.
In the preparation of agar cultures of alfalfa seedlings the seeds were
treated with a 0.25 per cent solution of mercuric chlorid and rinsed in
sterile distilled water. Three bacteria-free seeds were transferred to the
surface of soft mannit agar (0.7 per cent agar) in each tube.
The nitrates were added in solution to all cultures. Gram-molecular
quantities of potassium, sodium, calcium, and ammoniun nitrates
(Merck's) were weighed into sterile distilled water. These solutions
were prepared in such a manner that 5 c. c. contained 450 mgm. of
nitrate. In all nitrate solutions the nitrate radical, or anion, was
present in the same quantities, while the cation, or metal, was present
in varying quantities, depending upon the particular salt.
Jan. 28, i9i8 Nitrogen-Assimiloting BacteHa 187
Plate counts of all soil cultures were made by weighing 20 gm. (dry
weight) of the soil into a 200-c. c. water blank. From this suspension
all subsequent dilutions were made. Mannit agar ^ was used for the
plate counts in the cultures of Azotobacter and B. radicicola. Duplicate
plates were made for each dilution poured.
SOIL USED
Only one type of soil was employed, Miami silt loam obtained from
the Experiment Station farm. No chemical analyses of the soil were
made other than an estimation of its organic matter content, which was
approximately 2.75 per cent. The soil was neutral in reaction and its
nitrate content was approximately 1.5 mgm. of nitrogen as nitrate in
100 gm. of the dry soil.
ISOLATION OF AZOTOBACTER AND BACILLUS RADICICOLA
Azotobacter. — (i) Strain A was isolated from a silt loam soil. This
strain grew well on mannit agar, but produced no pigment after three
weeks' growth. (2) Strain B was isolated from a sandy loam soil.
This strain grew equally well on mannit agar and produced a brownish
black pigment within one week's growth. Both strains assimilated
practically the same amount of atmospheric nitrogen under laboratory
conditions.
BACiiyLUS RADICICOLA. — A stock laboratory culture of B. radicicola
was replated twice before taking the final culture. The nodule produc-
ing power of the organism was determined by inoculating bacteria-free
alfalfa seedlings (in soft agar). After sufficient incubation nodules were
produced in abundance.
influence; of nitrates on azotobacter
INFLUENCE OF NITRATES ON THE GROWTH AND REPRODUCTION OF AZOTOBACTER IN
STERILIZED SOIL
What effect do nitrates have on pure cultures of Azotobacter in ster-
ilized soil ? Do these salts cause a decrease in the numbers of the organ-
isms? Do they cause an increase in numbers? Or do they exert no
particular influence one way or the other ? It is difficult to believe that
the latter could be true, inasmuch as nitrates have such a profound
effect on higher forms of plant life. Such readily soluble and assimilable
substances as nitrates could hardly remain without affecting either an
increase or a decrease in the number of organisms existing in their
presence.
With the idea of determining what effect nitrates might have on
Azotobacter when grown in sterilized soil, the following experiments were
planned. In this work both strains of the Azotobacter (described on
* Fred, E. B. a i^aboratory manual of son, bacteriology, p. io8. Philadelphia and London, 1916.
Journal of Agricultural Research
Vol. XII. No. 4
p. 187) were employed and conditions governing the preparation and
incubation of the cultures were similar in the case of each strain. The
only variation was the periods used in incubating the cultures. Counts
were made after one and two weeks' incubation with strain A and after
one, two, and three weeks with strain B.
Table I.
-Influence of potassuim nitrate on the growth of Azotohacter {strain A) in
sterilized soil
Treatment
(nitrate
in 100
gm. of
dry soil).
Number of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
^
Mgm.
0
0
10
10
25
25
50
50
100
100
150
150
200
200
300
300
15,600
15, 600
15,600
15, 600
15, 600
15, 600
15, 600
15, 600
15, 600
15, 600
15, 000
15, 600
15, 600
15, 600
15, 600
IS, 600
825, 000
935)000
I, 500, 000
Per cent.
\ 100
} 170
I 523
} 2, 233
} I) 295
I 179
} 27
r °
r 315)000
L 360, 000
f I, 175, 000
Per cent.
> 100
} 348
4
c
4, 200, 000
5, 000, 000
20, 400, 000
18, 900, 000
II, 000, 000
II, 820,000
r 12, 350, 000
I 10, 750, 000
1" 27,750,000
} 3, 418
6
1 8,210
3
r 9, 000, 000
I 9) 150, 000
r 25, 000
\ 55, 000
f 0
I 0
f 0
I 0
} 2, 685
12
I) 575) 000
225, 000
250, 000
0
0
1 ^^
1 °
le
16
1 °
Table II. — Influence of sodium nitrate on the growth of Azotobacter {strain A) in
sterilized soil
Treatment
(nitrate
in 100
gm. of
dry soil).
Number of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
Mgm,.
0
0
10
10
25
25
50
50
100
100
150
150
200
200
300
300
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
13, 800
310, 000
225, 000
575, 000
430, 000
2, 850, 000
5, 800, 000
15, 200, 000
12, 750, 000
17,750,000
16, 200, 000
550, 000
400, 000
0
0
0
0
Per cent.
> 100
1 188
} 1,615
} s..
} 6, 335
} ■"
} »
} "
r 425, 000
\ 490, 000
/ 875, 000
Per cent.
2
(■ 100
\
1 191
c
/ 2, 250, 000
1
6
1 492
7
f 15, 500, 000
I 13) 300, 000
/ 9) 850, 000
I 15) 750, 000
f 690, 000
I 375) 000
/ 0
I 0
/ °
I 0
\
8
1 3, 150
0
1 2,800
10
I
12
1 117
1-2
\
14
1
IC
\
16
1
Jan. 28, 1918
Nitrogen- A ssimilaiing Bacteria
189
Table III. — Influence of calcium, nitrate on the growth of Azotobacter {strain A) in
sterilized soil
Treatment
(nitrate
in 100
gtn. of
dry soil).
Number of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
1
Mgm.
0
0
10
10
25
25
50
50
100
100
ISO
150
200
200
300
300
p p p p p p p p p p p p p p p p
OOOOOOOOOOOOOOOO
260, 000
330, 000
5, 800, 000
Per cent.
> 100
1 I, 966
} 3, 440
1 310,000
\ 260, 000
/ 975, 000
\ I, 090, 000
f 9, 200, 000
\ 8, 600, 000
r 13, 200, 000
\ 12, 600, 000
f 8, 750, 000
\ 8, 000, 000
f 2, 000, 000
I 2, 350, 000
/ 0
1 0
f 0
I 0
Per cent.
> 100
2
•2
I 362
5
6
10, 700, 000
0, 600, 000
1 3> 122
7
13,250,000 \
II, 600, 000 iJ ^' ^
6, 600, 000 1
6; 050; 000 1 ^'^44
^, qoo- coo |1
8
1 4, 526
0
10
} 2, 938
II
12
13
14
IS
16
3, 900, 000
0
0
0
0
1 1,254
} °
} =
} 763
1 °
One hundred and fifty gm. of soil (dry weight) were weighed into 500-
c. c. Erlenmeyer flasks and the nitrates added in solution, as indicated in
the following tables. At the same time i per cent of mannit was added
in solution and the moisture content was raised to approximately 18 per
cent. The flasks were allowed to remain at room temperature for one
day, when the contents were thoroughly mixed. The flasks and contents
v/ere then sterilized at 15 pounds' pressure for three hours. Upon cooling
they were inoculated with 5 c. c. of a suspension of the organisms in sterile
distilled water. The cultures were incubated at 28° C. and counts made
at the intervals already indicated. Mannit agar was used in pouring the
plates. Each number in the following tables represents an average of
duplicate plates. Tables I, II, and III show the results of the work vdth
strain A and Tables V, VI, and VII the results with strain B.
It will be seen at a glance that all three nitrates exerted an enormous
influence on the growth of the Azotobacter. The smallest concentration
did not appear to exert much influence either in increasing or decreasing
the number of Azotobacter. There was a slight gain, but it was not so
marked as that brought about by higher concentrations of nitrates.
When 25, 50, and 100 mgm. of nitrate were present in 100 gm. of soil,
very large increases were obtained in practically all instances. In one
instance sodium nitrate caused the greatest relative gain, but the most
consistent increase was produced by calcium nitrate. Beginning with
150 mgm. the number of Azotobacter began to decrease. This decrease
was especially noticeable in the cultures containing potassium and
sodium nitrates. At the end of the first week, Azotobacter organisms
190
Journal of Agricultural Research
Vol. XII, No. 4
were still found in the potassium-nitrate cultures where 200 mgm. were
present. However, at the end of the second week the organisms were
dead. The same concentration of sodium and calcium nitrates proved
even more toxic. No evidences were secured, indicating that these
organisms can resist concentrations in excess of 300 mgm. of nitrate per
100 gm. of soil.
The question may be raised in regard to the influence of sterilization
on the nitrate present in the soil. Does the prolonged heating in the
presence of soil organic matter reduce the nitrate? In order to study
this point, a few cultures were prepared similar to those already described.
They were subjected to sterilization under pressure of 15 pounds for two,
three, and five hours. Nitrate determinations at the end of these periods
failed to show any reduction. In the presence of i per cent of mannit
the nitrate content remained unchanged during sterilization.
From these results it is evident that small amounts of nitrate up to
150 mgm. of nitrate in 100 gm. of soil greatly increased the reproduction
of Azotobacter. In regard to the toxicity of higher concentrations,
sodium nitrate appeared to exert the greatest influence in this direction,
followed by calcium and potassium nitrates in the order named. The
results of the experiment are recorded in Table IV.
Table IV. — Inflttence of ammonium nitrate on the growth of Azotobacter {strain A) in
sterilized soil
Treatment
(nitrate
in 100
gm. OL
dry soil).
Nirmber of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
I
Mgm,
0
0
25
25
100
100
200
200
18, 500
18, 500
18, 500
18, 500
18, 500
18, 500
18, 500
18, 500
I, 400, 000
1, 050, 000
5, 600, 000
4, goo, 000
2, 900, 000
2, 600, 000
I, 100, 000
950, 000
Per cent.
'• 100
} 427
} 223
} 84
/ 975. 000
\ I, 100, 000
f 5, 000, 000
\ 3, 900, 000
/ 3. 95°. 000
\ 4, 100, 000
/ 875, 000
I 915) 000
Per cent.
\
2. . .
> 100
■2
"I
A
I 430
e. ..
} 388
6
7
I Q^
8
> 86
That the nitrate radical and not the combined metal was the causal
agent in the increase in the number of Azotobacter was indicated from
the results of the next test. Here ammonium nitrate was used.
It will be seen from the data of this experiment that ammonium nitrate
caused an increase in the number of Azotobacter when present in small
amounts. However, the increase in the presence of ammonium nitrate
was less marked than when equal quantities of the other nitrates were
used. Since the experiments with ammonium nitrate were not made at
the same time as the preceding experiments (discussed on pp. 189-190), it
is possible that conditions varied sufficiently to account for the less pro-
nounced results. When 200 mgm. of nitrate were present in 100 gm. of
Jan. 28, 1918
Nitrogen- A ssiviilating Bacteria
191
soil the number of Azotobacter showed a decrease. Apparently ammo-
nium nitrate is more toxic than potassium, sodium, and calcium nitrate.
However, the main point at issue seems fairly well established — namely,
that the increase in the number of Azotobacter is caused by the nitrate
radical and not by the combined metal.
Table V.
-Influence of potassiuin tiitrate on the growth of Azotobacter [strain B) in
sterilized soil
Treat-
ment
(nitrate
in
100 gm.
of dry
soil).
Number of organisms in i gm. of dry soil.
Culture
No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
After 3 weeks.
Rela-
tive.
I. . .
2
Mgm.
0
0
10
10
25
25
SO
50
100
100
150
150
200
200
300
300
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
12, 600
235. 000
Per cent.
\ 100
|l> 510
|2, 436
}i,34o
|l,320
} 851
} 373
1 °
f 112, 500
\ no, 500
f 2, 100, 000
\ 2, 250, 000
1 I, 575-000
\ 1,950,000
1 3, 250, 000
\ 4, 900, 000
r 4, 000, 000
\ 3, 500, 000
r 2, 000, 000
\ 2, 100, 000
r 800, 000
I 750, 000
f 0
\
Per cent.
\ 100
J
}i, 9SO
}i>58i
I3.65S
}i,838
} 695
\ 0
J
/ 116,000
l^ 117,000
r 875, oco
\ 1, 260, 000
r 1, 700, 000
1 1,325,000
/ 3, 525. 000
\ 2, 960, 000
r 2, 500, 000
\ 2, 900, 000
r 1, 500, 000
\ 2, 000, 000
r 650, 000
\ 700, 000
( I
Per cl.
\ 100
f 9^^
!-l,300
y, 783
12,317
|i, 502
} 580
1 °
3- • •
4. . .
5-- •
6. . .
7.. .
8.. .
9. . .
10. . .
11 . . .
12. . .
13. . .
14. . .
15.. .
16...
3, 750, 000
3, 300, 000
5, 750, 000
5, 700, 000
3, 100, 000
3, 200, 000
3, 200, 000
3, 000, 000
2, ZOO, 000
I, 900, 000
875, 000
880, 000
0
0
Table VI. — Influence of sodium nitrate on the growth of Azotobacter {strain B) in
sterilized soil
Treat-
ment
(nitrate
in
100 gm.
of dry
soil).
Number of organisms in i gm. of dry soil.
Cultu
No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
After 3 weeks.
Rela-
tive.
1. . .
2. . .
3-- •
A-
5-- •
6..
7--
8..
9. .
10. .
11. .
12. .
13- •
14. .
15- •
16..
Mgm.
0
0
10
10
25
25
50
50
100
100
• 150
• 150
200
200
300
. 300
15, 600
15, 600
15, 600
15, 600
15, 600
15, 600
15,600
15,600
15, 600
15, 600
15, 600
15, 600
15,600
15,600
15,600
15, 600
158,000
149, 000
I, 250, 000
990, 000
I, 765, 000
I, 825, 000
1, 875, 000
,2, 250, 000
2, 200, 000
I, 950, 000
165, 000
170, 000
0
0
0
0
Per cent.
1
>■ 100
} 727
}l. 165
}i,338
}i,35o
I 108
1 °
0
f 110,500
\ 126, 000
1 1,750,000
I 1,350,000
{ 6, 600, 000
1 5> 300, 000
r 2, 025, 000
1 3, 040, 000
r 2,775,000
\ 3, 200, 000
/ 530, 000
\ 785, 000
/ °
1 0
/ °
I 0
Per cent.
\ 100
(■I. 310
(-5. 029
|-2, 141
1-2,525
\ 556
} °
} »
/ 112,500
\ 115,000
r 5, 000, 000
\ 6, 600, 000
r 9, 150, 000
1 7, 150, 000
(15,950,000
\i4, 600,000
{ 5, Soo, 000
1 5, 250,000
1 3, 100, 000
\ 2, 750,000
f 0
I
{ I
Per cl.
> 100
} 5,097
} 7. 161
}i3, 423
} 4,860
} 2, 573
1 °
} »
27807°— IS-
192
Journal of Agricultural Research
Vol. XII. No. 4
Table VII.
-Influence of calcium nitrate on the growth of Azotobacter {strain B) in
sterilized soil
Treat-
ment
(nitrate
in
100 gm.
of dry
soil).
Number of organisms in i gm. of dry soil.
Culture
No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
After 3 weeks.
Rela-
tive.
I
2
3
4
5
6
7
8
9
10
II
Mgm.
0
0
10
10
25
25
50
50
100
100
150
150
200
200
300
300
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
22, 000
90s, 000
860, 000
23, 200, 000
19, 600, 000
17, 200, 000
19, 600, 000
11,800, 000
14, 000, 000
7, 500, 000
II, 000, 000
2, 550, 000
3, 500, ooo
107, 500
87, 500
0
0
Per cent.
> 100
|2, 423
|-2, 084
|l, 461
}i>o53
} 342
} "
} °
1 1,475,000
\ I, 460, 000
("28, 000, 000
136, 000, 000
(52, 000, 000
l43> 500, 000
("22, 500, 000
\20, 000, 000
112, 000, 000
{ 5,300,000
\ 6, 500, 000
f 2, 750, 000
I 3.225,000
{ 0
Per cent.
\ 100
U iSl
13.255
li,448
1 818
\ 402
1 203
1 °
f I, 130, 000
I 1,157,500
/34, 050, 000
(34, 600, 000
I29, 750, 000
\22, 250, 000
(■30, 400, 000
\29, 850, 000
r2i, 750, 000
\i8, 950, 000
\ 4, 800, 000
\
Per ct.
> 100
|3, 002
y, 273
|2, 633
}i, 780
• 420
1 ^^
f °
13
14
15
16
f 130, 000
\ 120, 000
{ c
A glance at the figures of Tables V, VI^ and VII shows that the small-
est concentration of nitrate used produced a much more marked relative
increase in numbers with strain B than it did with strain A. On the
other hand, the greater resistance of this strain to the higher nitrate
concentrations is clearly evident. In the potassium- and calcium-
nitrate cultures the organisms were present in an active state where
the nitrate was added in amounts equivalent to 200 mgm. of nitrate
in 100 gm. of soil. However, this same concentration of sodium nitrate
prevented the development of the Azotobacter. The first five concen-
trations of all three nitrates caused a very large increase in the number
of Azotobacter when compared with control cultures where no nitrate
was added. In one instance an enormous increase was noted after three
weeks' incubation in the presence of 50 mgm. of nitrate as sodium nitrate.
This increase far excelled that noted with other concentrations of the
same salt. The writer can offer no conjecture as to this occurrence.
Similar results were obtained by the writer in 19 14 (25) with a strain
of Azotobacter isolated -from a silt loam soil at the Pennsylvania Experi-
ment Station. It was found that soil and liquid cultures containing
small amounts of potassium, sodium, and calcium nitrates caused an
increase in the number of Azotobacter in pure culture compared with
control cultures containing no nitrate. An increasing concentration of
the nitrates continued favorable to the growth of the organism up to a
certain limit, but higher concentrations retarded its growth. Finally
a nitrate concentration was attained at which Azotobacter growth
altogether ceased.
Jan. 2S, i9i8 Nitrogen- Assimilating Bacteria 193
The results of the study of nitrates and their influence on Azotobacter
in sterilized soil show very clearly that small amounts of nitrate cause
a great increase in the number of Azotobacter cells. Higher concentra-
tions are not so favorable to the growth of the organisms, and the highest
concentrations studied prevented the development of the Azotobacter
in sterilized soil.
From a study of the results of these experiments, it seems that the
increase in number of Azotobacter in the presence of small amounts
of nitrate is a direct result of nuclear stimulation. Later studies to
be cited (pp. 205-208) show that nitrates exerted considerable influence
on the internal structure of the Azotobacter cell. It appears reasonable
to expect that the nitrate affected the nuclear structure in such a manner
that an increase in cell multiplication resulted. It seems probable that
the action of nitrate as a simple nutrient would be shown by a slower
increase in cell multiplication.
INFLUENCE OF NITRATES ON THE FIXATION OF NITROGEN BY .\ZOTOBACTER
It has been shown in the preceding paragraphs that the presence of
small quantities of nitrate in sterilized soil bring about a large increase
in the number of Azotobacter. This increase was noted in the case
of both strains of Azotobacter. It would be of interest to know whether
the increase in bacterial numbers was accompanied by a corresponding
increase in the amount of nitrogen assimilated.
The results secured by a few investigators indicate that in the presence
of combined nitrogen as nitrates the nonsymbiotic nitrogen-fixing
organisms will not fix atmospheric nitrogen. Stoklasa {44, p. 492-50J)
studied the influence of Azotobacter on sodium nitrate in aerobic and
anaerobic liquid cultures. He found only a small gain in organic nitro-
gen and from these results he concluded that in the presence of nitrates
Azotobacter could not assimilate atmospheric nitrogen. It has been
shown by Hanzawa {20) that in a hquid culture containing 12 mgm,
of nitrate (from potassium nitrate) in 100 c. c. of medium, a mixed
culture of Azotobacter fixed 5.25 mgm. of nitrogen. Under the same
conditions with 60 mgm. of nitrate present in 100 c. c. of medium he
found but 5.35 mgm. of nitrogen fixed. He concluded that nitrates
remained, as far as small quantities were concerned, almost without
influence on the amount of atmospheric nitrogen fixed by Azotobacter.
Some studies have been carried on with respect to the influence of ni-
trates on the nonsymbiotic anaerobic nitrogen-assimilating organism,
Clostridium spp. Bredemann (9) showed that ammonium nitrate in
solution caused a decrease in the amount of nitrogen fixed by species
of Clostridium. Pringsheim (40) grew cultures of C. atnericamim in
solutions containing potassium nitrate. He found that in the presence
of available energy the organism fixed some nitrogen when nitrate was
194
Journal of Agricultural Research
Vol. XII. No. 4
present but to a less extent than did control cultures containing no
nitrate.
From these results it appears that nitrates do not stimulate the nitro-
gen-assimilation of the nonsymbiotic nitrogen-fixing bacteria.
Inasmuch as nitrates in small amounts caused such an increase in
the number of Azotobacter in sterilized soil, it was thought advisable
to determine just what influence these salts exert on nitrogen fixation
by Azotobacter. Accordingly, experiments were carried out with
Azotobacter on agar films, in soil cultures and in solution.
Agar-film cultures. — In this work both strains of the Azotobacter
were used. One hundred c. c. of mannit agar were placed in liter
Erlenmeyer flasks and nitrates of potassium, sodium and calcium
added in varying quantities. The flasks and contents were sterilized
at ID pounds' pressure for 25 minutes, cooled, and inoculated with 10
c. c. of a suspension of the organism in sterile distilled water. The
flasks were incubated at 28° C. for three weeks. The weight of both
inoculated and uninoculated flasks was maintained throughout the
experiment by the addition of sterile distilled water. At the end of the
incubation period total nitrogen analyses were made. Because of the
high nitrate content dilute sulphuric-salycilic acid was added slowly
and carefully to prevent loss of nitrogen by the evolution of gaseous
oxids of nitrogen. The acid was allowed to react for a few days before
continuing the total nitrogen determination. The results of the experi-
ments are presented in Tables VIII and IX.
Table VIII. — Influence of nitrates on the fixation of nitrogen by Azotobacter {strain A)
on agar films
Cul-
ture
No.
3
4
5
6
7
8
9
10
13
14
IS
Treatment (nitrate in looc. c. of medium).
O
o
o
50 mgm. of NO3 potassium nitrate .
do
100 mgm. of NO3 potassium nitrate
. . . .do
50 mgm. of NO3 sodium nitrate. . .
....do
100 mgm. of NO3 sodium nitrate.
do
50 mgm. of NO3 calcium nitrate. .
....do
100 mgm. of NO3 calcium nitrate .
....do
Nitrogen contained in loo c. c. of medium.
Inoculated.
Found. Average.
Mgm.
13.00
12. 70
12. 60
18.50
18. 40
27. 60
27-75
18.65
18.30
27. CO
27.65
13-75
13-70
18.80
19-15
Mgm.
12. 80
} 18. 45
I 27. 70
} 18. 50
} 27-35
} 13- 75
} 18. 95
Uninoculated.
Found. Average.
Mgm.
4.0
4.0
4- I
7. 00
7. 20
16.80
15-70
7-50
7-30
15.00
15. 20
8.00
8. 50
14. 50
14-30
Mgm.
4-05
I 7.10
I 16.25
}'-
} 8-25
14.40
Nitrogen
fixed.
Mgm.
8
II
II
II
12
5
4
Jan. 2S, 1918
Nitrogen- A ssimilaiing Bacteria
195
Table IX. — Influence of nitrates on the fixation of nitrogen by Azotobacter {strain B)
on agar films
Cul-
ture
No.
Treatment (nitrate in looc.c. of medium).
Nitrogen contained in 100 c. c. of medium.
Inoculated.
Found. Average
Uninoculated.
Found. Average.
Nitrogen
fixed.
75 mgm. of NO3 as potassixxm ni-
trate
...do
150 mgm. of NO3 as potassium ni-
trate
....do
75 mgm. of NO3 as sodium nitrate
do
150 mgm. of NO3 as sodium nitrate
do
75 mgm. of NO3 as calcium nitrate
do
150 mgm. of NO3 as calcium nitrate
do
15-50
15-70
15.60
25. 20
25.40
36.40
36.90
25. 60
25. 70
37.60
37. 20
20. 10
19. 60
32.80
33-30
Mgm.
15.60
25-30
Mgm,.
}36
I 3i
Mgm,.
6. 40
} 13-85
I 23.60
I 13-00
} 25. 80
I 12.3s
} 24. 8s
Mgm.
9. 20
11-45
13-05
12.65
II. 60
7-50
8. 20
A glance at the results (Tables VIII and IX) shows that an increase in
nitrogen fixation occurred where potassium and sodium nitrates were
present, whereas a marked decrease in the total nitrogen content was
observed where calcium nitrate was used. Whether the calcium itself
is detrimental to an increase in organic nitrogen or whether it is the com-
bination of calcium with nitrate can not be stated. It is significant,
however, that this decrease in fixation of nitrogen was noted throughout
all the experiments where calcium nitrate was employed. It is very-
evident that calcium nitrate exerts some detrimental effect on the nitro-
gen assimilating properties of the organism.
There seems to be but a slight difference in the nitrogen-fixing ability
of the two strains studied. In the absence of nitrates the amount fixed
varies but little. Also in the presence of potassium and sodium nitrates
the relative increase in amount of nitrogen fixed remains about the
same. Calcium nitrate offers an exception where it is employed. The
detrimental effect seems to be more marked in the case of strain A than
with strain B. Strain A under normal conditions fixed slightly less nitro-
gen than strain B, so it may be possible that this strain is weaker.
The formation of pigment by the Azotobacter in the presence of the
nitrates is of interest. Strain A normally produced no pigment by the
end of three weeks' incubation. But when grown on the agar films in
the presence of nitrate a most marked pigment production appeared.
This pigment was especially noticeable in the presence of the calcium
196 Journal of Agricultural Research voi. xii, no. 4
salt. Since strain B normally produces a good pigment, the influence
of nitrate on this strain was not very marked. The relation of nitrates
to pigment formation will be taken up later (pp. 203-205).
From the results of the experiments with agar films containing various
amounts of nitrate, it seems apparent that potassium and sodium nitrates
in amounts of 50 and 100 mgm. of nitrate in 100 c. c. of medium cause a
small increase in the amount of nitrogen fixed. However, this increase
in fixation is not at all parallel with the increase in number of Azoto-
bacter caused by nitrates in sterilized soil.
It may be concluded that an increase in the number of Azotobacter in
sterilized soil as a result of nitrate stimulation does not mean a corre-
sponding increase in nitrogen fixation on agar films.
Soil cultures. — ^The conditions obtaining in these experiments were
strictly comparable with those heretofore cited dealing with the influence
of nitrates on Azotobacter in sterilized soil (pp. 187-193).
The fixation of nitrogen was studied in pure culture in sterilized soil
and in unsterilized soil. One hundred and fifty gm. of soil (dry weight)
were weighed into i -liter Erlenmeyer flasks, nitrates were added in vary-
ing amounts from 10 to 200 mgm., and i per cent of mannit was also
added. Triplicate flasks were prepared for each amount of nitrate studied.
The moisture content was raised to approximately 18 per cent and the
flasks allowed to remain at room temperature for one day. The con-
tents were then thoroughly mixed and a fine crumb structure produced.
The flasks for the experiments with pure cultures in sterilized soil were
immediately sterilized at 15 pounds' pressure for three hours. After
cooling, two of each set were inoculated with 5 c. c. of a suspension of
Azotobacter (strain A) in sterile distilled water. The remaining flask of
each set was not inoculated, but was incubated at 28° C. with the inocu-
lated flasks. The moisture lost by evaporation was replaced from time
to time by the addition of sterile distilled water. At the end of the incu-
bation period the soil was removed and spread out in thin layers and
allowed to dry. It was then thoroughly ground in a porcelain-ball mill
for one hour. At the end of this time all of the soil passed through a
loo-mesh sieve.
Soil cultures used in the study of the effect of nitrates on nitrogen
fixation in unsterilized soil were prepared in a similar manner, except
that the flasks were not sterilized. Previous to incubation a small
inoculum of Azotobacter (strain A) was added to insure the presence of
the nitrogen-fixing organism in the soil cultures. The proper moisture
content was maintained in the same manner as in the case of the pure
cultures in sterilized soil and the incubation period was the same for both.
The results are given in Tables X, XI, XII, and XIII.
Jan. 28, 1918
Nitrogen- A ssimilating Bacteria
197
Table X.
-Influence of sodium nitrate on the fixation of nitrogen by Azotobacter in
sterilized soil
Treatment
(nitrate in
100 gm. of
dry soil).
Total nitrogen in 100 gm. of dry soil.
Nitrogen
Culture No.
Inoculated.
Uninoculated.
fixed in
100 gni. of
Found.
Average.
Found.
Average.
dry soil.
I
Mgm.
0
0
0
0
0
0
10
10
10
10
10
10
50
50
SO
50
50
50
150
150
150
150
150
150
Mgm.
135-0
134. 0
132. 0
133-0
137-0
Mg7n.
\ 133- 7
[ 135- 0
136.6
[ 137- 0
[ 149- 0
[ 149- 2
\ 162. 3
[ 162. 5
Mgm.
132.0
131- 5
131-0
Mgm.
\ 131- 5
Mgm.
I
2- 7
I
2
I
2
1
3-5
:::::::::::
7,
137-0
136.0
137-0
136-5
137-5
137-0
149.0
149.0
149. 0
148.5
149-5
149-5
163.0
162. 0
162. 0
162. 5
163.0
162. 0
i 134- 0
133- 5
[ 134-0
[ 133- 7
■3,
2. 9
7,
a.
1
A.
f:::::::::::
[ 3-3
A
[..:
c
f 140. 0
137- 0
i 138- 5
f
} 138-5
e
10. 5
e
6
]
6
\ 10.7
6
[
7
7
f 152-0
\ 150- 0
I 152. 5
151-5
10.8
7
8
1
8
(:::::::::::
> II. 0
8
1
J
Table XI. — Influence of sodium nitrate on the fixation of nitrogen by Azotobacter in
unsterilized soil
Treat-
ment (ni-
trate in
100 gm. ol
dry soil).
Total nitrogen in 100 gm. of dry soil.
Nitrogen
Culture No.
Inoculated.
Uninoculated.
fixed in
100 gm. of
Foimd.
Average.
Found.
Average.
dry soil.
I
Mgm.
0
0
0
0
0
0
10
10
10
10
10
10
50
50
SO
SO
50
50
150
150
150
150
150
150
Mgm,.
132.0
135-0
13s- 0
132.0
134.0
134.0
137-5
138.8
138.8
137-5
137- 5
138.0
150.0
151. 0
150.0
149.0
149-5
150. 5
169. 0
167. 0
168. 0
167-5
168.0
168.5
Mgrit.
134. 0
• 133- 3
• 137- 8
137- 7
• 150- 3
■ 149- 7
168. 0
168. 0
Mgm^.
[ 130- 0
133- 5
I 132- 0
Mgm.
1 131-8
Mgm.
X
2. 2
I
2
1
2
1 '•'
•2
f 134- 0
\ ^33- 0
I ^33- 0
\ ^33- 3
5
4-5
•2
]
J
[ 4-4
A . . .
C
f 140. 0
{ 140. 5
[ 142. 0
i 140. 8
e
9-5
c
6 . . . .
1
6
,
8.9
6 . . .
7
f 148. 0
{ 154- 0
I 153- 5
1 151- 8
7
16. 2
7
8
1
8
\ 16.2
8
J
198
Journal of Agricultural Research
Vol. XII. No. 4
TABtE XII.
-Influence of calcium nitrate on the fiocation of nitrogen by Azotobacter in
sterilized soil
Treat-
ment (ni-
trate in
100 gm. of
dry soil).
Total nitrogen in 100 gm. of dry soil.
Nitrogen
Culture No.
Inoculated.
Uninoculated.
fixed in
ICO gm. of
Fotmd.
Average.
Found.
Average.
dry soil.
1
Mgm.
0
0
0
0
0
0
10
10
10
10
10
10
50
50
50
50
50
50
200
200
200
200
200
200
Mgm.
133- 0
133-6
133-3
133-0
134.2
133-4
137.0
137-0
136.5
136.5
137.0
137-5
148. 0
148.5
149.0
148.5
149.0
148. 0
173.0
173-0
174.0
173-5
173-0
174.0
Mgm.
^33- 3
133-5
136. 8
137. 0
148. 5
148. 5
173- 7
173- 5
Mgm..
131.0
131. 0
132. 0
Mgm..
1 ^3^-3
Mgm.
I
2. 0
I
2
1
2
\ 2. 2
2
J
•J
135- 0
134.0
135- 0
\ 134- 7
2
2. I
2
1
1 ''
A
C
140. 5
141. 0
140. 5
i 140. 7
e
7.8
e
6
1 ■•
6
6
7
163. 0
164. 0
164. 5
163.8
7
9.9
7
8
1
8
1 9-7
8
Table XIII.
-Influence of calcium nitrate on the fixation of nitrogen by Azotobacter in
unsterilized soil
Treat- _
ment (ni-
trate in
100 gm. of
dry soil).
Total nitrogen in 100 gm. of dry soil.
Nitrogen
Culture No.
Inoculated.
Uninoculated.
fixed in
100 gm. of
Found.
Average.
Found.
Average.
dry soil.
I
Mgm..
0
0
0
0
0
0
10
10
10
10
10
10
50
50
50
50
50
50
200
200
200
200
200
200
Mgm.
134-5
136.0
136-5
135-0
135-5
135-5
138.5
138.0
139.0
138.0
137.5
138.5
151- 5
152. 0
151. 0
150. 0
151-5
151. 0
177.0
178.0
176. 0
176.5
177.0
178. 0
Mgm.
\ 135- 7
[ 135- 3
138. 5
138. 0
[ 151.5
150. 8
177.0
[ 177- 2
Mgm,.
[ 134. 0
1 ^33- 5
I 132. 0
Algm.
\ ^33- 2
Mgm.
I
2-5
1
2
1
2 . . . . ...
2.1
2
J
•2
[ 133- 5
133-0
^33- 0
\ ^33- 2
?
5- 3
■t
4
1
A
\ 4-8
A
J
C
[ 140. 5
141. 0
L 141. 5
i 141.0
C
10. 5
c
6
1
6
9-8
6
7
f 164. 0
165. 0
[ 164. 0
[ 164. 3
7
12. 7
7
8
1
8
[ 12.9
8.
Jan. 28, i9i8 Nitrogen- Assimilating Bacteria 199
It will be seen at a glance that a greater relative increase in nitrogen
fixation in the presence of nitrates occurred in the soil cultures than on
the agar films. But in the latter instance the amount of nitrogen as-
similated in the absence of mistakes is far in excess of that assimilated
in the soil cultures under similar conditions. The amount of nitrogen
fixed in the soil cultures is surprisingly low, but as relative increases or
decreases are desired this does not materially influence the results.
The influence of sodium nitrate on the fixation of nitrogen by pure
cultures of Azotobacter in sterilized and unsterilized soil is brought out
very clearly in the figures of Tables X and XL In both cases, where
no nitrate was added, an equal fixation of nitrogen occurred. Where 10
mgm. of nitrate were added to 100 gm. of soil, slightly more nitrogen was
assimilated in the unsterilized soil than in sterilized. The reverse seemed
to be true when 50 mgm. of nitrate were added. But in the presence of
150 mgm. of nitrate, the fixation by the pure culture in sterilized soil
did not increase materially in comparison with that which occurred in
the 50 mgm. of nitrate concentration. Evidently the maximum fixation
under these conditions had been reached. The gain in the unsterilized
soil at the highest concentration of nitrate studied almost doubled the
amount fixed in the pure culture. It appears evident that the presence
of sodium nitrate causes a greater fixation of nitrogen in unsterilized soil
than it does under similar conditions in sterilized soil inoculated with
Azotobacter.
In the case of calcium nitrate, somewhat comparable results were ob-
tained. The fixation where no nitrate was added was equivalent to
that obtained in the controls for the sodium nitrate. Where nitrate
was added in amounts equal to 10 mgm. of nitrate in 100 gm. of soil, an
increased fixation was obtained in the unsterilized soil, but practically
no increase occured in the pure culture in sterilized soil. Fifty mgm.
of nitrate in 100 gm. of soil produced an increase in fixation. In the
highest concentration of calcium nitrate the difference in nitrogen fixed
between the pure culture in sterilized soil and unsterilized soil was not so
great as in the case where sodium nitrate was used.
In the sterilized soil where the two nitrates were present in equal
amounts it can be seen that more fixation took place in the presence of
sodium nitrate. The diilerence is not marked, but it exists neverthe-
less. It will be remembered that calcium nitrate had a detrimental
effect on nitrogen fixation by Azotobacter on agar films. However, in
soil cultures this same nitrate stimulated Azotobacter to an increased
assimilation of nitrogen. This difference is not suprising as it has been
shown repeatedly that bacterial activities in soil and in artificial cultures
are not always comparable.
From the results of the experiments performed with reference to the
influence of nitrates in soil on the fixation of nitrogen therein, it appears
200 Journal of Agricultural Research voi. xii. no. 4
evident that in pure cultures both sodium and calcium nitrates in the
amounts studied produced an increase in the amount of nitrogen fixed.
The sodium salt stimulated this process to a slightly greater extent than
did the calcium salt. In unsterilized soil nitrates exerted the same
action but to a more marked extent. The amount of nitrogen fixed
under these conditions was generally in excess of that fixed under similar
conditions in sterilized soil inoculated with a pure culture of Azotobacter.
Such large relative increases in total nitrogen in the soil in the presence
of nitrates would not normally take place under field conditions for here
no accumulations of nitrate occur in quantities sufficiently large enough to
influence this process.
Summing up all the experiments performed in relation to the influence
of nitrates on the fixation of atmospheric nitrogen by Azotobacter, it
appears that the increase in total nitrogen in the presence of these salts
is by no means comparable to the increase in the number of organisms
in sterilized soil under the same conditions. An increase in the number of
Azotobacter does not mean a parallel increase in the amount of nitrogen
fixed.
INFLUENCE OF AZOTOBACTER ON NITRATES IN SOLUTION
Attention has been thus far directed toward the influence exerted
by nitrates on the growth and nitrogen-assimilating power of Azotobacter.
The following points are now to be considered: Do the nitrogen-fixing
bacteria reduce nitrates to nitrites and ammonia ? Is there an increase
or decrease in the amount of organic nitrogen as a result of the presence
of nitrate in the medium?
Beijerinck and Van Delden (5) found that Azotobacter ckroococcum
reduced nitrate directly to ammonia. Stoklasa (44, p. 4^2-503)
studied the changes in a nutrient solution containing 0.2 per cent
of sodium nitrate inoculated with Azotobacter. He found under an-
aerobic conditions that the nitrate was largely reduced to nitrite and
ammonia and that a very small amount of organic nitrogen was formed.
Under aerobic conditions there v/as more nitrite formed than under
anaerobic conditions and very little ammonia or oganic nitrogen. He
concluded, therefore, that Azotobacter did not fix atmospheric nitrogen
in the presence of nitrates.
The following experiments were performed in an endeavor to answer
the questions raised in the initial paragraph of this section. To Erlen-
meyer flasks of 500-c. c. capacity, containing loo-c. c. portions of mannit
solution, sodium and ammonium nitrates were added in amounts equiva-
lent to 150 mgm. of nitrate in 100 c. c. of the solution. Nine flasks were
prepared for each nitrate and the same number for the controls containing
no nitrate. The flasks and contents were sterilized at 10 pounds pres-
sure for 30 minutes. After cooling, six of each set were inoculated, three
Jan. 28, 1918
Nitrogen-A ssimilating Bacteria
201
with strain A and three with strain B, and the remaining three were left
uninoculated to serve as controls. The flasks were incubated at 28° C.
for 21 days. The total weight was maintained throughout the experi-
ment by the addition of sterile distilled water from time to time. At the
end of three weeks the contents of each set of triplicate flasks were poured
together and 50-c. c. samples drawn for analysis. Nitrate ammonia and
total nitrogen were determined as given under "Methods." The results
are shown in Tables XIV, XV, and XVI.
Table ^1\ .^Influence of Azotobacter on nitrates in solution, giving the quantity of
nitrate lost
Treatment (ni-
trate in 100
c. c. of me-
dium).
Nitrate in loo c. c. of medium.
Cul-
Strain A.
Strain B.
ture
No.
Inoculated.
Uninoculated.
Nitrate
lost.
Inoculated.
Uninoculated.
Found.
Aver-
age.
Found.
Aver-
age.
Found.
Aver-
age.
Found.
Aver-
age.
lost.
Mgm..
0.00
.00
80.9
80.6
100.3
102. 1
Mgm.
> 0. 00
}8o.7S
>IOI.2
Mgm,.
f 0.00
\ .00
/iSO-4
liSi-3
/i49- 6
Uso. 0
Mgm.
?• 0.00
}i50.8
}i49-8
Mgm.
0.00
0—70.03
a— 48.60
Mgm.
{ 0.00
\ .00
f los- 6
\i05- 2
/1311
I130- 7
Mgm.
> 0.00
}to5.4
|i30.9
Mgtn.
( 0.00
\ .00
/150.4
USl-3
/149. 6
Uso.o
Mgm.
!• 0.00
}i5o.8
}i49.8
Mgm.
0.00
10-18
isogm.of NO3
as sodium ni-
trate
10-18
do
i>— 45-40
19-27
iSomgm.ofNOs
as ammonium
nitrate
do
«— 18. 90
1 strong NO2 r,eaction.
6 Medium NO2 reaction.
cSlightNO: reaction.
Table XV. — Influence of Azotobacter on nitrates in solution, giving the quantity of
ammonia produced
Treatment (ni-
trate in 100
c. c. of me-
dium).
Nitrogen as ammonia in 100 c. c. of medium.
Cul-
Strain A.
Strain B.
ture
No.
Inoculated.
Uninoculated.
Am-
monia
pro-
duced.
Inoculated.
Uninoculated.
Am-
Found.
Aver-
age.
Found.
Aver-
age.
Found.
Aver-
age.
Found.
Aver-
age.
pro-
duced.
1-9
1-9
10-18
Mgm.
0. 20
. 10
2. CO
1.80
13-90
13- 9S
Mgm,.
} 0. IS
} 1-90
}l3-97
Mgm.
/ 0.00
\ .00
|-.IO
I . 20
/ 13-90
I 13-90
Mgm,-
\ 0.00
|i3-90
Mgm.
0. IS
1. 85
-■07
Mgm.
f 0". 00
\ '.20
f 2.20
\ 2-40
/ 13-80
\ 13-80
Mgm.
> 0. 10
I 2.30
}i3-8o
Mgm.
f 0.00
\ .00
f-.io
\ .20
/ 13-90
I 13-90
Mgm.
}■ 0. 00
|i3-90
Mgm.
0. 10
ISomgm.ofNOs
as sodium ni-
trate
. do
2.25
19-27
19-27
ISomgm.ofNOs
asammonium
nitrate
do
. 10
202
Journal of Agriculiiiral Research
Vol. XII, No. 4
Table XVI. — Influence of Azoiobacter on nitrates in solution, giving the quantity of
nitrogen fixed
Treatment (ni-
trate in 100
c. c. of me-
dium).
Total nitrogen in 100 c. c. of medium.
Cul-
Strain A.
Strain B.
ture
No.
Inoculated.
Uninoculated.
Nitrogen
fixed.
Inoculated.
Uninoculated.
Nitro-
Found.
Aver-
age.
Fotmd.
Aver-
age.
Found.
Aver-
age.
Found.
Aver-
age.
gen
fixed.
1-9
Mgm.
5.00
5- 00
22.50
22.60
47-00
46.90
Mgm.
I S-oo
I22.SS
I46. 95
Mgm..
f 2.60
I 2.70
f 14.00
I 14.20
/ 43-20
I 42-90
Mgm,.
} 2-65
|i4. 10
}43-05
Mgm.
2.3s
8-45
3- 90
Mg-rtt.
/ 5-00
I S-io
r 28.00
1 27.80
/ 48. 10
\ 48. 20
Mgm,.
} 5- 05
I27.90
}48- IS
Mgm.
f 2. 60
\ 2.70
f 14.00
\ 14. 20
/ 43- 20
I 42-90
Mgm-
} 2.65
|i4. 10
|43-0S
Mgm.
2.40
lO-iS
10-18
isomgm.of NOa
as sodium ni-
trate
.. do
13-80
19-27
19-27
isomgm.of NO3
asammonium
nitrate
.....do
5-10
Table XIV showing the effect on the total nitrate content will be
discussed first. Strain A differed widely from strain B in its ability to
reduce nitrates. It will be noted that strain A reduces nitrate more
readily than strain B in the presence of both sodium and ammonium
nitrate. In order to determine the nature of the reduction of the nitrates,
qualitative and quantitative tests were made. The reduction of nitrates
by Azotobacter takes place with the formation of nitrites as shown in
Table XIV. Strain A effected a strong reduction of nitrate to nitrite
with both sodium and ammonium nitrate. Strain B also reduced nitrate
to nitrite, but to a lesser degree than did strain A.
An inspection of the data in Table XV indicates that the reduction of
nitrates ceased with the formation of nitrite, since no appreciable amounts
of ammonia were produced by either strain of Azotobacter.
In regard to the fixation of atmospheric nitrogen by these strains of
Azotobacter, it was found that nitrogen was assimilated both in the
presence and absence of nitrate. In the presence of nitrate there was a
large increase in the total organic nitrogen. Sodium nitrate stimulated
both strains, although strain B fixed the larger amount. Similar results
were obtained when the fixation of nitrogen on agar films was studied.
In the presence of ammonium nitrate the amount of nitrogen fixed was
considerably decreased, but the amount fixed was in excess of the control
cultures containing no nitrate. It seems evident that sodium and
ammonium nitrate in the amounts studied did not prevent the fixation
of atmospheric nitrogen. In fact, the presence of these salts seemed to
stimulate the process.
Under aerobic conditions both strains of Azotobacter studied caused a
reduction in the total amount of nitrate present in the solution. This
reduction may be accounted for in two ways: (i) The reduction of
nitrate to nitrite and (2) the assimilation of nitrate by the organisms.
Practically no ammonia was formed under the conditions of these experi-
ments. These results agree with those of Stoklasa. However, in con-
Jan. 28, i9i8 Nitrogen-Assimilating BacteHa 203
trast to the work of Stoklasa, both strains of Azotobacter assimilated
more atmospheric nitrogen in the presence of nitrates in solution than
in the absence of these salts.
INFLUENCE OF NITRATES ON THE PRODUCTION OP PIGMENT BY AZOTOBACTER
It has already been noted in the experiments dealing with the effect
of nitrates on the fixation of atmospheric nitrogen on agar films that
nitrates favor pigment production. This was true in the case of both
strains of the Azotobacter.
Moreover, it has been observed by other investigators that Azotobacter
when grown in the presence of nitrate will produce a darker pigment
than when grown in its absence. Beijerinck (4, p. 575) states that
Azotobacter in pure culture will form a dark-brown pigment in the
presence of glucose and a small amount of nitrate. Sackett {43) found
that nitrate caused an increase in pigment production by Azotobacter.
In media without the nitrate the pigment formation was materially
decreased and in some cases practically eliminated. He also noted that
the amount of nitrate present has a direct influence on the intensity of
the pigment formation. He found that when sodium nitrate was added
to a suitable medium to give a content of 0.0, o.oi, 0.03, 0.05,0.08, o.i,
0.3, and 0.5 per cent, with glucose used as the source of energy, the
organisms produced pigment. Streak inoculations were made, and
after 14 days' incubation he found that the maximum of color was
obtained at 0.05 to 0.08 per cent and that greater concentrations did not
increase the intensity of the brown-black pigment. From his results it
is evident that sodium nitrate caused an increase in pigment formation
by azotobacter.
In order to determine the possible effect of potassium, sodium, and cal-
cium nitrate on pigment formation with strains A and B, the following
experiment was performed.
Under normal conditions on mannit agar free from nitrate strain A
produced little or no pigment even after three weeks' growth. At the
end of this time dirty-yellow streaks occurred throughout the growth,
but no brown pigment was produced. However, with strain B at the
end of two or three weeks a decided brown to brown-black pigment was
produced in the absence of nitrate.
Agar slope cultures containing increasing amounts of potassium, so-
dium, and calcium nitrate, as indicated in Table XVII, were prepared.
These were inoculated with both strains of Azotobacter and incubated
at 28° C. for 10 days. Daily observations were made for first evidences
of pigment formation. In some of the cultures of strain A growing on
media containing calcium nitrate this pigmentation was observed as
early as 48 hours subsequent to inoculation. The following day pig-
mentation developed in strain B. The cultures on the potassium and
sodium-nitrate media began to show evidence of pigmentation in four to
six days. The final results, obtained after 10 days' incubation, are found
in Table XVII.
204
Journal of Agricultural Research
VoK XII, No. 4
1
0
0
,i4
'a
S H Ci HO
a 2 fl • • 2-^ •
^pg Qpq
ii
PQ
til
0
0
"S
IH
H
,Q
1
g
2
W
'd
30000 0^0
T^ • • • -"S ■
^
<u
(33
^
«
d
2
d
d
■+-»
K*
s
+3
M
a
'a
a
2
1
0
g ; c
S 0 0 0 0 0 o-Q
1
E
0)
c3
s
«
a
I
(U
en
5h
.a
'3
2
PI
c
2
«
0 S ^ OJ
2
^>3§
fi
pq
fi
S
I^
d
<j
2
d
0
a
■3
+->
^
i-i
42
'3
d
^
0
m
t:)
■^ c
Sc
c
0
ai ^-d -e'O.S'a'd 1
OT
0 OJ
P
V
2
^
• d
'r3
2
• : ^
s
3
^ p :
1
■^-^ 0 S c
c
> c
)
tu •^■y "Td .s "0 13 "Td i
(S
^
s
i! ^'o
•^
. 00 ij^ 00000
n
e Mwiooiooo
« H M C^ CO
2 °
,2
1
g.a
-2
^
^
O o
a
1 q; .
"3 v« O
H C) CO ''t lO^O t^CO
55^
1
Jan. 28. i9i8 Nitrogen-Assimiloting BacteHa 205
A general idea may be gained from Table XVII concerning the relative
increase in pigment formation in the presence of the nitrates. A study
of the table gives a fair idea of the relative differences in pigment pro-
duction.
Very interesting results were obtained with strain A. It will be seen
from Table XVII that no pigment was produced in the control culture
after 10 days, while in the presence of nitrates pigmentation was noted.
The intensity of the pigment varied with the increase of nitrate up to 150
mgm. Beyond 150 mgm. there was no increase. Potassium and sodium
nitrate did not exert such a decided influence on pigment production as
calcium nitrate. The latter salt produced an intense dark -brown to
brownish-black pigment.
In the case of strain B the influence of nitrate was not so pronounced
since this strain normally produced considerable pigment in the absence
of nitrates. Potassium and sodium nitrate caused a slight increase in
pigment formation. Here, again, the calcium salt brought about most
pronounced increase. However, the relative increase in pigment forma-
tion in strain B was not so pronounced as in strain A.
Where the nitrate was present, a much more spreading growth was
obtained. A heavy bacterial growth accumulated at the base of the slope
except in the two cultures in which the highest concentrations were used.
In the latter instances the accumulation was less than those in cultures
growing en media containing no nitrate. Although the original inocula-
tion could not be made absolutely uniform, so far as number of organ-
isms was concerned; nevertheless it was evident that on those slopes
containing 10, 25, 50, and 100 mgm. of nitrate in 100 c. c. of the medium
a much more abundant growth was obtained than on those slopes free
from nitrate. Here, again, it is seen, in a rough, comparative way, that
the smaller amounts of nitrates caused an increase in the number of
Azotobacter.
The results of this work on pigment production are quite in accord with
those of Sackett. Potassium, sodium, and especially calcium, nitrates
in varying amounts increase pigment formation by Azotobacter with an
increase in nitrate concentration. This effect is especially marked in
strain A, which under normal conditions does not produce any pigment.
INFLUENCE OF NITRATES ON THE FORMATION OF VGLrUTIN BODIES IN AZOTOBACTER
The presence of volutin bodies, or metachromatic granules, in Azoto-
bacter has been shown by Bonazzi (7). These substances, according to
Meyer {34, p. 238), are reserve food materials other than fat droplets,
glycogen, and similar substances reacting with iodin stain which occur
in the cytoplasm of the cells of various bacteria. With Millon's reagent
they give no reaction. He believes that these bodies are composed of
nucleic-acid compounds, but are not nuclear proteids.
2o6
Journal of Agricultural Research
Vol. XII, No. 4
In connection with the foregoing investigations concerning the influ-
ence of nitrates on pigment formation by Azotobacter, it was thought
that some results of cytological interest might be obtained in regard to
the effect of varying amounts of nitrates on the volutin bodies.
Slope cultures of mannit agar were prepared containing the different
nitrates as indicated in Table XVIII. These slopes were inoculated
with both strains of Azotobacter and incubated at 28° C. for 10 days.
At the end of this time each culture was stained and examined micro-
scopically. The following method was used for demonstrating the
presence of the volutin bodies. The organisms to be examined were
air dried on a glass slide and then fixed in the flame of a Bunsen burner.
The preparation was then flooded with a i to 10 aqueous solution of
methylene blue (Merck's) prepared by adding 10 c. c. of a saturated
aqueous solution of methylene blue to 90 c. c. of distilled water. The
stain was washed off after five minutes with a i per cent solution of
sulphuric acid and immediately rinsed in distilled water. The prepa-
ration was dried and examined with the oil-immersion objective. The
volutin bodies appeared within the cytoplasm as very dark blue dots,
the outline of the cell wall was a lighter blue, while the cell net work
was stained a very light blue.
Guignard's stain ^ was also used to demonstrate the presence of the
volutin bodies. Fresh smears on a glass slide were fixed over 10 per
cent osmic acid for three minutes. The preparation was then air-dried
and fixed to the slide by rapidly passing the latter a few times through
a Bunsen burner. The preparation was covered with the stain which
was allowed to react for five minutes. The stain was then washed off
with distilled water, dried, and examined with the oil-immersion objec-
tive. The outline of the cell as well as the net work wdthin-was stained
light purple. The granules within the cytoplasm were a reddish purple.
The results are given in Table XVIII.
Tabls XVIII. — Influence of nitrates on the formation of volutin bodies in Azotobacter
in 10 days
Treat-
ment
(nitrate
in 100
c. c. of
me-
dium).
Strain A.
Strain B.
Culture
No.
Potassium
nitrate.
Sodium
nitrate.
Calcium
nitrate.
Potassium
nitrate.
Sodium
nitrate.
Calcium
nitrate.
Mgm.
0
10
25
so
100
ISO
200
300
Present." —
do."
do.o
do a
Doubtful....
Present". . . .
do.a
do 6 .
Doubtful
Present". . . .
Present"
do."
Doubtful....
Present"
do.6
Present."
Do."
do."
do."
Do."
Doubtful
do."
do."
Do."
do 0
do. 6
do. ft
do. 6
Do."
6..
do "
do b
do"
.... do.b
do.fc
D0.6
... do.h ...
do.6
do.b
do.6
do.6
Do."
8
do.ft
do.6
do."
do.6
do.b
D0.6
"Representing an approximate average of two volutin bodies per cell.
b Representing an approximate average of four volutin bodies per cell.
* Guignard'sstain. Fifty c. c. of 2 per cent fuchsin in i per cent acetic acid; 40 c, c. of 0.2 per cent methyt
green in i per cent acetic acid; i c. c. of glacial-acetic acid. Distilled water was used in making the i per
cent acetic-acid solution.
Jan. 28, 1918
Nitrogen- A ssimilating Bacteria
207
It will be seen that all three nitrates exerted considerable influence
on the formation of volutin bodies. Not only was the number of bodies
increased, but also the size. The relative increase in size of the granules
was much more marked than was the numerical increase. In Azoto-
bacter cells grown on mannit agar containing no nitrate the number of
volutin bodies in each cell averaged about two ; in the presence of nitrate
four to five volutin granules were found. The greatest increase in num-
ber, as well as size, occurred where the nitrate concentration was highest.
With both strains sodium nitrate apparently caused the greatest increase.
This was true in the lower as well as in the higher concentrations. The
volutin bodies in strain B seemed to respond to the presence of nitrates
more noticeably than did those of strain A, especially in the presence of
potassium nitrate. It is evident that nitrates of potassium, sodium,
and calcium cause an increase in the number and size of volutin bodies
in Azotobacter cells.
Do these salts tend to hasten the appearance of these bodies, or do
they at first retard their development? The following experiment was
carried out in an endeavor to determine this point. Only sodium nitrate
was used, since this particular salt proved most beneficial to the forma-
tion of volutin bodies. Agar slopes were prepared containing the different
amounts of nitrate as indicated in Table XIX. The cultures were incu-
bated at 28° C. and examined daily for the presence of volutin bodies.
The methylene blue — i per cent sulphuric acid — method of staining was
employed. The results of the experiment are given in Table XIX.
Table XIX. — Influence of sodium nitrate on the rate of formation of volutin bodies. in
Azotobacter
Time.
Nitrate in 100 c.
c. of mediiun.
Strain A.
strain B.
0 Mgm.
25 Mgm.
100 Mgm.
300 Mgm.
0 Mgm.
2S Mgm.
100 Mgm.
300
Mgm.
Day.
Absent...
Present".
...do."....
...do."....
Absent.. .
Present".
...do."....
...do."....
Doubtful.
Present".
...do."....
...do.b...
Doubtful.
...do.b...
...do.b...
...do.b...
Absent.. .
...do
Absent...
Present".
Doubtful.
...do.b...
...do.b...
...do.b...
Dbtful.
Do.b.
3
4
Present".
...do."....
...do.b...
...do.b...
Dob.
Do.b.
" Representing an approximate average of two volutin bodies per cell.
b Representing an approximate average of four volutin bodies per cell.
A study of Table XIX shows that it is rather doubtful whether the
nitrate present tended to hasten the appearance of the volutin bodies.
No convincing evidence has been presented for or against this statement.
No granules were seen in the first day's growth of strain A, although
the next day they were present in all four cultures. In strain B more
convincing proof is furnished that the sodium nitrate hastened the
appearance of these reserve food substances. The volutin bodies were
not present in the control and lowest nitrate concentration cultures
the first day, but they were very noticeable in the culture containing
the highest concentration of nitrate and doubtful in the remaining one.
On the second day volutin bodies were present in all cultures grown on
27807°— 18 5
208
Journal of Agricultural Research
Vol. XII. No. 4
nitrate media, while the control culture was still free from them. The
third day showed the presence of volutin bodies in all four cultures.
Strain B offers the better proof that sodium nitrate tends to hasten the
appearance of volutin bodies in the cells of Azotobacter. Further
experiments were not made in an endeavor to determine what influence
nitrates might have on the cytology of the Azotobacter cell. The brief
studies reported here were made in connection with the pigment forma-
tion experiments, but do not bear any particular relation to them. The
increase in number and size of volutin bodies may bear some relation to
the increased amount of nitrogen fixed or assimilated by Azotobacter in
the presence of nitrates.
INIflvUENCE OF NITRATES ON BACILLUS RADICICOLA
INFLUENCE OP NITRATES ON THE GROWTH AND REPRODUCTION OF BACILLUS RADICICOLA
IN STERILIZED SOIL
One hundred and fifty gm. (dry weight) of the soil were weighed
into 500-c. c. E^rlenmeyer flasks and the nitrates added as indicated
in Tables XX-XXII. Duplicate cultures for each amount of nitrate
were prepared. One per cent of mannit (in 5 c. c. of distilled water)
was also added. The flasks were kept at room temperature for one
day and the contents then thoroughly mixed. The flasks were steri-
lized at 15 pounds' pressure for three hours. Upon cooling they were
inoculated with 5 c. c. of a suspension of Bacillus radicicola in sterile
distilled water. The number of bacteria in the inoculum was deter-
mined. The moisture content was then approximately 18 to 20 per
cent. The flasks were incubated at 28° to 30° C. and mannit-agar
plates poured at the end of one and two weeks. The results of these
experiments are given in Tables XX, XXI, and XXII, in which each
figure represents an average of duplicate plates.
Table XX. — Influence of potassium nitrate on Bacillus radicicola in sterilized soil
a Contamination,
Jan. 38, 1918
Nitrogen- A ssimilating Bacteria
209
Table XXI. — Influence of sodium nitrate on Bacillus radicicola in sterilized soil
Treat-
ment (ni-
trate in
100 gm. of
dry soil).
Nimiber of organisms in i gm. of dry soil.
Ctdture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
I
Mgm.
0
0
10
10
25
2C
15, 500
15. 500
i5» 500
i5> 500
15. 500
TC coo
I, 500, 000
1, 250, 000
2, 560, 000
3, 000, 000
6, 150, 000
5, 375, 000
4, 850, 000
5, 570, 000
2, 000, 000
1, 850, 000
1, 060, 000
835, 000
760, 000
725, OOO
250, 000
365, 000
Per cent,
f 100
> 201
1 418
} 378
> 140
} 69
} ^^
> 22
r 6, 750, 000
I 5. 950, 000
( 10, 000, 000
\ 12, 500, 000
/ 14, 650, 000
1 15. 700, 000
/ («)
\ 8, 500, 000
f I, 520, 000
\ I, 650, 000
f 850, 000
\ 940, 000
r 500, 000
\ 620, 000
f 150,000
\ 210,000
Per cent.
2
\ 100
5
4.
I 177
r
6
( 240
7
50 15. 500
50 15, 500
100 T c. con
8
} 134
0
1
10
100
150
150
200
200
300
300
-J7 J-
i5> 500
15, 500
15. 500
i5» 500
15. 500
15. 500
■ 15, 500
( ^5
II
12
I ^"^
XT.
14
1 8.8
le
16
1 2.8
a Contamination.
Table XXII. — Influence 0/ calcium nitrate on Bacillus radicicola in sterilized soil
Treat-
ment (ni-
trate in
100 gm.of
dry soil).
Number of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
I
Mgm.
0
0
10
10
25
25
50
50
100
100
' 150
150
200
200
300
300
p p p p p p p p p p p p p p p p
OOOOOOOOOOOOOOOO
960, 000
850, 000
3, 650, 000
3,940, 000
5, 500, 000
6, 700, 000
4, 000, 000
3, 500, 000
1, 200, 000
2, 050, 000
865, 000
I, 050, 000
375)000
260, 000
35. 000
47, 000
Per cent.
> 100
I 419
1 674
1 414
} 180
> 106
I 35
I 4-5
f 4, 675, 000
Per cent.
1
2
> 100
7
A
I 5,450^000 ■/ ^^4
c
6
I 14,700,000 / '^
/ 9.350,000 \\
I 8,670,000 / ^95
f 1,500,000 \
I 1,750,000 ij >55
/ 765,000,;!
\ 800 000 i ' '
7
8
0
10
12
13
14
/ 350, 000
\ 300, 000
f 25,000
\ 40, 000
} '■»
15
16
I .70
An inspection of all three tables reveals two marked differences from
the results obtained in similar woric with Azotobacter. First, it will be
noted that nitrates do not appear to exert such a marked stimulating
effect with B. radicicola as with Azotobacter. The numerical increase
due to the presence of the nitrate is clearly shown in the percentage
columns. Second, it will be noted that B. radicicola does not seem to be
so sensitive to higher concentrations of nitrates as does Azotobacter.
In all instances at concentrations equivalent to 300 mgm. of nitrate in
2IO
Journal of Agricultural Research
Vol. XII, No. 4
lOO gm. of soil the legume organisms were still alive, although present
in numbers far below those of the control cultures. In all Azotobacter
cultures no organisms survived this concentration.
No one nitrate produced an excessive stimulation in comparison with
the others. The calcium salt present as 150 mgm. of nitrate in 100 gm.
of soil at the end of the first week gave the greatest stimulation for con-
centrations of that amount. However, at the end of the second week
this concentration had caused a marked decrease in the number of or-
ganisms. In the case of all three nitrates the concentration representing
25 mgm. of nitrate in 100 gm. of soil produced the greatest stimulation.
This resulting stimulation also held true throughout the second vv^eek.
The decrease in number below those of the control cultures, due to in-
creasing concentrations of nitrate, began first in the presence of potassium
nitrate at 100 mgm. of nitrate per 100 gm. of soil, then with sodium
nitrate at 150 mgm., and lastly with calcium nitrate at 200 mgm. But
the number of organisms present in the soil cultures containing sodium
nitrate in amounts equivalent to 100 mgm. and calcium nitrate at 100
mgm. at the end of the second week was below those of the control
cultures.
It therefore appears from these results that small amounts of potassium,
sodium, and calcium nitrate stimulate the reproductive activity of B.
radicicola. Concentrations of nitrates greater than those amounts which
produced maximum stimulation cause a decrease in the number of or-
ganisms. The highest concentration of nitrate studied did not entirely
prevent the growth of the bacteria, but it reduced the number of organ-
isms far below those contained in control cultures where no nitrates
were added.
Ammonium nitrate was also employed. The soil cultures were pre-
pared as already described and inoculated with B. radicicola. The cul-
tures were incubated at 28° to 30° C. and counts were made at the end
of one and two weeks' time. The results of the study with ammonium
nitrate are given in Table XXIII.
Table XXIII. — Influence of ammonium nitrate on Bacillus radicicola in sterilized soil
Treat-
ment (ni-
trate in
100 gm. of
dry soil).
Number of organisms in i gm. of dry soil.
Culture No.
At begin-
ning.
After I week.
Relative.
After 2 weeks.
Relative.
I
Mgm.
0
0
25
25
100
100
200
200
S p p p p p p p
OOOOOOOO
850, 000
765, 000
2, 500, 000
3, 050, 000
I, 350, 000
I, 050, 000
700, 000
655, 000
Per cent.
> 100
} 343
} 148
} 84
f I, 365, 000
\ I, 400, 000
r 5, 060, 000
\ 4, 320, 000
r 1, 030, 000
\ 950, 000
r 63 s, 000
\ 605, 000
Per cent.
1 TOO
2
5
} 338
4
c
} -
6
7
} «
8
jaM. 28. i9i8 Nitrogen-Assimilating Bacteria 211
From the results as a whole it appears that it is the nitrate radical
and not the combined salt which causes the increase in the number of
B. radicicola when small amounts of nitrates are present. A stimulation
occurred, resulting in an increase in number which is quite comparable
to that obtained with potassium, sodium, and calcium nitrates. The
highest concentration of ammonium nitrate used did not appear to have
such an inhibiting effect as did the corresponding concentrations of the
three other salts.
Throughout the work with B. radicicola in sterilized soil compara-
tively low numbers of these organisms were found. Whether or not
this depression was due to toxic substances formed as a result of steriliza-
tion can not be stated. If this decrease in numbers as a result of the
presence of toxic substances is true, it is very evident that the detrimental
effect had not become materially lessened at the end of the incubation
period. However, in any event the validity of the outcome is not im-
paired, since comparative and not absolute data are of importance and
since in all probability the same conditions obtained throughout the
cultures.
It seems certain from the results of these studies on the effect of
potassium, sodium, calcium, and ammonium nitrates on the growth of B.
radicicola in sterilized soil that small amounts of nitrate stimulate the
growth of the organisms. It is also shown that B. radicicola is much
more resistant than Azotobacter to higher concentrations of potassium,
sodium, calcium, and ammonium nitrates.
INFLUENCE OP BACILLUS RADICICOLA ON NITRATES IN SOLUTION
The series of soil culture experiments just discussed served to give an
idea concerning the effect of nitrates on the legume organism. It was
found that in small amounts nitrates stimulated the bacteria to increased
reproduction. But no study was made as to the effect of Bacillus radicicola
on the nitrate. Does the organism break up the nitrate, reducing it to
nitrite or ammonia ? Does it cause any loss in nitrate when grown in a
solution containing that salt? Beijerinck {2, p. J62) as a result of
physiological experiments with B. radicicola, states that the organism does
not reduce nitrate. Prucha {41) also states that B. radicicola does not
reduce nitrates. However, Zipfel (49) found that B. radicicola will reduce
nitrates to nitrites but not to ammonia.
The following experiments, somewhat similar to those already cited in
relation to Azotobacter, were carried out in an endeavor to answer these
questions.
To twenty 500-c. c. Erienmeyer flasks containing 200 c. c. of mannit
solution, potassium, sodium, calcium, and ammonium nitrates were
added as indicated in Tables XXIV, XXV, and XXVI. Quadruplicate
flasks were prepared for each concentration of nitrate and for the control
cultures without nitrate. The flasks and contents were sterilized at
212
Journal of Agricultural Research
Vol. XII. No. 4
lo pounds' pressure for 30 minutes. After cooling, two of each set of four
flasks were inoculated with 5 c. c. of a suspension of B. radicicola in
sterile distilled water. The remaining two flasks of each set (uninocu-
lated) served as controls. The flasks were incubated at 28° C. for 2 1 days.
The total weight of the flasks was maintained throughout the incuba-
tion period by the addition of sterile distilled water from time to time.
At the expiration of the period of incubation the nitrate, ammonia, and
total nitrogen contents were determined as given under "Methods used
in experiments." The contents of the duplicate inoculated flasks were
poured together and 50 c. c. samples drawn for analysis. The same
procedure was followed in the case of the uninoculated flasks. The
results are given in Tables XXIV, XXV, and XXVI.
Table XXIV. — Influence of Bacillus radicicola on nitrates in solution giving the quantity
of nitrate lost
Treatment (nitrate in loo c. c. of medium).
Nitrate in loo c. c. of medium.
Cul-
ture
No.
Uninoculated.
Inoculated.
Nitrate
lost.
Found.
Average.
Fotmd.
Average.
None
Mgm.
0. 00
. 00
151- 4
151. 0
148.8
148.8
154.8
155-6
151- 4
151. 6
Mgm.
> 0. 00
}i5i- 2
|i48. 8
}i55-2
}i5i-5
Mgm..
( 0. 00'
\ . 00
/117-O
\li7-0
[ii4-4
\li4. 0
/ 76.6
I 76.8
fi42. 6
\i42. 6
Mgm..
> 0. 00
>ii7. 0
V114. 2
} 76.7
>i42. 6
Mgm.
.. do
0. 00
3
4
5
150 mgm. of NO3 as potassium nitrate .
do
-34-2
150 mgm. of NO3 as sodium nitrate . .
. . do
-34-6
7
8
150 mgm. of NO3 as calcium nitrate .
do
-78.5
9
150 mgm . of NO3 as ammonium nitrate
do
-8.9
Table XXV. — Influe?tce of Bacillus radicicola on nitrates in solution giving the quantity
of nitrogen as ammonia fanned
Treatment (nitrate in loo c. c. of medium).
Nitrogen as ammonia in loo c. c. of medium.
Nitro-
Cul-
ture
No.
Uninoculated.
Inoculated.
gen as
ammo-
nia
Found.
Average.
Found.
Average.
formed.
None
Mgm.
0. 20
. 10
. 10
. 20
. 20
. 20
.40
•30
13.90
13-95
Mgm..
} 0. 15
} ■ ^^
> . 20
} -35
} 13-92
Mgm.
j 0. 10
\ . 20
J . 00
\ . 10
f . 20
I .30
/ -30
\ . 10
f 13.80
I 13- 85
Mgm.
} 0-15
} '°^
} '^^
> . 20
} 13- 82
Mgm.
...do
0. 00
3
4
5
6
150 mgm. of NO3 as potassium nitrate .
.... do
— . 10
150 mgm. of NO3 as sodium nitrate . .
do
+.05
7
8
150 mgm. of NO3 as calcium nitrate.
do
--IS
9
JO
150 mgm. of NO3 as ammonium nitrate
do
-f.io
Jan. 38, 1918
Nitrogen- A ssimilating Bacteria
213
Table XXVI. — Influence of Bacillus radicicola on nitrates in solution giving the quantity
of nitrogen fixed
Cul-
ture
No.
2
3
4
5
6
7
8
9
10
Treatment (nitrate in 100 c. c. of medium).
None
....do
150 mgm. of NO3 potassium nitrate . .
....do
150 mgm. of NO3 sodium nitrate. . . .
....do
150 mgm. of NO3 calcium nitrate . . .
....do
150 mgm. of NO3 ammonium nitrate .
....do
Total nitrogen in 100 c. c. of medium.
Uninoculated.
Inoculated.
Found. Average. Found. Average.
MgTti.
2. 40
2.50
18. 00
17.90
16.80
17. CO
14. 00
13.80
40.50
41. 20
Mgm.
45
13
40,
95
90
Mgm.
3-3°
3-5°
18. 70
19. 00
19.30
19. 20
14. 60
14. 70
41.30
41. 70
Mgm.
} 3- 40
} 18. 8s
} 19-25
} 14-65
\ 41- 50
Nitro-
gen
fixed.
Mgm.
0-95
.90
2-35
• 75
•65
The data in Table XXIV show that a rather large reduction in the
total nitrate content took place. This reduction varied rather markedly
among the four different nitrates studied. The greatest reduction oc-
curred where calcium nitrate was used. Potassium and sodium were
next in order; the loss was almost the same for both salts. Ammonium
nitrate was last with but a very small comparative reduction in total
nitrate.
The question arises as to whether the nitrate is reduced to nitrite,
ammonia, or elemental nitrogen or whether the reduction in amount is
due to a natural assimilation of the nitrate by the organisms. The first
possibility was precluded when qualitative tests for nitrites were made
and none found. Table XXV reveals the fact that no ammonia was
produced. Table XXVI shows no loss in total nitrogen. Therefore it
seems obvious that reduction in total amount of nitrate present is brought
about by the assimilation of those compounds by the organisms.
An inspection of Table XXVI, which gives the results of the total
nitrogen determinations, shows that a slight fixation of atmospheric
nitrogen took place. This fixation is entirely possible, as will be shown
later when the influence of nitrates on the fixation of nitrogen is taken
up. In the presence of potassium, sodium, and ammonium nitrates the
amount of nitrogen assimilated is somewhat decreased. But in the
case of sodium nitrate a large increase in the amount of total nitrogen
seems to have taken place. This is interesting in the light of results
to be presented later.
From the results of the work on the effect of B. radicicola on nitrates
it may be concluded that the organisms do not reduce the nitrates to
nitrite or ammonia or elemental nitrogen under aerobic conditions.
214 Journal of Agricultural Research voi. xii. No. 4
INFLUENCE OF NITRATES ON THE FIXATION OF ATMOSPHERIC NITROGEN BY BACILLUS
RADICXCOLA
The ability of B. radicicola to fix atmospheric nitrogen in the absence
of the host plant has been studied by numerous investigators. From
the results of their work it seems fairly probable that the legume organ-
ism can fix nitrogen to a slight extent when growing in a nonsymbiotic
state. Beijerinck (j) was one of the earliest to make a study of the
possible fixation of atmospheric nitrogen by B. radicicola under these
conditions. He found that a small quantity, 0.91 to 1.82 mgm. of nitro-
gen was fixed per 100 c. c. of the medium,* Prasmowski (jp, p. 55) and
Berthelot (6) concluded as a result of their experiments that when the
organism was grown outside the host plant the gain in nitrogen was
small. The greatest gain in nitrogen was found by Maze (52) who re-
ported an increase of 23.4 mgm. of nitrogen per 100 c. c. of the medium in
1 6 days. lyewis and Nicholson {30) found by incubating the cultures for
a considerable length of time that a large increase in fixation occurred.
Bottomley {8) found that a pure culture of B. radicicola fixed approxi-
mately I mgm. of nitrogen in 15 days. Fred (77) made a study of the
possible fixation of nitrogen by the legume organism and found that it
fixed approximately 1.2 mgm. of nitrogen in 100 c. c. of the medium.
He found that on agar films a greater fixation occurred than when the
organisms were grown in a liquid medium.
A few investigators, however, found that no increase»r in nitrogen
occurred when B. radicicola was grown outside the host plant. Frank
{16) states that in a nitrogen-free medium the legume organisms did not
fix enough nitrogen to be accurately measured. Immendorf (25) also
found no increase in nitrogen when pure cultures of B. radicicola were
grown in soil containing a nitrogen-free solution.
It will be seen that the majority of investigators, especially the more
recent ones, found that a slight amount of atmospheric nitrogen was
fixed or assimilated by B. radieicola when grown outside the host plant
and on a medium suitable for its development.
It has already been shown that nitrates cause an increase in the num-
ber of B. radicicola when grown in pure culture in sterilized soil. Does
such an increase in the number of organisms necessarily mean an in-
creased fixation of nitrogen? Three experiments using agar films were
carried out in order to determine this point. Erlenmeyer flasks of
I -liter capacity containing^ 100 c. c. of mannit agar were used. Before
the medium solidified, the nitrates were added in the proportions indi-
cated in Table XXVI I. Six flasks for each different quantity of nitrate
were prepared, except in one case, as shown in Experiment II. The
flasks were plugged with nonabsorbent cotton and sterilized at 10 pounds'
pressure for 30 minutes. After cooling, three of each set were inoculated
with 5 c. c. of a suspension of B. radicicola in sterile distilled water.
The organisms had been growing on mannit agar at 28° C. for six days.
The flasks in Experiments I and HI (Table XXVII) were incubated at
Jan. 2S, 1918
Nitrogen- A ssimilating Bacteria
21
28° C. for three weeks and those in Experiment II for two weeks. The
moisture lost by evaporation in both inoculated and uninoculated flasks
was replaced from time to time by the addition of sterile distilled water.
At the expiration of the incubation period the total nitrogen was deter-
mined as given under "Methods used in experiments." The results of
the experiments are given in Table XXVII.
An inspection of the data reveals the fact that B. radicicola in pure
culture fixed a small amount of nitrogen when growing in a nonsymbiotic
state with no nitrate present. In the presence of nitrates there was an
increased fixation. Although the increase in total nitrogen is small,
because of the number of determinations made, it may be considered
as positive. The potassium and sodium salts seemed to be more effective
than the calcium nitrate, with one exception (Table XXVII, Experiment
I). It will be remembered that the latter salt appeared to depress nitro-
gen fixation by Azotobacter and the two former somewhat to favor it (p.
194-195).
Table XXVII. — Influence of nitrates on the fixation of nitrogen by Bacillus radicicola,
giving the increase in nitrogen
EXPERIMENT I
Culture
No.
Treatment (nitrate in 100 c. c. of medium.)
None
do
do
75 mgm. of NO3 as sodium nitrate . . .
do
do
150 mgm. of NO3 as sodium nitrate . .
do
do
75 mgm. of NO3 as calcium nitrate . .
do
do
150 mgm. of NO^as calcium nitrate . .
do
do
Total nitrogen in 100 c. c. of medium.
Uninoculated.
Inoculated.
Foimd.
Average.
Found.
Average.
Mgm.
Mgm.
Mgm.
Mgm.
4-5
1 4.7
4.4
\ 4-45
4-6
\ 4. 60
4.4
I 4-5
8.7
f II. 9
8.7
i 8. 70
\ II. 8
["•75
8.6
I 11.6
12. 5
r 14.9
12.7
>I2. 60
14.6
I 14-7
f 12.3
[14- 70
8.8
8.9
[ 8.90
12.8
>I2. 40
9.0
[ 12. I
13-3
f 14-5
13-1
|i3- 20
^4-o
[14. 10
13.2
[ 13-8
1
Nitrogen
increase.
Mgm.
o. 15
3-03
2. 10
3- 30
o. 90
EXPERIMENT 11
None
do
75 mgm. of NO3 as sodium nitrate .
do
150 mgm. of NO3 as sodium nitrate. .
do
75 mgm. of NO3 as calcium nitrate .
do
150 mgm. of NO3 as calcium nitrate .
do
4.90
4.90
8. 70
8.50
13-30
13.00
II. IS
II. 10
14. 70
(a)
}5-07S
I 9- 50
}u.35
}ii.65
}i5-25
o- 175
0. 90
1. 20
o- 525
o- 550
a Lost by breakage during sterilization.
2l6
Journal of Agricultural Research
Vol. XII, No. 4
Table XXVII. — Influence of nitrates on the fixation of nitrogen by Bacillus radicicola,
giving the increase in nitrogen — Continued
EXPERIMENT III
Culture
No.
9
lO
II
12
13
14
15
16
17
18
19
20
Treatment (nitrate in 100 c. c. of medium):
None
....do
....do
75 mgm. of NO3 as potassium ni-
trate
....do
....do
150 mgm. of NO3 as potassium ni-
trate
. . . .do
....do
75 mgm. of NO3 as sodium nitrate
....do
....do
150 mgm. of NO3 as sodium nitrate
....do
....do
75 mgm. of NO3 as calcium nitrate
....do
....do
150 mgm. of N03as calcium nitrate
....do
....do
Total nitrogen in
100 c. c. of medium.
Uninoculated.
Inoculated.
Found.
Average.
Found.
Average.
Mgm.
Mgm.
Mgm.
Mgm.
5.10
5-50
5- 10
5-07
j 5- 40
5-50
5.00
I 5-45
9-35
f 10.85
9-50
9-37
< 10. 90
> 10. 90
9-25
I 10.95
14.50
f 15-65
14. 20
[14.28
15-30
[ 15-45
14- 15
I 15- 40
8.50
9-85
8.30
8.38
\ 9- 90
9-83
«• 3.'5
i 9-70
12.35
f 12.95
12. 40
[12.33
j 13- 10
[ 13-03
12. 20
I 13-05
8.95
f 985
9. 10
\ 9- °i
\ 9- 90
[ 9-93
9. 00
I 10.05
13.90
f 14-40
13.80
13.80
\ 14-50
[ 14- 42
13.70
I 14-35
Nitrogen
increase.
Mgm.
0. 43
1-53
1. 17
I- 45
o. 70
o. 92
o. 62
It has been shown that, when nitrates are added in varying quantities
to sterilized soil, the number of B. radicicola are increased. Provided the
the organism can fix a small amount of nitrogen in the absence of nitrate
nitrogen, is it not possible that this increase in nitrogen fixation may be
due merely to the increase in the number of cells ? It seems that this is
true according to the results in Table XXVII. It appears probable that
the increase in nitrogen fixed in the presence of nitrates is very likely
because of an increase in the number of bacterial cells and not to any
physiological change brought about in the organism itself.
There was a marked increase in bacterial growth on the media con-
taining the nitrate compared with the same media free from nitrate.
The growth on the latter medium exhibited a normal, whitish watery
appearance, characteristic of this organism. On the cultures containing
nitrates a much more profuse growth occurred. In many instances a
pinkish tint was observed. This pigment production was especially
marked in the case of the culture containing the sodium salt. After the
first experiment had been completed, it was thought that possibly this
pigmentation was due to an impurity in the culture. Therefore the two
remaining experiments were made, using a subculture from the original.
Jan. 28, i9i8 Nitrogen- Assimilating Bacteria 217
This culture was plated three times, each plating being made from a well-
isolated colony. The final subculture was taken from a similar well-
isolated colony. However, pigment formation in the presence of nitrate
persisted in the two final experiments, showing clearly that some reaction
took place between the nitrate and the organism grown on the medium.
It is of interest to note that the pigment formation in the presence of
nitrate was observed in later work where the influence of nitrates on
nodule formation was investigated. Prucha (41) found that on agar
slopes of medium containing 0.5 per cent of potassium or calcium
nitrate, the growth of B. radicicola became opaque and that an iridescent
tint was produced.
Although the results of these experiments may vary somewhat among
themselves, taken as a whole it appears evident that B. radicicola may
fix a small amount of atmospheric nitrogen when grown without the
host plant and on a suitable medium. The addition of various amounts
of nitrates as indicated increased somewhat the amount of nitrogen
assimilated by B. radicicola.
INFIvUENCE OP NITRATES ON THE PRODUCTION OP GUM BY BACILLUS RADICICOLA
Since nitrates, especially in smaller amounts, cause an increase in the
number of B. radicicola in pure culture, it was thought advisable to
determine what influence these salts have on the production of gum. In
culture media favorable to the growth of B. radicicola these bacteria will
produce a gelatinous substance which is readily precipitated with 95 per
cent alcohol or acetone. Upon the addition of either of these precipi-
tants a fairly heavy, water-white, frothy gelatinous mass is formed
which soon rises to the surface of the liquid. Upon standing, this mass
contracts somewhat, and portions of it may fall to the bottom of the
liquid from which it has been precipitated.
Chemical analyses, according to Buchanan (jo), have shown that this
gum is a carbohydrate. Upon hydrolysis with 2 per cent sulphuric acid
and 15 pounds' pressure for one hour, Fehling's solution is reduced,
showing the presence of a sugar. The gum does not give proteid reac-
tions with the Millon, biuret, or xanthoproteic tests. Hence, the gum
is not protein in character; nor does it contain nitrogen in the combined
form. Clearly it is a nonnitrogenous body.
In the experiment undertaken to determine whether nitrates influence
the fonnation of gum only relative dififerences are noted. No attempt
was made to obtain quantitative results.
Erlenmeyer flasks of i -liter capacity containing 200 c. c. of mannit
solution were used. The cultures contained various quantities of nitrate
as indicated in Table XXVIII. Triplicate flasks for each amount of
nitrate were prepared. In this table these three flasks are represented
as "a," "b," and "c." After sterilization at 15 pounds' pressure for 25
minutes the flasks were cooled and inoculated with 5 c. c. of a suspension
2l8
Journal of Agricultural Research
Vol. XII, No. 4
of B. radicicola in sterile distilled water. The cultures were then
incubated at room temperature (approximately 25° C.) for eight weeks.^
At the expiration of the incubation period the contents of the flasks
were poured into hydrometer cylinders of equal depth and diameter.
One hundred and fifty c. c. of acetone were added to precipitate the gum.
After careful shaking, the cylinders were covered with inverted petri
dishes to prevent evaporation. At the end of 24 hours the amount ol
gum precipitated was observed. The relative amounts are recorded in
Table XXVIII.
Table XXVIII. — Influence of nitrates on the production of gum by Bacillus radicicola
Cul-
ture
Treatment (nitrate in loo c. c. of medium).
Relative production of gtun — precipitated by
acetone.
No.
Flask a.
Flask b.
Flask c.
None
Large
Very large .
Large
Very large .
Large
Large
Very large .
Large
...do
...do
Large.
Very large.
Large.
Very large.
Large.
Do.
2
3
4
5
6
75 mgm. of NO3 as potassium nitrate .
450 mgm. of NO3 as potassium nitrate .
75 mgm. of NO3 as sodium nitrate ....
450 mgm. of NO3 as sodium nitrate. . .
75 mgm. of NO3 as calcium nitrate. . .
. . .do
...do
7
450 mgm. of NO3 as calcium nitrate. .
...do
C 0 nsider-
able.
Con sider-
able.
From the results it is certain that the nitrates, especially in the smaller
of the two concentrates, caused a very considerable increase in the amount
of gum produced by B. radicicola. The nitrates of potassium and sodium
caused a production of more gum than did the calcium salt. It will be
remembered that in the experiments v^rhere the influence of nitrates on
the fixation of atmospheric nitrogen by B. radicicola was studied, less
nitrogen was fixed in the presence of calcium nitrate than in the pres-
ence of the other two salts. Here again the greater stimulative action
of potassium and sodium nitrates is emphasized.
Buchanan in his investigations on the formation of gum by B. radici-
cola has found that varying amounts of potassium nitrate in a 2 per
cent saccharose solution or in a 2 per cent saccharose-clover-extract
solution caused a slight increase in growth and in gum production.
It seems probable that the increased gum production in the nitrate
cultures is caused not only by an increase in bacterial cells but also
perhaps by an increased stimulation in the formation of gum by the cells
themselves. The relative increase in the amount of gum produced in
the presence of nitrates seems to be greater than the actual increase in
number of organisms brought about by the stimulating effect of the
nitrate. In the latter instance this stimulating effect has been deter-
mined in soil cultures only and so a fair basis of comparison can not be
Jan. 28, 1918 Nitrogen- Assimilating Bacteria 219
found. Had the influence of nitrates on the growth and reproduction
of B. radicicola been determined in liquid culture, as well as in soil
cultures, then a comparison could have been made. Furthermore, the
divergencies in the time element, eight weeks' incubation in the liquid
cultures and three weeks in the soil cultures, are such as to render futile
any attempt at correlation. It may be that the large formation of gum
was due to the prolonged incubation. A shorter period of three weeks
undoubtedly would show a relatively smaller amount of gum produced
as a result of the presence of the nitrate.
However, from the results of the experiment it is certain that potas-
sium, sodium, and -calcium nitrate influence the formation of gum by B.
radicicola. The three nitrates studied caused a large increase in the
amount of gum obtained by precipitation with acetone. Calcium nitrate
caused the least stimulation, but the difference was not large.
INFLUENCE OF NITRATES ON NODULE FORMATION
The results of numerous investigations have shown that nitrates
retard and oftentimes entirely prevent the formation of nodules on
leguminous plants when grown in soil or liquid cultures. Vines (45),
working with the horse bean, found that the use of large amounts of
nitrate in the form of potassium nitrate retarded nodule formation. He
concluded that a decrease in the amount of nitrates meant an increase
in the number of nodules. Woods (48) found that leguminous plants
assimilated more nitrogen when they were grown in the absence of
potassium and calcium nitrate than when thus supplied. His results
seem to indicate that nodule development was retarded somewhat by
these salts. Similar results were obtained by Frank (16). Nobbe and
Richter (37) in 1 902 grew soybeans in a rich garden soil and found upon
inoculation that a gain of 74.7 per cent of nitrogen occurred. However,
upon the addition of nitrates this gain was considerably reduced, the extent
of the reduction corresponding to the amount of nitrate added. About
this same time, Wohltmann and Bergen6 (47) using many different
types of soils, found that nodules were not formed on the roots of peas
when ammonium nitrate was added. Creydt (12) found that sodium
nitrate retarded nodule formation on yellow lupines when these legumes
were grown in soil. Fred and Graul (j^") found that very small amounts
of nitrates did not appreciably decrease nodule formation, but that
larger amounts proved detrimental and finally prohibited entirely the
development of nodules.
The presence of nitrates in culture solutions has also been found to
reduce and oftentimes to inhibit the formation of nodules on leguminous
plants. Marchal (31) concluded that alkaline nitrates in concentrations
of I to 10,000 in liquid cultures prevented the formation of nodules on
peas. Flamand (13) grew vetch and beans in a nutrient solution and
220 Journal of Agricultural Research voi. xii. no. 4
found that nitrates in the following amounts prevented nodule forma-
tions: potassium nitrate, i to 10,000, sodium nitrate i to 2,000, ammo-
nium nitrate i to 2,000, and calcium nitrate i to 2,000 and i to 10.000.
Hiltner's {24) experiments showed that 5 mgm. of nitrogen as potassium
nitrate per liter prevented nodule formation on peas.
In contrast to these experiments Bassler (i) claimed that results
obtained from his work indicated that no effect was noticed by adding
nitrates to lupines growing in quartz sand.
The question naturally arises whether this condition is due to the
weakening of the organism brought about by growth in a nitrated
environment and to a consequent impairment or entire loss of its infect-
ing power, or whether it is caused by some interreaction between the
salt and the plant root, tending to increase the latter's resistance to
the attack of this particular organism.
INFLUENCE OP NITRATES ON THE INFECTING POWER OP BACILLUS RADICICOLA
Some investigations have been carried out to determine what effect
nitrates have on the legume organisms themselves. Wilson {46) showed
that although nitrates inhibit the formation of nodules, the organisms
capable of producing nodules did not lose their vitality or nodule-pro-
ducing power when grown in the presence of nitrates. The results of
Prucha {41) are in accord with those of Wilson. He found that B. radi-
cicola does not seem to lose its infecting power when grown on media
containing nitrate. During the course of his work he found that potas-
sium and sodium nitrates inhibited the formation of nodules. Further
evidence that the organisms appear to retain their vitality in the pres-
ence of nitrates is produced by the results of Maze (jj, p. ly-i"/), who
showed that legume bacteria were able to fix a slight amount of nitrogen
when grown in a soil extract solution containing i per cent sodium nitrate.
Herke (22) states that potassium nitrate favors the growth of nodule
bacteria.
However, other investigators state that nitrates have a harmful
effect on B. radicicola. Laurent (29, p. 134) found that legume organ-
isms failed to grow in a pea or lupine decoction containing nitrate in the
form of potassium and sodium salts in amounts equivalent to i to 500
and I to 1,000. Moore {35) in his experiments demonstrated that nitrates
at I to 10,000 were sufficient to prevent nodule formation. He states
that B. radicicola loses its power of infection when grown in a medium
containing nitrates.
From the results cited it can be seen that there is some disagreement
as to the influence exerted by nitrates on B. radilcicola. In some cases
the organism seems to retain its vitality in the presence of nitrates,
while in others it appears to have become weakened. It must be ad-
Jan. 28, 191S
Nitroge n-A ssimilating Bacteria
221
mitted, however, that the evidence seems to favor the former contention —
namel)?-, that nitrates do not cause the bacteria to lose their nodule-
producing power.
In order to determine whether or not nitrates weaken these organisms,
the following experiments were made: Slopes of mannit agar (in test
tubes) containing various amountsof sodium and calcium nitrates as indi-
cated in Table XXIX were inoculated with B. radicicola. These cul-
tures were incubated at 28° C. for one week, when transfers were made
to fresh nitrate media and incubated at 28° C. for another week. At
the expiration of this time, three 4-day-old seedlings of alfalfa were
inoculated with three drops of a suspension of the organism in 5 c. c.
of sterile distilled water. The same slope cultures were incubated at
28° C. and used for all subsequent inoculations in this experiment.
The inoculated seedlings were placed in the greenhouse under cheese-
cloth covering. The temperature here during the daytime averaged
approximately 30° C. The seedlings were examined for the first appear-
ance of nodules and in no case did they appear before 18 to 20 days.
A total count of nodules on all plants was made at the end of 45 days.
Three subsequent inoculations were made under the same conditions.
In this way organisms in contact with nitrate for varying lengths of
time could be used. The results of the inoculation experiments are
given in Table XXIX.
Table XXIX. — Influence of nitrates on the infecting power of Bacillus radicicola
Cul-
Treatment (nitrate in loo c. c. of medium).
Number of nodules after 45 days.
ture
No.
Inoculated Inoculated
June 3. June 15.
Inoculated
July II.
Inoculated
July 17.
I
None
7
5
IS
3
4
5
6
7
9
8
II
9
4
6
3
5
5
II
8
7
3
8
4
6
8
9
3
3
2
3
4
5
6
7
8
15 mgm. of NO3 as sodium nitrate . . .
37 mgm. of NO3 as sodium nitrate . . .
75 mgm. of NO3 as sodium nitrate . . .
150 mgm. of NO3 as sodium nitrate . .
225 mgm. of NO3 as sodium nitrate . .
450 mgm. of NO3 as sodium nitrate . .
None
3
5
7
4
5
6
2
7
4
9
5
7
5
6
0
0
4
6
5
8
5
4
8
4
4
5
\
8
9
10
II
12
13
14
15
16
15 mgm. of NO3 as calcium nitrate . .
37 mgm. of NO3 as calcium nitrate. . .
75 mgm. of NO3 as calcium nitrate . .
150 mgm. of NO3 as calcium nitrate. .
225 mgm. of NO3 as calcium nitrate .
450 mgm. of NO3 as calcium nitrate .
Uninoculated
do
From the results given in Table XXIX it is very evident that under
the conditions of the experiment the legume organisms did not lose
their power of producing nodules when grown on a medium containing
222 Journal of Agricultural Research voi. xii. no. 4
varying amounts of sodium and calcium nitrates. The numbers of
nodules produced on the alfalfa plants by organisms grown on media
containing nitrate do not vary widely from those on the plants inocu-
lated v»^ith organisms grown on media containing no nitrate. Not only
did the organisms fail to lose their nodule-producing power, but from
all appearances their infecting power did not seem to be materially
weakened.
It therefore seems apparent that an explanation for the failure of
nodules to develop on leguminous plants in the presence of nitrates is
not found in the theory that the organisms producing these nodules are
weakened when grown in the presence of nitrates.
INPLUBNCB OP NITRATES ON ALFALFA ROOTS AND NODULE FORMATION
The next step taken would naturally be in the direction of a study of
the influence of the nitrates on the plant roots themselves in order to
determine whether or not they thus are made more resistant to the
attack of these organisms.
A review of the literature shows that almost nothing has been done
touching this phase of the question. Wilson {46), studying the effect of
certain salts on nodule production, states that possibly the salt has some
effect on the root, making it resistant to the attack of the organism.
Maze (55, p. 15-17), who also concluded that nitrates did not cause B.
radicicola to lose its infecting power, says that nodules do not develop
on roots of legumes when nitrates are present because the carbohydrate
in the roots is changed into protein material by the absorption of the
nitrate.
Alfalfa seedlings (Medicago sativa) growing in soft agar containing
potassium, sodium, and calcium nitrates, as indicated in Table XXX,
were used in this study. Quadruplicate tubes were prepared for each
amount of nitrate. The higher concentrations of the nitrate were not
used, since it was found that germination and subsequent growth were
considerably impaired in the presence of such large amounts. The tubes
with the mannit agar and nitrate were sterilized at 1 5 pounds' pressure
for 30 minutes. These were cooled and sterilized alfalfa seeds planted
as given under "Methods used in experiments." The tubes were
then placed in the greenhouse under cheesecloth covering and the seeds
allowed to germinate. Germination took place in all instances, although
it was retarded somewhat by the presence of the nitrate. At the end of
five days the first tube of each set was inoculated with three drops of a
suspension of JB. radicicola in sterile distilled water. Subsequent inocu-
lations were made as indicated in Table XXX. These were made at
different intervals in order to allow the roots of the seedlings to remain
for a longer time in contact with the media. It was hoped that in this
way an idea might be obtained as to the time when the root first became
resistant. The results are given in Table XXX.
Jan. 2S, 1918
Nitrogen- A ssimilating Bacteria
223
Tabl,E XXX. — Influence of nitrates on alfalfa roots and nodule formation
Culture
Treatment (nitrate in 100 c. c. of medium).
Total number of nodules in each tube of
seedlings inoculated after —
No.
S days'
growth.
10 days'
growth.
18 days'
growth.
22 days'
growth.
I
None
3
0
0
0
0
0
0
0
0
(^)
(^)
I
0
0
0
3
I
0
0
0
0
I
0
0
0
0
3
0
0
0
5
2
0
0
0
3
0
0
0
0
I
0
0
0
0
4
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
10 mgm. of NO3 as potassium nitrate
25 mgm. of NO3 as potassium nitrate
50 mgm. of NO3 as potassium nitrate
100 mgm. of NO3 as potassium nitrate ....
150 mgm. of NO3 as potassium nitrate . . . .
10 mgm. of NO3 as sodium nitrate
25 mgm. of NO3 as sodium nitrate
50 mgm. of NO3 as sodium nitrate
100 mgm. of NO3 as sodium nitrate
150 mgm. of NO3 as sodium nitrate
10 mgm. of NO3 as calcium nitrate
25 mgm. of NO3 as calcium nitrate
50 mgm. of NO3 as calcium nitrate
100 mgm. of NO3 as calcium nitrate
150 mgm. of NO3 as calcium nitrate
0
0
0
0
0
0
0
0
0
0
0
0
0
a Fungus contamination.
t> Plant died after few days' growth.
It will be seen that in a few instances where a high concentration of
nitrates occurred the development of the seedlings subsequent to ger-
mination ceased. This condition may have been due to too high a con-
centration of soluble salts or to inferior seed. However, losses were not
sufficiently serious materially to affect the outcome of the experiment.
In all cases the seedlings grown in agar without nitrate produced
nodules when inoculated with B. radicicola. A few nodules appeared on
seedlings in cultures containing the loAvest concentration of all three
nitrates. The number of nodules in these cases was less than in the
control cultures. No nodules whatever developed in any concentration
above lo mgm. of nitrate in loo c. c. of medium. Under normal condi-
tions in test-tube cultures the nodules make their appearance at about
18 to 20 days after inoculation. The incubation of all cultures was
extended 40 days after inoculation in order to make certain that no
further nodule development would take place.
The nonproduction of nodules was not due to the failure of the inoc-
ulum. In all cases an excellent inoculum growth was obtained, espe-
cially in the case where nitrate was present in the medium. Indeed, it
was so luxuriant that in many cases the organism grew in considerable
quantity far down into the root zone. In many cases where nitrates
were present the same pink coloration was produced that was discussed
under another caption, on page 216.
As has been already stated, seedlings of varying ages were inoculated
for the reason that it was thought that a more or less prolonged contact
of the roots with the nitrate in the medium might serve as an index to
27807°— 18 6
224 Journal of Agricultural Research voi. xii, no. 4
the time in the growth of the seedling when permanent resistance to
"'attack of the organisms was established. The results obtained do not
■ seem to indicate that seedling roots 18 to 20 days' old are any more
resistant to the attack of the organisms than are those that are younger.
Evidently if any reaction takes place between the nitrate and the plant
root it occurs during the very early stages in the development of the
plant.
These results seem to point to the conclusions that the nonformation
of nodules in the presence of nitrates is due not to a weakening of the
vitality of the organism, but to some reaction between the plant root
and nitrate. One naturally queries whether the plant root cells are
made more resistant to the bacteria seeking to gain entrance there or
whether the naturally occurring carbohydrate food supply to be used
by the organisms after gaining entrance is diminished by its conversion
into protein material by the absorption of nitrate? Further studies
were not made in an endeavor to solve this question.
INFLUENCE OF NITRATES IN SOIL ON ALFALFA NODULES AND ON THE REFORMATION
OF NODULES
Additional studies were made with nitrates in relation to their influ-
ence on nodules already formed and on the redevelopment of nodules
once removed from alfalfa plants. The experiments were carried out
in an endeavor to determine whether nitrates prevented an increase in
the number of nodules on plants possessing nodules and whether they
prevented the reformation of nodules when removed. Experiments
revealed clearly that removed nodules were replaced by new ones pro-
vided the plant was carefully replaced in the soil (soil with normal low
nitrate content) and the proper amount of moisture maintained.
In these experiments i -gallon earthenware jars were used. These
were filled to within an inch of the top with 1,800 gm. of soil of a low
nitrate content. Different amounts of the nitrates to be studied were
added in the quantities indicated in Table XXXI. Concentrations of
100 and 300 mgm. of nitrate in 100 gm. of soil were also used, but the
transplanted alfalfa seedlings were unable to withstand such excessive
concentration, with the result that all died within a week or ten days
after transplanting. Quadruplicate pots were prepared for each con-
centration of nitrate. The nitrates in solution were mixed with the
proper amount of distilled water which, when added to the pots, brought
the moisture content to approximately 20 per cent. The pots were then
allowed to remain undisturbed for one day at room temperature to
allow the water containing the nitrate to become well diffused through-
out the soil mass. Into two pots of each set were transplanted young
alfalfa plants from which the nodules had been removed. The two
remaining pots contained transplanted alfalfa plants with the nodules
left on and their location noted. The plants used in this experiment
Jan. 28, igif
Nitrogen- A ssimilating Bacteria
225
were removed from 'an alfalfa plot, the soil of which was a sandy loam.
Previous to transplanting the roots of the young plants were carefully
washed in running water and immediately transplanted. The pots
were kept well watered, and after two or three days they were removed
to the greenhouse. Here they were watered when necessary. Trans-
plantations were made on the 27th of June and the experiment termi-
nated on the 3d of August. The plants were removed from the pots,
the roots carefully washed and examined for the presence of nodules.
The results are presented in Table XXXI.
Table XXXI. — Influence of nitrates in soil on alfalfa nodules and on the reformation of
nodules
Nitrate in loo gm. of dry soil.
Treatment of
nodules.
Number of
nodules —
Pot No.
At
begin-
ning.
At
end.
A I
None ....
Removed
do
0
0
4
3
0
0
4
8
0
0
4
I
0
0
4
5
0
0
2
6
0
0
4
0
0
4
2
3
A
A 2
.do
A ^
do
Not removed . . .
do
8
A4
do
7
0
(a)
3
5
0
0
B I
B 2
25 mgm. of NO3 as potassium nitrate ....
. . . .do
Removed
do
B 7.
. do
Not removed . . .
do
B 4
do
Ci
C2
50 mgm. of NO3 as potassium nitrate
do
Removed
do
C z
..do
Not removed . . .
do
2
ci
do
I
D I
D 2 .
25 mgm. of NO3 as sodium nitrate
do
Removed
do
0
0
D ^ .
....do
Not removed . . .
do
2
D 4
.do
I
r: ^
E I
E2
50 mgm. of NO3 as sodium nitrateft
do
Removed
do
0
(a)
I
E^
do
Not removed . . .
do
eI:. . .
do
3
0
0
Fi
F 2
25 mgm. of NO3 as calcium nitrate
do
Removed
do
F 7
do
Not removed . . .
do
2
F4
....do
3
0
0
G I
G 2
50 mgm. of NO3 as calcium nitrate
. . do
Removed
do
G T.
. . do
Not removed . . .
do
7.
G 4
.do
I
" Plants died.
It will be seen in the control pots, where no nitrate was present (ex-
cept the small amount normally present in the soil at the beginning of
the experiment), that if the nodules were removed, new ones formed.
The location of the nodules before their removal was noted, and the
new ones were found to occupy the same place. However, when nitrates
were added to the soil no new nodules were formed. This statement
holds true for both concentrations of all three salts in all experiments.
226 Journal of Agricultural Research voi. xii. no. 4
Some interesting results were obtained where the nodules were not
removed. In the control pots an increase in nodule formation took place.
It can not be stated definitely whether the new nodules appeared as
a result of inoculation from the soil or whether the organisms had already
gained entrance to the roots before the plants were removed from the
field soil previous to transplanting. Nevertheless, it is shown that the
number of nodules increased as compared with the number present at
the time of transplanting. But where nitrates were added a reduction
in number occurred rather regularly throughout all the pots. In two
instances the number remained constant, in 10 it was reduced, and in none
was it increased. The calcium salt appeared to effect the least reduction
in number of nodules. Conclusions concerning the comparative in-
fluence of the three salts in this regard can not be drawn because of the
small number of determinations made. It is sufficient to note that
nitrates present in amounts equal to 25 and 50 mgm. of nitrate in 100
gm. of soil did not permit an increase in number of nodules, but rather
caused a decrease.
The conclusions drawn from the experiments relative to the influence
of nitrates on nodule formation are: (a) the presence of nitrates is
detrimental to the formation of nodules by alfalfa; (b) the nonformation
of nodules is not due to a weakening of B. radicicola yvhen grov/n in the
presence of nitrates; (c) some reaction takes place between the nitrates
and the plant root, thus preventing nodule formation; (d) nitrates in the
soil prevent the re-formation of nodules once removed and also cause
a decrease in the number of those already present.
SUMMARY
(i) Small quantities of potassium, sodium, and calcium nitrates
caused a great increase in the number of Azotobacter in sterilized soil.
Ammonium nitrate in the same quantities caused a less marked in-
crease. Higher concentrations were not so favorable to the growth
of the organisms.
(2) Potassium and sodium nitrates in the concentrationj studied
caused an increase in the amount of nitrogen assimilated by Azoto-
bacter on agar films. Calcium nitrate in the same amounts brought
about a decrease in the amount of nitrogen fixed to a point even below
' that representing the amount assimilated in the absence of nitrates. In
soil cultures nitrates of sodium and calcium caused an increase in total ni-
trogen, which was more marked in the unsterilized cultures than in those
cultures sterilized and inoculated with a pure culture of Azotobacter.
However, the increase in total nitrogen is not commensurate with the
increase in the number of Azotobacter noted under the same conditions.
(3) Under aerobic conditions Azotobacter in liquid cultures reduced
nitrate to nitrite, but not to ammonia. More atmospheric nitrogen was
assimilated in the presence of nitrate than in the absence of this salt.
Jan. 2S, igis Nitroge7i- Assimilating Bacteria 227
(4) Pigmentation occurred when potassium and sodium nitrates,
and especially calcium nitrate, were used with Azotobacter, the colora-
tion increasing with the concentration of the salt. This effect was more
marked in Azotobacter strains which produce little or no pigment in the
absence of nitrates.
(5) All three nitrates studied caused an increase in the number and
size of volutin bodies in Azotobacter cells. From all appearances
these salts also tended to hasten the development of these bodies.
(6) The number of Bacillus radicicola in sterilized soil was increased
by the addition of small quantities of potassium, sodium, ammonium,
and calcium nitrates. This increase was not so marked as in the Azoto-
bacter cultures. B. radicicola appeared to be much more resistant to
higher concentrations of nitrates than Azotobacter.
(7) B. radicicola under aerobic conditions did not reduce nitrates
in solution to nitrite, ammonia, or elemental nitrogen. The presence
of nitrates did not materially influence the small amount of atmos-
pheric nitrogen fixed under these conditions.
(8) When grown on agar films, B. radicicola fixed a small amount
of nitrogen, varying from 0.15 to 0.43 mgm. of nitrogen in 100 c. c. of
the medium. The addition of various amounts of potassium, sodium,
and calcium nitrates increased to a slight extent the amount of nitrogen
assimilated.
(9) In liquid cultures all three nitrates caused a large increase in the
amount of gum obtained by precipitation with acetone.
(10) The presence of large amounts of potassium, sodium, and cal-
cium nitrates proved detrimental to the formation of nodules on alfalfa.
B. radicicola did not appear to lose its infecting power when grown on
media containing varying amounts of sodium and calcium nitrates.
Alfalfa seedlings grown in the presence of large amounts of nitrate did
not produce nodules when inoculated wdth a viable culture of B. radicicola.
Nitrates in soil cultures prevented the re-formation of nodules once
removed and also caused a decrease in the number of nodules already
present.
LITERATURE CITED
(i) Bassler, p.
1895. SANDKULTURVERSUCHE UBER DIE STICKSTOFFASSIMILATION DER GELBEN
LUPINE IM STERIUSIERTEN UND GEIMPFTEN BODE.V BEX DARGEBOT
WECHSELNDER MENGEN VON SALPETERSAUREN SALZEN. In Jahresbef.
Agr. Chein., n. F., Bd. 18 fjahrg. 38), p. 131.
(2) BEIJERINCK, M. W.
1888. DIE BACTERIEN DER PAPILIONACEEN-KNOLIXHEN. In Bot. Ztg., Jahfg.
46, No. 48, p. 758-771. (Continued article.)
(3)
1891. OVER OPHOOPING VAN ATMOSPHERISCHE STICKSTOP IN CULTUREN VAN
BACILLUS RADICICOLA. In Verslag. en Meded. K. Akad. Weteiisch.
[Amsterdam], r. 3, deel 8, p. 460-475.
228 Journal of Agricultural Research voi. xn, No. 4
(4) Beijerinck, M. W.
1901. UEBER OLiGONiTROPHiLE MiKROBEN. In Centbl. Bakt. [etc.], Abt. 2,
Bd. 7, No. 16, p. 561-582, I pi.
(5) and Delden, A. van.
1902. UEBER DIE ASSIMILATION DES FREIEN STICKSTOFFS DURCH BAKTERIEN. In
Centbl. Bakt. [etc.], Abt. 2, Bd. 9, no. 1/2, p. 3-43.
(6) Berthelot, M. p. E.
1893. RECHERCHES NOUVELLES SUR LES MICRORGANISMES FIXATEURS DE L'
AZOTE. In Compt. Rend. Acad. Sci. [Paris], t. 116, no. 17, p. 842-849.
(7) BoNAZzi, Augusto.
1915. CYTOLOGICAL STUDIES OF AZOTOBACTER CHROOCOCCUM. In JoUT. Agr.
Research, v. 4, no. 3, p. 225-239. Literature cited, p. 238-239.
(8) BOTTOMLEY, W. B.
1909. SOME EFFECTS OF NITROGEN-FIXING BACTERIA ON THE GROWTH OF NON-
LEGUMINOUS PLANTS. In Proc. Roy. Soc. London, s. B, v. 81, no.
548, p. 287-289.
(9) Bredemann, G.
1909. bacillus amylobacter a. mt et bredemann in morphologischer,
PHYSiOLOGiscHER UND SYSTEMATiscHER BEziEHUNG. In Centbl. Bakt
[etc.], Abt. 2, Bd. 23, no. 14/20, p. 385-568, 13 fig., 2 pi. Literatur-
verzeichnis, p. 559-566.
(10) Buchanan, R. E.
1909. THE GUM PRODUCED BY BACILUS RADICICOLA. In Centbl. Bakt. [etc.],
Abt. 2, Bd. 22, No. 11/13, p. 371-396. Citations, p. 395-396.
(11) Chalmot, G. de
1894. THE INFLUENCE OF NITRATES ON GERMINATING SEEDS. In Agr. Science,
V. 8, no. lo/ii, p. 463-465.
(12) CrEydt, Bodo.
1915. UNTERSUCHUNGEN UBER DIE KALKEMPFINDLICHKEIT DER LUPINE UND
IHRE BEKAMPFUNG. In Jour. Landw., Bd. 63, Heft 2, p. 125-191, 6
pi.
(13) Drabble, Eric, and Scott, Daisy G.
1907. on the effect of ACIDS, ALKALIS, AND NEUTRAL SALTS ON THE FER-
MENTATIVE ACTIVITY AND ON THE RATE OF MULTIPLICATION OP YEAST
CELLS, hi Biochem. Jour., v. 2, no. 7/8, p. 340-349, i fig. Literature
P- 349-
(14) Fernbach, a., and Lanzenberg, A.
1910. DE L'aCTION DES NITRATES DANS LA FERMENTATION ALCOOLIQUE. In
Compt. Rend. Acad. Sci. [Paris], t. 151, no. 17, p. 726-729.
(15) Flamand, Henri.
1905. XJBER den EINFLUSS DER ERNAHRUNG AUP DIE ENTWICICLUNG DER
KNOLLCHEN DER LEGUMiNOSEN. In Centbl. Agr. Chem., Jahrg. 34,
Heft II, p. 738-740.
(16) Frank, B.
1892. DIE ASSIMILATION FREIEN STICKSTOFFS BEI DEN PFLANZEN IN IHRER
ABHANGIGKEIT VON SPECIES, VON ERNAHRUNGSVERHALTNISSEN UND
VON BODENARTEN. In Landw. Jahrb., Bd. 21, p. 1-44.
(17) Fred, E. B.
1913. A PHYSIOLOGICAL STUDY OF THE LEGUME BACTERIA. In Va. Agr. Exp.
Sta., Ann. Rpt., 1911/12, p. 145-173, fig. 34. Literature, p. 172-173.
(18) and Graul, E. J.
1916. THE EFFECT OF SOLUBLE NITROGENOUS SALTS ON NODULE FORMATION.
In Jour. Amer. Soc. Agron., v. 8, no. 5, p. 316-328. Literature cited,
p. 327-328.
Jan. 28, i9i8 Nitrogen-Assimilating Bacteria 229
(19) Greaves, J. E.
1916. THE INFLUENCE OP SALTS ON THE BACTERIAL ACTIVITIES OP THE SOIL. In
Soil Science, v. 2, no. 5, p. 443-480, 4 fig.
(20) Hanzawa, J.
I914. EINIGE beobachtungen UBER STICKSTOFF-BINDUNG durch azotobac-
TER IN STICKSTOFFARMEN UND IN STICKSTOFFREICHEN SUBSTRATEN.
In Centbl. Bakt. [etc.], Abt. 2, Bd. 41, No. 18/23, p. 573-576.
(21) HellriEgel, H., and Wilfarth, H.
1888. UNTERSUCHUNGEN UBER die STICKSTOFFNAHRUNG DER GRAMINEEN UND
LEGUMiNOSEN. In Ztschr. Ver. Riibenzuckerindus., Beilageheft,
Nov., 234 p., 6 pi.
(22) Herke, S.
i913. contributions on nitrogen fixation and nutrition op bacillus
radicicola and on bacterial tests op nitragin and azotogen.
(Abstract.) In Exp. Sta. Rec, v. 29, no. 8, p. 733. (Original article
in Kiserlet. Kozlem., v. 16, no. 3, p. 311-322, 1913. Not seen.)
(23) Hills, T. L.
the INFLUENCE OF NITRATES ON THE GROWTH OP AZOTOBACTER. Un-
published. Offered for publication in the Ann. Rept. of the Pa. Agr.
Exp. Sta.
(24) HiLTNER, L.
1900. UEBER DIE URSACHEN, WELCHE DIE GROSSE, ZAHL, STELLUNG UND
WIRKUNG DER WURZELKNOLLCHEN DER LEGUMINOSEN BEDINGEN. In
Arb. K. Gsndhtsamt., Biol. Abt., Bd. i. Heft 2, p. 177-222, pi. 3.
(25) Immendorfp, H.
1892. BEITRAGE ZUR LOSUNG DER " STICKSTOFFRAGE. " In Landw. Jahxb.
Bd. 21, p. 281-339.
(26) JosT, Ludwig.
1907. LECTURES ON PLANT PHYSIOLOGY. Translated by R. J. H. Gibson.
464 p., 172 fig. Oxford. Bibliography at the end of each lecture.
(27) Kayser, E.
I910. INFLUENCE DES NITRATES SUR LES FERMENTS ALCOOLIQUES. /nCompt.
Rend. Acad. Sci. [Paris], t. 151, no. 19, p. 816-817.
(28) Kossowicz, Alexander.
I914. UBER DAS VERHALTEN VON HEFEX UND SCHIMMELPILZEN ZU NITRATEN.
In Biochem. Ztschr., Bd. 67, Heft 4/5, p. 400-419.
(29) IvAURENT, Emile.
1891. RECHERCHEs SUR LES nodosite;3 radicales DES l:6gumineuses. In
Ann. Inst. Pasteur, annee 5, p. 105-139, 3 fig.
(30) Lewis, L. L., and Nicholson, J. F.
1905. soil INOCULATION. TUBERCLE-FORMING BACTERIA OP LEGUMES. Okla.
Agr. Exp. Sta. Bui. 68, 30 p., 8 fig.
(31) Marchal, Emile
1901. INFLUENCE DES SELS MIN^RAUX NUTRITIFS SUR LA PRODUCTION DES
NODOSIT^S CHEZ LE POis. In Coinpt. Rend. Acad. Sci. [Paris],
t. 133, no. 24, p. 1032-1033.
(32) Maz6, p.
1897. FIXATION DE L 'azote LIBRE PAR LE BACILLE DES NODOSITIES DES
l6guminEuses. In Ann. Inst. Pasteur, annee 11, no. i, p. 44-54.
{zz)
1898. LES MICROBES DES nodosit6s DES l6guminEuses. In Ann. Inst.
Pasteur, ann. 12, no. i, p. 1-25, i fig.
(34) Meyer, Arthur.
1912. DIE ZELLE DER BAKTERiEN. 285 p., 34 fig., I col.pl. Jena. Literatur,
p. 267-282.
230 Journal 0} Agricultural Research voi. xii. no. 4
(35) Moore, G. T.
1905. soil, inoculation for legumes; wit?i reports upon the successful
use of artificial cultures by practical farmers. u. s. dept.
Agr. Bur. Plant Indus. Bui. 71, 72 p., 10 pi.
(36) MUNTER, F.
I916. tJBER DEN EINFLUSS ANORGANISCHER SALZE AUF DAS WACHSTUM DER
ACTiNOMYCETEN. III. In Centbl. Bakt. [etc.], Abt. 2, Bd. 44, No.
24/25, p. 673-695, 9 fig.
(37) NoBBE, F., and Richter, L.
1902. UBER den EINFLUSS DES NITRATSTICKSTOFFS UND DER HUMUSSUBS-
stanzen AUF DEN iMPFUNGSERFOLG BEi lEguminosen. In Landw.
Vers. Stat., Bd. 56, Heft 5/6, p. 441-448.
(38) Pfeffer, W.
igo6. THE physiology op PLANTS, cd. 2. Translated by A. J. Ewart. v. 3.
Oxford.
(39) Prazmowski, Adam.
1891. die WURZELKNOLLCHEN DER ERBSE: II. TEIL. DIE BIOLOGISCHE BEDEU-
TUNG DER WURZELKNOLLCHEN. In Landw. Vers. Stat., Bd. 38, p.
5-56.
(40) PringshEim, Hans.
1914. ZUR STICKSTOFFASSIMILATION IN GEGENWART VON SALPETER. In
Centbl. Bakt. [etc.], Abt. 2, Bd. 40, no. 1/8, p. 21-23.
(41) Prucha, M. J.
1915. PHYSIOLOGICAL STUDIES OF BACILLUS RADICICOLA OP CANADA FIELD PEA.
N. Y. Cornell Agr. Exp. Sta. Mem. 5, 83 p. Bibliography, p. 79-83.
(42) RiTTER, G.
1909. AMMONIAK UND NITRATE ALS STICKSTOFFQUELLE PUR SCHIMMELPILZE.
In Ber. Deut. Bot. Gesell., Bd. 27, Heft 10, p. 582-588.
(43) SackETT, W. G.
1915. THE PIGMENT OP azotobacter chroococcum. In Proc. 35th Ann.
Meeting Soc. Prom. Agri. Sci., 1914, p. 80-88, 2 col. pi.
(44) Stoklasa, Julius.
1908. BEITRAG ZUR KENNTNIS DER CHEMISCHEN VORGANGE BEI DER ASSIMI-
LATION DES ELEMENTAREN STICKSTOFFS DURCH AZOTOBACTER UND
radiobacter. In Centbl. Bakt. [etc.], Abt. 2, Bd. 21, no. 15/16, p.
484-511.
(45) Vines, S. H.
1888. on the relation between the formation of tubercles on the
roots op leguminosae and the presence op nitrogen in the
SOIL. In Ann. Bot., v. 2, no. 7, p. 386-389.
(46) Wilson, J. K.
I915. PHYSIOLOGICAL STUDIES OF BACILLUS RADICICOLA OP SOY BEAN. (Ab-
stract.) In Science, n.s., v. 41, no. 1048, p. 180.
(47) wohltmann, Ferdinand, and bergen^,
1902. DIE KNOLLCHEN-BAKTERIEN IN IHRER ABHANGIGKEIT VON BODEN UND
DXJNGUNG. In Jour. Landw., Bd. 50, Heft 4, p. 377-395.
(48) Woods, C. D.
1892. the acquisition op atmospheric nitrogen by growing plants. in
Conn. Storrs Agr. Exp. Sta., 4th Ann. Rpt., 1891, p. 17-28.
(49) ZiPFEL, Hugo.
I911. BEITRAGE ZUR MORPHOLOGIE UND BIOLOGIE DER KNOLLCHENBAKTERIEN
DER LEGUMINOSEN. In Centbl. Bakt. [etc.], Abt. 2, Bd. 32, No.
3/5' P- 97-137- Literature, p. 136-137.
Vol. XII FEBRUARY ^, 1918 No. 5
JOURNAL OF
AGRICULTURAL
RESEARCH
CONXEMStXS
Page
New-Place Effect in Maize ------ 231
G. N. COLLINS
(Contribution from Bureau of Plant Industry)
Relation of the Variability of Yields of Fruit Trees to the
Accuracy of Field Trials - - - - - - 245
L. D. BATCHELOR and H. S. REED
< Contribution from California Agrjcaltuial Experiment Station)
Interrelations of Fruit-Fly Parasites in Hawaii - - 285
C. E. PEMBERTON AND H. F. WILI^RD
( Contribution from Buneau of Entomology)
PDBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICCITDRE,
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASI^IISIGTON, D. C.
WASHINOTON : OOVERNMENT PnillTlNO OTFICe Mei8
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
E:ARL F. KELLERMAN, Chairman
Physiologist and Associate Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Offite of Experiment Stations
CHARLES L. MARLATT
Entomologist and Assistant Chief, Bureau
of Entomotogy
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricultural Experinuiil
Station
H. P. ARMSBY
Director, Instiiutt of Animal Nutrition, Thf
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist and AssistattI
Dean, Agricultural Experiment Station of
the University of Minnesota
All correspondence regarding articles from the Department of Agriculture should be
addressed to Karl F. Kellermaa, Journal of Agricultural Research, Washington, D. C.
*Dr. Pearl has undertaken special work in connection with the. war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
J01N£ OF AGRIOmm RESEARCH
Voiv. XII Washington, D. C, February 4, 1918 No. 5
NEW-PLACE EFFECT IN MAIZE
By G. N. Collins, Botanist, Office of Acclimatization and Adaptation of Crop Plant? and
Cotton Breeding, Bureau of Plant Industry, United States Department of Agriculture
INTRODUCTION
Widely divergent opinions have been expressed regarding the ad-
visability of transferring seed from one region to another. With maize
(Zea mays) the transfer of seed is generally held to be disadvantageous.
Numerous experiments have shown that when seed of the same original
variety, but grown at two places, is planted side by side at one of the
places the results are in favor of the local seed.
The relative superiority of the locally selected seed has been so pro-
nounced that the securing of seed from distant localities, except where
grown for forage, has been discouraged. The natural result is to confine
the utilization of carefully selected strains to the localities where the
breeding is done. Caution in transferring seed is certainly desirable, but
the results of the experiments here reported indicate that two opposing
factors are involved, the relative importance of which must be determined
before generalizations are made.
The differences shown between the geographically separated lines of
the same variety when brought together in the locality where one of the
lines has been grown may be ascribed to three general causes:
(i) Cross-pollination. Varieties removed to different localities may
become crossed with other sorts, with the result that their character-
istics are changed in a greater or less degree.
(2) Different standards of selections. Changed conditions in con-
nection with the diversity that exists in all varieties bring slightly different
types of plants into prominence, and selection, either conscious or un-
conscious, results in a changed type.
(3) New-place effects. The more or less temporary changes that fol-
low a transfer to new conditions, caused by the novelty or "shock" of the
new environment without special reference to the nature of the change.
So far as known, new-place effect, which the experiments here re-
ported show to be a significant factor in the transfer of seed, has not
previously been considered with regard to maize.
Journal of Agricultural Research, Vol. XII, No. s
Washington, D. C. Feb. 4, 1918
lu Key No. G— 134
(231)
232 Journal of Agricultural Research voi. xii. no. 5
These experiments were originally planned with the idea that the seed
of a first-generation hybrid might be transferred to a distant locality
without showing the reduced yields thought to follow the transfer of
pure strains.
It was believed that with first-generation hybrid seed it might be a
matter of indifference where the hybrid was made, provided the parents
were suited to the localities where the crop was to be grown.* If found
to be the case, this would constitute an important addition to the ad-
vantages that follow the use of first-generation hybrid seed, making
possible a wider application of the results secured through the work
of skilled breeders.
The results indicate that hybrids made in one locality and grown in
another are not only at no disadvantage compared with the same hybrids
produced at the locality where the comparison is made but that the
introduced hybrid may even be superior. It further appears that the
same is true of pure strains and that, after the effects of cross-pollination
and selection have been eliminated from the problem, there is a residual
effect of the transfer to the new place that tends to increase rather than
to reduce the A^gor and yield of the plants.
NATURE OF THE EXPERIMENTS
It was planned to conduct the experiments at the following places:
Stockton, Kans. ; Victoria, Tex. ; and Lanham, Md. These places
represent a wide range of soil and climate. At Stockton, which is
in the eastern part of the semiarid Great Plains area, there is a fertile
friable soil, low rainfall, low atmospheric humidity, and a prevalence of
high winds. Victoria, located in the Gulf region of Texas, has a stiff
black clay soil of good fertility. The total rainfall is usually large and
the humidity high, though severe drouths are not uncommon. At
Lanham, a few miles north of Washington, D. C, the soil is sandy,
acid, and relatively infertile. The rainfall and humidity are usually
neither deficient nor excessive. In the discussion of results these locali-
ties will be referred to as Kansas, Texas, and Maryland.
Four varieties of maize were chosen for the experiment, as follows:
(i) Stockton, a white dent variety developed at Stockton, Kans.
(2) Strawberry, a large-eared Texas variety with red and white varie-
gated dent seed, well adapted to the conditions at Victoria, Tex.
(3) Hickory King, a strain of this variety grown in Virginia and well
suited to the conditions at Lanham, Md.
(4) Boone, a strain of "Boone County White," the seed of which was
obtained from Illinois.
In the spring of 191 2 seed of the four varieties were planted at each
of the three places named. The precaution was taken to mix the seed
1 COLUNS, G. N. THE VAI,ue OF FIRST-GENERATION HYBRIDS IN CORN. U. S. Dept. AgT. Bur. Plant
Indus. Bui. 191, p. 33. 1910.
Feb. 4. 1918 New-Place Effect in Maize 233
of each of the kinds so that the portions sent to the several localities
should be as nearly alike as possible. The order of planting was the same
at all the places, as follows: Every alternate row was planted to the
Boone variety, which was used as the male parent in making hybrids.
The seed of each of the four varieties, including the Boone itself, was
planted in the rows alternating with the Boone.
The Boone plants standing in the alternate rows throughout the field
were allowed to shed pollen. All others, including the interplanted
Boone, were detasseled, so that the only pollen shed was from the Boone
variety. Seed was saved from the detasseled rows only. This was of
four kinds, (i) Stockton X Boone, (2) Strawberry X Boone, (3) Hickory
King X Boone, and (4) cross-pollinated seed of the Boone. At each
locality the experiment was placed at a distance from all other corn.
In 1 91 3 it was planned to compare the behavior of the plants raised
from the seed produced at the three localities when grown at each of
these places. Although grown in the same field, no attempt was made to
compare the relative merits of the several hybrids, each hybrid together
with the pure-seed Boone constituting a separate experiment involving
only the comparison of the yield from the seed of the three localities.
Thus, there were four experiments to be made at each of the three
places. Since the arrangement was the same at all places, one
description will suffice.
To compare the Stockton X Boone hybrid from the three localities,
the seed from the several places were planted in adjoining rows, the
first row from the Kansas seed, the second from the Texas seed, and the
third from the Maryland seed. The series was repeated 10 times, making
10 distinct comparisons. A similar procedure was followed with the
three other hybrids and with the pure-seed Boone.
At Stockton, Kans., excessive drouth destroyed the entire corn crop.
Since no results were secured from Stockton, the behavior of the Kansas-
grown seed will be eliminated from the discussion of the results, which
will be confined, therefore, to the experiments conducted in Texas and
Maryland.
At Victoria, Tex., the rows were 100 feet long. The seed was drilled,
and the plants were thinned to about 2 feet in the row. When harvested,
a weighed sample of 20 pounds of ears was saved from each row. This
sample was thoroughly air-dried, after which it was again weighed to
determine the loss of water. The percentage of grain to cob was also
determined. No significant differences in water content or percentage of
grain were found in the crops from the seed from different localities, and
these determinations are, therefore, not discussed.
At Lanham, Md., the seed was planted in hills 3 feet apart, in rows
132 feet long. The plants were thinned to one stalk per hill. The
method of harvesting was similar to that of Victoria except that no
determinations of dry weight were made.
234
Journal of Agricultural Research
Vol. XII, No. s
At each locality the com from all the experiments was harvested the
same day; and the weight of ears, together with the number of plants,
was recorded for each row.
To avoid, so far as possible, differences due to inequalities of soil and
to obtain reliable averages, each pair of rows consisting of one row each
of Maryland- and Texas-grown seed was considered a separate test. The
relative behavior of the plants from the two sources of seed was deter-
mined by an average of all the comparisons, usually lo in number.
In Table I are given the yield in pounds per row and the yield per plant
from the several rows. Yields which stand opposite in the table are
from adjoining rows in the field.
Table I. — Behavior of plants from Maryland- and Texas-grown seed subseqtiently
planted in Maryland and Texas
STOCKTON X BOONE
Compared at Lanham, Md.
Compared at Victoria, Tex.
Variety and factor.
Yield
from
seed pro-
duced at
Lanham.
Vield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield
from
seed pro-
duced at
Lanham.
Yield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield per plant
Do
Pou
O
ids.
90
90
92
93
89
60
26
58
78
67
Pounds.
I- IS
I. 06
•93
.81
•74
•52
.61
•56
•78
•71
Per cent.
87.8
91.8
99.4
106. 9
109. 2
107. I
59-8
loi. 7
100. 0
97.1
Pounds.
0. 65
.65
.62
■63
•65
.62
.64
•59
.64
.64
Pounds.
0. 65
.66
•56
.81
•63
.69
.66
.62
.64
.66
Per cent.
100. 0
99.2
105. I
87. 5
loi. 5
94.6
98.5
97^5
100. 0
Do
Do
Do
Do
Do
Do
Do
Do
98.5
'
Average ....
•743
.787
96. o8±2. 75
•633
.658
98. 25 ±0.86
Yield per row
Do
30-5
32-5
33^o
25. 0
29-5
9.0
5-5
18.0
23-5
22. 0
31.0
41.5
32-5
29. 0
18.5
II. 0
19-5
18.5
26. 5
22. 0
99.1
87.8
100.8
92. 6
122. 9
90. 0
44.0
98.6
94.0
100. 0
32-5
31.8
32.0
32.8
34- 0
28.5
34-5
31.0
34-8
38.0
33- 0
32.5
32. 0
38.0
31-5
36.5
30-5
36-5
34-5
33- 0
99.2
98.9
100. 0
Do
Do
92.6
103.8
87.7
106. 2
Do
Do
Do
Do
91.9
100. 4
Do
Do
107. 0
Average . . . .
22
•85
25. 0
92. 99±3. 25
32-97
33- 80
98. 76±i. 64
STRAWBERRY X BOONE
Yield per plant .
Do
Do
Do
1.80
1. 81
I. 92
1. 61
I. 92
1.89
1.74
I. 64
96.8
105.9
104.9
107. I
0. 70
0.51
.72
.72
• 78
.84
.80
.84
II5-7
100. o
96- 3
97.6
Feb. 4, 1918
New-Place Effect in Maize
235
Tablet. — Behavior of plants from Maryland- and Texas-grown seed subsequently
planted in Maryland and Texas — Continued
STRAWBERRY X BOONE — Continued
Compared at Lanham, Md.
Compared at Victoria, Tex.
Variety and factor.
Yield
from
seed pro-
duced at
Lanham.
Yield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield
from
seed pro-
duced at
Lanham.
Yield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield per plant. . . .
Do
Do
Pounds.
1-83
I. 72
Pounds.
1.65
1-38
Per cent.
105. 2
III. 0
Pounds.
0.74
•74
•77
.69
-71
.70
Pounds.
0.82
•85
-74
.78
•65
.72
Per cent.
94.9
93-1
Do
93-9
104.4
98.6
Do
Do
Average ....
1.83
1.66
105. 15 ±1.09
•735
•747
99- 65 ±1.32
Yield per row
Do
75-5
63-5
71.0
68.0
62. 0
67. 0
71.0
64- 5
61. 0
62. 5
61.0
47.0
103. 07
99. 22
107. 58
104. 21
100. 81
"7- 54
39.00
42. 50
41.50
39.00
37.00
36.80
39-30
37.00
37.00
38.00
27.30
30-30
32.00
34-50
37-50
37-50
33-30
39.80
30.50
38.00
117. 7
116 8
Do
Do
112. 9
106 I
Do
Do
Do
99-3
99.0
108.3
96.4
109. 6
Do
Do
Do
Average ....
67.8
61. 2 105. 401b I. 80
38.70
34-05
106. 61 ± I. 82
HICKORY KING X BOONE
Yield per plant . . . .
Do
Do
Do
Do
Do
Do
Do
Do
Do
A.verage . . .
Yield per row
Do
Do
Do
Do
Do
Do
Do
Do
Do
Average . . .
1. 016
21.5
17-5
17-5
32.0
46. o
50. o
51-5
52.0
47.0
51.0
38.6
.60
.60
•97
^•15
1.08
1.50
1.24
1.36
1.30
1.038
18-5
18.0
40. 5
38.0
40. o
58-5
46. o
49.0
52.0
38-15
104. I
91.9
93^8
94.0
loi. 3
107.3
92-5
105. o
93-8
loi. 5
22±I. 42
lOI. 2
97.2
98.6
88.3
109s
III. I
93^6
106. I
97^9
99.0
100. 25 ±1.43
.606
31.00
29. 00
28. 00
32.30
31.30
28. 00
28.80
32.80
31-50
28. 50
30. 10
545
no. 9
loi. 7
100. o
103-9
104.5
104. 2
113. 1
99.1
no. 5
105.2
105. 3i±i. o
21. 00
23.00
25. 00
27-30
27.50
26. 00
23.00
24. 00
25-50
26. 50
24-87
119. 2
III. 5
105-7
108. 4
ic6. 4
103-7
III. I
II5-4
no. 5
103.6
109. 55 ± I. 13
236
Journal of Agricultural Research
Vol. XII, No. 5
Table I. — Behavior of plants from Maryland- and Texas-grown seed subsequently
planted in Maryland and Texas — Continued
BOONE
Compared at Lanham, Md.
Compared at Victoria, Tex.
Variety and factor.
Yield
from
seed pro-
duced at
Lanham.
Yield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield
from
seed pro-
duced at
Laaham.
Yield
from
seed pro-
duced at
Victoria.
Yield from seed
produced at
Lanham ex-
pressed as a
percentage of
the mean.
Yield per plant
Do
Pounds.
1.38
I- 15
1.08
1.30
I. 10
1-35
I. 22
1.30
I. 16
I. 14
Pounds.
1.36
I. 21
1.36
I. 26
I- 13
I. II
I- 51
1-35
I. 22
1.24
Per cent.
100. 7
97-5
88.5
loi. 6
98.7
109. 8
89.4
98. I
97-5
95-8
Pounds.
0.56
.60
■58
•63
.60
.69
.60
.62
Pounds.
0.78
•73
.68
.67
.64
.62
.68
■63
Per cent.
83.6
90. 2
92. I
96.9
96.8
105- 3
93^8
99.2
Do
Do
Do
Do
Do
Do
Do
Do
Average —
I. 218
1-275
97-76±o.9i
. 610
.678
94. 74±i-S4
Yield per row
Do
41-5
42.5
45-5
53-5
39-5
44-5
47-5
48.0
44.0
41. 0
45- 0
46.0
47-5
48.0
43- 0
41.0
S3-0
46. 0
47- S
49-5
95-95
96.05
97-85
105. 42
95-76
104. 09
94-53
102. 13
96.17
90. 61
30- 50
29.50
33-50
34.00
34.50
32.00
32-50
31-25
27. 00
32-30
32.00
30. CO
30.00
29-30
28. 00
32-00
106. I
95-6
102.3
106. 7
Do
Do
Do
107.0
104.7
107.4
98.8
Do
Do
Do
Do
Do
Average. .. .
44-75
46.65
97. 87 ±1.02
32.22
30.06
103. 5±i. 50
The yield of the plants from the Maryland- and Texas-grown seed is
made comparable by expressing the yield of the former as a percentage
of the mean yield of both. For example, in Table I where the behavior
of the Stockton X Boone hybrid is considered, the yield per plant of the
first row at Lanham, Md., which was from Maryland-grown seed, is shown
as 0.90 pound; the adjoining row from Texas-grown seed yielded 1.15
pounds per plant. The mean of the yield of these two rows is 102.5
pounds, and the yield from the Maryland-grown seed is 87.8 per cent of
this mean, the value given in the fourth column. The average of the 10
comparisons is, in this case, 96.08 per cent — that is, the yield of the
plants from the Maryland-grown seed averages 3.92 per cent below the
mean yield of this strain grown in Maryland. This expression for the
relative behavior in Maryland of the plants from the Maryland-grown seed
is to be compared with the results of the similar comparison made in
Texas, given in the next three columns. From these it is seen that in
Texas the yield per plant of the plants from Maryland-grown seed was
98.25 per cent of the mean yield of the strain.
Feb. 4, 1918
New-Place Effect in Maize
237
At both localties the Maryland-grown seed of this cross is inferior to
that produced in Texas, but the point to which attention is now directed
is that the inferiority is greater in Maryland than in Texas.
Table II contrasts the average behavior of each of the three hybrids
and the pure-seed Boone at the two localities. For example, when com-
pared in Maryland, the yield per row of plants from the Maryland-grown
seed of Stockton X Boone averaged 92.99 per cent of the mean yield
of the cross. In Texas the same comparison showed the average yield
of plants from Maryland-grown seed to be 98.76 per cent of the mean.
Thus, the plants from Maryland-grown seed averaged 5.77 per cent
higher in yield in Texas than in Maryland.
Table II. — Average behavior of Maryland-grown seed expressed as a percentage of the
mean of Maryland- and Texas-grown seed, igij
Factor and kind of seed.
Compared in
Maryland.
Compared in
Texas.
Comparative in-
increase of
Maryland-grown
seed in Texas.
Increase
divided
by prob-
able error.
Yield per plant:
Stockton X Boone
Per cent.
96. 08 ±2. 75
105. 15 ±1.09
98. 22 ± I. 42
97-76±o. 91
92. 99±3-25
105. 40±i. 80
100. 25 ± I. 43
97. 87 ± I. 02
Per cent.
98. 25 ±0.86
99- 65 ±1-32
105. 3i±i. 05
94-74±i-54
98. 76 ±1.64
106. 6lit:l. 82
109. 55±i. 13
103. 5o±i. 50
Per cent.
2. I7±2. 88
-5. 5o±i. 71
7. 09±i. 76
—3. 02 — I. 69
5.77±3-64
I. 2I±2. 56
10. 01 ± I. 58
5. 63 ±1.81
Per cent.
0-75
-3.22
4-03
— I. 69
1-59
•47
6.34
311
Strawberry X Boone
Hickory King X Boone ....
Boone . .
Yield per row:
Stockton X Boone
Strawberry X Boone
Hickory King X Boone. . . .
Boone
In comparing the yields both the yield per row and the yield per plant
were considered. In Texas the yield per row was much more dependent
on the number of plants in the row than in Maryland. In Texas, as the
number of plants increased, there was a pronounced tendency for the yield
per row to increase and the yield per plant to decrease, while in Maryland
an increase in the number of plants in the row was accompanied by only
a small increase in the yield per row, and there was an almost correspond-
ing increase in the yield per plant.
This difference in behavior at the two localties is not difficult to under-
stand. In Maryland the failure to secure a perfect stand was largely
the result of infertile spots in the field, and the same unfavorable condi-
tions which reduced the number of plants also reduced the yield of those
that survived. In Texas, the loss of plants was more the result of acci-
dental factors, which influenced the yield of the remaining plants only
by permitting them to take advantage of the increased space with a
consequent slight increase in the yield per plant. The method of planting
in Texas accentuated this difference, for with the plants close together
in the row the additional space resulting from a missing plant was
utilized more by the neighboring plants in the same row than by those
in adjoining rows. In Maryland, on the other hand, where the plants
were spaced equally in both directions, half the space made available
by a missing hill would be appropriated by the adjoining rows.
238
Journal of Agricultural Research
Vol. XII. No. s
Table III. — Stand of plants secured from Maryland and Texas grown seed compared in
Maryland and Texas
Variety.
Compared in Maryland.
Number
of plants
from
Mary-
land-
grown
seed.
Number
of plants
from
Texas-
grown
seed.
Stockton X Boone.
Do
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Average .
Strawberry X Boone.
Do
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Average .
Hickory King X Boone.
Do
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Average .
Boone. .
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Average.
34
36
36
27
33
15
21
31
30
2,2,
Stand of Maryland-
grown seed ex-
pressed as percent-
age of mean.
27
39
35
3(^
25
21
32
33
34
31
III. 5
96. o
loi. 4
85-7
113. 8
83-3
79.2
96.9
93-8
103. 1
42
37
35
40
37
35
3&
38
34
37
39
34
34
34
37
39
40
40
38
39
38
30
37
42
41
36
Z2>
39
37
38
36
36
31
30
42
?,3
37
39
37
36
40
33
38
35
38
38
37
35
34
39
40
96. 5 ±2. 50
106.3
93-3
102. 8
97-3
95-8
106.8
Compared in Texas.
Number
of plants
from
Mary-
land-
grown
seed.
50
49
52
52
52
46
54
53
54
59
100. 4±i. 86
97.1
104. 6
104. 8
93-
108.
103.
lOI.
lOI.
104. o
97-4
loi. 6zb.
95-2
98.7
109. I
103.8
97-
94.
105.
104.
98.
94.
100. i±i. 24
56
59
53
49
50
50
51
54
52
54
Number
of plants
from
Texas-
grown
seed.
44
48
49
52
54
45
51
58
50
47
54
49
58
54
57
46
54
50
51
49
57
47
50
53
46
59
54
50
53
42
38
41
46
44
45
51
47
53
Stand of Mary-
land-grown
seed expressed
as percentage
of mean.
99.0
100. o
95-4
105. I
102
92.
108.
94.
100.
108.3
100. 5±i. 20
102. 8
116.8
116. 5
108.8
104. 2
106. 4
106.3
102. 9
105. I
100. 9
37
40
44
48
52
46
54
42
50
48
36
44
47
45
47
47
41
51
107. i±i. iJ
108.6
109. I
105.4
104. o
loi. 9
98.9
97.1
116. o
100. o
98.9
104. o±i.3i
120.0
105.4
no. 5
109. I
109. 6
98.9
113- 7
99.0
108. 3 ± I. 72
Feb. 4, 1918
New-Place Effect in Maize
239
Ability to produce a stand may legitimately be considered one of the
manifestations of greater vigor. That it is a definite and positive factor
is shown in Tables III and IV, in which the relative stands are compared.
With all four kinds a comparison of the relative stand at the two local-
ities is in favor of the transferred seed. In the Boone variety the transfer
of seed has resulted in an 8 per cent increase of stand, a difference
nearly four times the probable error. Since the analysis of the compar-
ative stand of local and transferred seed shows that the differences are
not accidental, but are consistently in favor of the transferred seed, it
would seem that yield per row is a more reliable measure of comparative
vigor than yield per plant. Yield per row is the measure of the practical
results, and from this standpoint it is seen that all four strains showed
an increase in yield as a result of transfer of seed. In Texas, where
there was a definite tendency for an increased number of plants in a row
to reduce the yield per plant, yield per plant is obviously ill calculated
to bring out the real difference in vigor.
Table IV. — Average stand of plants secured from Maryland-grown seed expressed as a
percentage of the mean of Maryland and Texas-grown seed
Kind of seed.
Compared in
Maryland.
Compared in
Texas.
Increased
stand of Mary-
land-grown
seed in Texas.
Increase
divided
by prob-
able error.
Stockton X Boone
Per cent.
96. 5 ±2. 50
100. 4±i. 86
loi. 6±o. 98
100. i±i. 24
Per cent.
100. 5±i. 20
107. i±i. 18
104. o±i. 31
108. 3 ± I. 72
Per cent.
4-o±2. 77
6. 7±2. 20
2. 4±i-63
8. 2±2. 12
Per cent.
1.44
3- 04
1.47
3-87
Strawberry X Boone
Hickory King X Boone
Boone
EXPERIMENTS IN 1915 AND 1916
The results of the 191 2 and 191 3 experiments were so at variance with
current belief that it was thought best to obtain additional evidence
before publishing. A somewhat similar experiment was therefore
planned and carried out during the years 191 5 and 191 6. The same
varieties were used as in the previous experiment, but the localities
were changed by substituting Greenville, Tex., and Sacaton, Ariz., for
Victoria, Tex., and Stockton, Kans.
Crop failure at Greenville again limited the experiment to two locali-
ties: Lanham, Md., and Sacaton, Ariz. At Sacaton the temperatures
are high, and there is practically no rainfall during the growing season,
the crop being grown by means of irrigation.
To eliminate differences due to irregularities in the stand of plants, a
different system of planting was adopted. Seed from both localities
were planted in each hill, the seed from the two sources being identified
by their positions in the hill. At harvest the measurements were con-
fined to the hills which contained plants from both Maryland- and Ari-
240
Journal of Agricultural Research
Vol. XII. No. s
zona-grown seed. For all such hills the height of each plant was re-
corded with the total length of the ear or ears. In this way each hill
constituted a unit of comparison. The height of each plant was ex-
pressed as a percentage of the mean height of the two plants in the
same hill. These determinations were then averaged to secure an
expression of the mean behavior of the plants from each source of seed.
Length of ear was treated in the same way. Table V gives the results.
Unfavorable conditions so reduced the yields at Lanham, Md., that length
of ear was recorded for only three of the strains, and even for these
there was so much variation that the results are of doubtful significance.
They serve, however, to supplement the results on the height, with which
they are in accord.
Table V. — Average behavior of Maryland-grown seed expressed as a percentage of the
mean of Maryland- and Arizona-grown seed, igi6
Kind of seed.
Compared
in
Maryland.
Compared
in
Arizona.
Increase of
Maryland-
grown seed
in
Arizona.
In-
crease
Factor.
Produced in Mary-
land, 1915.
Produced in Arizona,
1915-
divided
by
prob-
able
error.
Height of
plants.
Do...
Do....
Do...
Do....
Do
Stockton X Boone
Boone X Stockton
Strawberry X Boone. .
BooneXStrawberry. .
Hickory King X
Boone.
BooneXSelf
StocktoaXBoone. . . .
do
Strawberry X Boone .
do
Hickory King X
Boone.
BooneX Boone
StocktonX Boone....
Hickory King X
Boone.
BooneX Boone
Per cent.
92.9±I-2I
9a.7io.85
89-2±i.74
9S-7ii-44
103. s±o. 53
108. 1 i 0.90
88.8i7-6
iia.4i2. 2
146. 3 ±6.0
Per cent.
98. 9± 0.38
99- 7 i 0.43
97- 1 ±0-58
IOC. 9 ±0.46
97- si 0.46
109. o±o. s6
I03.9i2. 7
I05.ii3.3
iS9-4i2-9
Per cent.
6. oil. 27
7-oio.9S
7-9ii-83
5-2±i.50
— 6. o±o. 70
o.9±i.o6
IS- lis. 1
— 7-3i4-o
13- lis- 7
Per ct.
4- 7a
7-37
4- 3a
3-47
-8.57
0.85
Length of
ear.
Do....
Do
BooneXStockton
Hickory King X
Boone.
BooneXSelf
1.9
1.8
2.3
At Sacaton, Ariz., in 191 5 reciprocal crosses were made with Stock-
ton X Boone and Strawberry X Boone, and these reciprocals were sepa-
rately compared with the seed grown at Lanham, where Boone was
used only as the male parent. In all of the six comparisons except one,
transferring the seed resulted in increased height ; and in all but one the
difference is almost certainly not the result of chance.
In 1915 the crosses in both Maryland and Arizona were made by hand
instead of by detasseling alternate rows as in 191 2. In gathering pollen
an effort was made to obtain pollen from as many plants as possible and,
so nearly as might be, in equal amounts from each plant. In like manner
selection of female parents was avoided so far as possible. In spite of
these precautions, it is evident that there would still be a measure of
selection. Some plants produce virtually no pollen, and many plants
fail to develop an ear. Furthermore, since the plants were thinned to a
stand of one in a hill from each locality, obviously weak plants being
removed, it would seem that here too there would be a tendency to retain
Feb. 4.1918 New-Place Effect in Maize 241
the types of plants best adapted to the conditions where the experiment
was tried. The entire effect of selection would be to favor the home-
grown seed, and that the transferred seed was not superior to the home-
grown in every instance may not be held to vitiate the cases in which
significant differences in favor of the transferred seed were observed.
The results indicate, however, that the stimulation is more pronounced
in some stocks than in others. Thus, in the 191 6 comparisons Boone X
Hickory King stands out as a conspicuous exception. In all other stocks
the transferred seed produced taller plants than the home-grown seed;
but with Boone X Hickory King, the home-grown seed exceeded the
transferred by 6.2 per cent, a difference not to be ascribed to chance,
being more than eight times the probable error. Of the three stocks
in which the yield was taken, Boone X Hickory King is also the only
one to show superiority for the home-grown seed. Taken alone, the dif-
ferences in yield could not be considered significant, but the agreement
with the results for height confirms the reliability of these results.
The insignificant increase in the case of Boone may be explained by
the fact that the Arizona-grown seed was more closely selected to fit
the Arizona conditions than were the other kinds. At Sacaton the pure-
seed Boone was obtained by selling. This procedure would restrict the
plants from which seed was secured to those able to produce both ears
and pollen under Arizona conditions. With cross-pollinated seed and
hybrids, plants that produced no ears would be represented in the progeny
as male parents.
DISCUSSION OF RESULTS
Three classes or degrees of new-place effects ("neotopism") have been
recognized by Cook S chiefly with reference to cotton: (i) Those in which
there is merely a stimulation of growth; (2) those in which there is also
a definite general change of the hereditary characteristics of the variety;
and (3) those in which the new conditions call forth a promiscuous
mutative diversity.
The results here reported give evidence under the first of these cate-
gories only. With respect to the more pronounced changes that fol-
low the transfer of varieties from the Tropics to a temperate climate
it may be said that many such changes do occur in maize, some of which
at least are inherited. For conclusive qualitative evidence on this point,
however, there is lacking definite information regarding the behavior
of the introduced varieties in their native countries.
Roberts- has pointed out that the striking effects which have been
ascribed to acclimatization in maize are to be referred either to cross-
pollination with native varieties or to the results of selection. The effect
of cross-pollination, which misled early investigators, has presumably
' Cook, O. F. aspects op kinetic evolution. In Proc. Wash. Acad. Sci.. v. 8. 1906, p. 336. 1907.
2 Roberts. H. F. acclimatization with reference to corn breeding. In ist Aon. Rpt. Kans.
Com Breeders' Assoc, [i9os]/o6, p. 60-64. 1906.
242 Journal of Agricultural Research voi. xii, no. s
been eliminated from recent experiments, but the effects of selection
are so pronounced and speedy that in experiments hitherto reported
any direct effect of the environment on the characters of the plants
would be completely masked. The characteristics of a maize variety
are altered readily by selection. When grown in a new locality for a
few years, even without conscious selection, the type may change rapidly;
and when brought back to the original locality, it is in reality a differ-
ent variety. The characters brought into prominence in the new locality
may render the stock less suited to the old conditions, though better
adapted to the new.
It would be very difficult, if not impossible, to eliminate completely
all selective action. Even when all seed is saved, those individuals or
types of plants which are best adapted to the conditions under which
they are grown will produce a greater proportion of the seed than will
the types which are less well adapted, and those least adapted may
produce no seed at all. In two localities where different conditions
prevail, the highest yielding plants — hence those contributing the largest
proportion of the seed — would presumably be of different types; and
when brought together and compared, we shou^ld expect to find a slight
advantage for the locally grown seed. Yet the results of the present
experiments indicate that the effect of selection during a single season
may be so slight as not to mask completely the opposing new-place effect.
Since new-place effect in maize seems to operate as a stimulus, it would
tend to obscure any lack of adaptation in newly introduced varieties.
The recognition of new-place effect may be said, therefore, to increase
rather than diminish the importance that must be assigned to adaptation.
As a result of the stimulation due to new-place effect, the cultivation
of an inferior strain might be extended as a result of its satisfactory per-
formance the first year following its introduction.
The stimulus that followed the transfer of seed in these experiments is
doubtless similar to the increased vigor imparted to many vegetables
by growing the crop in localities remote from the place where the seed
was produced. The economic utilization of increased vigor secured in
this way is usually confined to crops which are grown for the sake of some
part other than the seed. In cotton, for example, the increase of vigor
in the plants following a transfer of seed is often very pronounced,
although the crop of seed and fiber may be reduced. In maize, as a
result of the determinate habit of the plant, vegetative vigor and seed
production are more closely associated, so that the possibility of prac-
tical utilization seems greater.
CONCLUSIONS
Hybrids between the same pairs of varieties made at different local-
ities showed no decrease in yield as a result of transferring the first-
generation seed to a new locality. On the contrary, the change of
Feb. 4, 1918 New-Place Effect in Maize 243
environment seemed to act as a stimulus, with the result that the yields
were increased in all but one of the hybrids tested. One unhybridized
variety was included in the experiment, and this also gave slightly
increased yields as a result of being transferred to a new environment.
In 6 of the 10 ^comparisons the increase is too large to be ascribed to
experimental error and indicates that new-place effect should be taken
into consideration as a factor of production.
That significant increases may be secured by taking advantage of
new-place effect in maize should not be used as an argument in favor of
the general transfer of seed. There is no evidence that the importance
of using acclimatized seed has been overestimated. On the contrary,
the experiments show that new-place effect may often obscure the differ-
ences between acclimatized and unacclimatized seed when first com-
pared, and thus interfere with a full appreciation of the value of adap-
tation.
The investigations show the existence of a hitherto-neglected factor
in maize production, but much more extensive experiments are needed
to ascertain the extent and practical importance of this factor. The
existence of one definite exception indicates that the tendency to increased
vigor following a transfer of seed is not universal. The results also
indicate that adaptation in maize comes about through selection rather
than as a direct reaction to the environmental conditions.
RELATION OF THE VARIABILITY OF YIELDS OF FRUIT
TREES TO THE ACCURACY OF FIELD TRIALS^
By L. D. Batchelor, Professor of Plant Breeding, and H. S. ReED, Professor of Plant
Physiology, University of California, Citrus Experiment Station.-
INTRODUCTION
The value of the outcome of any trial depends upon the probability
that a similar result will be obtained if the trial is repeated. In recent
years the agricultural experiment stations of all countries have greatly
increased the number and size of their field trials. A casual examination
of such trials usually shows a wide range in the reliability of the results.
The yields of control plots in different parts of the same tract will often
differ as much among themselves as the yields of fertilized and un-
fertilized plots differ from each other. The purpose of this paper is to
present the results of a study of the variation in recorded yields of fruit
trees taken singly and in groups of various sizes, and especially to deter-
mine the effect upon variability of various combinations and repetitions
of unit plots.
It is recognized that the results of a single experiment are often
untrustworthy ; yet experimenters have published single results and have
based practical advice upon them. However well planned field trials
may be, the interpretations of the results can hardly be considered of
infinite reliability — that is, results which will invariably be obtained
when the trials are repeated. The best that can be done is to construct
the most probable results from the more or less varying observed results
of individual trials. If it is impossible to obtain perfect accuracy, it is
not impossible to fix the limits of error and thus to determine whether the
differences obtained are due to the treatments appUed or to unavoid-
able errors — that is, whether the differences are significant. A number
of trials are necessary before a reasonably reliable result can be obtained.
The average result of a series does not always represent the truth.
When averages are used, they should always be accompanied by their
probable errors, which are a measure of their reliability.
Before attempting to interpret the results of a plot experiment with
fertiUzers, it is necessary to know whether the differences observed are
any greater than those which might have occurred had none of the plots
been fertilized. The scientific method is to recognize the inevitable error
and, while reducing it by every possible precaution, at the same time to
> Paper 44, University of California, Citrus Experiment Station, Riverside. Cal.
' The writers wish to make acknowledgment of their indebtedness for aid and criticism to Prof. H. H.
Love, of Cornell University; Prof. E. B. Babcock, Dr. H. B. Frost, and other members of the University
of California.
Journal of Agricultural Research, Vol. XII, No. 5
Washington, D. C. Feb. 4, 1918
Iw Key No. Cal.— 14
(24s)
246 Journal of Agricultural Research voi. xn, No. $
measure its probable amount so as to make sure it is not likely to vitiate
our conclusions.
One of the chief difficulties in obtaining reliable results in field trials
is the natural variability of the material with which we are deaUng.
Crops are living organisms with inherent tendencies to vary, even though
it were possible to make their environmental conditions identical.
In agronomic experiments the number of plants taken is usually so large
that inherent variability ceases to be a factor of importance. In horti-
cultural experiments, however, where fruit trees are under observation,
the limited number of trees possible to include in a plot may make the
factor of inherent variability an important consideration.
Further variation is induced as a result of the many factors of the
environment which are beyond the control and possibly the recognition
of the experimenter. Some of these factors are independent; others
react upon one another. In designing a set of field trials, we try to
avoid, as far as possible, all secondary factors which may exert a
disturbing influence.
Lack of uniformity in both the physical and chemical characteristics
of the soil is one of the foremost factors causing variation in productivity
of plants. Apparently uniform surface soils may be underlain with a
heterogeneous subsoil. Differences also occur which are not evident on
a careful inspection of both the soil and the crops, but which are easily
measured by weighing the yields. In other words, the weighing machine
is more sensitive than the eye and reveals differences that mere inspection
can not detect.
The past treatment of the soil brings in variables the significance of
which may not be comprehended at the time a field trial is begun. The
persistent effects left by the application of stable manure on some of the
Rothamsted plots show how large a part is played by the past history of
the field. Plots which for 40 years have had identical treatment still
give different crop yields because of the effect of dressings of barnyard
manure applied at an earlier period.
Unequal prevalence of diseases and insects may bring about further
error in the results.
Besides the above sources of variation and possibly outweighing them
at times is the effect of season. No season is entirely normal; and it is
only when the experiment has been repeated for several years, or, in other
words, until "a fair sample of seasons" has been made, that any sort of
allowance can be made for seasonal effects.
As is shown by observations at Rot hamsted, it is not possible to estab-
lish a schedule of relative jdelds for a series of plots, even after several
years' comparison. In this case two grass plots were treated alike for
50 years; by taking the yield of one plot as the standard, the yield on
the other in the same season has been as low as 90 per cent and as high as
196 per cent.
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 247
Recent investigations, which are reviewed in succeeding pages, have
thrown light upon the nature of the variability inherent in experiments
conducted with groups of plants or of animals. So far as the writers
know, however, few studies have yet been published upon the uncer-
tainty with which one deals in attempting to carry on plot trials with
orchard trees. In connection with plans for extending the trial plots of
this Experiment Station, the writers have attempted to study the ques-
tion of variability of tree yields and to formulate some plan to determine
the probable reliability of the results. It is highly desirable to make
the tree plots as small as possible without sacrificing too much accuracy
on account of the large amount of land required for each individual.
A fruit tree is possibly more affected by environmental conditions than
an annual crop growing from seed to maturity in one season. The tree
roots penetrate more deeply into the soil and may be affected by soil
differences to a considerable depth. Climatic conditions during the rest-
ing period may have marked influence on the crop production of the trees.
Pruning also introduces variation. Uniform pruning is desirable;
yet different trees need different types of pruning, and more uniform
results will be obtained if this is recognized.
Individual trees have apparent idiosyncrasies in fruiting habits. A
tree may yield large crops of fruit in alternate years, with very light
crops in intervening years. The character of the stock upon which the
particular variety was budded to form the tree may profoundly influence
the type, habit, and productivity of the adult tree. Furthermore, we
can not overlook the possibility of errors in yields due to predacious
animals which devour or otherwise destroy more or less fruit.
Fruit trees therefore present opportunity for more variability than
would be expected in the growth of annual plants.
PREVIOUS STUDIES
The varying productivity of fruit trees has been called to our attention
by the writings of Macoun (1904),^ Munson (ipoy), Shamel (1912),
Hedrick (1912), Stewart (191J, p. 552-554), Fletcher {1913), Coit (Jpjo),
Whitten (19 15), Lewis and Vickers (19 15, p. 30-31), Barre {1915),
Gourley {19 15, p- 72-73), Kraus (19 16), and others. The above observa-
tions have been largely made in connection with plant breeding and
orchard management problems, and were not made to bear necessarily
on the accuracy of plot trials.
Holtsmark and Larsen {1906) were among the first to call attention to
the errors of field trials. They recognized the inevitable variation of
field results, and showed how it may be estimated by the use of the
standard deviation and the coeflficient of variability. They also showed
1 Bibliographic citations in parentheses refer to "Literature cited", pp. 382-283.
27808°— 18 2
248 Journal of Agricultural Research voi. xii. No. s
that the coefficient of variability decreases as the plots are enlarged, but
not proportionally to the size of the plot.
The limitations of field experiments were discussed by Carleton (ic^og),
who called the attention of experimenters to the various uses of control
plots, and to the general precautions necessary to obtain reliable results.
Hall {igog), Mercer and Hall (igii), and Hall and Russell (igii)
recorded extensive studies of the soil variations in experimental grounds
and the influence of size and repetition of plots upon accuracy. This
work was done largely with the yields of wheat, mangel, and hay crops.
The conclusions from the above work are that the error in field trials
diminishes as the size of the plot increases, but that the reduction is
small when the plot is enlarged to a size greater than one-fortieth of an
acre. The error may be further diminished by increasing the number of
plots similarly treated and scattering them about the area under experi-
ment; but there is not much to be gained by increasing the number of
plots above five.
Wood and Stratton (igio) sounded notes of caution concerning the
interpretation of experimental results. Frequency distribution is dis-
cussed from the point of view of its bearing on the reliability of averag-
ing results. The applications of the probable-error methods to questions
of sampling for analysis, to field experiments, and to feeding experi-
ments are illustrated. The probable error of field experiments was
investigated by two independent methods and found to be about 5
per cent of the mean yield. Tables are given showing the number of
duplicate plots or number of animals in a feeding trial which must be
employed to give any desired precision in the result. It is shown that
more accurate results may be obtained by employing large numbers of
small scattered plots than by using one large plot.
The estimation of errors in field-plot tests has been given consid-
erable attention by Lyon {igi2) and coworkers. It was shown that it
is not possible to establish a schedule of relative yields for a series of
plots, even after several years' comparison. Also, there seems to be
little gain by using plots larger than one-fiftieth of an acre in size when
the comparative yield of the crops is made the criterion. An area of
one-twenty-fifth of an acre of land distributed in four widely separated
plots, devoted to any one test, secures a much greater degree of accuracy
than the same area of land in one body. The probable error was reduced
from 4.5 to 2 per cent by such distribution.
Pickering (igii), from studies on apples and pears, concluded that
experimental plots should include 6 to 12 fruit trees. Precautionary
advice was also given concerning the measurement of results b)^ crop
production, foliage, and tree characteristics. In comparing the results
on the treated plots with the controls, instead of taking the average of
the controls, he prefers to plot these results out and to draw a smoothed
Feb.4.i9i8 Variability of Yields of Fruit Trees and Field Trials 249
curv^e through them, and then to compare the results of the experimental
plots with readings taken at corresponding points of this curve.
Wood {191 1) showed how the degree of reliance can be determined for
any set of experimental results by the use of the probable error. The
use of this constant was demonstrated in interpreting laboratory analy-
ses, as well as both plot and feeding experiments. Working with mangel
yields, the author calculates the number and size of plots required to at-
tain any desired precision, and working with the probabl eerror of live-
weight increase of sheep, tables are given showing the number of animals
required in an experiment to attain various degrees of reliability.
Several papers by Harris {1912, 1913a, 19131), and 1915) have drawn
our attention to several phases of the experimental error in field tests.
A measure of the variability of the soil productivity is obtained by
determining the correlation between the yields of ultimate small plots
and the yields of various groups of adjacent plots. The more nearly
this correlation approaches zero, the more homogeneous the soil. This
method of measurement does not seem to provide as definite a means
of obtaining a corrective term as the use of the coefficient of variability
and the probable error as used by Wood, Wood and Stratton, Mercer
and Hall, etc., or the contingency method of correction as used by
Surface and Pearl (1916).
Montgomery (1912) has also discussed the comparative variabiHty
resulting from increasing the size of the plot and from distributing small
ultimate plots over the area. The latter method was found to be more
accurate. In a subsequent paper {1913) the relative reliabihty of yields
of wheat planted in rows and in square blocks is discussed.
An exhaustive and discriminating discussion of the nature and magni-
tude of variability in the results of feeding experiments has been given
by Mitchell and Grindley {191 3). Much of their discussion is equally
applicable to experimentation with plants.
Olmstead (19 14) applied the method of least squares in calculating
the reliability of the yields of the mangel and wheat crop records of
Mercer and Hall, the potato records of Lyon, and the wheat yields of
Montgomery. The conclusions from this series of observations are:
The estimation of the probable error of a large number of small duplicate plots well
distributed in the area devoted to a field experiment indicates that the precision of
agricultural experiments can be increased by replicating the experiments on small
plats.
Coombs and Grantham (1916) have studied the variation in the
yields of rice and coconuts for one year, and discussed the range and
interpretation of the probable error. They showed that the yields from
any two single plots could only be significant when the difference
amounted to 22.8 per cent of the mean. They also introduced calcula-
tions to show the odds that any increase is a real increase and not a
probable error.
250 Journal of Agricultural Research voi. xii, No. s
The use of controls and repeated plantings in varietal tests was studied
by Pritchard {1916) in breeding work with sugar beets. His studies lead
to the conclusion that the practice of dispensing with control rows and
using the mean of all progeny rows as a standard of comparison appears
to be less accurate than the employment of frequent controls. However,
the employment of every alternate row as a control was not sufficient to
offset the variability in yield arising from irregularities of soil.
Stockberger {1916) discussed the value of a number of the common
methods for determining the normal yield of treated plots based upon
the yields of hops. Normal yields for various plots varied widely accord-
ing to the method of computation, the values in some cases differing from
the actual yield by as much as 40 per cent. Repetition brought about
a very marked reduction in variability, although with only five repetitions
the error is still relatively large.
The work of Surface and Pearl {1916) shows an advance in the refine-
ment of methods of conducting field trials. With the realization that
the use of frequent control plots often produces results far from satis-
factory, these workers have calculated by the contingency method the
probable yield of each plot of ground in their grain-testing series. This
calculated yield represents the most probable yield of each plot on the
supposition that they have all been planted with a h)'pothetical variety
whose mean yield is the same as the observed mean of the field. This
"calculated" yield may then be used as a basis for determining a cor-
rection factor, whereby each area must be given a handicap plus or
minus the actual yield, depending upon whether the plot in question is
calculated to be a low- or a high-producing area. This method of cor-
recting the soil variation is combined with four systematically repeated
plots of one-fortieth acre of each variety, and gives a high degree of
accuracy.
MATERIAL USED FOR STUDY
The studies to be reported in this paper deal with the variability of
fruit-tree yields. They are based upon the individual tree yields of
oranges {Citrus sitiensis), lemons (Citrus limonia) , walnuts (Juglans regia),
and apples {Malus sylvestris) from orchards which had received uniform
treatments for a number of years — indeed, so far as known, from the
time of planting the trees. The orchards were carefully examined, and
the records for all trees which were known to be abnormal from disease
or other apparent causes were eUminated. In place of the records of
trees thus eHminated the average yield of the eight surrounding trees
was substituted (assuming that the tree stood at the center of a square
block of nine trees).
This substitution is not entirely satisfactory, yet it was felt that it
was necessary in order to compute plots of homologous size and syste-
matic arrangement. The writers found, as a matter of fact, that there
is a very high degree of correlation in these orchards between the yields
Feb.4,i9is Variability of Yields of Fruit Trees and Field Trials 251
of individual trees and the average of surrounding trees. For example,
they found coefficients of correlation as high as 0.652 ±0.065 for the
Eureka lemons and 0.628 ±0.060 for the Arlington navels. According
to the formula given by Harris (1915), the correlation between individual
trees and eight adjacent trees in a plot of the Arlington navels is
0.576 ±0.04. In view of these results, the writers felt justified in using
this method of substituted values.
The fruit plantations herein discussed, to judge by the surface soil,
size, and condition of the trees, as well as their apparent fruitfulness,
appeal to the observer as uncommonly uniform. All the orchards
studied are situated in semiarid regions and are artificially irrigated
during the summer months. This fact is believed to be a distinct advan-
tage for the purpose of reducing the variability of one year's yield com-
pared with another, since it insures a fairly uniform water supply for the
soil and reduces one of the variants inevitable in nonirrigated localities.
All yields of the several fruit and nut plantations are given in pounds
per tree of the ungraded product.
DESCRIPTION OP THE PLANTATIONS
Navel orange (Arlington). — These records were of the 191 5-1 6
yields of one thousand 24-year-old navel-orange trees near Arlington
station, Riverside, Cal. The individual tree production is shown by
figure I.
The grove consists of 20 rows of trees from north to south, with 50
trees in a row, planted 22 by 22 feet. A study of the records shows
certain distinct high- and low- yielding areas. The northeast comer and
the south end contain notably high-yielding trees. The north two-thirds
of the west side contains a large number of low-yielding trees. These
areas are apparently correlated with soil variation. Variations from
tree to tree also occur, the cause of which is not evident. These varia-
tions, which are present in every orchard, bring uncertainty into the
results of field experiments.
In making their calculations this grove was divided by the writers into
imaginary plots of any size and shape desired. The yields of these plots
were then compared with one another and their variability ascertained.
The distribution of both the theoretical and actual yields of this grove
is shown in figure 10. The yields of the individual trees when plotted
according to their frequency give a skew curve of Pearson's Type I, since
the critical function
4(4^2-3^x)(2i82-3/3i-6) °-^5-
The distribution of the actual yields is shown on the figure by small
circles. The points for the theoretical curve were calculated by the
formula \ 1 / x 2
252
Journal of Agricultural Research voi. xii, no s
I
I
04
1
.a
w
H
•^
o
■(->
4)
^
IH
t*.
<u
'lU
eu
Si
•d
-«->
s
>
'rt
;>
TJ
a
bo
.a
I
a
I
Total
yield
per
row.
fOOvO « O t^OO «/^0 O>r0N N ^H miooo MSO C>N 0^O»t^vO C^O*N^O TOO «0 inr*^0
lOOO M 0» w t^O w MOO --tTtiriM r^ OnO r- »r( -ft « O O ro O^^O »At-N)HMro^^ioMr^Ol
Ov <o t^ looo ■<tV3 00 ■* O N ro w o -.tvo \0 ■* QO M N roOO H WM ■*'<fOitt^r-t^rOt-t^O M
M«N«M««NNHN«NC. «NC.«««««««««««M«««Nt'jn<-5«tO
o
rOM r>.M N O'O « «tnw r^OswvO W O 0\r>.r* OS (*5C0 « 0*cOO»^u^O t^w O t> O^ ^ri t^ Q
00 t^r^O 'too >Oi-i r^l>.i^NO u-^Tl■0^0^<^* O OO <n «0 w^O COOO t>.O«o*ON^^O00
»
O O N lOOO ■^•OOO tO« t^ \ri '^ t^ O Osw O <^0<S f<C0O Ov^f^i^ioroM t^O N tJ-moO tT
t^oo 0 »AMQOOO row cow r^ ^O u^urO fO'^w <n uic^oovim m o « O 0\ fOCO i^ »n lo ro w
00
u^w O-OvO O^OO 0»»0i>« low M\0 r^O 0 m OsOO O « -^t^roOvr^a »nOO \0 »0 »0 ro »0 m ^
WM«MWWMMMMMtHMWMMMWI-t tH M WHW MMMMHdM
t^
S|5|^5>|5 5^^K|S>o'|^a5Sv?M"&^<g^^SS8«S5^^5Saf:
■o
^^S^2.S^|K^^^o^§,'£|8,gg5"?3;KS8^3>S'S<<SSSJ§Sga
^
2 5<2 ? ^ ? S S ^ 2 2 -2 S S ^^ It- B^ S -2 ^ H ^ 3 J 5 ^^ ^^ S a H 5
T
?
°S22?H???SS^?'2t'lS"sS2S2°S^^2S = SSH^?^
^
0000i>0O>Ot>Ot~0wiH<7>00>Ot^t^00-*MM0'O'<t(N0i00000000\w00 0i0i«-*
O OvO r-OlO ■-■ «00 -^JJO t^t^'^iJt^OO -"tc-^OOOO -^tl OvOPl O OVOOCO •»«<» ^00 «w
::
N I- r» to mO M " nvO O O h h i/tO n «00 «Oii^, <»)t~mo»r<OoO'*>nO fO O O w 00
o
HTrr^00\>O00<^<^000«0tN-*f0r)-00>0N>ni-iCT.>0i^05.0OC0Mt»N«Q'*t^
'-22's:?°>:r;?2S'2?s^28?3:i'^^^:?s°°»^?'SJj??'Ss?s2;?s
o
c2s:c3K2^^||'ssg&s;^s>|s;s§«'^|^5st:^^?,'|^sSv^a>s
00
Tf M roOO O ■*« ^oc^O w lAOQO -^^ rol^iy-)t^« MvO f^jQ CO M O »^M wiin l^vO O -^t It
r-
^5S?^^SS^^HS2'^°^E'S52«H^l2 2?HS2H^8SS^?«^H
■c
5 2 c^l ^ §, §,^ ;5c^ i^ ^v? s 5 S.S v&<^ j3 ^1 ir=^% ^ « ^ g5 s s- ? ^^ a ^5
-
<X'5u^S!v&a3^^s;'&^<g&?f;jiifi?:s?^a^5o5^s,2g|225^
■>r
«
S^S^cSRScSS^aS^&S.SKS'S'SScZ'cSScS'^&S^^gvS^^^-Sgg
„
S ff^ ^ tl5 5 5'<2 3 5,5 <2 S ? 8; S"§« S &<g ^^' ^^ v^"^ K<S ^ 5^ ^ ^ 8 ;^ ? ^
"
<ssss^'^Rs;fi^5§s^^§a;5 5K5K^^2;§^^s.s^ffsi:?:?f7cg^
Row.
» « .;, 4 .Ao r;ad c^ o M 2 ;j j ;j-.-o ^« a 0 M j; jj 4 ^o rj-cc » o H j^ ^
Feb.4,i9i8 Variability of Yields of Fridt Trees and Field Trials 253
^
„
r^t
^
r4
r»
i^
in
r-
■«• 0
t-»
w
0\ -X
N
ex
m
m
t^
fO
**:
n
to t to ■* ■*
1^
„
t~cO
at
M
"
"
^
^
10
1/1
„
in
m
„
^
CO
r^
"
"
^
"
*^
w
«
CH
i~.
r^
„
in
r^oo 00
m
■n
•t
^
■<a
"
"
■^
*^
M
u
"
CI
^*
"
r*
0
„
„
I^vO
•<t TtoO
t^
„
■* •*
to
00
Oi
H
>o
"
"
"
H
t^
„
r^
„
M
M
m
TtOO 00
m
0 00
s-
"
"
w
^
CI
"
CI
r>.
M
in
^
^
^
<-)
„
0
M
Oc
0
„
0
•<t
(^
PI
n
t^
in
CO
1/
r*
8
r~-o
0
0
0
„
in
r-
0> 01 OvOO
r*
t
"
•^
^
t-t
M
*^
^*
C^
fH
„
•s
in
in
on
^
-n
„
0
m
to
00 -o
■D
0
"
M
1.1
^
^
C<
"'
CI
r^
fl
in
•fl-
„
IH
0
0
0
0
r*
t to t
to
M
"
w
"^
Ct
ct
c*
1^
ir,
1/1
CO
•fl-
„
r^
■*
m
-o
<n
W
"
"
*^
11
"
c*
"'
"
CI
^
00 %o
(-)
00
„
tn
f7\
in
0
„
too
0
■*
"
«
"^
"
"^
^
"J
"
t^
^
(10
?
fn
Oi
„
OV
>o
„
•*
„
0
r^
w
^
*^
*^
M
M
*^
c-*
*^
■S
m
0
Tf 0
Ov TfOO
Ol
„
m
0
^
00
■^
w
w
w
w
CI
(.■-1
t^
^
M
in
„
r^
ro
0
■0-
o\ 0
M
■TJ-
■*
"
"
W
"
**
C*
CI
«
CI
w
t^
0 -o
„
O
0
Ol
„
t^
„
0
«,
o<
^
fH
c
to
NO
n,
in
m
„
M
M
„
0
„
"
"
M
"
CI
u
CI
w
0
ct
CI
o
^
ir,
^
n
nn
•<J-00
„
00-0
^
r>
10
r^
f)
■0
"
W
w
^
^
*^
-o
„
rt
m
00
„
0
0
Oc
0
00
»
ao
^
^
i.A
^
"^
u
^
CI
"^
"2
rv
tn
„
r^
m
„
r^
„
0 0 NO *0
"
*^
^
"
"
"^
CI
"
w
w
^
„
m
0-0
m
„
w
Oi
t
S:""
0
u
a
•a
>.
3
H
:
^
:
3
:
r^co
■*
^
•^
<»
10
254
Journal of Agricultural Research
Vol. XII No.s
Navel orange (ANTEiyOPE Heights) . — The navel-orange grove later
referred to as the Antelope Heights navels is a plantation of 480 ten-year-
old trees planted 22 by 22 feet, located at Naranjo, Cal. The individual
tree records of this planting are shown in figure 2 and are the yields
obtained in 1916. The general appearance of the trees gives a visual
impression of uniformity greater than a comparison of the individual
tree production substantiates; however, the distribution of the yields
approximates closely the normal curve of errors, having a skewness of
only o.ooi ±0.037, ^s shown by figure 11.
Fig. 2. — Diagram showing the individual tree yield (in pounds) of the navel-orange
grove (Antelope Heights).
North
Yield per tree {pounds)
Row.
Total
yield
per
row.
4
5
6
7
8
9
10
II
12
13
14
IS
16
17
18
19
20
ai
22
23
24
25
26
27
28
29
30
31
32
33
Total yield
per row . .
130
120
200
ISS
250
300
22s
ISO
300
250
200
ISO
no
190
200
210
140
los
120
130
ISO
I9S
100
100
los
100
100
13 S
24s
200
22s
200
250
200
ISO
ISO
20s
225
200
2 SO
325
ISO
i8s
100
200
150
200
ISO
ISO
100
100
ISO
120
100
105
no
ISO
100
100
ISO
i8s
255
170
225
200
150
250
230
300
295
300
240
225
22s
290
17s
170
230
135
i6s
130
100
IIS
200
ISO
120
200
ISO
15s
200
ISO
270
230
280
250
SO
250
200
ISO
130
270
235
26s
200
200
200
210
200
I3S
175
ISO
160
135
100
no
135
ISO
100
150
175
185
265
150
1 75
100
ISO
250
300
3 SO
IIS
21S
170
300
275
280
320
290
210
275
225
100
170
200
200
180
150
18s
ISO
160
200
200
200
275
200
ISO
175
200
150
ISO
115
260
240
235
240
155
260
300
200
250
250
250
250
200
225
200
200
250
250
200
220
150
250
200
200
ISO
160
190
200
175
175
175
150
180
i8s
240
225
200
220
28s
340
170
130
185
220
200
250
225
225
210
215
220
190
150
225
200
185
22s
225
200
i5o
140
125
ISO
250
270
175
150
200
200
200
250
2SS
195
250
200
ISS
150
ISO
150
22s
200
300
200
200
200
ISO
150
150
I2S
200
ISO
100
ISO
ISO
100
175
175
200
200
250
210
270
250
250
225
250
ISO
210
i6s
I3S
170
205
250
I6S
190
160
225
300
200
I2S
225
150
125
200
200
22s
ISO
240
ISO
150
200
i8s
200
250
200
250
250
ii3SO
>45S
.285
,285
.475
, 240
.250
>370
, 190
.84s
>82S
.775
.835
,870
.64s
.345
.320
, 200
,28s
.145
I 210
.215
.075
.125
,280
,420
.730
.635
.005
. 170
,110
.025
.125
5. 460
5,385
6, 580 6,010 6,815
6, 930 6, 745
6, 23s 6, 27s 6,475 5. 70016, 175 6, 260
Valencia orange. — The Valencia orange grove is composed of 240
15-year-old trees, planted 21 feet 6 inches by 22 feet 6 inches, located at
Villa Park, Cal. Figure 3 represents this planting and the individual
tree yields which were obtained in 1916.
Eureka lemon, — The lemon yields were obtained from a grove of
364 23-year-old trees, located at Upland, Cal. Figure 4 represents the
individual tree yields of this planting. The records extend from October
I, 1915, to October i, 1916. The grove consists of 14 rows of 23-year-old
trees, extending north and south, with 26 trees in a row, planted 24 by
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 255
Fig. 3. — Diagram showing the individual tree yield (in pounds) of the \'alencia
orange grove.
Northwest
Yield per tree {pounds)
Row.
3
4
5
6
7
8
9
10
II
12
13
14
IS
16
17
18
19
100
375
37
63
75
100
SO
300
100
100
75
225
425
SO
437
400
3SO
175
3SO
400
Total yield ,
per row 14.087
375
325
75
175
75
212
136
200
I7S
ISO
125
22s
200
350
238
163
275
325
350
350
250
275
175
250
175
125
100
100
30
100
125
350
225
200
225
ISO
400
250
400
450
275
375
260
350
300
25
75
75
300
150
275
250
325
125
200
300
300
350
250
375
275
200
150
300
300
200
250
150
263
150
125
300
75
225
350
4SO
462
350
275
200
ISO
375
300
275
325
275
100
200
363
262
275
150
275
150
375
300
288
3SO
275
300
57363
200
225
25
325
300
325
250
200
350
250
150
300
425
225
225
400
412
100
212
350
175
230
275
300
275
250
100
275
300
275
225
300
275
250
225
350
363
300
S,3o6
225
275
450
250
375
250
190
200
337
363
312
262
375
250
300
350
375
300
350
275
6,064
27S
225
250
275
275
100
150
200
338
337
188
265
300
300
400
250
300
200
250
350
S.318
200
150
225
ISO
375
3SO
225
225
25
300
375
325
350
175
150
325
125
262
250
75
225
250
150
250
ISO
225
225
175
350
225
lOO
175
100
100
3>962
Total
yield
per
row
7625
.367
7S72
.113
.950
,087
,851
, 200
, 101
-737
,250
>077
.150
,600
,700
,663
•975
■ 140
-ISO
,850
59) 158
Fig. 4. — Diagram showing the individual tree yield (in pounds) of the Eureka
lemon grove.
North
Yield per tree {pounds)
Total
yield
per
row.
3
4
S
6
7
8
9
10
II
12
13
14
IS
16
17
18
19
20
21
22
23
24
2S
26
Total yield per
row
259
204
475
383
Sio
421
425
218
301
409
152
79
287
348
81
250
III
195
333
291
III
361
210
140
126
241
2S8
157
235
292
250
188
228
274
257
264
165
1x8
196
258
199
•146
165
162
156
172
189
114
245
197
229
263
227
305
274
190
277
167
206
176
164
123
150
170
161
144
106
223
13s
170
127
181
127
251
314
183
263
202
338
195
133
276
188
194
120
264
267
180
282
259
241
154
203
255
296
315
207
i8s
220
379
381
412
288
313
369
343
234
230
221
276
258
292
345
268
253
227
275
357
104
332
326
414
350
275
397
325
264
251
356
300
346
341
312
319
336
222
241
345
278
290
361
291
325
239
328
252
294
388
297
399
201
283
210
260
157
286
202
211
167
234
278
261
2.55
237
213
245
278
256
186
222
235
396
201
343
412
341
232
211
271
172
169
140
167
169
206
149
177
245
159
216
209
218
116
122
80
112
176
158
128
289
241
335
349
292
443
327
327
382
334
386
180
439
314
268
341
380
336
360
304
259
226
364
298
282
416
410
328
438
237
329
348
322
399
346
377
429
471
354
441
308
399
353
312
368
315
363
335
371
398
325
313
302
262
266
6,921 ,
' 4j 825
5.985'
)8,o63
6, 550 4, 867
8, 783 !
296
273
378
345
281
233
312
298
342
248
283
278
317
365
299
322
338
385
369
349
329
294
326
341
333
413
i 8,347
264
218
298
295
304
24S
300
281
238
279
281
322
307
301
279
338
318
3S8
358
402
307
322
410
283
280
237
7,82s
241
236
310
279
321
313
308
277
339
293
282
305
298
200
223
344
326
356
223
276
204
2S7
222
253
I3t
328
7,176
104
222
222
375
280
333
266
100
342
287
279
222
295
285
248
245
3"
294
288
338
244
236
276
292
277
316
6.977
,584
,611
,341
,231
.003
,217
,987
•557
,837
,848
.581
•157
,801
,061
,269
,767
,653
,69a
,560
-S70
-587
.776
.954
.040
.592
■ 383
98,658
256
Journal of Agricultural Research
Vol. XII, No. s
24 feet apart. This grove presents the most uniform appearance of any
under consideration. The land is practically level, and the soil is appar-
ently uniform in texture. The records show a grouping of several low-
yielding trees; yet a field observation gives one the impression that the
grove as a whole is remarkably uniform.
SUEDiyiNG WALNUTS. — The walnut-tree records used in the following
calculations were obtained during the seasons of 191 5 and 191 6 from a
24-year-old Santa Barbara softshell seedling grove, located atWhittier,
Cal. The planting is laid out 10 trees wide and 32 trees long, entirely
surrounded by additional walnut plantings, except on a part of one
Fig. 5. — Diagram showing the individual tree yield (in pounds) of the seedling
walnut orchard.
North
Yield per tree {pounds)
30.
31-
32.
Row.
172
164
74
122
120
160
170
95
6S
85
78
SI
17
7S
117
83
44
86
71
87
41
86
169
112
III
135
89
SO
lOI
106
34
104
69
179
188
122
144
155
83
176
SO
III
136
158
ISO
100
135
96
80
103
94
128
71
61
100
73
123
88
103
69
170
198
125
151
140
74
95
151
166
Total yield per row 2,664 3-052 3-037 2,936 2,625 2,849 2)7SS 2,904 2,545 2)689
113
III
40
80
108
74
135
82
66
195
100
112
71
95
68
35
105
180
124
32
192
130
58
170
54
80
138
113
35
207
78
112
129
121
93
61
89
72
76
82
57
6
47
65
120
48
142
113
147
112
123
115
75
148
39
167
3i
94
57
118
40
131
93
134
45
63
86
99
76
35
165
197
97
76
119
71
96
99
178
151
192
87
67
79
104
80
123
84
93
61
61
133
15
15
96
90
104
91
144
122
lOI
81
210
104
160
130
90
73
157
207
136
104
144
40
126
72
70
43
14
113
81
69
80
8s
31
106
77
160
131
113
94
109
87
"5
94
215
164
164
195
S4
52
107
30
8
60
25
73
S3
28
25
94
42
72
85
117
90
65
149
113
25
166
ISO
166
160
164
164
97
62
58
96
64
43
loi
83
50
65
13
75
97
105
83
87
39
131
126
148
129
128
235
201
Total
yield
per
row.
1,277
1, 116
1,246
811
896
828
1,016
970
771
677
842
815
612
629
603
858
914
1.252
1,080
1,117
1,022
1.386
950
1, 169
1. 031
1. 15s
1.475
1.538
28,056
side which is adjacent to an orange grove. The trees are planted on
the square system, 50 feet apart. Figure 5 gives the yield and arrange-
ment of these trees.
Jonathan apples. — ^The apple records ^ were obtained from a lo-year
old Jonathan apple orchard located at Providence, Utah. The surface
soil of this orchard is very uniform to all appearances except on the
extreme eastern edge, where the percentage of gravel increases slightly.
The trees are planted 16 feet apart, east and west, and 30 feet apart
north and south. Figure 6 gives the yield and arrangement of these trees.
1 The authors wish hereby to express their appredation of the kindness of the Utah Bxperiment Station
in furnishing these records.
Feb. 4, 1918 Variability of yields of Fruit Trees and Field Trials 257
RANGE OF INDIVIDUAIv TREE YIELDS
The extremes of individual tree productivity are shown in Table I.
This indicates a wide range of variation even in oranges, lemons, and
apples, which are clonal varieties. The greatest range, however, is in
the yield of the seedling walnuts. The variations set forth in this table
may seem excessive to workers with annual crops. To those familiar
with the variation in tree crops, however, this will be recognized as only
the normal variation which occurs in most fruit plantations growing on
apparently uniform soil, as mentioned in a previous section. Wide
variation in the vigor of the rootstock, as well as variation in soil pro-
FlG. 6. — Diagram showing the individual tree yield (in pounds) of the Johnathan
apple orchard.
Yield per tree (pounds)
Row.
Total yield per row 7, 082 8, 070 8, 469
387
187
337
42s
42s
100
325
I2S
375
300
512
162
275
175
312
I2S
300
250
350
187
300
200
250
187
112
SO
187
162
250
200
112
437
450
212
437
350
400
450
487
312
375
462
387
400
375
200
175
15°
300
187
100
200
287
125
100
ISO
550
175
575
400
425
350
337
262
350
312
212
250
50
362
312
375
250
250
450
5°
262
387
337
337
175
200
287
187
362
400
S50
450
375
575
375
437
412
250
325
37
286
387
350
ISO
312
250
261
350
375
362
362
337
32s
375
SO
200
212
350
300
250
52s
337
SCO
212
550
525
425
537
47S
187
262
375
1 75
7S
287
375
325
42s
312
450
32s
437
87
SO
437
100
550
362
375
137
S50
32s
450
337
500
300
275
337
337
375
375
300
87
100
42s
200
100
300
200
275
ISO
7S
8,334
337
500
475
287
437
337
350
475
437
437
437
375
412
437
500
50
450
262
312
275
350
312
225
200
150
275
325
100
437
375
300
362
375
350
337
300
250
225
212
475
300
337
75
100
425
225
375
187
225
150
125
287
250
Total
yield
per
row.
2,747
2,324
3,161
3,048
3,387
2,348
3,236
2,561
3,324
2, 223
2,087
2,559
2,935
2,150
2.S74
1,662
2,548
1. 912
2,562
2,448
1,873
2,236
1,724
1,862
1,473
1,174
68. 284
ductivity, may have been instrumental in causing such a variation in
yield.
Table I. — Range of variability in crop production 0/ fruit and nut trees
[Extremes and range expressed as percentages of the mean yields of the respective plantations]
Kind of fruit.
Naval oranges (Arlington) . .
Naval oranges (Antelope
Heights)
Valencia oranges
Eureka lemons
Seedling walnuts
Jonathan apples
Mean yield.
Pounds.
i37±i-6
i86±i. 7
246 ±4. 3
270±2. 9
86±i. 6
304 ±5- 6
Extreme yield.
Per cent.
9. I to 246. 3
. 4 to 193. 8
. I to 188. o
. 4 to 180. 3
. 4 to 276. I
- 3 to 193. 2
Range.
Mode.
Percent
237.2
Pounds.
129.8
160. 4
182.9
186. I
270.8
147.9
261 7
180. 9
299.9
75-6
345-1
Skewness.
o. 143 ±0.025
. 001 ±
• 25o±
•359±
. 269 ± . 049
•329± -055
037
053
043
258
Journal of Agricultural Research
Vol. XII, No. 5
The biometrical constants for the several plantations are given in
Table II. The oranges, lemons, and apples, as might be expected, show
less variability than the seedling walnuts. The coefficient of variability
of the clonal varieties ranges from 29.72 to 41.23 per cent. This total
range of only 11. 51 per cent shows a marked similarity of the extent of
variation.
Table II. — Variability in yield of the different individual fruit trees
Kind of fruit.
Naval orange (Arling-
ton
Naval orange (Ante-
lope Heights)
Valencia orange
Eureka lemon
Seedling walnut
Seedling walnut
Seedling walnut, aver-
age
Jonathan apple
Acre-
Total
Crop.
age
num-
Mean yield
per
ber of
per tree.
tree.
tree^
Pounds.
191S
0.011
1,000
137. 6±i. 2
1916
.on
495
186. 2 ±1.7
191S
. on
240
246. 3 ±4. 3
191S
.013
364
270. 7 ±2- 9
191S
•OS7
280
99-8±i.9
1916
.057
280
77-6±l. 7
(1915
1
]and
i -057
280
86.4±i-6
[1916
1
1914
.oil
224
303.9±s-6
Standard
deviation.
Pounds.
54- 42 ± 0.8a
55- 33 ± I- 19
97- 84±3- 01
81.38i2.03
47-77±l-36
41-94±I-I9
40. io± 1. 14
125. 30±4. CO
Coefficient of
variability.
39- 55 ±o- 68
29. 72±o. 69
39- 72 ± I- 40
30. o6±o. 8i
47. 86± I. 64
53- 91 ±1-92
46. 41 ±1-58
4i-23±l. S2
Probable error.
Pounds
per tree.
Percent-
age of
mean.
26.67
20.05
26.79
20. 28
32-28
31-30
27-81
The probable error, expressed in pounds of fruit per tree, is the greatest
in case of the Jonathan apple, amounting to 85 pounds, while the Valencia
orange and the Eureka lemon fall to 66 and 55 pounds, respectively.
Such probable errors, expressed in pounds per tree, are not comparable,
however, unless the mean yields are approximately the same. The
probable error expressed as a percentage of the mean is therefore added
to Table II to make it more easily compared with tables of other writers
who have seen fit to use this constant rather than the coefficient of
variability.
METHODS OF CALCULATING VARIABILITY
The yields of the various fruit plantations have been studied, with
trees singly and combined into plots of various sizes. The coefficient of
variability and probable error have been used as the basis of comparison
in most cases.
Plots of different sizes necessarily have varying mean yields per plot;
therefore the coefficient of variability is more readily interpreted than
the standard deviation. The probable error may only be used with
accuracy in cases where the number of variants is relatively large and
their distributions normal.
VARIABILITY OF ORCHARD PLOT YIELDS
EFFECT OF INCREASING THE NUMBER OF ADJACENT TREES PER PLOT
The first point studied was the effect of increasing the number of
adjacent trees per plot, measured by the coefficient of variabihty. Based
on the theory of random samphng of variables, the average production of 10,
or even 5, trees should be a more typical sample of the orchard than that
Feb.4.i9i8 Variability of Yields of Fruit Trees and Field Trials 259
of I tree. The reduction of the coefficient of variability by combining a
number of adjacent trees in a plot would, however, be expected to fall
short of the theoretical reduction, because such a combination may have
a tendency to group trees of similar productivity together. Gradual soil
variation from one side of the plantation to the other, or irregularities of
the field which are larger than the area taken up by a single tree, will tend
to bring about a correlation between the yields of adjacent trees.
For practical purposes the two more or less antagonistic sources of
variation between plots may be arranged in two groups :
1. Those which may cause the variations to become greater as the size
of the plot increases — ^for example, variation in soil productivity.
2. Those which may cause the variations to become less as the size of
the plot increases — for example, variations in inherent producti\-eness
of the trees. This may depend to no small degree on the variation in
vigor and character of growth of the rootstock. Measured by crop pro-
duction, it may be practically impossible with grafted or budded trees to
separate by mere inspection the variation which may be caused in
inherent qualities of the bud from those of the rootstock on which it is
propagated.
The reduction of the coefficient of variability in the several plantations
as a result of increasing the number of adjacent trees per plot is shown in
Table III. The acreage per plot is recorded for sake of comparison with
similar work by agronomists, where the size of the plots studied has been
dependent entirely on acreage rather than number of plants to the plot.
Other biometrical constants are likewise included for ease of comparison
with above-mentioned studies.
The effect of increasing the number of adjacent trees per plot on
reducing the coefficient of variability between the plots of all the fruit
crops studied is shown in the summary of Table III, and figure 7 shows
the same thing graphically. The curves show a marked similarity be-
tween the varieties of fruits and agree quite closely in demonstrating
that there is little to be gained in including more than eight adjacent
trees in a plot. As a rule, there is a rapid reduction in the coefficient,
as progress is made from a i-tree to an 8-tree plot. Increasing the plot
above eight .adjacent trees shows only a comparatively small reduction
of the coefficient of variability. In fact, the reduction is not significant
when the probable errors ^ are considered. The Antelope navels and
apples show a reduction slightly less than the probable error between
a 4-tree and an 8-tree plot. Again, the lemons show an apparently
exceptional reduction when the i6-tree plot is compared with the 24-tree
plot. The same is true of the 8- and i6-tree plots of apple trees. These
exceptions are in part explained, where they concern the larger plots,
' The probable error of the difference between two averages A i and A 2, of which the probable errors
El and £2 are known, is the square root of the sum of the squared probable errors; or probable difference
of /li-/l2=±y£i+£j
26o
Journal of Agricultural Research
Vol. XII, No. s
by the fact that in a given area, as the size of the plot increases the
number of plots necessarily decreases, and thus lessens the reliability
of the comparisons. Thus, these exceptions may be due to chance in a
small population and might not hold true with a larger number of variants.
Table III. — Effect of increasing the number of adjacent trees per plot
Kind of tree.
Navel oranges (Ar-
lington)
Do
Do
Do
Do
Do
Navel oranges (An-
telope Heights)
Do
Do
Do
Do
Do
Valencia oranges
Do
Do
Do
Do
Do
Eureka lemons.
Do
Do
Do
Do
Do
Num-
ber of
trees
per
plot.
Seedling walnuts.
Do
Do
Do
Do
Do
Jonathan apples. .
Do
Do
Do
Do
Do
Num-
ber of
plots.
,000
500
250
125
60
40
495
247
125
61
30
20
240
120
60
30
364
280
140
224
112
56
28
14
Acre-
age per
plot.
o. on
. 022
.044
.088
.176
. 264
. on
. 022
.044
.088
.176
. 264
.011
. 022
.044
Mean yield
per plot.
Pounds.
137- 6i 1.2
275. 6i 2.9
551. 6i 7.4
1, 100. 8i 19. 2
2, 220. oi 56. o
3.343-8i 92.5
186. 2i 1. 7
373- oi 4-2
742. 7i 10. o
1,460. 7± 26. 7
2,973. 3 ± 68.4
4,450. oi 121. 6
246.3 i 4.3
487. 92 i 10. O
991- 7 i 23. 9
1,966. 7 i 59-4
176 3,813.3 ii58.6
264 5,880.0 i243. o
•013
. 026
. 211
•317
.057
. 114
. 228
.456
. 912
1.368
.oil
. 022
.044
.088
. 176
. 264
270. 7 i 2. 9
544. 3 ± 6. 8
1,081. o ± 17. 3
2, 172. o i 46. 7
4,395-o ii2i. 7
6,692.0 ii53. 5
86.4 i 1.6
178. 7 i 3. 6
353- 6 i 8. 6
717. I i 20. 7
1,409. 4 i 58. O
2,154.6 il03.3
303-9 ± S-6
6og. 8 i 12. 3
I, 210. 7 i 28. 8
2,414.3 i 76.9
4,864.3 ii75-S
7,277.8 i3l3-9
Standard
deviation.
Pounds.
54. 42i 0.82
95. 6oi 2. 03
173- 89i 5- 24
319. ooi 13.60
642. 6oi 39. 58
867. 70i 65.43
S5-33± I- 19
97. 2ii 2.95
164. 37± 7.07
309. 05 i 18.90
555- 55 i 48.38
806. 54i 86.01
97- 84i 3- 01
161. 8oi 7- 04
275. 2oi 16. 92
482. 6oi 42. 02
910. 58iii2. 13
i,l40.ooil7i.90
81. 38i 2. 03
135- 40i 4- 78
245. 35 i 12. 27
448. 50i 33. 01
827. ooi 86. 07
820. 30i 108. 50
40. io± 1. 14
62. 40i 2. 52
106. 40 i 6. 06
181. 24i 14. 60
354- 40i 41.00
508. ooi 73.05
125. 30i 4- 00
193. ooi 10. 44
319. 40 i 20. 40
603. 38i 54.38
973. 68 i 124. II
1,396. ooi 221.90
CoelBcient
of
variability.
39.55±o. 68
34.68io. 82
31. 52ii.o5
28. 98ii.30
28. 95 i I. 92
2S-95±2.oS
29. 72 ±0.69
26. o6±o. 84
22. I3il. 00
21. i6ii. 35
18. 68ii.68
18. ioil.91
39- 72 ±1.40
33- i6ii.6i
27. 75ii.84
24. 54i2. 26
23. 88i3. 10
I9-39±3.03
30. 06 io. 81
24. 88 i 0.93
22. 70il. 19
20. 65 i I. 57
18. 82 i 2. 02
12. 26ii. 65
46. 41 i I. 58
34-92ii. 57
30. 09ii. 86
25. 27i2. 16
25. I5i3. 07
23- S8i3. 53
41. 23il. 52
31.65i1.87
26. 38il. 80
24. 99i2. 39
20. 02 i 2. 65
19. i8i3. 16
Probable error.
Pounds. f^J-
' centage
of
mean.
per
tree.
26.67
23-39
21. 26
19.55
19.53
17- SO
20.05
17-58
14-93
14.27
12.60
12. 21
26.79
22.37
18.72
16.5s
16. II
13-08
20. 28
16.78
IS- 31
13-93
12.69
8.27
31.30
23- ss
20.30
17.04
16.96
15.90
27.81
21- 3S
17.79
16.86
13.50
12.94
SUMMARY
Coefficient of variability.
A^-era
tiou ol
^'e reduc-
coetiicient
Ntmi-
ber of
trees
Navel oranges.
Valencia
oranges.
Eureka
lemons.
Seedling
walnuts.
Jonathan
apples.
Average.
of variability by
increasing num-
ber of adjacent
trees per plot.
per
plot.
Arlington.
Antelope
Heights.
In-
crease
from —
Average
reduc-
tion.
39. 55 io. 68
34.68io.82
31.52i1.05
28.98i1.30
28.9si1.92
25-9S±2-o8
29. 72 io. 69
26. ooi 0.84
22. 13 i I. 00
21.l6il.35
i8.68ii.68
18. ioii.91
39-72ii-40
33,. l6ii.6i
27. 75ii.84
24. 54i2. 26
23.88i3. 10
i9.39i3-03
30. o6io. 81
24.88io.93
22. 7oii. 19
20.65i1.57
18. 82 i 2. 02
12.26i1.6s
46. 4iii.58
34.92i1.57
30. 09ii. 86
25. 27i2. 16
25-l5±3-07
23-s8i3-S3
4l-23il.S2
31.65il.87
26.38il.80
24.99i2. 39
20. 02 i 2. 65
19. l8i3. 16
37-78io. 52
30. 89io. 55
26. 76 io. 62
24. 27io. 77
22. 58ii.oi
19. 74ii.o8
2
4
8.
16
24
1 to 2
2 to 4
4 to 8
8 to 16
16 to 24
6. 89io. 76
4. 13 i 0.83
2. 49io.99
i.69ii.27
2.84±I<4S
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 261
The averages of all six fruit plantations show that there is a rapid
reduction of the coefficient of variability until the 8-tree plot is reached,
but from then on the reduction is less in comparison with the probable
error. (See Table III, summary, and figure 8.)
'^2
"M
38
36
3^
32
3C
23
2e
\22
K
\36\
\7.
\30
\2S
V''
V) 22
\^20
\ /6
^3S
3^^
32
30
23
26
2^
22
20
/3
\^o
^2
38
36
3^
32
30
28
26
2^
c/OA^^r///^// ,^^/:>/.£'
^£'£'£>z//^(^ /y^^/v<yr
/2^
S /6 2^^
/^•^ ^ /6 2^^
Fjg. 7. — Graphs of the reduction of the coefficient of variability by increasing the number of adjacent trees
to the plot.
Although the average reduction of the coefficient between the 8-and
i6-tree plots is doubtful when compared with the probable error, this
reduction in all six cases is constant — that is, the variation occurs in
26:
Journal of Agricultural Research
Vcl. XII, No. s
one direction only, and therefore has more significance than is indicated
by a mere comparison of the averages.
EFFECT OF SYSTEMATIC DISTRIBUTION OF PI.OTS OVER THE AREA STUDIED
The importance of distributing plots over the experimental area is
more or less obvious, and has been dwelt upon by many writers. Its
value arises from the fact that the soil varies over the area, and it is
better to have similar-sized plots on both high- and low-yielding areas
than to have them
solely on one or the
other kind of soil.
The method should be
of special value on
areas which vary
rather uniformly in
one direction.
Increasing the num-
ber of trees to the plot
k^^" \ \'^'\^ in scattered units of
^^y- ,s> T«L _ either four or eight
trees gives a more
typical sample of the
productivity of the
total planting than
the same number of
adjacent trees. In
scattering the plots
throughout the area
studied, they were sys-
tematically repeated.
For example, if there
were loo plots in all to
be grouped in pairs,
the first and fifty-first,
and the second and fifty-second were united, and so on through the series.
If a' quadruple series was desired, the first, twenty-sixth, fifty-first, and
seventy-sixth plots were combined.
Table IV shows the results of scattering 4- and 8-tree plots, respectively,
in the plantations studied. Figure 8 illustrates the reduction of the
coefficient of variability by increasing the number of trees to the plot in *.
both 4- and 8-tree scattered units, compared with the average coefficient
of variability for the several fruits by increasing the size of a plot from i
to 24 adjacent trees, together with the theoretical curve calculated from
the mean coefficient of variability of all the i-tree units.
A comparison of the curve for adjacent trees and those for scattered
units shows at once the marked decrease in favor of the scattered units.
Fig. 8.
—Graphs of the reduction of the coefficient of variability by
increasing the number of trees to the plot.
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 263
Table IV. — Effect of increasing number of trees per plot in scattered units
FOUR TREES IN A UNIT
Total
num-
ber of
trees
per
plot.
Coefficient of variability.
Unit
No.
Navel oranges.
Valencia
oranges.
Eureka
lemons.
Seedling
walnuts.
Jonathan
apples.
Arlington.
Antelope
Heights.
Average.
I. . . .
2. . . .
3...-
4....
S
6....
4
8
12
16
20
24
31.52i1.00
18. 59±o. 82
15. ilio. 81
10. 76 io. 66
11. isio. 76
9. 89 i 0.74
22. I3il. 00
l3.o6io. 85
li.07±o.88
9. S2±o.87
S- 74±o. 58
6. 38±o. 72
27.75±i-8o
16. 04ii. 43
16. 48ii. 81
10. 39±i. 29
9.28±I. 28
13. 16 i 2. 02
22. 7oii. 10
14. 76il. 07
I2.6iii. 16
8. Olio. 81
9. 48ii. 07
S-38io.66
30.09ii. 8s
12.36i1.01
14. I7il. 41
8. 59 i 0.99
6. 97 i 0.89
10. OS ± 1.4s
26. 38 i I. 80
15. 89±i.47
12.04i1.37
8. 44ii. 08
7. 7S±i-i3
6. i2±o. 97
26. 76 i 0.60
15. I2io. 47
13. 58io. S3
9. 29io. 40
8. 40 i 0.40
8. 49 i 0.49
EIGHT TREES IN A UNIT
28. 98ii.30
17. 49±I09
II. 78io. 89
14- 85 i I. 30
10. O9io. 96
21. i6±i. 35
24. 54i2. 26
20. 6sii. 57
9. 2iio. 83
15. 22il. 92
10. 98ii. 17
7. 44io. 84
X4.34±2. 21
II. 7oii. 44
8. 87ii. 14
8.72±i.57
7-56ii. 16
8. 6oi I. 24
5- 24ii. 02
7. 8i±i.32
25. 27i2. 16
10. 94il. 27
11. 3S±I. 63
6.68ii. 13
II. oii2. 01
24. 99i2. 39
13. 20il. 69
10. 99il. 7S
10. 54il. 90
4-94±I-0S
24. 27io. 77
12. 84io. s6
II. 27io. 63
9. S4±o. 57
7-9S±o.S4
The 4-tree unit apparently gives a greater degree of accuracy than the
8-tree unit with the same total number of trees. This point is clearly
shown by the curves. In considering a total number of either 8, 16,
or 24 trees the curve for the 4-tree unit approaches more nearly the
theoretical curve than the curve of the 8-tree unit. With a total num-
ber of 24 trees, for example, the 4-tree units would be scattered about
regularly in six different places, while the 8-tree units would be located
in three different places. The larger number of unit plots thus gives a
more typical sample of the area than half as many units with twice the
number of trees in a unit.
In combining both the 4- and the 8-tree units in the regular scattering
of ultimate plots, an attempt was made to throw both high- and low-
yielding small plots into a combination, although a systematic distri-
bution was maintained. The fact that the curve of the 4-tree units
drops below the theoretical in one place indicates that this attempt
was successful. An ordinary regular scattering of the ultimate plots
might not approach the theoretical curve of random sampling so closely
if a knowledge of the relative productivity of the soil was not available
before arranging the plots. A 4-tree unit might not be practical in
tests of cultural methods, fertilizer, or irrigation trials. In such trials
an 8-tree plot is usually the smallest practical unit. In the case of
walnuts, however, which should be planted at least 50 by 50 feet, a
4-tree unit in a single row would allow for a space 50 feet wide by 200
feet long, and, if guard rows were planted between the experimental
trees, the plots would be 100 by 250 feet, a very practical size upon
which to handle orchard machinery. For the trial of rootstocks, prun-
ing experiments, variety tests, etc., the 4-tree plot is a practical-sized
27808°--18 3
264 Journal of Agricultural Research voi. xii, no. s
unit and could be expected to give more reliable results if repeated at
four regularly-placed intervals than either two 8-tree units, or 16 ad-
jacent trees — that is, such a regular scattering of the several units
which make up the combination plot reduces the error of the final com-
parisons which is caused by the variation in soil productivity.
The fact that marked soil variations occur which tend to make ad-
jacent trees or adjacent plots yield alike, even on soils which were
chosen because of their apparent uniformity, is well shown by the work
of Harris (19 13). The criterion for the measurement of such variability
proposed by this author is the coefficient of correlation between neigh-
boring plots of the field.* Applying this to the Arlington navel oranges,
the writers have calculated the correlation between the yield of the
8-tree plot as the ultimate unit, and the yield of the combination of
four such adjacent plots and it was found that
r= +0.533 ±0.085.
This shows a marked correlation, indicating a pronounced hetero-
geneity in the soil of this grove, influencing fruit production.
However, when we calculate the correlation between the 8-tree plot as
the ultimate unit and the yield of the combination of four such system-
atically scattered plots, it is found that —
r= -f o.i37±o.i20
This coefficient is practically equal to its probable error and can be
regarded as significantly zero. This is merely another means of calcu-
lating the value of scattering a 32-tree plot in four ultimate plots of 8
trees each rather than selecting 32 adjacent trees.
DEGREE OF ACCURACY EXPECTED WITH A PLOT OF A GIVEN SIZE
Assuming, for example, that experimental plots have been laid out
in the navel oranges (Arlington) with a total of 32 trees to the plot in
four scattered units of eight trees each, the question might logically be
asked, "What differences in the yields of such plots can safely be at-
tributed to differential treatment as different methods of irrigation or
fertilization, and what may probably be due to mere chance because of
soil heterogeneity and the fluctuating variation of the trees?"
Table IV shows a coefficient of variability of 14.85 ±1.30 in this
plantation laid out in 32-tree plots of four scattered units of eight trees
each. The probable error,^ then, in this example, that such a plot of
32 trees is typical of the area in question, is 14.85X0.6745= ± 10.02
• The formula used is , 2
r_ . <[S(C P)-S(P^)]lm[n(n-j)]}-p
where />== yield of an individual plot; ot= number of larger plots, each made up of n contiguous ultimate
units; Cp= yield of the larger combination plots; S= summation of the yields of all the ultimate or com-
bination plots of the field.
* The probable error of a single variant of a population may be defined as that departure from the mean
on either side, within which exactly one-half of the variants are found. Expressed as a percentage of
the mean, it is determined by multiplying the coefficient of variabihty by 0.6745.
Feb. 4, 1918 Variability of Yields of Fruit Trees and Field Trials 265
per cent of the mean production. That is, the chances are even that any
plot as described, of 32 trees, will fall within ±10.02 per cent above or
below the true mean. The chances are equally as good that such a plot
will not fall within the accuracy of ± 10.02 per cent of the mean. In
comparing two such plots, both with the same probable error of ± 10.02
per cent, the probable error of such a comparison will be greater than
the probable error of one — that is, it will be equal to ±10.02 per cent
xV2=± 14-17 per cent. Therefore, if plots undergoing differential
treatment vary from each other by only ±14.17 per cent of the mean
of the plantation, half the time such differences in yield may be due to
the treatment, and half the time they may be due to casual variation.
It is clear, then, if fertihzer or irrigation experiments laid out in such
plots differ from each other by only 14.17 per cent of the mean produc-
tion of the total area, we are not assured beyond an even chance that the
difference is a real difference due to the factors which are being experi-
mented upon. So slight an assurance can hardly be expected to be
sufficiently reliable to prompt a farmer to purchase fertilizer, to change
his method of irrigation, or to undertake any new business; much less
will this assurance justify an experimenter in drawing conclusions from
the result of a field trial.
Our present knowledge of orchard fertilization in the arid West will
hardly justify any assumption on our part more reliable than an even
chance that one fertilizer will produce an increased yield of fruit com-
pared with another, or even cause an increase over an untreated plot.
The same thing may be said in comparing different methods of irrigation.
In most cultural trials we would therefore be comparing two results
where the difference may occur in either direction. (Tables V and VI. )^
Table V. — Table of odds for differences which may occur in cither direction
Difference be-
Difference
tween two re-
Odds against such
from the mean
sults in terms
difference occurring
in terms of
of probable
under uniforoi condi-
probable error.
error of each
result.
tions.
I. 00
I. 41
I to 1
1-25
1.76
3 to 2
1.44
2.03
2 to I
I. 71
2.41
3 to I
I. 90
2.68
4 to I
2. 00
2.83
9 to 2
2.05
2.87
5 to I
2. 50
3-53
10 to I
2-93
4- 13
20 to I
3.00
4. 24
22 to I
3.20
4-51
30 to I
4. 00
5.66
140 to I
4.90
6-93
1 ,00c to I
5. 00
7.07
1,350 to I
' Tables V and VI are taken from the writings of Wood (igii) who in turn adopted thexn "from one of
the standard reference books on astronomy. "
266
Journal of Agricultural Research
Vol. XII, No. s
Tabi<E VI. — Table of odds for differences which may occur in one direction only
Difference be-
Difference
tween two Re-
Odds against such
from the mean
sults in terms
difference occurring
in terms of
of probable
under uniform condi-
probable error.
error of each
result.
tions.
I. OO
1.41
3 to I
I. 25
1.76
4 to I
1.44
2.03
5 to I
1.58
2.23
6 to I
I. 71
2.41
7 to I
I. 81
2.55
8 to I
I. 90
2.68
9 to I
2. 00
2.83
10 to I
2.48
3-50
20 to I
2. 70
3.81
30 to I
2.89
4.07
40 to I
3.00
4.24
44 to I
3-03
4. 28
50 to I
3-44
4-85
100 to I
4. 00
5.66
290 to I
5.00
7.07
2 , 700 to I
On assuming that a lo-to-i chance is a reasonable assurance, the ques-
tion logically arises, "What difference between any two plots must be
manifested for this degree of confidence that the difference is due to
treatments applied?" By referring to columns 2 and 3, Table V, one
may find the difference in terms of the probable error which is necessary
between two results to obtain this degree of reliability. Here it can
be seen in column 2 that there must be a difference 3.53 times the prob-
able error to give the odds of 10 to i in column 3 against such a difference
occurring under uniform conditions. Thus we find in this example that
the difference between two 32 -tree plots in four scattered units must be
at least 14.17 per cent X 3.53 = 50.02 per cent of the mean production,
to give the assurance of a lo-to-i chance that the difference is due to
fertilizer, irrigation, or whatever factors are under consideration.^
Even with this difference, conclusions based on such results obtained in
this navel orange (Arlington) grove may be correct 10 times out of 11
and wrong once out of 11 times.
On turning now to the 32-tree plot of adjacent trees with navel oranges
(Arlington), it is seen the probable error is 16.42 per cent. To pro-
ceed as before, 16.42 per cent, the probable error of one plot, X V^ =
23.22 per cent, the probable error of the difference; 23.22X3.53 = 81.97
per cent. Therefore a difference between two such plots of 81 .97 per cent
of the mean of the total area would be necessary to give the assurance
that such differences are real 10 times out of 11 and due to pure chance
1 If comparisons were being made between two radically different treatments which were known to pro-
duce different effects in fruit production, such as irrigation compared with dry farming, or the use of large
quantities of stable manure on light soils, compared with no manuring, then reference should have been
made to columns 2 and 3 in Table VI, which are based on the differences occurring in one direction only.,
Feb. 4, 191 8 Variability of Yields of Fruit Trees and Field Trials 267
only once out of 11 times; yet this grove was chosen because of its ap-
parent regularity, for it has been judged sufficiently uniform for plot trials.
The point might be justly raised that the small number of plots in-
volved in the above calculations are not sufficient to give the laws of
chance a fair opportunity of asserting themselves. On laying the area out
in plots of adjacent trees there were 30 plots, while made up of scattered
units there were 31 plots. Figure 9 shows the distribution of these plots,
together with the theoretical cur\^e, which was caclulated for the scat-
tered-unit curve. The scattered-unit curve closely approaches the theo-
retical normal curve
of errors, and there- I ' ■"
fore reliance can be
placed upon its
probable error.
In the case of the
plots of adjacent
trees, however, the
30 units are not suf-
ficient to give the
laws of chance fair
play. Table VII
sums up the results
of the foregoing cal-
culations, adding ex-
treme and mean
yields of the two dif-
ferent types of plots
and the theoretical
probable error. The theoretical probable error based on the theory of
random sampling for a hypothetical 32-tree plot is the probable error of
one tree 26.67-^ -^32 =4-71 P^'^ cent. This is readily calculated from
the distribution of the yields on a one-tree unit, the curve of which is
shown by figure 10. The large number of trees involved, even though
the distribution is not normal, justifies the use of the probable error as a
minimum probable error. Based upon the theory of random sampling,
two hypothetical 32-tree plots with a probable error of 4.71 per cent
should show a minimum difference of (4.71 X V 2 X 3.53) = 23.51 per cent
to give an assurance of a lo-to-i chance that such a difference is real
and not due to casual variation. Therefore, if the calculations in Table
VII based on adjacent trees can not be fully relied upon because of the
small number (30) in the population and because their distribution is not
normal, we may at least reasonably expect that the necessary difference
between two such plots will fall between the theoretical 23.51 per cent
and 81.97 with a practical certainty that it will be greater than 50.02 per
cent of the mean.
2 ^ -F ^ e 7- e
Fig. 9. — Graphs of production, 32-tree plot, navel oranges (Arlington).
Scattered in four 8-tree units.
Adjacent»32-tree units.
268
Journal of Agricultural Research
Vol. XII, No. s
Tabl^ VII. — Comparison of the reliability of a plot of 32 adjacent trees with that of a J2-
tree plot of four scattered units of 8 trees each. Navel oranges {Arlington)
E.Ktreme yields
per plot.
Mean yield
per plot.
Standard
deviation.
Probable
error per
plot (per-
centage
of mean).
Difference necessary
to give lo-to-i as-
surance.
Plot.
Percent-
age of
the mean.
Observed
yield.
Pounds.
2,500 to 7,500
3,000 to 6,000
Pounds.
4,367±I3I
4,484± 81
Pounds.
i,o63±93
666±57
±16. 42
± 10. 02
±81.97
±50.02
Pounds.
±3.s8o
±2.243
Theoretical (based on ran-
±4- 71
±23-51
± 1 1 03s
On turning now to the navel oranges (Antelope Heights), which the
calculations show to be the most uniform planting of any observed, the
question might arise
as to what degree of
reliability may be ex-
pected in comparing
two i6-tree plots.
Table VIII shows a
comparison between
such plots made up of
adjacent trees and
plots composed of
scattered units of four
trees each together
with the theoretical.
A much greater range
is found with the plot
of adjacent trees, while
the necessary differ-
ence between two
plots for a 10- to- 1 as-
surance is 62.94 per
cent. With plots
made up of scattered
units the necessary
difference is 32.05 per
cent. The hetero-
!0 ^ 10
f\j W^ K
\
Fig. 10.— Curve of yields of individual trees, navel orange (Arlington).
geneity of this soil is clearly shown by both the increase in range
and the increased probable error when plots are composed of adja-
cent trees. This block of trees, however, appeals to the observer as
unusually uniform and would be considered desirable for plot experimen-
tation, the fluctuation in the productivity of the trees approaching closely
Feb. 4, 1918 Variability of Yields of Fruit Trees and Field Trials 269
the normal curve (see fig. 11); the mean production per tree is 186.2 ±
1.7 pounds, standard deviation 55-33±ii9, coefficient of variability
29.72 ±0.69 per cent, probable error 20.05 per cent of the mean, skewTiess
o.ooi ±0.037. Never-
^ r
theless, if devoted to
plot experiments, a
difference between
two plots of 16 adja-
cent trees each, of
even 62.94 per cent of
the mean production,
might be due to differ-
ential treatment 10
times out of 11 and
due to casual varia-
tions of soil and trees
once out of 11. The
calculations sound a
note of warning
against drawing con-
FiG. II.— Curve of yield of individual trees, navel orange (Antelope
Heights).
elusions between such plots if the differences are less than 50 per cent of
the mean production of the plantation, provided we wish to have such
conclusions as dependable as a lo-to-i chance.
Table VIII. — Comparison of the reliability of a plot of i6 adjacent trees with that of a
plot of 16 trees of four scattered units of 4 trees each. Navel oranges {A ntelope Heights)
Plot
Extreme yields
per plot.
Mean yield
per plot.
Standard
deviation.
Probable
error per
plot (per-
centage
of mean).
Difference necessary
to give lo-to-i
assurance.
1
Percent- ^, ,
age of 0^?:"'^'l
the mean. y^'-^-
1
Pounds.
1,800 to 4,000
2,400 to 3,600
Pounds.
2,973±68
2,96o±36
Pounds.
556±48
282±25
±12.61
± 6.42
±62. 94
±32-05
Pounds.
±1.871
± 949
± S-oi
±-'5-03
± 746
An example taken from the Jonathan apple orchard may well be con-
sidered. Suppose it is desired to know the necessary difference which
must exist between plots of 16 trees each to give us the reliance of a lo-
to-i chance that it is due to differential treatment (a) when the plots are
made up of adjacent trees, (b) when the plots are made up of two scattered
units of eight trees each, and (c) when the plots are made up of four
scattered units of four trees each.
The probable errors in the above cases, a, b, and c, are ± 13.49, ±8.90,
and ±5.69 per cent, respectively. By proceeding as before it is found
that the necessary difference for a lo-to-i chance is as follows: (a)
270
Journal of Agricultural Research
Vol. XII, No. s
±67.35 per cent; (b) ±44.44 per cent; (c) ±28.42 per cent. Table IX
summarizes the results. It seems probable, therefore, that a difference
between two 16 adjacent tree plots of less than 50 per cent of the mean
production should be considered with caution before attributing it to
differential treatment. The scattering of the units of the plots increased
the accuracy very decidedly, four units giving more accurate comparison
than two. Even with this scattering, differences of less than 30 or 40
per cent are well within the realm of chance. The apparent cause for the
4-unit plot having a probable error less than the theoretical is accounted
for by the fact that the variation in productivity of the soil was known
when the distribution of the units was made. The results might not have
approached so closely to the theoretical if the distribution had been
decided upon before harvest.
Table IX. — Comparison of the reliability of a plot of 16 adjacent trees with that of two
units of 8 trees each and of four units of 4 trees each. Jonathan apples
Extreme yields
per plot-
Mean yield
per plot.
Standard
deviation.
Probable
error per
plot (per-
centage
ofmean)-
Difference necessary
to give lo-to-i
assurance.
Percent-
age of
the mean.
Observed
yield.
Pounds.
3,000 to 6,500
3,900 to 6,100
3,800 to 5,400
Pounds.
4,864±i76
4,857±ii5
4,87i± 74
Pounds.
974±I24
641 ± 82
4II± 52
±13-49
± 8-90
± 5-69
±67-35
±44-44
±28.42
Pounds.
±2,158
± I , 384
Theoretical (based on
± 6-9S
±34- 70
±1,687
The large probable errors which are apparently always present with
plot trials of fruit trees emphasize the importance and value of obtaining
individual tree records of experimental orchards before differential
treatment is started. The probable errors will likely differ somewhat
from year to year, and possibly be further influenced by the advanced
age of the trees. Nevertheless, if but one or two years ' records of mature
trees are available before differential treatment is commenced, at least
some idea can be obtained of the casual variation of the plots — that is,
limitations can be placed beyond which observed differences in plots
may be due to chance rather than to the factors under experimentation.
In the absence of such previous records, the employment of frequent
controls or standard treatments may be indicative of the probable error
of the entire area.
REIyATlON OF THE SHAPE OF THE PLOT TO THE VARIABIUTY OF THE
COMPARATIVE YIELDS
The shape of small plots may be of great importance when cultural
operations are considered.
Lyon {igi2) found no satisfactory evidence that long and narrow
plots are less likely to error than square plots when no control plots are
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 271
used, but when controls are placed every second or third plot in the
series, the evidence is in favor of the long and narrow plots. The use of
every second or third plot in the series as a control plot obtains greater
accuracy than when no controls are used and the average of the field is
considered the normal yield for all plots.
A long plot is much more economical of time and labor than a square
plot containing the same number of trees. If the orchard must be irri-
gated, a square plot containing nine trees will require three standpipes
instead of one and also three times as many irrigation furrows.
Arguments have been advanced from time to time in favor of both
the linear and the square plot. The advocates of the former have
urged its reliability on two points: first, if the soil or other conditions
change in a direction parallel to the plot, it will contain both high- and
low-yielding areas, and the average will correct one or the other error;
second, if conditions change in a direction more or less perpendicular to
the plots, each plot will vary from its adjoining plot because of its shape
by such a minimum that intercomparisons are more reliable. On the
other hand, the advocates of the square plot have claimed that the
arguments apply equally well to their case, provided the plot be made
small enough.
In view of the importance of the shape of the plot for cultural opera-
tions, the writers have investigated the variability of plots of various
sizes and shapes in three of the experimental orchards. The results
(Table X) are for unrepeated plots and for one year's yield of fruit,
except the apples, which are for two years. Considerable interest lies
in the computation on the 9-tree plot in the total navel-orange grove,
because it is based on a large population (i,cxx) trees) and because the
soil is known to vary in a general way from northeast to southwest.
Comparisons were made between square plots containing 9 trees and
linear plots consisting of a single row of 9 trees extending in the
north-south and in the east-west direction. It is interesting to see how
closely the coefficients of variability coincide in the three cases. The
differences between the coefficients are only a fraction of the probable
errors. Therefore on this lot of i ,000 orange trees there is no difference
between a square and a linear plot of 9 trees, so far as the reliability
of comparative yields is concerned. (The deviations were taken from
the mean of all the plots of the grove.)
Similar comparisons were made on a selected block of 256 trees in this
same grove. The coefficient of variability of the trees taken singly is
34.47 ±1.14, which indicates that these trees were a fairly typical sample
of the entire grove. The block was divided into plots of various shapes
containing 16 trees. With one exception there is little difference in the
variability of the plots, regardless of their shape. This exception is
found in the linear plot of i by 16 trees extending from east to west,
which has a lower coefficient than any other arrangement. In the block
Journal of Agricultural Research
Vol. XII, No. 5
of trees chosen there is a gradual variation of soil in this direction, and
these plots therefore include both high- and low-yielding trees in about
equal proportions, the difference from plot to plot being small, as shown
in Table X.
TabIvE X. — Effect on variability of changing the shape of plots
Kind of trees.
Total navel-orange grove (Arlington)
Do
Do
Portion of navel-orange grove (Arlington),
256 trees
Do
Do
Do
Do
Do
Walnuts, 1915
Do
Do
Do
Do
Do
Do
Apples, 1914-15
Do
Do
Do
Do
Do
Do
Number
of trees
per plot.
16
16
16
16
16
4
4
9
9
9
24
24
16
16
24
24
Shape of plots.
3 by 3
I by 9 N.-S. .
I by 9 E.-W.
1 by I
4 by 4
2 by 8 N.-S. .
2 by 8 E.-W.
I by 16 N.-S.
1 by 16 E.-W
2 by 2
I by 4
3 by 3
I by 9 N.-S. .
I by 9 E.-W .
4 by 6
1 by 24
2 by 4
I by 8 N.-S. .
1 by 8 E.-W.
4 by 4
2 by 8 N.-S. .
4 by 6
I by 24
Coefficient of
variability.
28. i8±l. 48
28. 92 ± I. 46
28. 36±i. 43
34- 47 ± I- 14
16. 77 ±2. 05
17. i7±2. 10
iS.39±i.88
16. 36 ±2. 00
io.09dbi. 31
31. 09±i. 93
30.40±i. 89
23. 94±2. 32
26. 23±2. 57
22. 46±2. 16
21. 62±3. 25
6. 56±i. 00
19. 32±I. 81
18. 7o±i. 74
21. 35±2. 18
15- i3±i-97
16. 4i±2. 15
i3-75±2.36
12. 8i±2. 20
Changing the shape of plots of walnut trees had little effect upon the
coefficient of variability except in the case of the larger plots. The
coefficients vary only slightly from one another, regardless of shape, in
the case of the 4- and 9-tree plots. In the 24-tree plot, however, there
is a great difference between the 4- by 6- tree plot and the i- by 24- tree
plot in favor of the latter. It should be borne in mind, nevertheless,
that a plot of 24 walnut trees is an abnormally large plot, and in the
example just referred to it is a plot 250 feet by 350 feet, compared with
one 100 feet by 1,250 feet. This may take into consideration marked
variation in productivity, even on apparently uniform soil.
In the case of the apple trees there is little difference between the
variability of plots of equal size but different shapes, even in the case
of the 24-tree plots. Regarding the question of the shape of the plot,
it therefore appears that in most cases there is no difference in the varia-
bility of a linear or a square plot.
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 273
proportional to their distance from the plot. This would obviously be
In the case of a large number of trees, where a linear plot is long
enough to include both high- and low-yielding areas in each plot, there
appears to be an advantage in favor of the hnear plot. The great advan-
tage of the hnear plot in most cases is the economy of cultural operations.
Another advantage will be discussed in the following section on the
interpolation of control plots.
A carefully conducted fertilizer experiment with orchard trees requires
a guard row between adjacent plots in order to eliminate the possibility
of the fertilizer's affecting the margin of the next plot, since there are
many observations which show that tree roots extend considerable dis-
tances and often reach the lines of adjoining rows (Hedrick, 1914; Ballan-
tyne, 1916). It is obvious that it is more reliable to employ guard rows
rather than to divide cultural treatments midway between tree rows.
The number of guard trees required for square plots is smaller than for
a linear plot containing the same number of trees. A 9-tree plot in the
form of a square requires 7 guard trees; in the form of a linear plot it ■
requires 1 1 guard trees. In the former case seven-sixteenths of all trees
are in the guard rows, in the latter eleven-twentieths. As the size of the
plot increases, the difference becomes greater. It thus becomes a ques-
tion of the extent to which one is willing to go in enlarging the size of the
linear plot before the increase in the number of trees in guard rows offsets
the economy in cultural operations thus obtained.
USE OF CONTROL PLOTS
INTERPOLATED CONTROL PLOT
Agronomists are in the habit of using every third, fourth, or fifth row
or plot in the experimental tract as a standard from which the normal
yield of any intervening treated plot may be calculated. The nature of
the crops and the cultural methods used commend this system of arrange-
ment.
This method, like others, has its advantages and disadvantages. After
the results are obtained there is still a need for a proper method of com-
parison. There are several different methods of estimating the ' ' normal ' '
yield of any plot. The "normal" (A^) may be estimated by the formalu
Q-fQ-f- ...Cn
n
which is simply the mean of all the control plots in the area. If the
soil of the area were uniform and all variations in the yield of the controls
were purely chance variations, this method would give a satisfactory
result. Again N may be estimated from the yields of the two nearest
control plots. For example, if every third plot is a control and the ar-
rangement is Cj, A, B, Cj, and so on, the normal for A would be
yiC^ + y^C^- In this way the yields of the controls receive weights in-
versely proportional to their distance from the plot. This would ob-
274 Journal of Agricultural Research voi. xii. No. s
viously be satisfactory, provided there was no difference in the amount
of variation between the plants on different plots and that the soil varied
uniformly in one direction.
The first method mentioned computes the ' ' normal ' ' (N) for the whole
area; the second, for a locus on that area. These normals may be com-
bined to represent the resultant of both general and local conditions.
Thus, the formula N=}4 (C+ ^Q + J^Cj) indicates that N is the mean of
the values for N computed by the two preceding formulas. This assumes
an equal value for the adjacent control plots and the mean of all control
plots. Its use as N brings up the calculated yields of plots on low-
yielding areas and reduces the same on high-yielding areas. In the case
of cereals there is usually small chance for difference in the yielding powers
of a plot and its nearest control on account of their proximity; but in
the case of orchard trees situated some distance apart there may be
greater soil changes between adjacent plots, and consequently a marked
difference in yield, aside from the effect of treatment, between a plot and
its nearest control plots. The introduction of the mean of all control
plots might be expected to introduce a stabilizing factor. In the formula
^2 (C+^Ci + J4C2) the mean of all control plots has equal weight with the
normal derived from the nearest controls. Since the soil over any but
very small areas may not be uniformly variable, it might seem more
logical to weight the normal derived from the nearest controls more
heavily than that derived from the mean of all control plots, and to
combine the two. This has been done by Olmstead (1914) and others,
making the formula ^^^ 1 L h — ^^ which p^ and p2 are constants
Pl'p2
arbitrarily chosen. Stockberger {1916) found satisfactory results by
assigning the values pi=i and p2=3-
The method used by the Office of Cereal Investigations, of the Bureau
of Plant Industry, is K(c+Q). which employs half the sum of the mean
of all control plots and the yield of the nearest control plot as the normal
for any given plot.
Stockberger {1916) compared the relative precision of these formulas
in computing the normal yields of plots of hops. Using the yields of
six years he obtained the greatest precision from the formula
P, + P2
though no formula maintained the same relative rank throughout the
six years. It would appear that there is no way of determining in
advance the formula best suited to any particular case, at least not
until more applications of the different formulas have been studied.
The five formulas above stated have been tested on the Arlington
grove of navel oranges. The grove was parceled into linear plots of
10 trees each. Each alternate plot was designated as a guard row and
Feb.4.i9i8 Variability of Yields of Fruit Trees and Field Trials 275
discarded from the calculation. This left 32 "treated" plots whose
"normal" yields were computed. The arrangement of these plots and
their yields are given in Table XI.
Table XI. — Arrangement and yields of lo-tree plots of the navel-orange grove
{Arlington)
[Yield expressed in tens of pounds]
108G 84*
150* 88B
112 control 119*
124* 146 control
121F 151*
159* 174C
142E 132*
148* 182 D
139 control 166*
153* 174 control
177D 180*
179* 169E
176C 131*
188* 162F
183 control 183*
178* 147 control
203B 188*
182* 168G
206A 223*
199* 201H
219 control 191*
191* 228 control
80 control
116*
105*
152 control
88H
184*
115*
155A
117G
164*
94*
154B
109 control
160*
118*
163 control
94F
147*
76*
122C
83E
142*
97*
137D
104 control
III*
lOI*
131 control
99D
148*
117*
154E
looC
165*
112*
134F
74 control
III*
lOI*
114 control
112B
109*
107*
123G
98A
78*
108*
97H
80 control
80*
108*
106 control
108H
137*
no*
135A
Computing the normal yields of these plots by the five formulas above
described gives the results shown in Table XII.
Table XII. — Value of different formulas for computing the coefficient of variability and
probable error of yields of oranges in lO-tree plots in Arlington grove
Formula used.
standard
deviation.
CoelEcient of
variability.
Probable
error ( per-
centaKC of
mean).
I C,4-Co +Cn.
Pounds.
3(>3±3^
I7i±i4
2i9±i8
i8o±is
233±20
26. S5±2. 39
12. 52±I. 07
16. or±i. 38
12. 93 ± I. 11
16. 78±i.46
17.91
8.44
n
2. 2^(7 j_izc
, l4(C+^4C, + ]/iC^)
10.80
. i>,C-\-t>o(-AC, + i4C-,)
9. 00
11.32
c U(C-\-C,)
* Guard row.
276 Journal of Agricultural Research voi. xii. No. s
These figures show that there is less error in this case by the use of
formulas (2) and (4) and that there was very little difiference between
these two.
This question of a "normal" yield of a plot depends obviously upon
the portion of the population chosen as a standard and upon the method
of calculation. Investigators differ in the choice of both of these factors.
An illustration may be given to show the results of calculating the
normal from different standards. The coefficient of variability of the
lo-tree plot of Arlington navel oranges was computed, taking the devia-
tions from (a) the mean of all plots, (b) the mean of all control plots,
and (c) the "normal" calculated by formula 2. The results are shown
in Table XIII.
TabIvE XIII. — Deviation of yields from mean of area compared with deviation from lo-iree
linear control plot
For-
mula
No.
Deviation.
Coefficient of
variability.
Deviation taken from mean of all plots
26. 00±2. 30
26. 55 ±2. 39
12. 5±i. 10
Deviation taken from mean of control plots . . . ....
3
Deviation taken from normal calculated from nearest control
Since no differential treatments had been given these plots, the mean
of all plots differs little from the mean of all control plots, the respective
means being 1,376 and 1,367 pounds. The variability of plots calcu-
lated from these two standards is not significantly different. In the
case of the deviation taken from the normal yield, however, there is a
very significant decrease in variability, since there are discontinuous soil
variations not recognizable in the general average of the area which are
taken into account by this formula.
CHANCE ARRANGEMENT OF CONTROL PLOTS
A weakness of this system of comparison with interpolated plots lies
in the possibility that the series of control plots may not be representa-
tive of the area. The plots chosen for controls may be on soil superior
or inferior to that of the intervening plots. There are indications that
this possibility may be more real than one would expect from purely
random sampling.
A few computations will show the extent to which the different methods
of choosing control plots may affect the results. The Arlington grove
records were recomputed, shifting the control plots back one row. The
yield of the first control plot (see Table XI) was 1,051 pounds instead of
800, and the last was 1,160 instead of 1,520 pounds. The first arrange-
ment will be termed "arrangement A," the second "arrangement B."
The mean yield of all control plots is not greatly changed by the different
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 277
arrangement. The mean of the "A" control plots is 1,367 pounds, that
of the "B " control plots is 1,389 pounds, yet, as shown in Table XIV, the
coefficient of variability is increased from 12.5^1.1 in "arrangement A"
to 18.2 ±1.6 in "arrangement B" for single plots, with similar increases
for the repeated plots.
Table XIV. — Effect of two methods of arranging control plots
Coefi&cient of variability.
Arrangement of plots of lo trees.
Repeated once.
Repeated twice.
Repeated
three times.
Repeated
four times.
Arrangement A ^Arlington). . .
Arrangement B (Arlington). . .
12. 52 ±1.07
18. 2S±I. 59
7. 53 ±0.89
16. o7±i. 97
7. 68±i. 17
12. 46±i. 91
4-33±o-73
8. 7S±i.5o
The records of Stockberger's hop yields have been computed in the
same manner. The coefficient of variability of the plot yields in his
arrangement is 16.71 ±3.86. Moving the control plots down one — that
is, using plot A as c^ and so on — produces a coefficient of variability of
1 5.93 ±3.55. There is obviously no difference between these two values.
The problem was also investigated by the use of the 8-tree plot lemon
records. In this case three possibilities were tried, with the first, second,
and third plots in turn as c^ and every subsequent third plot as a control
plot. The coefficients of variability for the different arrangements were
21. 7±2. 21, 22.5±2.29, and 23.8i2.82. In view of the probable errors
of the coefficients, there seems to be no real difference in the result of the
different arrangements in this grove.
Since a decided difference was found in one case out of the three studied,
it would seem that there is a rather high probability that significant
differences may result from different arrangements of control plots.
VARIABIIvlTY IN THIS YIELDS OF MORE THAN ONE YEAR
It is often assumed that the mean vield of two or more years is less
variable than the yield of one year.
We have had opportunity to study the variability in the yield of 60
navel-orange trees over a period of several years. The data were kindly
furnished us by Mr. A. D. Shamel, of the Bureau of Plant Industry.
The trees in question had been selected for their uniformly good produc-
tion and the individual yields recorded for six years.
The figures presented in Table XV show that the variability of yields
fell off distinctly after one year, but the reduction was negligible after
the yields of two years were combined. It will be noted that the coeffi-
cient of variability of the single trees even for one year is notably low.
Considering the records of single trees, the average of six years' records is
not less variable than the average of two.
278
Journal of Agricultural Research
Vol. Xn, No. s
Table XV. — Comparative variability of yields of navel-orange trees through a period of
six years
Coefficient of variability of the yield of —
Class.
I year.
2 years.
3 years.
4 years.
5 years.
6 years.
Individual trees
Plots of 10 trees each
22. 8±I. 5
3.9±o.8
15. o±o. 9
2. 9±o. 6
i4-7±0-9
2. 7 ±0.5
13. 3 ±0.8
2. 6±o. 5
13. 7 ±0.9
3-4±o. 7
14. 6±i. 0
3.9±o-8
The coefficient of variability for the lo-tree plot is remarkably low
for the reason that the trees themselves are so uniform and only a small
area of ground is involved. It is no surprise, therefore, to find that the
coefficients of variabiUty are nearly equal, calculated from one to six
years. The probable errors are relatively large and it is difficult to
assert that there is any real difference.
The walnut yields can be used as additional data for the study of this
question. Table XVI shows the coefficient of variabihty of the 191 5
and 1 91 6 yields and their total, with both i- and 8-tree units. By con-
sidering the individual tree as a unit, the total yield for the two years
was less variable than the 191 6 yield, the difference equaling three times
the probable error. The variability of the 191 5 yield is practically
equal to that of the total. On considering an 8-tree plot as a unit, there
is a difference between the coefficients for 191 5, 191 6, and the total of
the two years respectively ; however, the observed difference is less than
three times the probable error and its significance may be somewhat
doubted. Apparently the mean of two years' yields in this case is less
variable than one year's yield.
Table XVI. — Comparative variability of yields of seedling walnut trees through a
period of two years
Class.
Individual trees
Plots of eight trees each.
Coefficient of variability of the yield.
191S yield.
47- 9±i-6
30. 0±2. 6
1916 yield.
53-9±i-9
Mean of
191S and 1916
yields.
46. 4±i. 6
25. 3±2. 2
Further studies on the comparison of the variability of yields through
several years were made, from data published by Hedrick (igii). Table
IV of the bulletin cited gives the yearly yields of individual apple trees
from 1902 to 1 910, inclusive, upon which our computations are based.
Four differential fertilizer treatments have been given to eight plots
of five trees each, each treatment being duplicated on nonadjacent plots.
There are three nontreated plots serving as controls.
The variability of the individual trees was computed on the 1 5 control
trees.
Feb.4.i9i8 Variability of Vields of Fruit Trees and Field Trials 279
The variability of plots is based on hypothetical plots made up of one
tree from each of the treated and one from each of the untreated plots.
These hypothetical plots are therefore made up of an equal number of
trees having similar treatments. The variability of the 10 plots thus
obtained was computed for the single year 1910, and for the sum of two,
three, four, and seven years. Table XVII gives the coefficients of varia-
bility for single trees and for 5-tree plots.
Table XVII. — Comparative variability of yields of Baldwin apple trees through a period
of seven years «
Coefficient of variability of the yield.
Class.
I year (1910).
2 years (1909
and 1910).
3 years (1908-
1910).
4 years (1906,
1908-1910).
7 years (1902,
1903. 190S. 1906,
1908-1910).
Individual trees . .
Plots of five trees
each
37-3±5-2
18. 5±2.9
33. 6 ±4. 6
16. 8±2. 6
32- 5 ±4- 4
18. I ±2. 8
32. 2 ±4. 4
18. 2 ±2. 8
34- o±4- 7
2i.7±3-4
"Records taken from Hedrick (rpir, p. 172-174).
It is interesting to note how slightly the variability of the yields is
decreased by combining two or more years. If regard is paid to the
probable errors, it can not be said that there is any real difference. In
other words, one year's records of the yields of these apple trees seem to
be as reliable for variation studies as those for several years.
It seems, therefore, that the continuation through several seasons
may not so materially decrease the variability of tree 5aelds as one might
expect. This has a direct bearing on the reliability of the major portion
of the calculations of this paper which are based on the variability of the
yields of one season. It should be kept in mind, however, that these
studies do not concern the relative yield of one plot compared with another,
but rather deal with the total variation from the mean of the yields of
all the plots. As cited before from the work of several experimenters,
the relative productivity of a group of plots may not be fully determined
even after a period of years, whereas the tree yields from Hedrick, Shamel,
and the data of the writers indicate that a measure of the variability for
one year of a group of trees divided into plots, may be very representative
of the mean variability for several years.
SUMMARY
(i) The present paper is the result of a study of the nature and extent
of the casual variability of yields of fruit trees under field conditions and
its bearing on the reliability of plot trials.
(2) Studies have been made upon the variability of the yields of orange,
lemon, apple, and walnut trees. The orchards studied were selected on
27808°— 18 4
28o Journal of Agricultural Research voi. xii, no. s
account of uniformity of treatment and appearance, yet the variability
in productivity was considerable. The coefficient of variability for the
yield of individual trees of the clonal varieties ranged from 29.27 ±0.69
to 41 .23 ± 1 .52 per cent, but for the individual seedling walnuts, the coeffi-
cient was somewhat higher, reaching 53.91^:1.92 per cent. The varia-
bility of these tree yields approaches the normal curve of errors. This
variability may be assumed to be the result of "casual" factors which are
beyond the control and possibly the recognition of a careful experimenter.
(3) The effect upon variability of combining trees into plots of various
sizes and shapes has been investigated. As the number of trees per
plot is increased, the coefficient of variability decreases. The coeffi-
cient of variability does not decrease, however, in proportion to the in-
creased number of trees per plot. In most cases there is little gained in
accuracy by increasing the plot to include more than eight adjacent trees.
(4) One of the great causes of variability in yields appears to be the
heterogeneity of apparently uniform soil. While a combination of a
sufficient number of adjacent trees into a plot will overcome largely the
fluctuations of individuals, nevertheless the plots may not sufficiently
include both high- and low-yielding areas to give a typical average.
Greater reliability may be secured by a systematic repetition and dis-
tribution of plots through the experimental area. A consistent gain in
reliability resulting from this method of repetition is shown by the use
of several different methods of computing the variability.
The coefficient of variabiHty for an average plot of 16 adjacent trees
was 22. 58 ±1.01, while 16 trees in four scattered ultimate plots each of
four trees have a coefficient of variability of 9.29^0.40. The larger the
number of units in a combination plot the more typical is the sample of
the area obtained. A i6-tree plot can be expected to give more reliable
results if divided into four equal plots and repeated at four regularly
placed intervals than can either two 8-tree plots, or 6 adjacent trees.
The same principle holds true for larger units. A given number of unit
plots will give a greater accuracy than half the number of units with
twice as many trees per unit.
Four repetitions of an ultimate plot reduced the coefficient of variability
to a point considered practical for cultural operations. Further repeti-
tions, though reducing the coefficient in less degree, did not appear to
justify the additional number of trees required. A minimum of 8 to 10
trees is required for plots involving cultural experiments. In the case of
rootstock, pruning, or variety trials, twice as many plots each contain-
ing half as many trees might be used to obtain greater accuracy.
The fact that marked soil variations occur which tend to make adjacent
trees or adjacent plots yield alike, even on soils which were chosen
because of their apparent uniformity, is well shown by applying the
formula proposed by Harris {191 5) for measuring the coefficient of correla-
tion between neighboring plots of the field. Applying this to the Arlington
Feb.4,i9i8 Variability of Yields of Fruit Trees and Field Trials 281
navel oranges, the writers have calculated the correlation between the
yield of the 8-tree plot as the ultimate unit and the yield of the com-
bination of four such adjacent plots, and it was found that
r= +0.533 ±0.085.
This result shows a marked correlation, indicating a pronounced
heterogeneity in the soil of this grove influencing fruit production.
However, when the correlation between the 8-tree plot as the ultimate
unit and the yield of the combination of four such systematically scattered
plots was calculated it was found that
r= +o.i37±o.i20.
This coefficient is practically equal to its probable error and can be
regarded as significantly zero.
(5) In the computations made by the writers emphasis is also laid
upon the nature and magnitude of the probable error. It is shown in
several cases that the probable error of comparison between plots may be
so large that relatively large differences must be evident between treated
and untreated plots for a reasonable assurance that it is due to the
factors being experimented upon. With the plots of 16 to 32 adjacent
trees which were studied, a difference of from 62.94 to 81.97 per cent of
the mean production would be necessary in order to obtain chances of 10
to I that the results were due to differential treatment and not to casual
variation in the productivity of the trees. With the same number of trees
in scattered units, a difference of 28.42 to 50.02 per cent would be necessary
for the same odds. It seems probable, therefore, that a difference between
two tree plots of less than 50 per cent of the mean production should
be considered with caution before attributing it to differential treatment.
(6) The relation between the shape of a plot and its variability was in-
vestigated by making comparisons betvveen square plots and linear plots
containing the same number of trees. Except in the case of large plots, the
difference in the variability of plots of different shapes was insignificant.
(7) In any method of field experimentation where a standard of com-
parison is desired the theoretical or "normal" yield of a plot is a question
of importance. By the use of certain formulas the "normal" yield may
be computed from control plots. As a standard, one may use the average
yields of the control plots of the entire area, or of the nearest control
plots, or a combination of the two. In cases studied, the coefficient of
variability was reduced 50 per cent by calculating the normal yield from
the nearest controls in place of using the mean of the entire area. The
employment of every alternate row as a control plot was not sufficient
to offset the variability due to soil heterogeneity.
(8) Computations made on the yields of orange, walnut, and apple
trees for several consecutive years showed little annual fluctuation in
their variability. One or two crops may not show greater variability
than the average of six or seven crops.
282 Journal of Agricultural Research voi. xii. No. s
LITERATURE CITED
Ballantyne, a. B.
1916. FRUIT TREE ROOT SYSTEMS. Utah AgT. Exp. Sta. Bui. 143, 15 p., 5 fig.
BarrE, H. W.
1915. APPLE breeding. In S. C. Agr. Exp. Sta. 28th Ann. Rpt. [i9i4]/i5, p. 13-17.
Carleton, M. a.
1909. LIMITATIONS IN FIELD EXPERIMENTS. In Proc. 30th Ann. Meeting Soc.
Prom. Agr. Sci., p. 55-61.
CoiT, J. E.
1910. THE RELATION OF ASEXUAL OR BUD-MUTATION TO THE DECADENCE OF CALI-
FORNIA CITRUS ORCHARDS. In Proc. 37th Fruit Growers' Conv. Cal.,
P- 32-39-
Coombs, G. E., and Grantham, J.
1916. FIELD experiments AND THE INTERPRETATION OP THEIR RESULTS. In Agr.
Bul. Fed. Malay States, v. 4, no. 7, p. 206-216, i fig.
Fletcher, S. W.
1913. eliminating unprofitable trees from the apple orchard. in proc. 33d
Bienn. Session Amer. Pomol. Soc, p. 138-145.
GOURLEY, J. H.
1915. STUDIES IN FRUIT BUD FORMATION. In N. H. Agr. Exp. Sta. Tech. Bul. 9,
79 p., 5 fig., 4 col. pi., 13 charts. Bibliography, p. 77-79.
Hall, A. D.
1909. THE EXPERIMENTAL ERROR IN FIELD TRIALS. In JouT. Bd. Agr. [London],
V. 16, no. 5, p. 365-370-
and Russell, E. J.
1911. FIELD TRIALS AND THEIR INTERPRETATION. In Jour. Bd. Agr. [London],
Sup. 7, p. 5-14, 2 fig.
Harris, J. A.
1912. ON THE SIGNIFICANCE OP VARIETY TESTS. In Science, n. s., v. 36, no. 923,
p- 318-320.
1913a. AN ILLUSTRATION OF THE INFLUENCE OF SUBSTRATUM HETEROGENEITY
UPON EXPERIMENTAL RESULTS. In Science, n. s., v. 38, no. 975, p. 345-346,
I fig.
1913b. SUPPLEMENTARY NOTE ON THE SIGNIFICANCE OP VARIETY TESTS. In
Science, n. s., v. 37, no. 952, p. 493-494.
I915. ON A CRITERION OP SUBSTRATUM HOMOGENEITY (OR HETEROGENEITY) IN
FIELD EXPERIMENTS. In Amer. Nat., v. 49, no. 583, p. 430-454.
Hedrick, U. p.
1911. IS IT NECESSARY TO FERTILIZE AN APPLE ORCHARD? N. Y. State Agr. Exp.
Sta. Bul. 339, p. 153-195. 7 fig-. 4 pl-
1912. pedigreed NURSERY STOCK. N. Y. State Agr. Exp. Sta. Circ. 18, 8 p.
1914. A COMPARISON OF TILLAGE AND SOD MULCH IN AN APPLE ORCHARD. In N. Y.
State Agr. Exp. Sta. Bul. 383, p. 249-281, i fig., 6 pl.
HoLTSMARK, G., and Larsen, B. R.
1906. tJBER DIE FEHLER, WELCHE BEI FELDVERSUCHEN DURCH DIE UNGLEICHARTIG-
KEiT DES BODENS BEDINGT wERDEN. In Landw. Vers. Stat., Bd. 65,
Heft 1/2, p. 1-22.
Kraus, E. J.
1916. SOMATIC SEGREGATION. In Jour. Hered., v. 7, no. i, p. 3-8, i fig.
Lewis, C. I , and Vickers, H. A.
1915. ECONOMICS OP APPLE ORCHARDING. In Ore. Agr. Exp. Sta. Bul. 132, 104 p.,
14 fig.
Feb. 4, 1918 Variability of Yields of Fruit Trees and Field Trials 283
YON, T. L.
1912. SOME EXPERIMENTS TO ESTIMATE ERRORS IN FIELD PLOT TESTS. In ProC
Amer. Soc. Agron., v. 3, 1911, p. 89-114, 5 fig.
Macoun, W. T.
1904. INDIVIDUALITY OF FRUITS. In Exp. Farms [Canada] Rpt., 1903, p. 102-104.
Mercer, W. B., and Hall, A. D.
1911. THE EXPERIMENTAL ERROR OF FIELD TRIALS. Ill Jour. Agr. Sci., V. 4, pt. 2,
p. 107-132, 10 fig.
Mitchell, H. H., and Grindley, H. S.
1913. THE ELEMENT OF UNCERTAINTY IN THE INTERPRETATION OF FEEDING EXPERI-
MENTS. In 111. Agr. Exp. Sta. Bui. 165, p. 459-579, 8 fig. Bibliography,
P- 578-579-
Montgomery, E. G.
i912. variation in yield and method of arranging plats to secure com-
PARATIVE RESLTLTs. In Nebr. Agr. Exp. Sta. 25tli Anai. Rpt., [i9ii]/i2,
p. 164-180, 4 fig.
I913. EXPERIMENTS IN WHEAT BREEDING: EXPERIMENTAL ERROR IN THE NURSERY
AND VARIATION IN NITROGEN AND YIELD. U. S. Dept. Agr. Bur. Plant
Indus. Bui. 269, 61 p., 22 fig., 4 pi.
Munson, W. M.
1907. ORCHARD RENOVATION. In Maine Agr. Exp. Sta. Bui. 139, p. 53-56.
Olmstead, L. B.
i914. some applications of the method of least squares to agricultural
EXPERIMENTS. In ]ouT. Amer. Soc. Agron., v. 6, no. 4/5, p. 190-204, i fig-
Pickering, S. U.
1911. EXPERIMENTAL ERROR IN HORTICULTURAL WORK. In Jour. Bd. Agr. [lyondon].
Sup. 7, p. 38-47.
Pritchard, F. J.
I916. THE USE OF checks AND REPEATED PLANTINGS IN VARIETAL TESTS. In
Jour. Amer. Soc. Agron., v. 8, no. 2, p. 65-81, 3 fig.
Shamel, a. D.
1912. BUD SELECTION as A MEANS OF IMPROVING CITRUS AND OTHER FRUITS. In
Ann. Rpt. Amer. Breeders' Assoc, v. 8, 1911, p. 497-503.
Stewart, J. P.
1913. FERTILIZATION AND CULTURE METHODS IN APPLE ORCH.\RDS. In Pa. Agr.
Exp. Sta. Ann. Rpt. 1911/12, p. 497-563.
Stockberger, W. W.
1916. relative precision op formula for calculating normal plot yields.
In Jour. Amer. Soc. Agron., v. 8, no. 3, p. 167-175.
Surface, F. M., and Pearl, Raymond.
1916. A METHOD OF CORRECTING FOR SOIL HETEROGENEITY IN VARIETY TESTS. In
Jour. Agr. Research, v. 5, no. 22, p. 1039-1050, 4 fig. Literature cited,
p. 1050.
Whitten, J. C.
1915. BUD SELECTION FOR INCREASING YIELDS. In Mo. Agr. Exp. Sta. Bul. 131,
p. 479-480.
Wood, T. B.
191 1. THE INTERPRETATION OF EXPERIMENTAL RESULTS. In Jour. Bd. Agr. [Lon-
don], Sup. 7, p. 15-37, 2 fig.
and Stratton, F. J. M.
1910. THE interpretation OF EXPERIMENTAL RESULTS. In JoUT. Agr. Sci., V. 3,
pt. 4, p. 417-440, 10 fig.
INTERRELATIONS OF FRUIT-FLY PARASITES IN HAWAII
By C. E. Pemberton, Assistant Entomologist, and H. F. Willard, Fruit-Fly Quar-
antine Inspector, Mediterranean Fruit-Fly Investigations, Bureau of Entomology,
United States Department of Agriculture
INTRODUCTION
The introduction and ultimate establishment of four species of larval
parasites of the Mediterranean fruit fly {Ceratitis capitata Wied.) in
Hawaii, wherein exist ideal conditions for the rapid and unchecked
development of the host throughout the year, has opened an exceptional
opportunity for an investigation of all phases of the work of these para-
sites, not only in their relations to the host but also to one another, and
has made possible the determination of many points of unusual interest to
biological students of insect parasitism and of particular value to ento-
mologists dealing with considerations relating to general parasitic control
of insect pests. ^
The information herein presented is given, not as a final decision or
positive argument against the introduction of many parasites of a single
pest, but to reveal the actual need for careful biological studies of para-
sites, especially in their interactions upon one another, before general
ntroductions or liberations can be intelligently undertaken.
Entomologists, detailed in the past for researches in insect parasitism
in foreign countries, have customarily adopted the policy of assembling
all available species of primary parasites of the insects under investiga-
tion, with the final intention of conveying them all to the home country
for propagation and liberation. The chief caution of these workers has
usually been the elimination of all secondary parasites from the material
prepared for shipment. Admirable results have often been achieved.
However, after the introduction of several species, through one or suc-
cessive importations, few, if any, considerations have been given to the
possibility of detrimental results arising from interference with the action
of one parasite of primary importance and great prolificness by another
of less value and proved inferiority.
The important bearing that a preliminary and detailed knowledge of
parasite habits may have upon the general question of parasite importa-
tions has already been well directed to the attention of entomologists by
Dr. L. O. Howard. He states ^ that—
It is unwise and most unpromising to attempt heterogeneous and miscellaneous
importations of parasites without careful study of the host insect on its home ground
1 For a history of these parasitic introductions and a discussion of climatic and host relationships favor-
jng parasitic increase, see Back, E. A., and Pemberton, C. E., the mediterranean fruit fly in
HAWAII. U. S. Dept. Agr. Bull. 536, 119 p., 21 pi., 24 fig. 1918.
s Howard, L. O. the practical use op the insect enemies op injurious insects. In U. S. Dept.
Agr. Yearbook 1916, p. 282. 1917.
Journal of Agricultural Research, Vol. XII, No. 5
Washington, D. C. Feb. 4, 1918
jr Key No. K— 60
(28s)
286 Journal of Agricultural Research voi. xii, No. s
and in its natural environment throughout the whole range of its existence and a
similar biological study of its parasites and natural enemies under such conditions.
Some results of the recent work in Hawaii most strongly bear this out.
Sufficient evidence has been obtained to throw serious doubt upon the
assumption, often accepted, that the greater the number of species of
parasites associated with the host, the greater the chances for its
control. It is felt that the following data quite definitely indicate, at
least in some cases, that better results may be obtained by a method of
judicious selection of desirable species for introduction rather than by a
wholesale and indiscriminate procedure. If 90 per cent of all individuals
of an insect pest are destroyed by a single species of parasite, is it wise
to attempt further control by bringing in other species, until it is known
by positive and careful experimentation in the laboratory and field that
these new species will not interfere with or check the normal activities
of the first species ?
In May, 191 3, Prof. F. Silvestri succeeded in bringing two species of
opiine parasites of the fruit fly into Hawaii. One, Opius humilis Sil-
vestri, he brought from South Africa, and the other, Diachasma tryoni
Cameron, was secured in Australia. Both were soon established in the
Kona coffee district of the island of Hawaii. By 191 5 it had become
clearly evident that O. humilis was often parasitizing from 60 to 90 per
cent of all of the fruit-fly larvae developing in the coffee cherries. D.
tryoni steadily but slowly increased and in time exhibited a capacity for
occasionally parasitizing 50 per cent or more of the host larvae. Here
it is obvious that overlapping or duplication in parasitism was occurring.
Clearly some fly larvae were being stung by both species of parasites and
frequently to a very considerable extent.
CANNIBALISM AMONG THE PARASITES
Early in 1916, Dr. E. A. Back, of the Bureau of Entomology, while
examining the contents of some parasitized fruit-fly material from the
field, observed under the microscope a larva of the parasite Diachasma
tryoni attacking one of the parasite Opius humilis. A suspicion of dis-
advantageous consequences arising from complications attending the
interactions of these parasites led Mr. C. L. Marlatt to assign to the
writers an investigation of this subject.
Careful microscopical examinations of large numbers of fruit-fly larvae
and pupae, collected from localities where both species of parasites were
known to be well established and actively working together, soon revealed
one striking fact. In the majority of cases where fruit-fly larvae had been
parasitized by both Diachasma tryoni and Opius humilis the latter was
killed and the former developed to maturity. O. humilis is killed purely
by wounds and lacerations inflicted upon it by the long, curved, sickle-
like mandibles of the newly hatched larva of D. tryoni. This larva
Feb. 4, 1918 Interrelations of Fruit-Fly Parasites in Hawaii 287
attacks the young of O. humUis, usually at a point a few segments back
of the head, the point of contact being clearly seen in Plate 12.
Its mandibles open wide and snap into the body of the attacked larva
spasmodically and with remarkable quickness. Often the entire opera-
tion of broadly opening and closing the mandibles may be almost imper-
ceptible to the eye. It may move its entire body quickly, the caudal
tip may be curled beneath the body and extended again suddenly, and
the mandibles may be repeatedly opened and closed until successfully
grasping the Opius larva. Besides possessing unusual powers for inflict-
ing injury to other parasitic larvae about it, it may avoid counterattack
through ability to move quickly and through the protection afforded
the entire ventral surface of the body by a thick mass of serosal, cellular
material that accompanies the larva when it emerges from the egg, and
which remains with it during its entire life in the first instar. This mass
of cells may be seen clinging to Diachasma larvae of the first instar in
Plate 10.
The newly hatched larva of Opius humilis possesses mandibles which
are also long and pointed, as shown in Plate 12, A, B. These may
be used to good advantage when the larva is successful in bringing them
in contact with individuals of its own or of other species of parasites.
The larva, however, is sluggish, moves much less quickly than that of
Diachasma tryoni, is protected ventrally by a much thinner, less adhesive
mass of serosal cells, is much less capable of quick and powerful move-
ment of the mandibles, and usually holds the body in a somewhat hori-
zontal and exposed position. These deficiencies seem to explain its
inability to avoid destruction by larvs of D. tryoni or to offer successful
counterattack when the two are lodged within the same host larva.
During the examination of nearly 3,000 fruit-fly larvae or pupae, para-
sitized in each case by the two species of opiines, the dead or dying and
often struggling Opius larvae were frequently dissected from the tissues
of the host while still tightly clasped in the mandibles of the Diachasma
larvae. Plate 10, A, is reproduced from a photomicrograph of a larva
of O. humilis actually within the grasp of an active, living larva of
D. tryoni. In this particular case the operation of removing the two
larvae from the host, placing them upon a microscope slide in water,
and covering them with a cover glass failed to separate them, and the
Diachasma larva ultimately died with its mandibles deeply embedded
in the body of the dead Opius larva in the exact position as shown.
Plates 12 and 13, A, B, C, show larvae of O. humilis in various stages of
laceration and distortion just as they were removed from fruit-fly larvae
or pupae in which were also one or more larvae of D. tryoni.
Extensive laboratory experiments have exactly confirmed the results
of the first series of field observations. The aggressive, cannibalistic
period of activity of the larvae of these parasites is during existence in
Journal of Agricultural Research voi. xii, no. s
the first instar, and particularly during the early period of this stage,
before the body becomes engorged and swollen with food. Plate ii, B,
which shows a mature first-instar larva of the parasite Diachasma tryoni,
is interesting in this connection when compared with the other illustra-
tions of newly hatched Diachasma larvse of this instar. The head is
unchanged in size, but the body is greatly distended after two days of
feeding preparatory to molting to the second instar. In all of the illus-
trations the enlargement is the same. It is obvious that the enlarged
and somewhat rigid body of the well-fed larva permits much less free-
dom of movement than is possible shortly after hatching. The body is
first elastic, flexible, and capable of quick and effective action. This
change in the size of the body may occur in from two to two and one-half
days.
The molt to the second instar still further incapacitates the opiine
larva for carrying on further cannibalistic action. The mandibles are
then small, soft, and are almost imperceptible, even under high magnifi-
cation, because of their transparency, being wholly unfitted for active
use except in the separation of the semiliquid media in which the larva
lies after the host has formed the puparium. The strong and heavily
chitinized head of the opiine larva in the first instar is thus entirely
discarded upon the molt to the second instar, and all further cannibalism
ceases. The helpless condition after the first molt is well suggested in
Plate 1 1 , A, which shows a larva of Diachasma tryoni in the second instar.
During 191 6 a microscopical examination of the contents of a total
of 2,925 parasitized fruit-fly larvse and pupae definitely showed that the
parasites readily oviposit in the same host larva more than once and
exhibit no discernible instinct of selection of parasitized or unparasitized
larvse. All eggs so deposited hatch. Thus, fly larvse commonly opened
were found to contain several eggs or larvae of the two species of para-
sites and sometimes a third species (Diachasma jullawayi) , a later intro-
duction.
Here w^as certain evidence of a complicated overlapping or duplication
of parasitism. In no single instance were two parasite larvse ever
observed to develop to maturity in the same host larva, except in the
case of the chalcid Tetrastichus gifjardianus (p. 292). All but one were
killed while still in the first instar, or occasionally before hatching. At
times from 8 to 10 opiine larvae were found within the host larva. Only
one would mature and, as a rule, if a larva of Diachasma were one of the
number, it survived all others.
If several eggs of Diachasmu tryoni or D. jullawayi alone are deposited
in an individual host larva over a period of two or three days, the last
parasite larva to hatch stands the best chance for destroying the others
and maturing. The case is identical with the Opius humilis However,
a well-grown and fully-fed larva of D. tryoni or D. jullawayi, if still in the
Feb. 4. i9i8 Interrelations of Fruit-Fly Parasites in Hawaii 289
first instar, seems entirely able to destroy larva of O. humilis that are
newly hatched and unencumbered with a body engorged with food.
All opiine larvae within an individual fruit-fly larva or pupa may some-
times be fatally wounded and no parasite develop. This is not frequent.
Often the deposition of 8 or 10 parasite eggs into a single fly larva causes
its death. In such cases the parasite eggs usually hatch and the resulting
larvae die within a short time.
Cases occurred in which as many as 10 dead larvae of Opius humilis were
dissected from a fruit-fly larva, together with a single, vigorous, active
larva of Diachasma tryoni. This does not necessarily mean that all of
the 10 larvae were killed by the latter. No doubt some of the former
larvae destroyed each other, but it clearly shows the superior aggressive
and defensive power of the larva of D. tryoni. Many cases have been
observed in which a larva of O. humilis was badly cut and distorted from
attack by a larva of D. tryoni. Some such cases are shown in Plates
10, 12, and 13. Occasionally the body may be found entirely severed from
the head. These extreme cases are no doubt caused by reattack upon the
larva a day or more after it has died and has become somewhat softened.
In most cases the death of the larva seems to be caused by the first grasp
or pinch of the attacking larva. A single perforation in the body wall
should be sufficient to cause the death of the larva in a short time.
Occasionally an opiine larva will destroy eggs of its own kind when it
occurs in the same individual host with the eggs. In this manner mature
embryos are sometimes very much distorted and almost unrecognizable.
This is not frequently seen.
Cool weather materially retards the development of the opiine &gg,
particularly in the case of Diachasma tryoni and D. fullawayi. At such
times many cases have been observed in which larvae of Opius humilis
have developed to the second instar before an egg of a species of Dia-
chasma, which had been deposited into the same host larva harboring
O. humilis, had hatched. Upon hatching the small, active Diachasma
larv^a quickly destroyed the large, bulky Opius larva. This unusual con-
dition has been observed only in January in the cool, elevated, coffee
districts on the Island of Hawaii.
From over 2,900 cases where parasitized fruit-fly puparia have been
opened, in no single instance has a case been observed in which the host
pupa was fofmed. A host lar\'a once parasitized quite readily forms
into a normal puparium when sufficiently developed, but the presence
of a single small opiine egg within its body invariably prevents any
further development. The puparium is formed, the histolysis of the
tissues is completed, and here all development of the fruit fly ceases.
The broken-down and liquid medium thus prepared within the puparium,
in which the parasite larvae may move about and feed, enables them to
reach all portions within and easily to come in contact with any other
parasitic individuals that may occur there with them.
290 Journal of Agricultural Research voi. xii. No. s
Thus, the inevitable tendency of every individual opiine lan^a upon
hatching is to destroy every other parasitic larva in its domain, whether
it be one of its own kind or of another species. This would appear to be
an infallible instinct and one of great consequence in Hawaii in the
development of the opiine parasites now present.
SUPPRESSION OF OPIUS HUMILIS BY DIACHASMA TRYONI AND D.
FULLAWAYI •
The parasites Diachasma tryoni and the closely related Diachasma
fullawayi, by virtue of this larv^al instinct, coupled with further en-
dowed superior body characters, have been responsible for the great
suppression of the parasite Opius humilis. The last-named species is
more prolific and hardy than either of the two others and is more gener-
ally efficient than both combined. By their association with the 0. humilis,
they have worked a detriment by reducing the total extent of parasitism
to a point below that to which it is capable of exerting alone. The evi-
dence of such suppression, gained from microscopical examinations of
fruit-fly larvae and pupae secured from various fruits in Hawaii during
191 6 and 191 7, may be expressed as follows:
From April 16 to May 10, 1916, a dissection was made of 757 fruit-fly
pupae, freshly secured from coffee collected in the Kona coffee district of
the Island of Hawaii. From this total, 345 were parasitized by only
Opius humilis, 90 contained living larvae of Diachasma tryoni together
with dead larvae of O. humilis, 9 contained living larvae of O. humilis
together with dead larvae of D. tryoni, i was parasitized by only D.
fullawayi, 5 contained living larvae of D. fullawayi together with dead
larvae of O. humilis, 2 contained living Opius larvae together with dead
larvae of D. fullawayi, 57 contained living larvae of only D. tryoni, and
248 were not parasitized. Here it is seen that from the total of 757 pu-
paria, 106 cases of duplication in parasitism occurred in which a species
of Diachasma was found in the same puparium with one or more indi-
vidual larvae of O. humilis, and that in 95 of these the latter was killed
and the former survived.
A further series of microscopical examinations of the contents of fruit-
fly puparia, freshly secured from coffee fruits in this same district in
January, 191 7, during the coolest part of the year, strongly confirmed
the results of the previous year. Six hundred and twenty-seven puparia
were opened and examined. Of these a total of 343 were parasitized by
Opius humilis alone, 67 were parasitized by only Diachasma tryoni, 2 con-
tained only larvae of Diachasma fullawayi, 129 contained living larv'ae of
D. tryoni together with dead larvae of O. humilis, 8 contained dead larvae
of D. tryoni together with living larvae of O. humihs, 4 contained living larvae
of D. fullawayi together with dead larvae of O. humilis, 2 contained dead
larvae of D. fullawayi together with living larvae of O. humilis, and 72
puparia were unparasitized. These results are again very significant.
Feb. 4, 1918 Interrelations of Fruit-Fly Parasites in Hawaii
291
From the total of 627 puparia 143 cases are noted in which an overlapping
in parasitism occurred, wherein the puparia in each case contained larvae
of O. humilis in combination with larvae of a species of Diachasma, and
in 133 of these cases the former was destroyed.
The collection of extensive data on the percentage of parasitism ex-
erted by the fruit-fly parasites in Hawaii over a period of three years
gave abundant proof that the parasite Diachasma iryoni was most active
during the warmer months of the year. This increase in activity and
abundance paralleled a reciprocal decrease in the abundance of Opius
humilis. The reverse was true during the remainder of the year, when
the former species rapidly decreased and the latter ascended to a place
of first importance. The data presented in Table I most positively reveal
the extent of fluctuation in the comparative abundance of O. humilis and
D. iryoni, the effectiveness of O. humilis being clearly at its maximum
during the spring, when the abundance of D. iryoni is at its lowest,
owing to the accumulated effect of the cool winter months.
Table I. — Comparison of seasonal abundance of Optus humilis and Diachasma tryoni
Locality.
Honolulu, Oahu
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Do
Kona District, Hawaii
Do
Do
Do
Do
Do
Number of
Number of
Percentage
Date of collec-
Diachasma
Opius hu-
of
tion of host, a
tryoni
milis emerg-
Diachasma
emerging.
mg.
tryoni.
Mar., 19 1 6
30
I 000
2.9
Apr., 1916
1,200
9,778
10. 9
May, 1916
499
2,127
19. 0
June, 1916
2,303
998
69.8
July, 1916
1,786
549
76. s
Aug., 1916
2,286
649
77-9
Sept.,1916
4,514
1,139
79-9
Oct., 1916
6,772
2, 061
76.7
Nov., 1916
4,451
I, 206
78.9
Dec, 1916
2,605
1,602
61. 9
Jan., 19 1 7
1,406
679
67.4
Feb., 1917
558
I, lOI
33-6
Sept., 1915
330
118
73-6
Dec, 1915
210
274
43-3
Mar., 1916
85
440
16. I
Apr., 1916
756
3,031
19.9
Aug., 1916
271
77
77-9
Jan., 19 1 7
558
4,749
10.5
Percentage
of Opius
humilis.
97. I
89. I
81.0
30.2
23-5
22. I
20. I
23-3
21. I
38.1
32.6
66.4
26. 4
56.7
83- 9
80. I
a For similar data for 1915, see Back, E. A., and Pemberton, C. K., the mediterranean fruit
FLY IN HAWAII. U. S. Dept. Agr. Bui. 536, 119 p., 21 pi., 24 fig. 1918.
From September to December, 1916, a microscopical examination was
made of the contents of 618 fruit-fly puparia obtained about Honolulu
from coffee, guavas, and the winged kamani (Terminalia catappa). Of
this total, 55 puparia were parasitized by only Opius humilis, 331 were
parasitized by only Diachasma iryoni, 35 contained only larvae of D. fulla-
wayi, 96 contained living larvae of D. tryoni together with dead larvae of
O. humilis, 4 contained dead larvae of D. tryoni together with living larvae
292 Journal of Agricultural Research voi. xii, no. 5
of O. humilis, 6 contained living larvae of D. jullawayi together with dead
larvae of O. humilis, i contained a dead larva of D. jullawayi together with
a living larv^a of O. humilis, and 90 were unparasitized. Here again is
striking evidence of positive suppression of the parasite O. humilis by the
other parasites, particularly by D. tryoni. Of the 618 puparia examined,
the O. humilis occurred 107 times in combination with a species of Dia-
chasma and won the struggle for existence in only 5 cases.
The foregoing would tend to explain the very noticeable fluctuations
in the comparative abundance of the two principal species of parasites
at differenct seasons of the year, as shown in Table I. It is seen that the
extent of parasitism by Opius humilis is greatly influenced by the abun-
dance or scarcity of the parasite Diachasma tryoni, and, as elsewhere
discussed, the abundance of this latter parasite in Hawaii is very much
dependent upon seasonal conditions. This causes a seasonal rise and
fall in the effectiveness of 0. humilis. In the summer and autumn of
the year the ascendancy of the D. tryoni causes a great reduction in the
abundance of O. humilis. During the winter and spring seasons the
reduced activity of the Diachasma permits a rapid increase in parasitism
by O. humilis. This is particularly true in the elevated Kona coffee
district, where the winter temperatures are somewhat below those about
Honolulu.
The results of the first series of dissections of fruit-fly larvae and puparia
on the island of Hawaii in January, 1916, led to further laboratory ex-
periments in Honolulu on a comprehensive scale, duplicating as closely
as possible the field conditions. Unparasitized fruit-fly lars^ae were ex-
posed within fruit to the attack of both Opius humilis and one or both of
the species of Diachasma for a few hours. The fruit-fly puparia thus
obtained were usually all parasitized, and no loss in time resulted from
examinations of unparasitized material. In this manner 393 cases were
obtained in which fruit-fly larvae were parasitized by both O. humilis and
D. tryoni. In 387 of these the Opius larvae were all killed and the Dia-
chasm.a survived. In only 5 of the puparia did the former succeed in
overcoming the latter and developing.
Out of yj cases in which fruit-fly larvae were parasitized by both Opius
humilis and Diachasma jullawayi, in only 2 did the Opius develop.
In the remaining 75 the Opius larvae were all killed by the Diachasma
larvae.
SUPPRESSION OF OPIUS HUMILIS BY TETRASTICHUS GIFFARDIANUS
The fruit-fly parasite Tetrastichus gifjardianus Silvestri, a late introduc-
tion into Hawaii, has proved decidedly destructive to any of the opiines
when occurring in the same fly larvae or puparia with them. This
chalcid, seemingly of importance, has, after a two years' trial in Hawaii,
given but small promise of accomplishing any perceptible control of the
Feb. 4, 1918 Interrelations of Fruit-Fly Parasites in Hawaii 293
fruit fly. It is occasionally bred out from fruit-fly material secured in
the field. Several times during dissections of puparia from the field its
larvae have been found in combination with larvae of Opius humilis or of
one of the species of Diachasma. Though soft, sluggish, and armed with
small, inconspicuous, blunt mandibles, it nevertheless survives the opiine
larvae by sheer force of numbers and consequent rapid absorption of
food. When ovipositing, this chalcid usually places about 10 eggs in
the host at one insertion of the ovipositor. The opiines deposit but one
egg, remove the ovipositor, and look for further larvae to attack. A
single oviposition, then, by a T. giffardianus into a host larva already
parasitized by an opiine, places about 10 individuals of the chalcid with
I of the opiine. If the host larva has already received several opiine ovi-
positions, usually only one individual is alive, as already shown. The larvae
of T. giffardianus exhibit no cannibalistic tendencies, and do not destroy
each other. Many of the chalcid larvae are then killed by the opiine
larvae, but seldom all of them. The opiines ultimately die, and one or
more of the Tetrastichus larvae develop. The death of the Opius or Dia-
chasma larvae results usually from starvation or suflfocation or possibly
by the absorption of toxic excretions of the Tetrastichus larvae. Certainly
the chalcid larvae inflict no visible bodily injury on the opiine larv^.
In view of the demonstrated ineffectiveness of this chalcid and of its
capability for surviving the Opius humilis when occurring in fruit-fly
larvae with it, it is here considered a detrimental introduction because of
interference with the work of the latter.
OPIUS HUMILIS, THE MOST PROLIFIC OF THE INTRODUCED
PARASITES
From the foregoing summaries of the data it is obvious that the
parasite Opius humilis is killed in the larva stage in almost every instance
in which its larvae become associated in a host larva with any one of the
three other species of introduced fruit-fly parasites and that the percent-
age of cases of such duplication is large. Biological studies of all of these
parasites, so far as two and three years' adaptation in a new country may
show, have indicated quite clearly that under Hawaiian conditions
O. humilis is the most prolific of the four species in all seasons of the year
and that none of the others show particular abilities for reaching larvae of
the fruit fly that are not as easily accessible to the attack of O. humilis.
The very considerable activity of the parasite Diachasma tryoni and
the occasional heavy parasitism by it has made necessary the establish-
ment of careful proof that it is less prolific than Opius humilis, before it
can be maintained that the introduction of the former is a detrimental
one. The unimportance of the other two parasitic species eliminates all
present need for discussing them any further.
From what has already been shown, we know positively that the
parasite Diachasma tryoni has a clear field in Hawaii for unchecked
294 Journal of Agricultural Research voi. xii, No. s
reproduction. Opius humilis, though present everywhere, does not
hinder its activities. Extensive records kept in Hawaii during a period
of three years on emergences from more than 100,000 fruit-fly puparia
from many sections of the islands and from all available types of fruits
have clearly shown that the average maximum degree of parasitism by
D. tryoni does not exceed or equal that of O. humilis. This alone is suffi-
cient proof of superior prolificness on the part of the latter.
In no instance have fruits of all varieties been found harboring larvae
of the fruit fly that are less frequented by Opius humilis than by Dia-
chasma tryoni. The ability of the former to find the host is, then, equal,
if not superior, to that of the other parasite. In no case has evidence
been found to show that the longer ovipositor of the D. tryoni is really
an advantage over that of the O. humilis, whose ovipositor is less than a
third as long. The comparative difference is considerable, but the dif-
ference when considered in fractions of an inch is really small.
Preliminary records of individual females of both species to determine
the total number of eggs deposited by an individual in a lifetime show no
superiority of the Diachasma tryoni over Opius humilis in this respect.
The life cycle of Diachasma tryoni is consistently longer than that
of Opius humilis. From 28,410 records on the length of the life cycle of
the former, secured during 191 6 and 191 7, it is almost invariably found
to be from two to four days longer than that of the latter during most of
the year, and in the winter months a great number of the Diachasma
individuals, hibernating in the larva stage, extended the cycle to from
one to six months longer. From 22,889 cases under observation in 1916
and 1 91 7 on the life cycle of O. humilis, in no single case has an individual
ever been known to so hibernate or extend the length of the life cycle
beyond the average for more than a few days. Cool weather and drouth
seem most favorable for inducing this hibernating tendency in the
larvae of the D. tryoni. Of 3,077 cases under observation by the junior
author in January, February, and March, 191 7, in which fruit-fly larvae
had been parasitized by a Diachasma, a total of i ,404 cases occurred in
which the parasite went into hibernation as a mature larva. This seems
to explain the great reduction in abundance of D. tryoni in the field in
winter. Its capacity as a parasite in the winter months is thus strik-
ingly less than that of O. humilis.^
From the standpoint of longevity, all experiments so far show no great
superiority of one species over the other, except as noted above in regard
to hibernation. Individual adults of both Opius humilis and Diachasma
tryoni have been kept alive for about four months.
' Credit is here due Mr. J. C. Bridwell for valuable suggestions offered in connection with the study
of the hibernation of these parasites.
Feb. 4, 191S Interrelations of Fruit-Fly Parasites in Hawaii 295
The relative proportion of females to males in these two species of
parasites is interesting. Of 26,975 individuals of Diachasma tryoni,
reared from material collected in the field in 191 6 and 191 7, 16,845, or
62.4 per cent, were males. Of 10,843 individuals of Opius humilis
similarly reared from material collected in the field in 191 6 and 191 7,
6,128, or 56.5 per cent, were males. Here, again, the advantage, if any,
lies with O. humilis.
In view of these several facts relating to the comparative prolificness
of the parasites Opius humilis and Diachasma tryoni, it appears that the
former is superior to the latter species or to any of the other introduced
parasites. Thus, when a host larva is parasitized by both O. humilis
and D. tryoni, the latter survives, and in so doing produces an individual
less prolific than would have been the case had the O. humilis been per-
mitted to develop. This seems to point to a certain, distinct loss. If
one species of larval parasite when working alone parasitizes 60 per cent
of the host, and another species not strikingly different from the first and
working the same in every known respect parasitizes 40 per cent of the
host when working alone, there is no reason to assume that both com-
bined can exceed a parasitism of 60 per cent. All overlapping by the
species capable of only 40 per cent parasitism can only serve to reduce
the total effect to a point below 60 per cent of parasitism.
CONCLUSION
Sufficient evidence has been presented to prove the superiority of the
parasite Opius humilis over the other introduced fruit-fly parasites in
Hawaii and demonstrates the decided restraint operated over it by the
unfailing cannibalistic activities of the larvae of Diachasma tryoni in
particular and of the other parasites in part. Knowing the capacity of
O. humilis for parasitizing from 80 to 90 per cent of the larvae of the fruit
fly in favorable localities, such as the large Kona coffee belt on the island
of Hawaii, the writers here maintain that detrimental results to a certain
extent have arisen from the liberation in Hawaii of parasites other than
O. humilis that attack the larva of the fruit fly. The total parasitism
has simply been reduced in value to that of a parasite of secondary value.
It is hoped that the present analysis of the interrelated activities of
the imported fruit-fly parasites in Hawaii may serve to stimulate greater
discrimination in the selection of parasites proposed for future intro-
duction.
27808°— 18 5
PLATE lo
Diachasma tryoni:
A. — Freshly hatched larva with its mandibles actually embedded in the body of a
newly hatched but dead larva of Opiiis humilis. Xioo.
B. — ^Newly hatched larva with its mandibles closed, showing ventral serosal material
surrounding the body and the two gill-like appendages on the first body segment.
X loo.
(296)
Interrelations of Fruit-Fly Parasites
Plate 10
A
B
Journal of Agricultural Research
Vol. XII, No. 5
Interrelations of Fruit-Fly Parasites
Plate 1 1
Journal of Agricultural Researcli
Vol. XII, No. 5
PLATE II
Diachasma tryoni:
A. — Lateral view of larva in the second instar, showing particularly well the fat-
body of the host recently taken in as food. X loo.
B. — Lateral view of a 2-day-old larva engorged with food and about to molt, showing
the enlarged and stiffened body. X loo.
PLATE 12
Opius humilis:
A, B. — Dead larva in first instar; killed by first-stage larva of Diachasma iryoni,
showing cut on body made by the attacking larva and mandibles extended in final
death struggle. X loo.
C. — Dead larva in first instar; killed by first-stage larva of Diachasma tryoni. The
point of attack is here clearly seen. The body contents have been apparently with-
drawn from the lower portion of the body by the attacking larva. X loo.
D. — Dead larva in first instar: badly lacerated and distorted by attack of first-stage
larva of Diachasma tryoni. In this case the dead larva was probably destroyed while
n the embryouic stage and a few hours prior to the hatching of the egg. X loo.
Interrelations of Fruit-Fly Parasites
Plate 12
Q
o
■ '^'^•■^^'■,
Journal of Agricultural Research
Vol. XII, No. 5
Interrelations of Fruit-Fly Parasites
PLATE 13
^^^H
H|
Journal of Agricultural Research
Vol. XII, No. 5
PLATE 13
Opius humilis:
A. — Dead larva in first instar, with body shriveled and twisted through attack by
first-instar larva of Diachasma tryoni. X 100.
B. — Dead larva in first instar; killed by first-instar larva of Diachasma tryoni. Here
the larva had been feeding and developing for several hours before being attacked by
the Diachasma larva. The body contents can be seen protruding from an inflicted
wound on the seventh and eighth body segments. X 100.
C. — Dead larva in first instar; killed by first-stage larva of Diachasma tryoni. This
is the appearance of the dead Opius larvae most commonly seen. The body is pinched
by the mandibles of the Diachasma larva in the first or second body segment back
of the head. X 100.
D. — Healthy, living larva in first instar. X 100.
E. — Healthy, uninjured, living larva in first instar. X 100.
ADDITIONAL COPIES
Of THIS PTJBUCATION MAY BE PBOCtTRED FBOM
THE SUPERINTENDENT OF DOCUMENTS
GOVEBNMENT PRINTraG OFFICE
WASHINGTON, D, C.
AT
20 CENTS PER COPY
Stjbscbiption Pbice, $3.00 Per Year
Vol. XII KEBRU ARY 11, 1918 No. 6
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTE^NXS
T^mfi
Water Eztractions of Soils as Criteria of their Crop^Pro-
ducing Power -------- 297
JOHN S. BURD
(Contribution tram California Agricultural Experiment Station)
Effect of Season and Crop Growth in Modifying the Soil
Extract - - - - - - - - - 311
GUY R. STEWART
(Contribution from California Agricultural Experiment Station)
The Freezing-Point Method as an Index of Variations in
the Soil Solution l)ue to Season and Crop Growth - 369
D. R. HOAGLAND
(Contribution from California Agricultural Experiment Statton)
PUBUSHED BY AUTHORHy OF THE SECRETARY OF AGRICULTURE.
WITH THE COOPERATION OF THE ASSOCUTION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINGTON, D. C-
'.'A^HINaTO^<:oOVERNUENTPmKTIf<a OFFICE ; lOr;
.lp'«;MWK.--*««BI»*iVM1Pftl(f^=«»««»*»»«"»''
0" k...^ )^
V
l:
EDITORIAL COMMITTEE OF THE
UNITED pf^Tt'^i I^lARf^EfW d^ AND
THE IlS^OCiA^IIM cAiu!^ AOl
COLLEGES AND EXPERIMENT STATIONS
fl^URAL
FOR THE DEPARTMENT
KARL F. KELLERMAN,XHAiRMAk
Physiologist and Associate Chief, Bureau
of Plant IndusUtv
EDWIN W.ALLEN
Chief, Office of Experiment Stations
CHARLES L.liiARtA'Ift'
EntomoloQist and AssisimU C^,, Bureau
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricultural Experiment
H. p. ARMSBY
Dit^ftor, Institute of Animal Nutrition, The
Pennsylvania State College
E. M. :^p:Efi^MAN'"
Botanist, fla^ jPaShphgist ftnd^ssfstv!^
Dean, Agricutiiiral Experiment Sta^pn of
the University of Minnesota ■ -■ - ■ ■
All correspondence regarding articles from the Department of Agriculture should be
addressed to K^l JF, Kellerjpaii, Joiup^ pf Agri<;ultut4 Resmch, W^
*Dr. Pearl has undertaken special work in cpn^ectioii with ti|e war epj!?rgency;
the^ore, until further "notice all cbrre^hierice regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
JOMAL OF AGRlOimm ISEARCH
Vol. XII Washington, D. C, February ii, 1918 No. 6
WATER EXTRACTIONS OF SOILS AS CRITERIA OF
THEIR CROP-PRODUCING POWER
By John S. Burd,
Chemist, California Agricultural Experiment Station
THE PROBLEM . V vuKfe
Given sunlight, suitable moisture and physical condition, the limi- ^'-^f ANiCAl
tation on the power of soils to produce crops is variously ascribed to the ^^^Oi^,
following general causes:
1. Lack of capacity of the soil to supply the plant roots at all times
with watery solutions of the essential elements in proper concentrations,
2. Presence of toxic substances.
3. Lack of physiological balance in dissolved soil constituents.
It is clear that any one or a combination of these conditions may be
the cause of low production, even though the remaining conditions may
be entirely satisfactory. The instances where the second and third in-
hibitive causes operate to limit production are perhaps quite numerous.
It may even be that, because of frequent occurrence, they are of as much
or more practical importance than incapacity of the soil to supply the
plant with proper nutrients. Nevertheless we are disposed to regard
such instances as special cases and to lay greater stress on the capacity
of the soil to deliver up solutes to the growing plant. Studies involving
this capacity are obviously more fundamental, in that they have to do
with an important function of all soils.
WATER-EXTRACTABLE SUBSTANCES IN SOILS
When we consider this function, we naturally turn to water as the
logical agent for the removal and determination of the soluble substances
in soils. The application of water as a solvent requires the use of a suffi-
cient amount in proportion to the soil to cause it to come into equili-
brium with the true soil solution and thus to insure the complete removal
of all dissolved matters when the soil suspension is filtered. Water in
excess unquestionably removes not only the substances already in solu-
tion but additional quantities somewhat proportional to the amount of
water used.^ Furthermore, v\e are told that the absorption of solutes
by plants is related to the concentration of the soil solution and not to
1 HOAGLAND. D. R. THE FREEZING-POINT METHOD AS AN INDEX OF VARIATIONS IN THE SOn, SOLUTION
DUE TO SEASON AND CROP GROWTH. In Jour. AgT. Research, v. 12, no. 6, pp. 369-395, 9 fig. 1918.
Journal of Agricultural Research, Vol. XII, No. 6
Washington, D. C. Feb. n, 1918
Ik Key No. Cal.— 15
(297)
298 Journal of Agricultural Research voi. xii, no. 6
the absolute amounts present in the soil. We heartily concur in this
opinion; but the impossibility of removing for investigation the true and
unmodified soil solution, even from soils whose optimum water content
is high, is too well recognized to require discussion. It is true that by
the method of Bouyoucos and McCool ^ we can determine the total con-
centration of the soil solution, but are left in the dark as to the concentra-
tion of the individual solutes in that solution.
In the present paper, therefore, our data are presented in terms of
the amounts of solutes extracted from soils and not in terms of concen-
trations. We shall endeavor to use this material in such a way as to
justify certain conclusions as to the presence or absence of power on the
part of soils to supply the needs of crops and to maintain optimum
concentrations in the soil solution. Hoagland,^ in this laboratory, has
shown that, while the total amounts of material extracted from soils by
an excess of water are invariably greater than those contained in the soil
solution, they are of the same general order of magnitude (as 2 is to 4
or 5), depending on the type of soil. Obviously, then, if the amounts of
soil constitutents extracted from cultivated soils can be shown to be
relatively high or relatively low, we may legitimately infer that the
amounts in the soil solution are correspondingly high or low. We do
not mean to imply that figures so obtained would necessarily indicate
the existence of an adequate supply of any element, because even low
figures might constitute adequacy. The question of adequacy or in-
adequacy may or may not be answered by such data, but it would cer-
tainly reflect the relative magnitudes of the present supply although it
might not indicate the soils' power of renewal.
Data presented by Stewart ^ in figures 8 to 20 show that normal seasonal
fluctuations in the amounts of the essential elements extracted at different
times by water (i to 5) from a cultivated soil (cropped or uncropped) are
likely to be as great as the variations between different soils. He also
shows that soils under crop (barley) invariably contain smaller amounts
of nitrate, potassium, calcium, and magnesium than their uncropped
duplicates, apparently reflecting the inability of soils to maintain their
initial concentrations and at the same time supply the needs of growing
plants. Since the fluctuations in the amounts of water-soluble substances
in soils are known to be quite large and the effect of withdrawal important,
it follows that the limitations of soils can best be shown by data obtained
during the period in which the withdrawal is actually taking place.
A comparison of the charts referred to above shows that, even though
good (uncropped) soils may contain considerably more solutes than poor
1 Bouyoucos, G. J., and McCool, M. M. The freezing point method as a new means of measuring
THE CONCENTRATION OF THE SOn, SOLUTION DIRECTLY IN THE SOIL. Mich. Agr. Exp. Sta. Tech Bui.
24, p. 592-631, 2 fig. I916.
* HOAGLAND, D. R. Op. cit.
3 Stewart, G. R. the effect of season and crop growth in modifying the sou, extract. In Jour.
Agr. Research, v. 12, no. 6, pp. 311-368, 24 fig., pi. 1918.
Feb. II, 1918 Water Extractions of Soils and Crop Production
299
(uncropped) soils, both good and poor soils are reduced to the same
general level at the time the crop is growing. The condition of soils under
crop is epitomized in Table I.
Table I. — Water extractable matters in soils^^ under crop. Average during period ^ of low
nitrate content
[Results expressed as parts per million of soil]
Productivity.
Good.
Medium.
Poor.
Soil No
I
2
5
6
8
II
14
4
7
10
8
13
45
3
5
6
64
9
5
7
54
Nitrate (NO3)
Phosphate (PO4)
Basic ions
(K+Ca+Mg)
5
7
65
5
5
71
4
17
108
5
7
54
5
12
62
4
23
72
4
II
76
6
9
130
5
7
69
5
13
41
o For a detailed description of soils, see Stewart, G. R. Op. cit.
b Period covered: Soil 7, twelfth to eighteenth week after planting; soils 12 and 14, eighth to eighteenth
week after planting; all others, tenth to eighteenth week after planting.
We observe in Table I a general uniformity in the magnitudes of the
various solutes present at the period of depletion in soils of varying pro-
ductive capacity. Apparently crops first reduce the supply of solutes
until these approach a comparatively low and (with a few exceptions)
fairly uniform level. Subsequent withdrawals must further reduce the
existing supply or be made good by new material coming into solution.
WITHDRAWAL BY CROPS
In the second year's experiments reported by Stewart we installed
eight additional containers of soil i , planted to barley, and harvested and
analyzed the crop from time to time. This particular soil was chosen
because it was known to be one of our most productive soils and the
amounts withdrawn by the crop would presumably represent the usual
draft to be expected from the crop under consideration. The data
obtained are presented in Tables II and III. In scrutinizing these data
it must be remembered that we are not attempting to show the require-
ments of plants or their nutritive habits, but merely their effect on the
soil.
If we now compare these data with the water-soluble content of the
soil, we may develop the approximate relation between the demand of
barley and the supply in a productive soil. The relation is expressed in
the column headed "Ratio." Literally interpreted, the ratio for each
Ion expresses the number of day's supply contained in the water extract
during the several periods covered. Thus, in the case of nitrate we see
that there was never less than a nine days' supply in the soil during any
period of the growing season. If the plant is capable of absorbing all of
300
Journal of Agricultural Research
Vol. XII, N'o. 6
the nitrate present in the water-soluble condition, we have direct evidence
that the water-soluble nitrate need only be renewed once in nine days even
at this critical period, while at all preceding and subsequent periods the
rate of renewal may be much slower. If we apply this same reasoning to
all of the ions, we develop the fact that there are always present in this
particular soil, in a condition capable of solution in water, at least —
Nine days' supply of nitrate.
Twenty-five days' supply of phosphate.
One hundred and forty-four days' supply of potassium.
Two hundred and sixty-five days' supply of calcium.
Seventy-six days' supply of magnesium.
Table II. — Rate of zvithdrawal by crop « as shown by periodic harvesting of soil I
iWithdrawals computed to parts per million of soil — 50 plants and i ,800 pounds of soil to the unit containerl
May I (planted)
Jtme 12
June 26
July 12
July 24
August 7
August 21
August 28
Nitrate (NO3). 6
Grams
per
plant.
•405
(?)
. 520
•594
. 629
.664
.694
Parts per million
of soil.
Entire.
24.85
31.90
36.40
38.40
40. 80
42. 50
Per
day.
0-59
Phosphate (PO^).^
Grams
per
plant.
o. 000
.038
•115
. 120
• 144
. 184
• 194
Parts per million
of soil.
Entire.
2.32
7. 02
7-35
8. 10
8.82
II. 20
II. 90
Per
day.
0-055
. 024
.054
.051
Potassium (K).
Grams
per
plant.
O. 000
• ^33
(?)
(?)
.223
•233
. 276
.294
Parts per million
of soil.
Entir e.
Per
day.
8. 131 o. 194
13-70
14.30
16. 90
18. 00
133
043
186
157
Date.
May I (planted)
June 12
June 26 ,
Jtily 12
July 24
August 7
August 21
August 28
Calcium (Ca).
Grams
per
plant.
o. 000
. 012
•034
(?)
. 041
(?)
.048
•043
Parts per million
of soil.
Entire. Per day.
o. 712
2. 092
2-530
2. 940
2.650
o. 017
. 016
. 014
Magnesium (Mg).
Grams
per
plant.
o. 000
.013
.031
.036
(?)
.044
•049
•051
Parts per million
of soil.
Entire. Per day.
0.816
1. 920
2. 190
2. 670
2. 980
3. 140
o. 019
.079
. 019
017
, 022
023
o Analyses based on varying numbers of plants: June 12, 69 plants; June 26, 42 plants; July 12, 42 plants;
July 24, 29 plants; August 7. 28 plants; August 21, 38 plants; August 28, 87 plants.
6 Nitrate (NO3) and phosphate (POi) computed from the amounts of nitrogen and phosphorus in the
crop.
The entire time the crop was in the ground was 119 days. The time
to elapse after the period of greatest depletion was 35 days for nitrate,
63 days for phosphate, 77 days for potassium, 63 days for calcium, and 63
Feb. II, 1918 Water Extractions of Soils and Crop Production
301
days for magnesium. It would appear that there is actually present in
the water-soluble condition at the period of greatest depletion more than
enough potassium, calcium, and magnesium to supply the entire subse-
quent withdrawals of the crop at the same rate, and that any renewal or
increase of these amounts should be unnecessary if the plant is capable
of utilizing the entire supply present.
Table III.
-Comparison of water-extractable matters in soil I (cropped) with daily draft
by plant
Nitrate (NO3).
Phosphate (PO<).
Potassium (K).
Weeks.
Parts per
million of
soil.
Ratio.
Weeks.
Parts per
million of
soil.
Ratio.
Weeks.
Parts per
million of
soil.
Ratio.
In
soil.
Daily
draft.
In
soil.
Daily
draft.
In
soil.
Daily
draft.
1-6, inclusive .
7-10, inclusive
11-12, inclusive
i3-i4,inclusive
69
4
3
8
O.S9
.14
117
16
9
57
18
1-6, inclusive . .
7-8, inclusive . .
9-10, inclusive.
11-12, inclusive
13-14, inclusive
15-16, inclusive
17th
7
8
10
3
8
5
0.055
.336
.024
.054
.051
. 170
. 100
127
24
417
55
157
29
1-6, inclusive . .
7-12, inclusive.
13-14, inclusive
15-16, inclusive
17th
28
29
34
38
0.194
•133
•043
.186
• 157
144
217
790
204
1
i 1
Calcium (Ca).
Magnesium (Mg).
Weeks.
Parts per
million of
soil.
Ratio.
Weeks.
Parts per
million of
soil.
Ratio.
In
soil.
Daily
draft.
In
soil.
Daily
draft.
28
26
18
21
1.6^7
1-6, inclusive
IS
6
14
9
20
0. 019
.079
. 019
.0x7
. 022
• 023
789
. 098] 265
.016 Iji:5
. 014 I, 500
76
17th
17th
1
It is evident that nitrate and phosphate are in a class by themselves,
in that the supply must be renewed at a much more rapid rate than any
of the other ions; but, inasmuch as this soil was highly productive, it
must have possessed the power to replenish the soil solution as required.
But we have shown heretofore that there is very little difference in the
absolute amounts of solutes between good and poor soils under a crop of
barley. It follows that figures of the same order of magnitude as those
just stated would be obtained by similar computations from the poor soils.
We conclude, therefore, that in soils in any degree fit for agricultural
purposes (with the possible exception of very abnormal types, such as
peat, etc.), the potentially soluble matters are sufficient in amount to meet
the requirements of crops; but it does not follow that the plant is capable
of drawing upon this supply at the concentrations corresponding to
these amounts.
302
Journal of Agricultural Research
Vol. XII. No. 6
CONCENTRATION AND RATE OF FORMATION OF SOLUTES
Since even our poor soils appear to be able to furnish adequate amounts
of all solutes at the most critical periods, we can only ascribe nutritive
difficulties of soils to inability on the part of the plant to absorb this po-
tentially soluble supply. The rate of absorption by any plant may be
limited by the possible supply and by the concentration of that supply,
but we have seen that in all cases which have come under our observation
the supply is adequate at all times. We should therefore either measure
the concentration of the soil solution itself or evolve some other means
of estimating the relative power of soils to maintain an optimum concen-
tration. Our data do not permit us, unfortunately, to estimate the
concentration of individual ions in the soil solution, but we may per-
haps get an idea of the relative power of soils to maintain or restore a
proper concentration by consideration of the data on soils with and
without crop. We present in Table IV the seasonal averages of the
water-extractable substances in cropped and uncropped soils.
Table IV. — Water-extractable matters in cropped and uncropped soils.
averages
[Results expressed as parts per million of soil]
Seasonal
Constituent.
Yield in bushels per acre .
Soil No
Nitrate :
Uncropped.
Cropped ....
Difference. .
Phosphate:
Uncropped .
Cropped . . . .
Difference. .
Potassium:
Uncropped.
Cropped . . . .
Difference. .
Calcitiin:
Uncropped.
Cropped. .. .
Difference. ,
Magnesium :
Uncropped.
Cropped . . . .
Difference.
Productivity.
131
36
95
40
31
9
34
24
10
23
13
10
120
87
55
50
5
37
19
18
15
7
146
30
116
18
17
47
14
91
58
Z2>
24
17
7
Good.
8s- 9
141
43
Z3
27
6
47
28
19
87.0
180
28
152
17
12
5
49
31
18
66
34
32
15
10
5
154
27
127
30
26
67
19
51
29
22
13
7
6
96
23
73
13
Aver-
age all.
S4. 7
137
30
107
14
52
40
12
51
30
21
17
10
7
Feb. II, 1918 Water Extractions of Soils and Crop Production 303
TabIvE IV. — Water-extr actable matters in cropped and uncropped soils. Seasonal
averages — Continued
Constituent.
Yield in bushels per acre
Soil No
Nitrate:
Uncropped
Cropped
Difference
Phosphate :
Uncropped
Cropped
Difference
Potassium:
Uncropped
Cropped
Difference
Calcium:
Uncropped
Cropped
Difference
Magnesium :
Uncropped
Cropped
Difference
Productivity.
Medium.
141
33
108
45
40
5
66
18
26
69.4
118
34
31
22
9
70. 6
130
39
91
49
30
19
Aver-
age all.
129
35
94
9
10
— I
42
31
66
44
16
56. 2
88
30
58
43
37
6
32
25
7
14
4S-8
54
18
36
44
35
9
83
23
60
30
20
10
41
24
17
Aver-
age all.
75
24
51
31
25
6
39
28
Examination of uncropped soiIvS. — ^We regard the magnitudes shown
as the resultant of the combined effects of previous withdrawals, the time
which has elapsed since the soil was last depleted, and the rate at which
the soil is capable of replacing solutes removed from the soil solution. A
soil might show high figures simply because previous withdrawals by
crops had been relatively low or remote in time. Nevertheless it is
significant that the average content of the more productive soils for each
constituent is relatively considerably higher than the average for the poor
soils. We infer from this that high figures ordinarily, but not necessarily,
indicate a relatively high rate of elaboration of solutes on the part of the
soil. A conspicuous example of an exception to this would be the case of
a so-called alkali soil, where large accumulations of solutes might have no
relation to the present elaborate power of the soil.
Examinations of cropped soils. — While all averages of the good soils
under crop are higher than the averages of the poor soils, the magnitudes of
the differences are quite small. Apparently the crop by withdrawal or
indirect effects on the soil tends to reduce the initial amounts of solutes
304 Journal of Agricultural Research voi. xn, no. 6
(and presumably their concentrations) to the same general level in both
good and poor soils. It seems probable that, when certain minima are
reached, the plant can not absorb sojutes, and subsequent withdrawal
then depends upon the capacity of the soil to elaborate additional solutes
as rapidly as the plant requires them. The figures themselves, however,
can not give a direct measure of this capacity, but small variations may
possibly reflect important differences in the rate at which solutes are
elaborated by different soils.
EquivaIvEncs of bases. — If we compare the figures for the individual
basic ions in good and poor soils (both in the cropped and uncropped
condition), we find numerous instances in which some one or more ions
are lower in the good soils than in the poor soils. These discrepancies
need not concern us if we regard the figures merely as a means of getting
at the power of soils to replenish the soil solution. In fact, this power is
more likely to be reflected by the aggregated soluble matters than by
individual constituents, and preferably by the aggregate of those con-
stituents of similar chemical properties. We propose, therefore, to con-
solidate our data for the basic ions in one figure (by addition) for subse-
quent consideration. If we further consider all of our data with reference
to the seasonal requirements of a good crop, we can perhaps indicate why
some soils are more productive than others.
In Table II we gave the amounts of the chemical elements found in a
good crop of barley produced on soil i , computed to parts per million of
soil. These figures were 42.50 p. p. m. of nitrate, 11.90 p. p. m. of phos-
phate, 18.00 p. p. m. of potassium, 2.9 p. p. m. of calcium, and 2.98 p. p. m.
of magnesium; or in round numbers, 42 p. p. m. of nitrate, 12 p. p. m. of
phosphate, and 24 p. p. m. of total bases. We assume these figures to
represent the probable requirements of a good crop of barley on all soils
and propose to examine our data with reference to the capacity of the
various soils to meet this requirement. In Table V we combine all
pertinent data available.
Nitrate. — The figures for excess in the good and medium soils show
an extraordinary discrepancy between the amounts of nitrate in the
uncropped soil and the amounts found in the cropped soil plus the with-
drawal of a good crop. We can not satisfactorily account for this dis-
crepancy, which appears to be ca,used in part by an inhibition of nitrifi-
cation due to the presence of the crop, but which also represents an actual
loss of nitrate from the cropped soil. Since the soils were kept in tight
containers, there could be no loss from drainage; and neither denitrifica-
tion, reduction to ammonia, nor the possible loss of ammonia by way of
the plant appeals to us as an adequate explanation. We can only sug-
gest that the presence of a crop may cause such a change in the biological
environment of the soil that the nitrogen of nitrates is stored in insoluble
(protein) forms in the soil. For our present purposes, however, it is
only necessary to point out that such losses appear to be a necessary
Feb. II, 1918 Water Extractions of Soils and Crop Production
305
incident to the production of a good crop. If the soil is not capable of
sustaining such losses, it is extremely probable that crops will suffer from
a lack of nitrates.
Table V. — Water-extractable matters with reference to seasonal withdrawals of a good crop
[Expressed as parts per million of soil]
Constituent.
Productivity.
Good.
Medium.
Poor.
Soil No
I
2
5
6
8
II
14
4
7
10
30
42
9
18
42
12
Nitrate (NO3)
36
42
33
42
30
42
43
42
28
42
27
42
23
42
33
42
34
42
39
42
23
42
(c) Suni(a+6)
78
131
75
120
72
146
85
141
70
180
69
154
65
96
75
141
76
118
81
130
72
88
60
54
f>f
8^?
63
55
74
56
no
85
31
66
42
49
16
—6
i8
Phosphate (P04):
6
12
5
12
17
12
8
12
12
12
26
12
II
12
10
12
7
12
13
12
6
12
6
12
12
12
(c)Smn(a+6)
18
7
17
5
29
18
20
8
24
17
38
30
23
13
22
II
19
7
25
II
18
7
18
6
24
((/) In uncropped soil
13
II
12
11
12
7
8
10
83
24
II
12
14
II
12
II
Bases (KCaMg):
68
24
76
24
108
24
65
24
75
24
84
24
128
24
73
24
61
24
73
24
61
24
50
(c) Sum (a+6)
92
97
100
107
132
162
89
lOI
99
130
108
131
107
113
152
155
97
lOS
85
112
97
89
85
72
74
80
5
7
30
12
31
23
6
3
8
27
-8
— 13
6
When we turn to the poor soils, we see that they all had in the un-
cropped condition more than enough nitrate to supply the actual amounts
withdrawn by a good crop; but all were incapable of sustaining addi-
tional losses of the magnitudes, which we must regard as normal.
Phosphate. — The figures for all soils under crop plus the requirements
of a good crop are invariably greater than the amounts in the uncropped
soils. If the figures for these latter represent in each case the maximum
amount of soluble phosphate the soil is able to hold in that condition,
it is quite clear that either the plant absorbs insoluble phosphate, or the
good soils replace the soluble phosphate as rapidly as it is required by
the plant. The latter explanation appears the more probable. But the
computed deficiencies of the poor soils are no greater than those of the
good soils, nor are the soluble phosphate contents of the former less than
many of the latter. The rate of solution of phosphate in the good soils
must be very high and we can find no reason to conclude that the poor
soils are in any way inferior in this respect.
Basic ions. — The amounts of basic ions in cropped soils plus the
requirements of a good crop are usually less than the amounts in the
uncropped soils. We note two exceptions to this in the poor soils Nos.
3o6 Journal of Agricultural Research voi. xii, no. 6
3 and 9. For these soils to have supplied the requirements of a good
crop and at the same time to have maintained their general level of
concentration would have necessitated that more bases come into solu-
tion. It is not impossible that they would have responded to the demand
for bases, just as they would probably have responded to the demand
for phosphate. The important difference between these poor soils and
the remaining soils is that the latter would never have been called upon
to furnish larger amounts of bases than the uncropped soil shows ability
to supply.
CRITERIA OF FERTILITY
The growth of a crop reduces the average nitrate content of soils to a
comparatively uniform level in soils of all degrees of productivity (see
Tables IV and V). The figures for nitrate in uncropped soils are always
higher than the known withdrawals by plants, but may not always be
sufficiently high to supply these withdrawals and certain other inherent
losses. The relative ability of soils to meet these losses may be inferred
from the amounts of nitrate in the uncropped soils.
Except in a few soils containing large amounts of soluble phosphate,
the growth of a crop does not reduce this ion to any considerable extent.
(The small differences shown approach the magnitude of experimental
error.) Furthermore, the figures do not show that good soils possess the
power of renewing the soluble phosphate more rapidly than the poor
soils. Figures for soluble phosphate can not be considered to reflect the
relative power of soils to supply the plant.
The growth of a crop reduces the basic-ion content of soils, but the
amounts remaining are still far in excess of the crop requirements. It is
not improbable, however, that their concentrations are in some cases
falling below the optimum requirements of plants. If this be the case,
comparison of the basic solutes of cropped soils may indicate deficiency
in this respect. The differences between the basic ion contents of cropped
and of uncropped soils are usually, but not always, greater than the
demands of crops. These differences therefore express the relative
power of soils to supply the crop requirements and to maintain the
concentration equivalent to the amounts shown by the cropped soil,
without drawing upon greater amounts of solutes than the uncropped
soil indicates capacity to supply.
Test op criteria. — If we arrange our soils in the order in which they
possess the various characters to which attention has been called, we
may be able to bring out salient differences between good and poor soils.
The order in which the soils occur in each character is frequently deter-
mined by very small variations in absolute amount of solutes, but it must
be remembered that these are based on considerable numbers of analytical
determinations and are probably not vitiated by experimental errors.*
1 Stewart, G. R. Op. cit.
Feb. II. 1918 Water Extractions of Soils and Crop Production 307
Very small differences at critical points may reflect significant differences
in the performance of soils (fig. i).
All of the poor soils appear among the three lowest in two out of three
characters. If we examine the medium soils, we find that Nos, 4 and 10
are among the three lowest in one character. If we draw a line over the
three lowest characters as a tentative indicator of the existence of sub-
optimal conditions, we find that five of our six medium and poor soils
fall below in one or more characters; that the remaining medium soil,
No. 7, approaches it in one character; that none of the good soils fall
below in any character. It is true that several of the good soils, No.
14, 6, and i, approach the line very closely, and we do not wish to be
understood as attaching too much importance to slight differences in the
Fig. 1. — Graphs showing soils arranged with reference to yield and important characters.
magnitudes of these characters. It would appear, however, that the
yields of soils 7, 3, 12, and 9 are closely correlated with their nitrate
content, but the mediocre yields of soils 4 and 10 can only be accounted
for by a reference to other characters.
We conclude that the nitrate content of uncropped soils is the most
valuable single criterion for appraising the crop-producing power of
soils; that the amount of basic ions in cropped soils is indicative of the
extent to which soils tend to maintain their concentrations when
subjected to depletion by crops; that the differences in basic-ion content
between cropped and uncropped soils may reflect the ability of the soil
to meet the demands of the crop without dangerous diminution of con-
centration and without drawing upon the reserve (iasoluble) supply
of the soil.
3o8 Journal of Agricultural Research voi. xii, no. 6
CONCLUSIONS
The evidence presented in the early part of this paper indicates that
there is always present in soils, in a condition permitting ready solution
in water, enough of the more important chemical elements to supply
the immediate needs of crops. It is hardly conceivable that substances
in this condition do not represent a potentially available supply. Inas-
much as this supply never entirely disappears, even in the case of
nitrates, it would seem that there is no such thing as a lack of available
nutrients in soils which are at all productive, but that a plant may still
be unable to satisfy its requirements if the concentration with reference
to the individual ions falls below certain minima. It is furthermore
highly probable that the optimum concentration varies with every soil
in accordance with the physicochemical system present in the soil solution.
Slight differences in the character of this system may modify in a marked
degree the power of a plant to absorb solutes, so that even if we were
able to obtain and analyze the true soil solution, we would not necessarily
be able to say that any figure for individual ions constituted inadequacy.
Attention has been called to certain characters as reflecting the com-
position of the soil and its power to produce crops. Inasmuch as these
involve three variables to which it is impossible to assign definite rela-
tive weights,anexactcorrelation of productivity with the figures presented
is not to be expected. Nevertheless, the correlation between the general
magnitudes of the figures presented and the crop-producing powers of
the soils studied is sufficiently close to justify the belief that they give
expression to the relative power of soils to produce crops, although they
are not an exact measure of that power.
We believe that the evidence obtained is sufficient to justify the hope
that we may be able to predict, within reasonable limits, thte relative
crop-producing powers of soils by comparing their figures expressing
these characters with similar data derived from soils whose productive
power is known. Before such a method is generally applicable, how-
ever, it will be necessary to study the behavior of many soils with numer-
ous type crops. This is quite feasible if the various characters can be
developed without the enormous number of analytical determinations
involved in the present experiments. We believe that this can be
accomplished by substituting for our figures, representing the sums of
the basic ions, figures obtained for total soluble salts, or preferably
direct determinations of the concentration of the soil solution by some
such method as that presented in the preceding paper.^ It is quite
certain that we shall never have a precise measure of soil fertility until
soils are studied concurrently in the cropped and the uncropped con-
dition and under strict control. The reasons are obvious in that data
from the soil under crop can not indicate its latent power, and data
'HOAGLAND, D. R. Op. cit.
Feb. II, 191S Water Extractions of Soils and Crop Production 309
from the uncropped soil taken alone do not take into account the fact
that the solutes in the cropped soils can not be reduced below certain
minimum and probably variable limits.
In the present paper we have dealt entirely with chemical criteria
because they afiford the most convenient expression of the results of the
activities of the soil, chemical, physical, and biological. We do not
wish to be understood as minimizing the importance of biological stu4ies
because we regard living organisms as being the most important single
agency, through the formation of nitrates and carbonic acid, in modify-
ing the soil solution. While biological studies have a most important
bearing on the proper treatment of soils, the resultant effects of all
activities upon these heterogeneous mixtures can only be developed by
the actual growth of crops and observation of their effects.
EFFECT OF SEASON AND CROP GROWTH IN MODIFY-
ING THE son. EXTRACT
By Guy R. Stewart/
Assistant Chemist, California Agricultural Experiment Station
HISTORICAL REVIEW
The first studies on the water-soluble material of soils were inspired
by the results of the investigations of the absorption of plant nutrients
carried on by Thompson (55)^ and by Way (60). Their epoch-making
discoveries in the year 1850 aroused a widespread discussion of the
manner in which the essential compounds might be held in the soil.
Liebig (59) was greatly impressed by this work and carried out a
series of investigations in which he studied the absorption of calcium
phosphate and potassium sulphate. He concluded that, since the
phosphate and potassium radicals were so readily absorbed by soils,
very little of these essential nutrients could be present in the soil water.
The results of lysimeter and drainage studies confirmed him in his belief,
and he proposed the theory that the plant roots must be able to draw
nourishment directly from the soil particles. At the same time Grouven
(22) had analyzed the extract from three soils obtained by percolating
6,000 c. c. of distilled water through 2,000 gm. of soil. He attempted
to relate these figures to the amount of material that would be brought
into solution by the season's rainfall.
Eichhom {14) also studied a soil near Bonn, Germany, in an attempt
to obtain a solution which would approximate the moisture existing in
the soil. He added 36.5 per cent by weight of water, and allowed it
to stay in contact with the soil for 10 days. He concluded that the
soil contained all water-soluble nutrients necessary for raising a crop.
This view was severely criticized by Wunder (63), who held with
Liebig {39) that the soil could not furnish sufficient water-soluble ma-
terial. Schumacher (52) repUed to Wunder (65), upholding Eichhorn's
views. Gradually, through the work of Peters (45), Jarriges (31),
Ulbricht (38), Hoffman (28), Wolff (62), and Cossa (13), it became
evident that the water extract from soils contained the major plant
nutrients.
1 The writer desires to make acknowledgment of the assistance of the Division of Soil Technology, of
the California Experiment Station, in selecting the soils employed in the investigation and in perform-
ing the physical analyses reported in Table II. The writer is also indebted to Mr. A. W. Christie, of
this station, for the performance of ammonification and nitrification studies and to Messrs. H. E. BiUings,
A. W. Christie, and J. C. Martin, of this Station, for assistance in portions of the analytical work per-
formed in 1916.
* Reference is made by number (italic) to " Literature cited," p. 364-368.
Journal of Agricultural Research, Vol. XII, No. 6
Washington, D. C. ^^^- "• ^9i8
I- Key No. Cal.— 16
(31O
312 Journal of Agricultural Research voi. xii, no. 6
The existence of appreciable quantities of water-soluble phosphates
was the point longest in doubt, though the results of Heyden {25) and
Schulze {51) established this fact satisfactorily.
The most notable of the early investigations and those which have
had the greatest influence on modern work were performed by Schloesing
(46, 47). His method consisted of treating 30 to 35 kgm. of soil with
an artificial rain and then collecting the first portions of clear solution
which ran through. This he believed to be identical with the actual
soil solution, and his was the first attempt to obtain it in an unaltered
condition. Schloesing showed the presence of all the principal elements
in this soil extract, and his procedure is still occasionally used in Euro-
pean work. Schloesing, jr. {4.8, 49), has continued this portion of his
father's work, devoting special attention to water-soluble phosphates,
and has concluded that there are differences in the phosphate content
of various soils, and also that there is almost enough soluble phosphate
present in most soils to supply an average crop.
The first work performed in the United States on the water-soluble
material of soils was that undertaken by King (36- j8) , and his coworkers
at Wisconsin. This was largely devoted to studies of the nitrate con-
tent of cultivated field soils, and was later extended to studies of the
total salt content by the use of conductivity methods. The results
obtained in this work may be considered to have been merely prelim-
inary to King's more extensive investigations performed in the Bureau
of Soils of the United States Department of Agriculture {34, 35).
Before this later work of King's appeared, Whitney and Cameron (61)
issued a publication from the United States Bureau of Soils which has
attracted more attention than any other single paper dealing with water
extracts. In it they gave the amounts of phosphoric acid, nitrates, cal-
cium, and potassium found in the water extracts of both good and poor
soils under varying conditions. The extracts employed for these analyses
were obtained by stirring 100 gm. of soil in 500 c. c. of distilled water
for three minutes. After standing for 20 minutes the liquid was decanted
into a cylinder of Brigg's (6) filtering apparatus and forced through
unglazed Pasteur-Chamberland filters under pressure. They concluded
that practically all soils gave essentially identical solutions and that
even where only a small quantity of one constituent was present, it was
sufficient for the growth of a fair crop if the mechanical condition of
the soil was good.
It was concluded also that the soil moisture was practically a saturated
solution of the mineral substances present in the soil. Consequently, as
fast as salts were removed by the plant further quantities were quickly
dissolved, thus keeping the solution at nearly the same concentration
throughout the growth of the plant. One of the most significant facts
claimed to have been shown by the investigation was that the equilib-
Feb. II, i9i8{ Effect of Season and Crop Growth on Soil Extract 313
rium of the solution quickly readjusted itself, at least as quickly as the
plant disturbed it by withdrawing nutrients.
It was therefore believed that the controlling factors in fertility were
moisture and the physical condition of the soil, and not fertilizers or
plant nutrients.
This was an entirely new viewpoint in soil investigations and has
proved extremely stimulating to other workers, though it should be
stated that Cameron {10-12) has somewhat modified his conclusions in
his later writings.
The first portion of the work of King {34), which had been carried on
concurrently with that of Whitney and Cameron (<5i) in the Bureau of
Soils, appeared a year later. It consisted of three papers, which were
published privately by the author, while the remaining three were issued
from the Bureau of Soils {35).
This work constitutes the most extensive investigation so far carried
out on the water extracts of soils. The methods employed were the
same as those used in the previous investigation of Whitney and Cam-
eron {61). The analyses in the preliminary work were performed on the
fresh samples of field soil, but later in the investigation, despite the large
error involved, oven-dried samples were employed.
The work of the first season was largely of a preliminary nature and
was principally carried out at Goldsboro, N. C. Additional samples
were taken in Georgia, Virginia, Maryland, New Jersey, Pennsylvania,
and Wisconsin. In the second season the work was more intensive and
consisted of a study of eight soil types in the four States of North Caro-
lina, Maryland, Pennsylvania, and Wisconsin. The crops used were
cotton, peas, beans, corn, and oats.
Applications of fertilizer consisting of 5, 10, and 15 tons of manure
and 300 pounds of guano were made to two crops, com and potatoes.
Analyses of the water extracts of the soils from these varied crops were
made from three to six times during the season, as well as numerous
extracts of the plant saps.
In general, the results and conclusions drawn were diametrically
opposed to those of Whitney and Cameron. Relatively large amounts
of nutrients were found to be either actually in solution or in such form
that they entered into solution when diluted with distilled water. It
was also shown that the application of fertilizers materially increased
the amounts of salts recovered from the soils. The largest amounts of
salts were, as a whole, found where the yields were largest, and the same
results were obtained from the examination of the plant sap.
The influence of farm manures was found to increase not only the crop
but also the amounts of soluble salts which could be recovered from the
soil.
27809°— 18 2
314 Journal of Agricultural Research voi. xii. no. 6
At this same period Gola {18-21) published the first of his ecological
and chemical studies on the relation of the soil solution to the natural
plant environment. He attempted to obtain, by a method founded on
that of Schloesing (46,47), an extract which would be similar to the
actual soil solution. The soil was broken up so that it would pass
through a 2-mm. sieve and was placed in a glass cylinder, 25 cm. high
and 4.5 cm. in diameter. A gentle rain of distilled water at the rate of
25 to 30 mm. per square centimeter per 24 hours, was then allowed to
fall upon it. After a period of time, which varied with the soil, drops of
solution issued from the lower tubulure, and the process was allowed to
continue till 25 to 50 c. c. of "pedolytic" solution had been collected.
Gola then subjected the saturated soil to pressure and collected what
be called the "pedopiezic" solution.
The total solids, and in some cases the total colloids, were determined
in these solutions. From the figures so obtained Gola divided up the
possible habitats of various plants into some 32 groups which he believed
to be controlled by the solution naturally occurring in the soil. In gen-
eral, he concluded that the relation between the soil and the organs of
absorption of the plant was controlled by the osmotic pressure of the soil
solution. High concentrations, and, especially, rapid changes in the
solution were likely to be harmful to plants, though many may grow
successfully in weaker solutions than were normal for them. The prin-
cipal factor determining the habitat of plants was the concentration of
the solution and, to a lesser extent, its chemical composition.
Snyder (55, 54), at Minnesota, carried on a short series of investiga-
tions in which he studied the absorption of nutrients from soil extracts
which were added to sand cultures, believing this condition comparable
to the absorption from the soil water. He arrived at conclusions op-
posed to those of Whitney and Cameron (61) and also criticised their
results from theoretical considerations.
Gedroitz (16), working in Russia, announced that the concentration of
the soil solution fluctuated so rapidly that it was impossible for any
water extract to give any indication of the character of a soil.
Mitscherlich (42) issued an extremely valuable contribution to the
study of the water extract. His method of procedure was essentially
different from those previously discussed. He employed for the extrac-
tion carbon-dioxid-saturated water at 30° C, the maximum temperature
which would probably be encountered in the soil. The proportions of
soil and water varied from i to 5 up to i to 30. At least two dilutions
were used in each study, generally i to 10 and i to 25. The soil and
water were placed in a thermostat, with a stirring apparatus running
into the center of the flask. Carbon dioxid was passed in constantly
and the extraction allowed to continue for ii^ hours.
From the data obtained by these varied extractions Mitscherlich
graphically estimated the amounts of additional material which were
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 315
dissolved by the water extract over and above that portion which was
actually in solution. This was a very valuable differentiation. His
conclusion that these graphs could be directly produced and would fol-
low the concentration of the actual soil solution is discussed by Hoag-
land (27) in connection with experimental data bearing on this point.
Mitscherlich was able by his procedure to distinguish between fertil-
ized and unfertilized soils and also between various grades of fertilizer
application. An excellent feature of his work was the complete and
careful estimation of the factor of error involved for all determinations
and its influence on the final result.
Ishcherekov (29) published the first of numerous attempts, which are
still continuing, to obtain the soil solution by the use of various reagents.
Briggs (6) and Briggs and McCall (7) had previously obtained small
amounts of solution from the soil by centrifugal force and capillarity
when the soil contained moisture slightly in excess of the optimum,
Ishcherekov now attempted to obtain the solution from a soil which was
close to saturation by pouring a layer of 0.5 cm. of ethyl or methyl
alcohol over the soil and assumed that the first portion of clear solution
which ran through was the actual soil solutio^i.
The same author later (50) reported a series of studies in which he
used the methods of the U. S. Bureau of Soils and from which he drew
conclusions which were practically in entire agreement with those of
Whitney and Cameron {61).
Engels {15), on the other hand, reported a series of studies in which
he used distilled water, carbon-dioxid-saturated distilled water, and 2
per cent citric acid and concluded that the citric acid was the most
satisfactory reagent to estimate the soluble material in soil.
Maschhaupt and Sinnige {41) conducted an investigation in which
carbon-dioxid-saturated water and 2 per cent citric acid were employed
and concluded that carbon-dioxid-saturated water was preferable.
Lyon and Bizzell (40) have attempted to estimate the density of the
soil solution indirectly by determining the relation of the dry matter
formed to the transpiration observed. They conclude that the addition
of f ertiUzer caused an increase in the density of the solution and obtained
confirmatory evidence by measuring the density of the soil solution with
the Wheatstone bridge.
Van Suchtelen {17, 59) has announced a modification of Ishcherekov's
(29) procedure in which he uses paraffin oil instead of alcohol. He
claims to obtain the soil solution in unaltered form, though the full
details of his method have not yet appeared. As the first announcement
of this method appeared in 1912, and the last statement of its pros-
pective full publication was made by Morgan {43) in 191 6, it is to be
hoped that it may soon appear in its entirety.
31 6 Journal of Agricultural Research voi. xii. no. 6
Ballenegger (2) has used the methods of the Bureau of Soils, as well
as determinations of electrical conductivity, in the study of 75 typical
Hungarian soils. He concluded that the character of the water solu-
tions may be used to differentiate the various types of soil. The soils
investigated varied from the poor gray forest soils to the very fertile
''at fold" soils.
Toulaikov {56, 37) believed, like Gola (19), that the osmotic pressure
of the soil solution was the important factor in plant growth. He
found the optimum to be a pressure of three atmospheres and that the
growth of wheat was benefited by an increase up to that point.
Pantanelli {44) has attempted to study the concentrations of the
solutions of soils from Tripoli by determining the electrical conductivity
of the liquids obtained by percolation. He was able by this procedure
to differentiate between several classes of cultivated and virgin soils.
Hall, Brenchley, and Underwood (25) have reported a noteworthy
investigation in which the soils from the Rothamstead experiment plots
were used. Solutions were prepared using 20 kgm. of soil and 35 kgm.
of water, and wheat and barley plants were grown therein. From the
analysis of these solutions and the growth of the plants in them it was
concluded that —
The composition of the natural soil solution as regards phosphoric acid and potash
is not constant, but varies significantly in accord with the composition of the soil
and its past manurial history. Within wide limits the rate of growth of a plant varies
with the concentration of the nutritive solution irrespective of the total amoimt of
plant food available. When other conditions such as the supply of nitrogen, water
and air are equal, the growth of crops will be determined by the concentration of the
soil solution in phosphoric acid and potash; which, in its turn, is determined by the
amount of these substances in the soil, their state of combination and the fertilizer
applied.
These authors' did not find any toxic effect on soils which had grown
wheat and barley for even 60 years. Growth in the soil solutions agreed
with the growth in culture solutions containing equivalent amounts of
phosphoric acid and potash.
In a series of nutritive solutions of various degrees of dilution the
growth varied directly, but not proportionately, with the concentration
of the solution.
Finally, the authors concluded that the results of the investigation
restored the earlier theory of the direct nutrition of plants by means of
fertilizers and nullified the theories advanced by Whitney and Cameron
(.61).
Bouyoucos and McCool (5) have proposed an ingenious method for
determining the concentration of the soil solution directly by means of
the freezing point. Use has been made of this procedure in the present
investigation by Hoagland, and its application and limitation are dis-
cussed in a separate paper (27).
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 317
Bogue (j) has published a brief investigation in which he leached
four soils with water and also studied their absorptive capacity for
potassium and phosphates. From his work he agreed with Whitney
and Cameron that the composition of the soil moisture is not influenced
by the chemical composition of the soil, but instead is dependent on the
mechanical texture of the individual soil.
Jensen {32) reported a series of observations on eight sugar-beet plots
at Rocky Ford, Colo. The methods of investigation employed were
those of the U. S. Bureau of Soils {50). The plots receiving composted
manure showed nearly twice as much water-soluble potash in the surface
foot as did any of the other plots. The seasonal averages of this element
were not appreciably influenced by any other fertilizer treatment. There
was a decrease noted in the quantity of water-soluble potash from the
middle of May till the middle of July. After that time the quantity
increased to approximately the amount that had been found earlier in
the season. None of the treatments resulted in a marked increase of
soluble phosphates, and the variation in this compound was less than
in any other element measured.
Jordan {33) has recently published the results of an investigation in
which nine soils were analyzed by complete analysis, acid extraction
with hydrochloric acid of 1.115 specific gravity, 10 days' leaching with
water, N/200 and A^/25 hydrochloric acid, and five hours' extraction
with the last three solutions. Vegetation experiments were conducted
with the soils in the greenhouse during two years. At the close of this
time it was impossible to establish any relationship between any of the
elements determined and the crop-producing capacity. There appeared
to be a slight relationship between the total soluble matter in the soil
and crop production, but it was not consistent in all cases. The final
conclusion was that no method had been developed by which the fer-
tility of a soil could be measured through laboratory investigation.
Harris and Butt (24), working in Utah, have studied the effect of
varying amounts of irrigation water upon the development of nitrates
and soluble salts in cropped and fallow soils. They found notable
differences between the cropped soil and the fallow duplicate, but did
not find that these differences were related to variation in the crop yield.
SCOPE OF THE INVESTIGATION
The foregoing discussion indicates the contradictory nature of the
results already obtained. Two of these moot points especially open to
further study are :
1. The relationship between the soluble soil nutrients in cropped and
uncropped soils.
2. The relationship between the soil extract and the crop produced
thereon.
These questions are the subject of the present study.
31 8 Journal of Agricultural Research voi. xii, No. 6
METHOD OF ATTACK
It is evident that in any soil a large number of complex factors deter-
mine production. These factors, besides influencing the final crop
obtained, can also conceivably modify the condition of the plant food
in the soil itself. It becomes very clear then that a chemical study of
the soil solution can hope to obtain a fair degree of success, only if a
number of these modifying factors are subordinated or made comparable.
Prominent among such factors are the influence of physical texture,
climate, and moisture. To reduce the effect of physical texture, two
types of soils were chosen: silty clay loams and fine sandy loams. The
representatives of each class were selected with as uniform a physical
texture as possible. Comparisons can therefore be made between the
soils within each class, with the assurance that the physical factor is
reduced to the minimum.
The effect of climate was made uniform by transporting exceptionally
large samples of surface soil to Berkeley. The samples were then
sifted and placed in containers which will be described later. Moisture
conditions were made as uniform as possible by adding just sufficient
distilled water to keep the soils at their optimum content.
SELECTION OF THE SOILS
The soils used in the investigation were chosen according to the map-
ping of the Bureau of Soils of the United States Department of Agri-
culture. The silty clay loams were all chosen from the Yolo series.
Of these, three were taken from the Sacramento Valley and three from
the Santa Clara Valley. The past history of all these soils was very
different, as shown in Table I. The crops grown on them were equally
divided between orchard crops (prune, almond, peach) and field crops.
None of the orchards had received any special treatment. Among the
field soils, the sample from the University Farm at Davis was the only
one which had been manured, though its cropping was no more varied
than No. 4 from the Santa Clara Valley. Soil 3 from Yolo had a less
varied history than any other in either set. This soil had been under
cultivation since about i860, and, except for two years in sugar beets,
191 1 and 1912, it had been steadily cropped with wheat or barley each
year. Its past production was stated to be good, fully the average for
that section, though no exact record had been kept of the yields.
Feb. II. 1918 Effect of Season a^id Crop Growth on Soil Extract 319
Table I. — Classification and history of the soils used in this investigation
Soil
No.
Soil series and type.
Origin.
Crop grown.
Past treatment.
I
Yolo silty clay
Sacramento Val-
Field crops. . .
Early planting of grain;
loam.
ley (University
Farm, Davis).
1909-1911, barley;
1912, fallow and ma-
nure; 19 13-14, barley.
do
Sacramento \ al-
Almond or-
Formerly grain; almond
orchard 12 years old.
ley (Yolo).
chard.
3
do
do
Barley
Planted about i860;
since then barley and
wheat, except sugar
beets in 1911-12.
4
do
Santa Clara Val-
Field crops . .
Originally grain; later
orchard; several years
ley (San Jose).
alfalfa; three years
field crops.
5
do
do
Prunes
Originally grain; prime
orchard about 20 years
old.
6
Yolo clay loam . . .
Santa Clara Val-
ley (Lawrence).
Peaches
Originally grain; peaches
for 8 years; heavy
crop, about 12 tons per
acre.
7
H a n f 0 r d fine
Southern Cali-
Oats
Originally grain; about
sandy loam.
fornia (Arling-
ton).
1890 put into alfalfa
for 13 years; potatoes
2 years, alfalfa 4 years,
oats 5 years; yield of
oats, 4 tons of hay per
acre.
8
Fresno fine
San Joaquin Val-
Seedless
Originally grain; 14
ley (Fresno).
grapes.
years in Sultanina
(Thompson Seedless)
grapes. Production
about 2 tons of raisins
per acre for last 6
Q
Kimball fine
Southern Caiifor-
Navel or-
25 years in oranges; pre-
sandy loam.
n i a (Re d -
lands).
anges.
viously bare land. A
great variety of fertili-
zers had been used.
10
T e j u n g a fine
Southern Califor-
Peaches
Originally 15 years in
sandy loam.
nia (San Fer-
nando Valley).
prunes; now 10 years
i n peaches; small
amount or mantue the
only treatment.
II
Madera fine
San Joaquin Val-
Navel or-
Orange trees about 15
sandy loam.
ley (Kearney
Park).
anges.
years old.
12
Arnold fine
San Joaquin Val-
Oats
In cultivation about 40
sandy loam.
ley (foothills).
years; early crops
largely wheat; last foiu-
or five years biennial
crops of oats; alternate
year summer fallow.
13
Unnamed fine
Mendocino Coim-
\ irgin
Very shallow soil, about
sandy loam.
ty coast.
I foot in depth under-
lain by clay subsoil.
14
St an dish fine
Honey Lake re-
do
Desert soil, small shrubs
sandy loam.
gion.
and weeds, natural
growth.
320 Journal of Agricultural Research voi. xii. no. 6
The fine sandy loams had an even more varied past history. Three
of these soils were from southern California. One of these was from a
Redlands orange grove which had been treated in the past with a great
variety of fertilizers. A second was from a peach orchard of the San
Fernando Valley, and the third from the Riverside area had been de-
voted to field crops.
Three fine sandy loams were obtained in the San Joaquin Valley. Two
from the vicinity of Kearney Park, in oranges and Sultanina (Thomp-
son Seedless) grapes, had an excellent record for past production.
The third from the foothills near Oakdale was purposely chosen from a
body of Arnold fine sandy loam where the growth of oats was unusually
small, though the soil was normal in depth and drainage. The last two fine
sandy loams were virgin soils. One of these (No. 13) came from the coast
of Mendocino County, and was an extremely poor, shallow soil. In
fact, it later developed that this soil was very acid; therefore no de-
ductions have been drawn from the results obtained with it. The other
virgin sample was obtained from the Honey Lake area. This soil was
considered to be rather in the class of desert soil, but was not especially
deficient in organic matter.
The soils of both types were selected to include as many past treat-
ments and crops as possible. They had, of course, been exposed to very
different climatic influences. This is an unavoidable factor in Cali-
fornia, with its great diversity of local conditions. It is believed that
these various influences have been largely neutralized by the cropping
of the first season, 191 5, under uniform conditions. Greater emphasis
is therefore attached to the results of the past year (1916).
The attempt was also made to choose soils which, though of uniform
physical texture, would have a very different crop-producing power.
In this way it was hoped that some light might be thrown on the relation
between yield and water-soluble nutrients.
In Table II are given the moisture equivalents, hygroscopic coeffi-
cients, specific heat, and mechanical analyses of the 14 soils. Moisture
equivalents are obtained by the method of Briggs and Shantz (<?);
hygroscopic coefficients by the method of Hilgard (26). Specific heat
was determined by heating 20 gm. of soil in a tin-foil container until it
reached a temperature of approximately 100° C. It was then lowered
into an insulated tank of water of known temperature and volume. The
amount of heat contained in the soil was measured by the increase of
temperature of the water. The mechanical analyses were performed
by the method of the Bureau of Soils (9). It should be noted that the
analyses as reported were performed on the same samples which were
used for the chemical analyses. These samples had been passed through
a i-mm. sieve. This was done in order that the chemical and physical
analyses should be made on identical portions. The percentage of gravel
by this screening process is also stated, so that the mechanical analyses
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 321
may, if desired, be recalculated to the usual form. A study of these
figures shows that the members of each group of soils in physical texture
are as similar as was desired. From these analyses Mr. C. F. Shaw, of
this Station, states that the silty clay loams would be accurately classed
as —
clay loams with a very high percentage of fine sand and silt, and that with the excep-
tion of No. 6 they would function as very tmiform silty clay loams. No. 6 would be
classed as a clay.
The fine sandy loams are considered to be quite uniform in texture, so
far as porosity, ease of root penetration, and behavior under cultivation
are concerned. The moisture equivalents are very different, so that at
first glance it would appear that the plants would be unequally sup-
plied, especially in the fine sandy loams. It will be pointed out later,
however, that this view does not consider the fixed or "unfree" water
of the soil. The amount of moisture, therefore, which is available to the
plants, as shown by the wilting coefficient, and also the amount which
affects the concentration of the soil solution, can not be judged from
the moisture equivalent alone.
Table II. — Moisture equivalents, hygroscopic coefficients, specific heat, and mechanical
analysis of experimental soils
MOISTURE EQUIVALENTS
Test No.
Soil
I
2
3
4
5
6
7
8
9
10
II
12
13
14
Test I
Test 2
Test3
Test 4
27-51
27-27
27-08
27- 21
25-23
25-06
24-93
24-77
27.44
27. 10
27- 26
27.69
23-62
23-95
24-53
24.01
24. 12
23.29
24-38
23-75
30.57
30.09
30.24
30.22
13-61
13-45
13.06
13-10
7-77
7-94
8.24
8.30
9-95
9.27
9-73
10-53
18.41
17-98
18.76
16. 29
16.00
16.61
16. 24
9. IS
10.00
9-72
10.01
18- 57
19.02
18.22
18.43
17-84
17-45
17-99
Average...
27.28
25.00
27.37
24-03
23.88
30.28
13-31
8-06
9-87
18.38
16.28
9-72
18.56
17.99
HYGROSCOPIC COEFFiaENTS
Test I
Test 2
7-9S
7.92
6.09
6.36
8-33
8.08
S-49
S-77
8.11
8.14
8.82
8.84
2.83
2-94
I- IS
I. 20
1-57
1-49
2.93
2-95
3-14
3-18
1.85
1-68
1-31
1-3°
4-72
4-76
Average. . .
7-94
6.23
8.20
5-63
8.12
8.83
2.89
1.17
1-53
2-94
3.16
1.77
1.31
4-74
SPECIFIC HEAT
Test I
Test 2
0-2122
.2095
0-2057
. 2042
0-193
. 200
O-1901
.2008
0- 2246
.2245
0.243
.242
0.1969
.1876
0.1758
.1661
0.230
.234
0- 184
.188
0. 2480
. 2501
0- 1807
• 1656
0- 1933
.i860
0-224
• 230
Average...
.211
.205
.1965
•1954
.2245
.2425
.1918
.1709
.232
.186
.2490
•1731
.1897
.227
322
Journal of Agricultural Research
Vol. XII, No. 6
Table II. — Moisture equivalents, hygroscopic coefficients, specific heat, and mechanical
analysis of experimental soils — Continued
MECHANICAIv ANALYSES
[Averages of two analyses)
Kind of soil.
Soil
I
2
3
4
5
6
7
8
9
10
II
12
13
14
Fine gravel
( 1-2 mm.)-
o.oo
0.00
o-oo
0.00
O-OO
0.00
0.04
o-oo
0.016
o-oo
0.00
0.00
0.04
0.00
Coarse sand
(i-i-snun.)
• 15
.60
.60
2.09
1.84
•746
2.82
1-95
9-986
1-57
2-77
4.99
7-77
3.09
Medium
sand (0.5-
0.25mm.) .
Fine sand
•23
1-50
1.04
1.24
1.70
-574
2.87
3-19
6.308
2-94
2.818
4-38
J:4^55
4^05
(0.25-0.010
mm.)
1.69
8.00
3-53
5-47
9.61
1.786
15.60
21. 22
21-730
16.62
20.945
23.27
13-49
24.54
Very fine
sand (o.oi-
0.05 mm.) .
32-62
31- 26
26.28
30.32
32-73
21-775
53-67
50-63
43-67
38.76
39.34
46.24
IS- 70
38.94
Silt (0.0s-
0.00s mm.)
42.68
36-97
38.12
37-99
30.52
37-415
16.01
17-57
10.87
26.25
19-075
9.61
34-72
15.55
Clay (0.005-
omm.)
24.17
22.47
28.47
25.08
24.72
39-380
9-77
S-S6
8.19
13-37
14.62
9.44
12.9s
13.87
Sum of
percent-
ages
101.54
100.80
98.68
102- 19
101. 12
101.676
100.58
loo- 12
100.77
99-51
99. 568
97-93
99.22
100.04
Gravel r e -
moved by
previous
screening. .
. II
.09
•39
6- IS
3-38
.11
3-64
•99
6.41
-59
2. 12
1-57
I- IS
2. 21
In choosing the soils in the field the following plan was adopted.
One soil of each type was taken as the typical sample. Then each suc-
ceeding member of that group was chosen by comparison with a small
sample of this type. This same small sample was carried while all the
work was being done on that group, and constant comparisons were
made when there was ground for doubt. In this way it was believed
that each member of a group would approach very closely to the other
members in general physical texture.
The samples in each case were taken from the top foot of soil. Pre-
liminary borings with a soil auger were made in order to ascertain whether
the subsoil was free from hardpan or other abnormal factors which might
influence the surface soil. The amount of each of these samples was
2>^ tons. The soil was shipped in new, heavy burlap sacks holding ap-
proximately 100 pounds each. Every attempt was made to prevent
undue drying or exposure to sunshine after the soil was sacked.
Upon the arrival of the samples in Berkeley they were immediately
passed through X-iiich concrete sieves to obtain a uniform physical tex-
ture. This work was performed with all possible expedition, and it is
beUeved that they had a normal bacterial flora when placed in the
containers. None of the soils, except No. 14, was actually air-dry when
finally prepared for use. The ammonifying and nitrifying powers of the
soils have been determined as a test of normal biological activity. This
work was performed at the beginning and close of the season of 191 5
and at the end of the growing season in 191 6. The tests were carried out
Feb. n. 1918 Effect of Season and Crop Growth on Soil Extract 323
by the well-known tumbler method. All the soils showed a satisfactory
capacity for ammonification.
At the close of the season of 191 5, soils 12 and 14 were noticeably
lower in nitrifying power than the others of the group. At the close of
1 91 6, soil 12 alone showed a significant lower range in nitrification. With
these exceptions the soils gave results which indicated normally active
bacterial floras. The detailed data are not considered sufficiently il-
luminating to justify their inclusion in this paper.
SOIL CONTAINERS AND INSTALLATION
The soil containers selected were 30 inches wide, "60 inches long, and
18 inches deep. They were made of No. 24 galvanized iron and were
thoroughly coated with asphaltum varnish. When filled, each con-
tained approximately i ,700 pounds of soil. The design of the containers
I X /4- iron s>{rao -y
Perforated '/z' pipe for sub irrigation,
hofe-i 2" ap>ort.
Fig. 1. ^Design of soil containers.
is shown in figure i . It will be observed than an outlet is placed in the
bottom at one end. This outlet not only prevented the accumulation
of excess water, but also gave additional aeration to the lower soil.
The escape pipes may possibly be seen in the illustration of the wire
house inclosing the containers (PI. 14, A).
Distilled water was the only moisture used at any time. This was
added partly between the rows of grain by a long-spouted sprinkling
can and partly by subirrigation through the two perforated pipes running
from end to end of the containers. So far as possible, only enough water
was added at one application to keep the soil at optimum moisture. It
was found possible to observe the moisture condition quite closely by
drawing out a core of soil with an i8-by-^-inch cheese trier, and then
replacing the soil.
The containers numbered 28, 2 for each soil, and were arranged on
level mud sills in two duplicate sets, as indicated in the accompanying
diagram (fig. 2). By the above arrangement the smallest possible
324
Journal of Agricultural Research
Vol. XII, No. 6
external surface of each container was exposed to any temperature
change. These changes were further minimized by surrounding the set
of containers with a boxing, i8 inches high, placed 6 inches from the
outside of the entire group. This 6-inch space was then filled compactly
with local soil. The insulation furnished by this arrangement was excel-
/&-&■
/©•-6"
Fig. 2. — Diagram of the arrangement of the soil containers.
lent. No difference in growth was observed in any portion of the con-
tainers.
Protection from birds was assured by a framework, 4^ feet in height
above the containers, covered with i -inch-mesh wire netting. The general
arrangement of the two sets is shown in Plate 14, A.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 325
During the growing season selected for these experiments, from May
to September (the first season, however, being from June to October),
very little rain falls in Berkeley. To protect the soils from flooding by
any heavy showers that might occur, waterproof canvas covers have
been provided and have been put over the wire houses in a few
emergencies. During the season of heaviest rains these covers are kept
on, and the containers are never flooded or subjected to leaching by
rain. In this season between crops the soils are watered with dis-
tilled water at intervals, so that they are close to the optimum at all
times. It is, of course, true that this is not comparable to the seasonal
changes to which field soils are subjected. On the other hand, such
seasonal changes of rainfall and drouth are never comparable for any two
seasons or any two places. Many orchard soils in this State, by con-
stant cultivation and irrigation in the summer, retain a very regular
moisture supply in the soil. The moisture conditions of this experiment
are, in fact, very similar to those found in such cases. The orchard, in
addition, is subjected to occasional large excesses of water from rainfall.
After filling the containers with the prepared soils, the excess portions
of sifted soil were stored in the set of tightly covered bins shown in
Plate 14, B. This soil was available for supplying the small portions
removed by sampling and also furnished a stock for additional studies.
ANALYTICAL PROCEDURE
In the season of 191 5 the analytical methods used were those outlined
in United States Department of Agriculture Bureau of Soils Bulletin 31
{50). With pure solutions containing only a single compound the de-
terminations obtained by these methods are very satisfactory. Before
the close of the season considerable doubt was felt in regard to certain
of the results obtained. This applied especially to calcium, potash, and
phosphate. Before entering upon the work of the season of 191 6 a series
of studies were outlined to test the accuracy of the methods which had
been in use. The preHminary work previously performed by the pro-
cedures in Bulletin 31 had all been carried out with solutions which con-
tained only one pure salt of the radical to be determined. Under these
conditions the results were extremely accurate.
In the studies now undertaken three sets of solutions were prepared
to simulate as closely as possible the highest, average, and lowest con-
centrations ordinarily obtained in soil extracts. The salts used were
monocalcium phosphate, calcium nitrate, magnesium sulphate, potas-
sium chlorid, and ammonium chlorid. The solutions were made up with
distilled water which had stood in ordinary glass bottles and so con-
tained small amounts of sodium and silica, though not of course com-
parable to the amounts which occur in water extracts from soils.
326 Journal of Agricultural Research voi. xii, No. 6
Quadruplicate portions of these solutions were analyzed without further
treatment. Large aliquots from the same stock bottles were also passed
through new and old Pasteur-Chamberland filters, and some portions
were treated with G. Elf carbon black. These treatments were included
to check the possible effect of absorption of elements by the filter candles,
as well as the effect of the carbon black which was used to decolorize
soil extracts containing organic matter.
Table III gives the results of the analyses of the untreated and of the
filtered solutions. It will be seen that the determinations of ammonia
and nitrate are satisfactory throughout. The methods used are essentially
those almost universally applied in sanitary water analysis (z) and with
the proper precautions are not believed to be open to criticism. The re-
sults for potassium, calcium, sulphate, and phosphate are seen to be
extremely inaccurate in the lowest concentration. Of these the calcium
is by far the most questionable. The error in the determination of this
element is so large that the results for calcium obtained in 191 5 have con-
sequently been discarded. Even with the potash and phosphate the
percentage of error is very large in some determinations. In the higher
range of concentrations the results are much closer throughout, and if it
were possible to confine all the work done to such solutions there would
perhaps be no objection to the methods employed. In many solutions
smaller concentrations of one or two elements are frequently encountered.
In such a case the error may be 50 per cent or more, either plus or minus.
Such a variation can readily obscure any significant change in concentra-
tion. It became necessary to obtain a more reliable procedure for potas-
sium, calcium, and phosphate, all of which are extremely important
elements in any study of soluble plant nutrients.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 327
Table III. — Analysis {in parts per million) of dilute, average, and strong solutions
by the methods outlined in United States Department of Agriculture Bureau of Soils
Bulletin ji
Ca
Mg
SO4
P04
NH3
NO3.
K.
(turbidi-
(colori-
(turbidi-
(colori-
metric).
metric).
metric).
metric).
Solution and treatment.
.a
■d
d
•d
.a
•a
.a
•d
01
.a
-s
•a
u
c
a
1
>>
g
>.
a
H
&
a
v..
1
0
0
0
01
0
0)
0
0>
0
0>
0
0>
0
(D
s
&>
01
0>
01
-l-l
0>
-*->
0
ja
(U
ja
a
ja
t<
J3
V
.a
V
j:
01
j:
Oi
H
0
H
0
H
Q
H
Q
H
0
iri
Q
H
Q
fo. 50
0.48
4.00
4.08
I. 2
S-72
i-S
1.76
0.50
z.
1.99
2.72
I.O
2-5
Dilute, untreated
1 .50
.50
.48
.48
4.00
4.00
4.08
4.00
I. 2
I. 2
1.92
1.92
S
s
1.68
3- 80
•so
•SO
.46
•49
1.99
1.99
2.68
3^6o
I.O
1.0
2.4
1-6
•50
.48
4.00
4. 16
1.2
1.80
s
2.08
•50
•44
1.99
2.60
I.O
1.9
f -50
•45
4.00
4. 16
I. 2
3-4°
5
.48
•SO
•44
1.99
2.68
I.O
1.6
Dilute, filtered through Pasteur
1 .50
•45
4.00
4.08
1.2
3^6o
S
1.84
•SO
•44
1.99
2. 60
I.O
!•?
filters
• 50
,40
4. 00
4.08
4.08
1. 2
3- 20
3-12
5
5
1.24
1.60
•SO
•SO
.46
•44
1. 99
2. 76
I.O
1.9
1-7
•50
•4S
4.00
I. 2
1.99
2.72
I-O
Dilute, treated with carbon black
2.88
and filtered through Pasteur
. -5°
•43
4.00
3-92
I. 2
2. 12
5
1.44
.50
•56
1.99
I.O
!•?
filters
•SO
/.50
I. SO
fi-SO
.40
.42
4.00
4.00
3-92
3^68
I. 2
I. 2
2.72
3^24
S
5
1.28
1.92
•SO
•SO
•S6
•54
1.99
1.99
2.80
2.60
I.O
I.O
1.6
Dilute, passed through new Pas-
^•S
teur candles
.40
1.66
4.00
20.00
3-84
19-2
1. 2
2.64
4- 30
5
S
I. 08
•SO
2.50
•49
2.36
!• 99
3-68
I. 0
1-7
6.0
6.00
8.00
9.8
9.00
5-0
Average, untreated solu-
J I- SO
I. 58I20. 00
19.6
6. GO
S^io
5
7-70
2. 50 2.41
9.8
8.8
5.0
5^2
tion
|l-SO
I. SO
I. 58120.00
20. 0
6. 00
S-88
6. 24
5
5
7.502.50
7.902.50
2.30
2.36
9.8
10. 0
S-o
S-o
S-2
5-0
1.58
20.00
20.0
6.00
9.8
8.8
1.50
1. 61
20.00
19.2
6.00
6. 64
5
7.902.50
2.66
9.8
9.6
S-o
4.8
Average solution filtered
J I- so
1. 61
20.00
18.8
6.00
8.69
_
S
8.00^2.50
2.61
9.8
9.2
5.0
4.0
through Pasteur filters ....
1.50
1. 61
20.00
19.2
6.00
8.40J 7
S
7.60|2.so
2.61
9^8
8.8
S-o
4.6
I. SO
t.64
20.00
18.8
6.00
8.00
7
S
7. 70 2. 50
2.66
9^8
9.2
S-o
S-o
Average solution treated
with carbon black and
,i-SO
1.64
20.00
18.6
6.00
6.00
7
5
8.00 2.50
2.61
9.8
13-4
S-o
4-3
filtered through Pasteur
filters
1.50
I. 64
20.00
18.6
6.00
6.00
7
5
8. 20 2. 50
2-43
9^8
II. 8
5-0
4.6
Average solution passed
through new Pasteur fil-
.1.50
1.64
20.00
ig. 2
6.00
6.96
7
S
8.302.50
2.30
9.8
II. 8
S-o
4.8
ters
I- SO
(3.00
1.64
3-07
20.00
38.00
18.0
37-6
6.00
12.00
6.56
7
S
0
7.802.50
16. 60,5.00
2.41
5-31
9-8
19-7
II. 8
20.0
5-0
10. 0
4-7
10.66 15
10.4
Strong solution, untreated
I3.00I2.94
38.00
38.00
37^6
38.4
12.00
12.00
10. 66 15
13- 12,15
0
0
16.40 5.00
16. 40 5.00
4.80
4-74
19.7
19.7
19.0
18.2
10. 0
10. 0
10.8
13.00
2.94
10.4
3-0O
2.94
38.00
38.4
12.00
1 1^76 15
0
i7.oo'5.oo
4-93
19.7
19.0
10. 0
10.2
[3.00
2.86
38.00
38.4
12. 00
12.32 15
0
16. 60 5.00
4-93
19-7
19- S
10. 0
9.8
Strong solution passed through
3-00
3-07
38.00
37-6
12.00
12.49 15
0
16. 80 5.00
4.80
19-7
18.2
10. 0
9-2
Pasteur candles
3- 00
2.90
38.00
37-6
12.00
I3^s6i5
0
16. 40 5.00
4-86
19.7
18. s
10. 0
9.2
3- 00
2.94
38.00
38.4
12.00
li.6o IS'O
16. oo's- 00
4^74
19-7
18.7
lO.O
9-4
The determination of magnesium, though it is the same procedure as
used for phosphate, has, at least in this series, given more accurate
results. The same standard solution is used in this method as in the
phosphate determination, and has a value approximately one-fourth as
great when expressed in terms of magnesium, a fact which would decidedly
tend to reduce the percentage of error in the final result. The same
colorimetric procedure for magnesium is still in use, as no more satis-
factory method for small amounts of this element has so far been
developed. The sulphate determination is seen to be more accurate
than that of calcium, but the error in low concentrations is from 35 to
85 per cent. This procedure has therefore been dropped and the sulphate
determinations are not included in the 191 5 charts.
In planning the work for 1 916 it was evident that a smaller number of
more accurate determinations were to be preferred to a larger number
on which little reliance could be placed. With this idea in mind, work
328 Journal of Agricultural Research voi. xii. no. 6
was undertaken in which larger aliquots than those recommended in
Bulletin 31 were used. The results obtained by the use of larger portions
of solution yielded, in general, more reliable figures. If larger quantities
of solution were to be used, a great deal of the argument in favor of the
colorimetric and turbidimetric procedures disappeared. The small por-
tions of solution required for these methods have always been considered
one of their great merits. In the soil extracts which contained the
largest quantities of plant nutrients, if larger aliquots were taken, it
was very clear that the ordinary volumetric and gravimetric procedures
could be used. It then became important to learn whether the standard
methods of determination could not be applied to the weaker solutions
if a carefully standardized technic were adopted. In the work under-
taken along this line the dilute and average solutions previously men-
tioned were used. As a result, it is believed that a more satisfactory
procedure than the colorimetric has been developed for phosphates,
calcium, and potash. The methods used are quite well known, but a
great deal of work was done to establish the exact conditions for accurate
results with the dilute solutions employed.
The description of the detailed procedures used follows.
Phosphate. — Evaporate two portions of soil extract of 200 c. c. each in a 200-c. c.
porcelain casserole. This size of casserole stands ignition excellently. Add a few
drops of hydrochloric acid diluted i to i before the above solution reaches dryness
to aid in decomposing soluble silicates. Ignite the dry residues in the casserole
over a Meeker burner at a moderate temperature till a grayish white residue is
obtained. Cool. Takeup the residue with loc.c. of nitric acid diluted i tog. Cover
with a watch glass and digest on the steam bath for approximately 10 minutes to
insure complete solution. Filter into a 200-c. c. Erlenmeyer flask. Wash the
casserole with two more portions of 10 c. c. each of hot nitric acid diluted i to 9.
Wash the dish, and filter with small portions of hot water. Keep the total volume
of solution to about 50 c. c. Cool. Add a few drops of methyl orange indicator
and neutralize rapidly with concentrated ammonia. Bring just to acid reaction
with concentrated nitric acid. Add 2 c. c. of saturated ammonium nitrate. Place
the flasks and a quantity of properly acidified ammonium-molybdate solution ^
in a water bath at 50° C. When all the solutions have reached this temperature,
add 5 c. c. of molybdate to each determination. Keep at 50° C. for K tour. Remove
from the water bath, and filter at once on prepared asbestos felts. Wash with cold
distilled water till free from acid by the usual tests. ^ Transfer filter felt to the same
flask, using approximately 25 c. c. of distilled water free from carbon dioxid. Add
15 c. c. of sodium hydroxid of which i c. c. is equivalent to o.i mgm. of phosphorus-
pentoxid and observe carefully whether the solution of the yellow precipitate is
complete. Titrate the excess alkali with hydrochloric acid of the same strength,
using phenolphthalein as indicator. Calculate results to either elementary phos-
phorus or the phosphation, as desired.
Calcium. — Evaporate 200 c. c. of water extract to dryness in a 200-c. c. casserole.
Ignite at a moderate temperature over a Meeker burner till a grayish white ash is
obtained. This step is desirable to remove traces of organic matter, even though the
'Wiley, H. W., ed. ofpiciai, 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.), p. 4. 1908. Reprinted in 1912.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 329
extracts may be decolorized with carbon-black. Take up with 20 c. c. of hot distilled
water acidified with hydrochloric acid diluted i to i. Digest for a short time on the
steam bath to insure total solution. Filter and wash the dish and the fimnel with
hot water. Do not permit the total volume of solution to exceed 50 c. c. Make
faintly alkaline with ammonia, heat to boiling, add 5 c. c. of 10 per cent ammonium
chlorid and 5 c. c. of saturated ammonium oxalate. Cover solutions with watch
glasses, place on steam bath at moderate temperatiue (about 60° C), and allow to
stand overnight to insure complete precipitation. Filter on quantitative filter paper.
Wash till free from chlorids, using successive small portions of hot water. Dissolve
the precipitate in 50 c. c. of hot sulphuric acid diluted i to 4, receiving the acid and
hot water subsequently used for washing the filter paper in the beakers used for
precipitation. Titrate the hot solution with N 1I200 potassium permanganate, i c. c.
of which is equivalent to o.i mgm. of calcium.
Potash. — Measure 200 c. c. of water extract into a 200-c. c. casserole, acidify with
0.5 c. c. of sulphuric acid diluted i to i, and evaporate to dryness on the steam bath.
Heat gently on an asbestos hot plate and keep the casserole constantly in motion to
prevent spattering imtil the excess of sulphuric acid has volatilized. Then ignite
over a Meeker burner until a white or gra^dsh white residue is obtained. Cool and
take up the residue in successive small portions of hot distilled water. Filter through
5.5 cm. quantitative filter paper, using 30 to 40 c. c. of water. Acidify with three or
four drops of hydrochloric acid diluted i to i. Add 8 to 12 drops of platinic-chlorid
solution and evaporate to a pasty consistency on the water bath ; care shoxild be exer-
cised to avoid overheating. Cool. Take up the residue in 5 c. c. of 95 per cent
alcohol, observe that an excess of platinic chlorid is present. Tritiu-ate the precipitate
carefully with a rubber "policeman" and filter at once on a carefully prepared and
weighed Gooch crucible. Wash with 95 per cent alcohol till the filtrate is colorless,
then with 30 to 40 c. c. of ammonium chlorid, 100 gm. to 500 c. c. of water, which
has been saturated with potassium platinic chlorid. Wash once with 80 per cent
alcohol and then thoroughly with 95 per cent alcohol. Dry in an oven at 100° C.
and weigh. Calculate the weight of potassium.
With the use of the known solutions previously prepared and the
above methods, the following results (Table IV) are typical of the ac-
curacy obtained.
27809°— 18 3
330
Journal of Agricultural Research
Vol. XII. No. 6
Table IV. — Analyses of dilute and average solutions by the methods used in this
investigation
K.
Ca.
PO4.
Solution and treatment.
Theory.
Deter-
mined.
Theory.
Deter-
mined.
Theory.
Deter-
mined.
Dilute solution (untreated)
P.p.m.
I. 2
I. 2
j 1. 2
I. 2
I. 2
I. 2
P.p.m.
1.05
I. OS
I. 21
•97
I. CO
I. 22
P.p.m.
1.50
1.50
1.50
1.50
1.50
1.50
P.p.m.
I. 62
1.47
1.65
I. 60
1-45
I. 60
P.p.m.
I. 0
I. 0
I. 0
I. 0
I. 0
I. 0
P.p.m.
I. 12
I. 01
I. 20
•94
I. 10
•97
Mean
I. 2
1.08
1.50
1.56
I. 0
I. 06
Average solution (untreated)
( 6.0
6.0
6.0
6.0
6. o
. 6.0
6.05
6. 20
6.05
6.60
5-95
6.50
7-5
7-5
7-5
7-5
7-5
7-5
7.91
7.90
7. 62
7.64
7.66
7.80
S-o
S-o
S-o
5-0
S-o
S-o
5.06
5.02
4.90
S-I5
5.10
5-19
Mean
6.0
6. 22
7-5
7-75
5-0
5-07
Percentage of error, dilute solution :
Mean . .
10. 00
19. 00
3-6
10. 0
4. 00
10. 00
3.00
5-4°
6. 00
Maximum
20. 00
Percentage of error, average solution:
Mean
I. 40
Maximum
3-80
The determinations on each element are given on six duplicate portions
of solution. In addition, the mean percentage of error is calculated, as
well as the maximum error for any single determination. It will be ob-
served that the agreement with the theoretical result is satisfactory for
such dilute solutions. The potassium determination shows the greatest
discrepancy, with a mean error of 10 per cent and a maximum of 19 per
cent on the dilute solution. On the average solution the results are
much closer, ranging from 3.6 to 10 per cent. With calcium the error
on both the dilute and average solution is even smaller, 4.0 and 3.0 per
cent for a mean result, and 10 and 5.4 per cent for a maximum. The
phosphate determination shows a slightly larger fluctuation; in the
dilute solution from 6 to 20 per cent, but the higher figure was only
obtained in one analysis. The stronger solution gave greater accuracy
than was obtained in any of the other work.
Taken as a whole, it is believed that the number of determinations is
sufficiently large to show the small probability of the occurrence of the
maximum error. The mean error will therefore be taken as the usual
figure to be applied in the interpretation of the subsequent results.
All data bearing on the seasonal studies which follow will be considered
in the light of their accuracy as affected by the mean and maximum
error.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 331
EXTRACTION PROCEDURE
It has been pointed out in an earlier portion of this discussion that
several extraction procedures have been applied to the study of the water-
soluble material of soils. In Europe the methods of Schloesing {46, 47)
and Mitscherlich (42) are largely favored, though the work of the U. S.
Bureau of Soils (61) has some followers. This latter system of extraction
is practically the only one which has been used in this country. The
desire to have the result of this investigation somewhat comparable to
previous work was therefore an argument in favor of this general pro-
cedure. This would not have held had there been any closer method of
approximating the actual soil solution, for the medium v/hich nourishes
the plant should always be our final object. As before noted, it is not
known that this actual solution has ever been obtained. It then be-
comes necessary to use some arbitrary method of extraction, and from
the data at hand in beginning this work the general procedure of the
U. S. Bureau of Soils was chosen.
In all the seasonal studies which are recorded later only fresh soil
was used for the analytical studies. The great change which soil under-
goes in drying as seen from the work of King (55), outweighs any con-
siderations of expediency. King himself took the contrary view, but a
study of his own figures shows how large and how variable were the
changes between different soils caused by drying the samples previous
to analysis. His adoption of this modification was influenced by a
desire to carry on work at several stations from one central laboratory.
In the present investigation the soils had been brought together, so no
such reason existed. It has been desired throughout to perform all
work on the fresh soils and to show what differences existed in them.
The uniform procedure of extraction is as follows :
Place the weighed portion of fresh soil in a large mortar and mix with a small
amount of distilled water till it becomes a thin homogenous paste. Then add the
rest of the distilled water, making in all five times the weight of the soil and mix
the whole for exactly 3 minutes. Then transfer to wide-mouthed bottles all the
solution and suspended soil that will pour off and allow it to stand for 40 minutes.
At the end of this period pour the solution without mixing into cylinders of a Briggs
filtering apparatus, connected to the pressure system, and apply compressed air at
approximately 100 pounds' pressure till filtration is complete. Discard the first
portion of the solution, about 50 c. c, to avoid any dilution by the moisture held in
the Pasteur-Chamberland candles and also any small absorption by the filter from the
solution.
After filtration is complete wash the filter candles and cylinders in tap water, rinse in
distilled water, and pass 600 to 800 c. c. of distilled water through each candle to wash
out any remaining portions of solution. After each weekly set of filtrations is com-
pleted, thoroughly dry the washed candles in an air oven at ico° C. to prevent the
development of bacterial flora in the outer pores of the tubes. The efficiency of
filtration of the candles is gradually impaired by continual use, owing to the deposition
of clay and fine silt in the unglazed porcelain of which it is composed. This can be
partly remedied by igniting the dried tubes in a mufile furnace for }A hour or more.
^^2 Journal of Agricultural Research voi. xii. No. 6
Then cool the candles and pass distilled water through them to remove charred
organic matter and salts formed by the process of ignition.
In the work carried out in 191 5, with the analytical methods of the
Bureau of Soils, 150 gm. of fresh soil constituted the sample treated
with 750 c. c. of distilled water. In 191 6 the modified procedure which
has been described required a larger volume of solution. Accordingly
340 gm. of soil has been treated with i ,700 c. c. of distilled water. This
volume of solution is the maximum quantity that can be filtered in two
cylinders of the filtering apparatus. These two portions of solution
are afterwards united and furnish a volume sufficient for the determina-
tion of phosphates, calcium, potash, nitrates, magnesium, and total
solids, if desired. A moisture determination was in each case performed
on the sample of fresh soil, and the result used in correcting the analytical
data to bring it to a uniform basis.
In this connection it should be pointed out that the form in which
the final results are to be stated is a matter of some importance. Mitscher-
lich (42), King {35), Whitney and Cameron (<5i), all expressed their
results on the basis of percentage or parts per million of the dry soil.
Gola {18), on the other hand, referred his to the concentration of the
solution obtained, which he considered to be representative of the soil
solution. Inasmuch as it is the soil solution which is our final object,
it would be natural to refer the analytical results to the actual soil
moisture determined in each sample. Desirable as it is, it is not believed
that we possess the necessary information in regard to the state of the
soil moisture to do this correctly. In discussing the results on the
moisture equivalents given in Table II we have referred to the unfree
water present in soils.
The work done by Briggs and Shantz (<?) on the wilting coefficient of
soils has shown the variations which occur among different soils, espec-
ially where they are of diverse types. Their work demonstrates clearly
that a large amount of moisture is unavailable to the plant in soils
which are high in colloidal material. Recently Bouyoucos {4) has
called attention to the correlation which exists between the moisture
held in unavailable form in the soil, as determined by the above method,
and the unfree moisture determined by the dilatometer method. He
has also compared the results obtained by these two procedures with
the determination of the moisture which fails to freeze. A remarkable
agreement is found to exist. Working with some 14 soils used in the
present study Hoagland (27) has found that in all the silty clay loams
a large percentage, varying from 13 per cent to 18 per cent, does not
freeze. In the fine sandy loams the percentage is much smaller, from
4 per cent for soil 8 to 8 per cent for soil 10. The extent of these varia-
tions and their significance is discussed at greater length by Hoagland
(27). If this moisture is held, either loosely chemically combined or
physically absorbed or both, it is at least possible that it is not available
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 333
for the solution of the soil nutrients. In such a case the actual available
solution carrying plant nutrients would not be very different in the
complex silty clay loams from that which exists in the fine sandy loams.
It has therefore been decided to express all results on the basis of the
dry soil until further evidence is obtained as to the condition of the
moisture in the soil.
AGREEMENT OF DUPLICATE EXTRACTIONS
Before the results obtained by any extraction procedure can be con-
sidered, it is necessary to know the limits of the accuracy which it is
possible to achieve. The first important point in this connection is
the possibiUty of obtaining accurate duplicate extractions. A number
of sets of duplicate extracts have been made from samples of fresh
field soils. The results on four typical soils, two silty clay loams and two
fine sandy loams, are given in Table V. It will be observed that the
agreement is well within the limits of the mean factor of error given
in Table IV for all the elements except calcium. It will be remembered
that the factor of error for this element ranged from 4 to 10 per cent.
With the two fine sandy loams the variations between the two duplicates
is between 10 and 15 per cent. This result would readily be accounted
for by the solubility of the calcium compounds present in the soil and
indicates that it is necessary to obtain far larger changes with this
element than any others studied before the result can be considered
significant. It must be emphasized, however, that duplicate results
such as those given above can only be obtained if every step in the
procedure chosen is carefully followed.
Table V. — Typical duplicate extractions of silty clay loams and fine sandy loams
[Results expressed as parts per million of dry soil. Ratio of soil to water 1:5]
Soil.
NO3.
PO4.
K.
Ca.
Mg.
No. I silty clay loam
32.5
4-3
3-8
33-8
34-8
16. 7
16.3
10.3
9.9
Do
Mean
32.9
4. I
34.3
16.5
10. I
No. 5 silty clay loam
27.6
29. 0
13- 9
29-5
30. 0
57-1
63.0
14.8
15.0
Do
Mean
28.3
13- 6
29.7
60. 0
14.9
No. 7 fine sandy loam
26.3
26.3
3-6
4. 2
18.0
18.6
37-5
42. 0
IO-5
Do
II. 4
Mean
26.3
3-9
18.3
39-7
No. 8 fine sandy loam
22. 6
22. 2
II. 7
10. 6
Zi- 7
30. 6
35- I
30.0
8.1
8.S
Do
Mean
22. 4
II. I
32.1
32.5
8.3
334
Journal of Agricultural Research
Vol. XII, No. 6
EFFECT OF TEMPERATURE AND CARBON DIOXID
The next point to be investigated was the possible effect of compara-
tively rapid changes of temperature. This was an important consider-
ation in studying fluctuations in soluble nutrients within a short period
of time. Such fluctuations were found to occur. It was extremely
desirable to know whether the change in solubility was due merely to the
physical effect of the increased temperature or was a secondary influence
due to increased activity of the bacterial flora. Such activity would
result in the production of appreciable quantities of carbon dioxid and a
possible increase in soluble soil compounds.
A series of studies were undertaken to investigate these various
effects. A silty day loam soil of excellent productivity was used. One
container of this soil had been kept fallow all season; the other had
grown a crop of barley. Both samples contained practically identical
quantities of moisture, slightly below the optimum for this soil.
One portion of each soil was placed in a bacteriological incubator at
29° C. A second was kept in a refrigerator at 6° for two hours, a third
was treated with carbon dioxid for 24 hours, a fourth was air dried at
20° for 48 hours, and a control portion was kept for 24 hours untreated.
The results are given in Table VI.
Table VI. — Effect of loiv temperature, increased temperature, air-drying, and carbon'
dioxid on water-soluble nutrients
[Results expressed as parts per milUon of dry soil]
No.
sA (Yolo silty clay loam,
No. 5B (Yolo silty clay loam, un-
cropped).
cropped).
mX
■S
•S--i
•t3
Treatment.
■-.2
Si,
•3
1%
1
•3
0
0
0
0
01 r;
•3 S
■3
1
0
0
'1%
Z
PL,
U
a
H
>=.
"A
^
Pk
1^
0
H
291
254
404
"A
70.7
68. 4
62. 1
17- S
14- S
22. 9
39-9
4.-i-5
37-9
83-3
83-3
68.3
472
459
547
195
189
319
277
270
228
139-5
143-2
127. 2
20.5
13-0
22.4
52.8
67-3
49-7
103.9
97-7
70-3
676
626
751
385
^72
48 hours air-dry
347
24 hourscarbon dioxid at i s ° C —
72-5
18. 1
.;2.6
113- 2
805
252
.S53
150-7
21.8
52-8
120. 9
918
31b
602
24 hours untreated at is" C
68.4
20. 2
47-5
75- S
447
195
252
135-8
17.2
5«-3
97.0
b32
273
359
The data in Table VI must first be considered by applying the mean
factor of error for the several determinations. This correction shows
that the effect of the temperature treatments in the incubator and re-
frigerator has been negligible. It is therefore probable that sudden wide
changes of concentration of water-soluble nutrients are due to other
factors than changes in the solubility of plant food caused by tempera-
ture alone.
The air-drying to which the two samples were subjected was carried
out at 20° C. without exposure to sunlight. It was consequently less
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 335
rigorous than the similar treatment which surface soil received in the
field. Even so it is seen to have produced several striking changes.
The nitrates and calcium are slightly decreased, but the greatest differ-
ence is observed in the total solids. Here we see a significant increase.
This is further shown by the volatile solids to be entirely due to a gain
in soluble organic matter. Such an increase in soluble organic materials
would suggest that the efifect of air-drying on the soil minerals would
largely be of a secondary nature. Any cultivation or stirring into
the soil mass of this air-dried portion of the soil would result in placing
this soluble material in circulation in the soil solution. It would thus
become available as food for the biological life of the soil, and may be one
of the factors tending to produce the large liberation of nutrients after
cultivation.
In the final treatment of the two soil samples carbon-dioxid gas was
frequently passed through the jars containing the soil during the 24-
hour period. This resulted in a striking liberation of soluble mineral
compounds. There was also a slight increase in the production of ni-
trates in comparison to the untreated soil, but it is very noticeable that
the phosphates remained practically constant. This is significant. It
will be pointed out later that phosphates are the only one of the plant
foods with which the solution of each individual soil appears to be satu-
rated. The amount of nonvolatile solids liberated by this treatment is
over 200 p. p. m., and shows that the efifect of the carbon dioxid e^Jtends to
other compounds than those indicated. From this study it seems prob-
able that the effect of the carbon dioxid present in the soil air and dis-
solved in the soil solution is the most important influence that renders the
soil minerals soluble.
NATURE OF THE WATER EXTRACT
The above series of results establish the accuracy of the extraction pro-
cedure which has been employed. They do not throw any light on the
nature of the solution obtained. All of the minerals present in the soil
possess a small but appreciable solubility. Any soil extract will there-
fore contain not only the material which is actually in solution or very
readily soluble but also an extra amount derived from the soil minerals
themselves. This latter portion of the soil extract may in some cases
be different for the same soil. It is a well-known principle that when a
compound is treated with a solution containing a common ion its solu-
bility will be repressed. If we now turn to the graphs showing the nutri-
ents extracted from the same soils, cropped and uncropped (fig. 3-6), it
will be observed that there are striking differences, which will be dis-
cussed later. In the cropped soil there is much less soluble material,
and the solution used to extract it will more nearl}^ approach distilled
water than in its duplicate, where nothing has been removed by the
336
Journal of Agricultural Research voi. xii, no.6
ERA. PRY SOIL
500
450
>
^ ^''
SOIL
350
4
^^^
300 ^^^''
^00/
POTAS^IV^ ..CK.)
150
100 _,.---•""'
- ' '
'''
50 ,.--'"
/ -
i!0?i.5 5 10 i'o
1
40
_ . P H. p 5 F H.AT.&....f-'5^?-^>
1
60
RPiTlO 50!L:VATEII
Fig. 3. — Graphs of the nutrients extracted from soil 4 by varying the ratios of soil to water. Calculated to
parts per milUon of dry soil.
F.PA.
500
PRY
50IL
450
^
400
350
300
50IL5
£50-
200
/<^^"
^,-^^
^^<^
_TOT.f.A^)^-^-^-
ISO
100 J^
:^
__,.o.TH/^X^-^^-
SO ^,.
y
1
5
1 1
10 ^0
1
eo
RATIO :5oil;vatek
Fig. 4. — Graphs of the nutrients extracted from soil 5 by varying the ratios of soil to water. Calculated to
parts per million of dry soil.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 337
p.p.n.
50O
PRY iOlL.
45fi
400
XI LQ
350
300
250
2.00
150
~
n^— -
...^^^^^^^^
-- °
100/
1
5
.X9T ?>AVyj^- SAL
. ,
7y^ob?n.f^.lX'..^y.9^).
1
10
1
10
1
40
80
RATIO
50IL
:VATEK
FlO. 5.— Graphs of the nutrients extracted from soil 8 by varying the ratios of soil to water. Calculated to
parts per million of dry soil.
T.T.n. PKY301L
500
/
//
450
O"^
f
400
50IL 10
/
/
350
V
500
4'
£50
20Q
I50;
^^^
'^"^'^^ yo^^-^:>^^--'""'^'
100
^-'=^=^^^^^.— ---
- - ""'
.'r^r-'^'j^"'
•^ty-
'-^ .
.xno^^^-^^-^^-""
Q*.
_^ .-..---•—•"*
'^hW,
1
1
10 20
40
ea
RATIO 501L
IVATER
Pig. 6. — Graphs of the nutrients extracted from soil lo by varying the ratios of soil to water. Calculated
to parts per million of dry soil.
338
Journal of Agricultural Research
Vol. XII, No. 6
plant. It may therefore be expected that in the solution from this soil
more material derived from the soil minerals will be soluble than in the
uncropped extract. The effect of this will be to make the differences
which actually exist less striking and evident than is really the case.
The evidence given by the freezing-point method, which has greatly de-*
veloped the above theory, is discussed by Hoagland (27).
The suggestion was made by Burd that it might be possible to obtain
some conception of this repressive effect by extracting a cropped soil
with a diluted extract from its uncropped duplicate. Two soils were
used for this purpose: Soil 5, Yolo silty clay loam, and soil 10, Tejunga
fine sandy loam. In each case an extract was first prepared of an un-
cropped reserve portion of the soil and of another portion upon which
a crop had been grown. A determination of total solids was immediately
made. The solution from the uncropped soil was then diluted with
distilled water, so that this solution added to that obtained from the
weaker cropped soil would again be equal to the strong solution. The
cropped soil was extracted with this diluted solution. The results of
this extraction are compared in Table VII with the amounts obtained
by the usual treatment with distilled water plus the quantities already
present in the diluted extract.
Table VII. — Extraction of two cropped soils with diluted extracts from duplicate bin soils
[Results expressed as parts per million of dry soil]
Soil.
Total
solids.
Fixed
solids.
Volatile
solids.
FO*.
Ca.
Mg.
K.
Cropped soil 5 extracted with di-
lute extract from reserve bins . . .
Nutrients extracted from same soil
by distilled water (1:5)+ nutri-
ents contained in the dilute ex-
tract
1,686
1,575
1,163
962
412
15-4
25-7
187.0
195.0
41-5
46. 4
66.1
Rr. 6
Depression of solubility
439
10.3
8.0
4.9
15-5
Cropped soil 10 extracted with di-
lute extract from reserve bins . . .
Nutrients extracted from same soil
by distilled water (1:5)4- nutri-
ents contained in the dilute ex-
tract
426
589
157
298
269
291
8.8
16. 4
32.1
55-8
6.6
8.8
55-4
76.8
Depression of solubility
163
141
22
7.6
23-7
2. 2
21.4
With the Yolo silty clay loam 5 there was seen to be a distinct de-
pression of phosphates and a slight depression of potassium, both of
which were greater than the maximum factor of error. The depression
of calcium was no more than the average factor of error. The effect
was especially seen in the amount of fixed solids which was held out
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 339
of solution, but the total solids were slightly increased as organic matter
replaced the nonvolatile solids.
With Tejunga fine sandy loam 10 there was a distinct depression
throughout. Both solids and all individual constituents were decreased
in quantity.
This evidence, together with that obtained by Hoagland (27) using
the freezing-point method, clearly showed that the differences which
may be found between cropped and fallow soils by the extraction pro-
cedure was always less than that which actually exists.
In the same paper (27) Hoagland has also shown, both by special ex-
traction studies and by calculation from the amounts of total solids
contained in the i-to-5 extract, that the relation of the actual soil solu-
tion to this water extract is extremely close. The amounts of material
in the i-to-5 extract are approximately two to five times as great as in
the soil solution. The range of concentration shown by these extrac-
tions is therefore undoubtedly too high. This does not nullify the
differences between cropped and uncropped soils which are shown to
exist by this procedure. Hoagland 's results show that, if anything,
these differences are actually much greater. A decrease in the range of
concentrations would therefore have no more effect than to change the
scale upon which the seasonal curves have been drawn.
EFFECT OF VARYING THE PROPORTION OF SOIL AND WATER
It was important to learn what relation the conventional i-to-5 extract
bears to other possible ratios of soil and water. Reference has been
made to the work of Mitscherlich (42), in which he employed a series of
varied extractions of soil with water saturated with carbon dioxid.
This idea was now applied. Four soils, two silty clay loams and two
fine sandy loams, were extracted with proportions of soil to water vary-
ing from I part of soil to % part of distilled water up to i part of soil to
80 parts of distilled water.
The results in parts per million of dry soil have been plotted in figures
3 to 6. It will be observed that with each soil the proportions of com-
pounds extracted in the lower ranges of concentration were essentially
the same. It was not until a range of concentrations of soil to water
greater than i to 10 were reached that a distinct change in the extrac-
tion of nutrients took place. These results therefore warranted the
employment of the i-to-5 extract as a conventional procedure.
SEASONAL STUDIES OF THE WATER EXTRACT
FIRST SEASON, 1915
In the first season's work, in 1915, both containers of each soil were
sown to barley. Half of the soils were planted on June 14, and the
remainder one week later. At the same time the first soil samples were
340 Journal of Agricultural Research voi. xii. No. 6
taken with the iS-inch cheese trier previously mentioned. From 9 to
12 triers of soil were required for a sample of approximately 600 gm.
The holes left by removing these samples were filled with reserve soil
from the bins.
It was felt that this procedure might be open to some objections,
owing to the fact that the soil in the bins was partially air-dried. In
any one season 10 sampling periods is the maximum number, and the
amount removed would not exceed 6 kgm. This amount was less than
0.8 per cent of the total soil present. The error involved was therefore
smaller than it was possible to detect by any available methods.
In order to minimize this influence, the containers were always sampled
by a definite system, so that no two samples were taken from the same
spot twice in the same season. The process of sampling was performed
with great care. There was no breakage of plants and no evidence of
injury from the small roots cut by the cheese trier. The growth obtained
has been extremely satisfactory, both in height and vigor. The plants
were spaced at the usual field distance, 6 inches apart each way. In
1 91 5 one seed was planted in a place, and a small number of extra seeds
were started simultaneously in the same soil. These extra plants were
then transplanted into any spaces where the seed failed to sprout. An
even stand was obtained in all except two duplicate containers, soils
3 and 5. The following season the writers were able to obtain a per-
fectly uniform growth by planting three or four seed in each place, and
then thinning to one plant of average size. The seed used was furnished
by the Division of Agronomy and was a pure strain of Beldi barley.
Before planting, it was treated with formalin to prevent smut; seed of
uniform size were selected.
A slight difficulty has been experienced toward the close of each
season from the tendency exhibited by the crops on several soils to lodge.
This has been corrected by a system of supporting wires.
In 1 91 5 the crop matured, and was harvested from October 21 to
November 6. This was a later date than commercial grain crops are
grown in California, but it is believed that the results were strictly com-
parable among themselves. The crop yield is given in Table VIII and
is expressed as total air-dry crop and grain in grams. The grain is also
calculated as pounds and bushels per acre. It will be observed that in
all the soils except No. 3 and 5 the agreement of the duplicates was
excellent. These two soils, it will be remembered, were those in which
the growth was least uniform.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 341
Table VIII. — Crop yield of 191$
Soil.
Yolo siltv clay loam :
No. lA
No. iB
No, 2A
No. 2B
No. 3 A
No. 3B
No. 4A
No. 4B
No. 5 A
No. 5B
Yolo clay loam:
No. 6 A
No. 6B
Hanford fine sandy loam:
No. 7A
No. 7B
Fresno fine sandy loam :
No. 8A
No. 8B
Kimball fine sandy loam :
No. 9A
No. 9B
Te junga fine sandy loam :
No. loA
No. loB
Madera fine sandy loam :
No. iiA
No. iiB
Arnold fine sandy loam:
No. 12A
No. 12B
Standish fine sandy loam :
No. 14A
No. 14B
Average variations between
duplicates
Total
yield of
air-dry
grain
and
straw.
Gm.
2, 092
1,904
2,038
1,649
1,421
1,083
1,331
1,187
1,863
1,742
2, 287
2, 182
1,115
1,059
2, 136
2, 022
I, 066
1, 096
1,401
•1,337
2, 020
2, 194
796
775
1,364
1,359
Gm.
73°
73°
605
548
474
333
534
459
529
420
739
705
432
363
827
780
405
495
500
742
774
242
214
390
403
Grain
(pounds
per acre).
5,614
5,614
4,653
4,215
3,646
2,561
4, 106
3,530
4,069
3,230
5,684
5,422
3,323
2, 792
6,361
6, 000
2,984
3,115
3,807
3,846
5,707
5,953
1,862
I, 646
3,000
3, 100
Grain
(bushels
per acre).
93-6
93-6
77-5
70.3
60.8
42.7
68.5
58.8
67.8
53-8
94-7
90.4
55-4
46.5
106. o
100. o
49-7
SI- 9
63-5
64. I
95-1
99.2
31.0
27.4
50.0
51-7
Variation
from
maximum
yield of
grain.
Per cent.
II. 7
II. 7
26. 9
33-7
42. 6
59-7
35-4
44- 5
36.0
49.2
10. 7
14.7
47-7
56.1
5-7
53-1
51.0
40. I
39-5
10.3
6.4
70.8
74. 2
52.8
51.2
Differ-
ence
between
dupli-
cates.
O. O
6.8
17. I
9. I
13-2
4.0
8.4
5-7
2. I
0.6
3-9
3-4
1.6
5-8
In order to compare the two seasons' crops on a uniform basis and
also to allow for the mean and maximum difference in duplicates, all
results of the yield of grain have been calculated to the percentage of
variation from the highest yield. The figures given for the duplicate
containers have been substracted and expressed in a separate column.
The difference between soils 3A and 3B was the maximum, amounting
to 17. 1 per cent, while that of soils 5A and 5B was 13.2 per cent. The
mean figure given by the above was 5.8 per cent. These differences
have been expressed diagrammatically in figure 7. Here it is seen that,
by taking the maximum variation between duplicates as the standard,
soils 8, 11,6, and i have practically an equally high production. Soil 2
is significantly lower, and the others range lower, down to soil 12.
342
Journal of Agricultural Research
Vol. XII, No. 6
In figures 8 to 20 it is seen that the graphs given on the duplicate con-
tainers agree excellently. In all the soils except No. 9, 12, and 7 there
was an appreciable increase of potassium, which occurred from four to
six weeks after the time of planting. In all except the same soils, and
in addition No. 14, there was a liberation of nitrates which, like the
potassium, began to fall soon after the crop was rapidly developing.
The method employed for calcium had proved so inaccurate that the
results were excluded from the final charts. It was known that the
potassium method had a high factor of error in the lower concentrations.
A definite increase of nitrates had been shown in the earlier periods of
growth. It was very clear that before any interpretation of these figures
could be made it was necessary to have not only more accurate methods
100 8A
ot,h'jKjl^ liiio
5jElft;^o;i 1916
Fig. 7. — Graphs of the yield of grain in 1915 and 19:6, expressed as a percentage of the maximum yield.
The heavily shaded portions in 1915 represent difference between duplicates.
of determination but also a comparison between the conditions which
were found in the planted soil and those which would have occurred in
the same soil had the crop not been present.
The first season's work therefore established the following facts:
(i) That a satisfactory duplicate crop could be grown under the con-
ditions employed; (2) that the extraction technic followed yielded prac-
tically duplicate graphs for the elements extracted. The period of a
year's treatment under uniform conditions had also tended to bring the
soils into normal relation to each other. It had given an opportunity
for the soils to recover from the sifting and handling which they had
received and be more comparable to a varied group of field soils exposed
to the same climatic conditions.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 343
j^ ii ^ ^1 'I i^ i<.^
Fig. 8. — Graphs of the seasonal studies of the water extract of soil i, Yolo silty clay loam.
344
Journal of Agricultural Research
Vol. XII, No. 6
SO//L 2/^
so
so
TO-
SO/A 2B
»^ *| *| 5| «!^ 't'! ifX |ii!
Hi
5| <o| §1 <V^ ^^ !?"( \
Fig. 9. — Graphs of the seasonal studies of the water extract of soil 2, Yolo silt clay loam.
Feb. IX. 1918 Effect of Season and Crop Growth on Soil Extract
345
j>o
eo
70
eo
so
l\
'fO
1 \
JO
' V^' ''
^o
/o
/^■•An
-^^^^^^^^,
so
eo
TO
SO/A 33
C/?OP B/9/?A£:y
^
^ ^ =^ O ^ t S b ^
SO/L Oi^
/9/e
%t tt t\t \t\ m
SO/L 3 s
C/?OP /VOA/£
nTiiriiniTp
Fig. lo.-Graphs of the seasonal studies of the water extract of soil 3. Yolo silty clay loam.
27809°— 18 4
346
Journal of Agricultural Research
Vol. XII, No. 6
/9/S
I % % % % %
^ M^ ?l ^ «^ (!^
Fig. II.— Graphs of the seasonal studies of the water extract o fsoil 4. Yo'o silty clay loam.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract
347
sSO//. ^y^
^e2^ts^^^^_
I ^^ I i 1^ 1^ 1^: i
^ ^l >| ^ ^1 ^1 ^1 ^^
Fig. 12.— Graphs of the seasonal
studies of the water extract of soil 5, Yolo silty clay loam.
348
Journal of Agricultural Research
Vol. XII. No. 6
.SO/L s/i
C/?0/=> B/^/?L£Y
... /y/>gvg3it//Y/%y
SO/^ SB
C/?0/=> B/^/?L£y
/3/S
<\3 ;' ^' <\i ^ -^ -M ") 5; -vi &fi <o
r^ u II u § § ^1 §
^. — ? —
§g 5i^ ^ 5S| ^
.—Graphs of the seasonal studies of the water extract of soil 6, Yolo clay loam
Feb. II. 1918 EJJect of Season and Crop Growth on Soil Extract 349
ao-.
70-:
J^^^M^
■ ■- / \
/ "^•^---.
Por^ssiU^M- •
. ^fA6A/CS/(//T ^ f) |_.
I
9S^cyw7j^^^
'— '">r'^^.?'^-"/^-i-dir
1^- y 1^ |5 1^ |5 1^ i|5^
%^ ?| 5| ^? ?^ ^f i^ ^?
Fig. 14.— Graphs of the seasonal studies of the water extract of soil 7, Hanford fine sandy loam.
350
Journal of Agricultural Research
Vol. XII. No. 6
--aj5^5
I ^ ^ ^ ^ i^ ;o
^ ^ ^ ^ ^ '^ <>i »! O O
so/L as
1 I ^ ^ $ I n ^ ^
k k
30/L a/i
C/?Of^ B/^/?^£y
/9/6
\ §5 \% SS ^ 55 % % |5i
^ *»? »? *| 5| 5^ 5l| !t^ S|
FlO. IS. — Graphs of the seasonal studies of the water extract of soil 8, Fresno fine sandy loam.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 351
so
so
70-
eo
■50
/9/e
5^ ^^
5)^ i)^ b^ 5^ i!^ b^ l§S
»^ ^ ^1 ^1 1^ 1^ il
1?^
so-
eo--
ZO--
60--
.30- ■
so-
20
/o
o
.A
/9/6-
/y
s
1 1 8 i I § ^i I
^^ H *| ^1 ^^ !^^ ^^ i^^
Fig. 16.— Graphs of the seasonal studies of the water extract of soil 9, Kimball fine sandy loam.
352
Journal of Agricultural Research
Vol. XII. No. 6
W3J
^ ^ ^ 5^ 55 k- k:^ ^•
i ^ ^
so/z. /o^
%
>SO/L /OS
C/?0/=> A/OAZ/T
,"^^^4^
<^-"^/0rt(^-"
^^Ci-J ^.'^Z- —
^ ^^ i^ ^ ^ fi ^ '^^ iS;
S «^ *v li 1^ «\ 1^ *^ %
'^ *.| »| ^ ^1 1| ^^ ^ ^1
Fig. 17. — Graphs of the seasonal studies of the water extract of soil 10, Tejtinga fine sandy loam.
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 353
SO/L ///I.
I 1^ ^vl 1^ 1: ^: ^'v S ^S b !^ a
^ n| >| S| *| ^ 5i| ^ *| ^ ^ i^
Fig. i8.-Graphs of the seasonal studies of the water extract of soil ii, Madera fine sandy loam.
354
Journal of Agricultural Research
Vol. XII. No. 6
so
TO
60
SO
fO
JO
20
/O
SO//L ZP.^
..—-^SOis^ji^, .por^s^'iT^—
•SO/L /2'0
II ^ q ii 1^ 1^ ^? p ^^
SO/L/^^
% ^1 ^1 I? ^1 f^
Fig. 19. — Graphs of the seasonal studies of water extract of soil la, Arnold fine sandy loam.
Feb. II, i9i8 Efject of Season and Crop Growth on Soil Extract 355
§5 ^X *S *S ^jj 5Cnj :to ■*to irt&j
S5 ^^ „^ 10^ <b^ 5^ 5,^ 5S ||
SO/A MS.
C/?0/=> A/OA/E.
/sue
Fig. 20.—
g 12 g sj 1^ 15 i^ r
^ ^ ?| 1^ ^^ ^ 5^ ^^
Graphs of the seasonal studies of the water extract of soil 14. Standish fine
sandy loam.
356 Journal of Agricultural Research voi. xii, no. 6
SECOND SEASON, 1916
In the season of 191 6 one container of each soil was again planted to
the same strain of Beldi barley, and the duplicate was at all times treated
in the same manner, except that no crop was grown upon it. One week
before planting, the top soil to a depth of 8 or 10 inches was forked up
and put in an excellent state of tilth. The soils were again sampled
every two weeks from the time of planting in May until the crop was
harvested in August. During the succeeding fall and winter samples
were taken at approximately 8-week intervals.
The analyses of the water extracts were performed by the modified
methods which have been previously outlined. During the latter
portion of the sampling season cooperative work was performed by
Hoagland on the samples, using the freezing-point method.
An extremely uniform stand of barley was obtained in all containers.
Shortly after sprouting, the plants were thinned to one vigorous seedling.
The growth throughout the season was steady and vigorous. A
diagrammatic representation of the height at each soil-sampling period
is given in figure 21. It will be seen that the most rapid period of
growth was from the fourth to the tenth week. By that time the
plants had almost attained their full height, and the heads of grain
were beginning to be formed. At the close of the twelfth week the
maximum growth in height had been attained. Five weeks' additional
time was required for complete development and ripening. It was
interesting to note that in these studies, where moisture was never a
limiting factor, the growth in height was extremely uniform, even though
the dry matter produced did show great differences.
The results of the crop yield in total crop and grain are given in
Table IX. The grain yield has again been calculated to pounds and
bushels per acre, and the percentage of variation from the maximum
yield is shown.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 357
5QLL5
I/1CHE5
5£1
I
VK5. PLA/{TL]7 ^ 4 6 6 10 12. 14 t^EACHtP A/^X. HT.
Fig. 21. — Graphs of the growth of crops in height, season of 1916.
Table IX. — Crop yield in igi6
Soil.
Yolo silty clay loam :
No. lA
No. 2 A
No. 3 A
No. 4A
No. 5 A
Yolo clay loam , 6A
Hanford fine sandy loam, 7A. . .
Fresno fine sandy loam, 8 A
Kimball fine sandy loam, qA. .. .
Tejunga fine sandy loam, loA. . .
Madera fine sandy loam, iiA. . . .
Arnold fine sandy loam, 12A . . .
Standish fine sandy loam, 14A . .
Total yield
of air-dry
grain and
straw.
Grams.
1,474
1,530
970
1.252
1,652
1.523
1,414
I. 719
946
I, 266
1.547
970
1,464
Grain.
Grams.
655
439
549
690
670
541
679
357
551
625
390
630
Grain
(pounds
per acre).
5. T92
5,038
3,377
4.223
5.307
5,153
4, 161
5, 222
2, 746
4,238
4,807
3, 000
4,846
Grain
(bushel.s
per acre).
85
69
87
45
70.
80,
50. o
80.8
Variation
from maxi-
mum yield
of grain.
Per cent.
2. 2
5-0
36.4
20. 4
. o
2.8
22. 5
1.6
48. 2
20. I
9.4
43-4
8.6
358 Journal of Agricultural Research voi. xii.no.6
The figures obtained in the preceding season for the mean and maxi-
mum variation between duplicates will be used in order to find w^hat will
constitute a significant difference in yield. By taking the mean varia-
tion as the standard, soils 5 A, 8A, lA, 6A, and 2A are to be considered
equally productive, while iiA and 14A are slightly lower. If, however,
the consen/ative figure of the maximum variation between duplicates is
used, it will be seen that No. iiA and 14A will also fall in the group of
the highest yield. These relations are presented diagrammatically in
figure 7.
It will be seen that six soils were significantly lower than the group of
highest productivity. Of those No. loA, 4A, and 7 A were practically
equal, while 3A, 1 2 A, and 9A were distinctly low in yield.
The results of the water extractions are again presented in the form of
graphs. In figures 8 to 20 the plot marked "A" is in each case the planted
portion, while the uncropped duplicate is called "B." Some extremely
striking differences are exhibited by these graphs.
In the planted soils the water-soluble nutrients at the beginning of the
season, either remained on practically the same level or increased slightly
for the first four weeks. Then, without exception, the nitrogen com-
menced to decrease rapidly and was followed in a smaller degree by the
calcium and potash, and very slightly by the magnesium. The con-
trast shown by the unplanted soil was equally uniform in nature, though
variable in the extent of the effect. In almost all the soils the effect of
the cultivation was to cause a considerable liberation of soluble nutrients
and, though there was later shown to be a depression from this high
figure, yet the general range of all the nutrients except the phosphates
continued to be higher in the uncropped soil.
The soils which did not show this stimulation and liberation of nutri-
ents were No. 3B, 9B, and 12B. These were the least productive soils
of the group and also had the lowest range of soluble nutrients.
It is extremely significant that this same period of high soluble nutri-
ents in the uncropped soils corresponds to the period of lowered nutrients
in the cropped duplicates. It was also the period, as may be observed
in figure 21, in which the plants were making their most rapid growth.
The one compound which did not exhibit this liberation of excess
nutrients in the uncropped soil was the phosphate. There were large
differences between the amounts present in various soils, but for any one
soil the amount was practically constant in the cropped and uncropped
plot. It was the only soil nutrient studied which behaved in this manner.
All the above differences between the planted and unplanted soils
which have been noted are greater than can be accounted for by the
probable factor of error which has been determined in an earlier portion
of the study. The maximum figure for any nutrient in the concentration
found is 10 per cent, and that applies only to potassium and calcium.
On applying this correction plus or minus to the periods of greatest de-
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 359
pression, it is clear that this will not account for the differences noted.
With phosphates the probable maximum is 4 to 6 per cent, while with
magnesium it would vary from 6 to 10 per cent. All these corrections
fail to change significantly the figures recorded.
At two periods during the season of 191 6 daily studies of the water
extract were made with soil 8, and at one period with soil i. The result
of these tend to show that considerable fluctuations may take place from
day to day. The results of these two studies are plotted in figures 22
to 24. It will be seen that the fluctuations in general occur simulta-
neously in both containers and so will not alter the relationship of the
graphs for the planted and unplanted soils. They show, however, that
small differences in the range of the graphs can not be considered sig-
f.F.A.-PRY iOlL
100
90
A Soil 15
(SO
A / \ A
70
60
50ILfA
y \ .'hnoi.
50
40
30 \ A
10 y-
0 ^r-r
yY..
v^*""-— ^ '\F04.
AvQ. 7 0 9
10 II 11 14 15 It 17 l\
AvQ.7 6 9 10 11 12 1415 % IT U
Fig. 22. — Graphs of the daily studies of the water extract of soils lA and iB, season of 1916.
nificant. The general range of the graphs is believed to represent an
average figure, but small variations between soils should always be dis-
regarded. This would also indicate that an expression of the differences
in water-soluble material can only be given by a series of related obser-
vations.
DISCUSSION AND DEDUCTIONS
In the foregoing studies it is believed that the limitations as well as
the possibilities of the experimental methods have been pointed out.
It is earnestly desired that no deductions should be drawn which are not
conservatively justified by the data presented.
It should be noted that the results have been obtained on 13 soils
which comprise only two distinct soil types. These types are, however,
36o
Journal of Agricultural Research
Vol. XII, No. 6
P.PA PRY 50iL
100
90
501L 6A
m
60
50
\
40
3Q
/
V
20
A A
1
iQ
I
•J''
501L SB
CaLCIV/A (Co.')
>^^- - POTA551Vy<\ ( K)
N / . /
\ /V
\
"■--. PH03PHATL (VOi). .,.■•*■■•——■— sj^--'---^-v
^ I I I I I I I I t I I i
Jsjnr. \l 15 II 15 16 17 19 m UliaU £6
/)lTROqL/MC/lrt03J
I I I I I I I I I I I I I
Jv/tE 12 13 U 15 16 17 10 19 ZO a 22 i3 IH
-P04
— Alg
I
lb
Fig. 23.— Graphs of the daily studies of the water extract of soils 8A and 8B, July, 1916.
V.Yt'
\. I7KY 501L
100
90
eo
50IL6A
70
60
,\
1 \
/ \
50
K I \
l\ 1 ^Ca
40
A / \/
^^' \ 'h^ .
30
y ., \ -'a
....-•■.,/ /\^.K..._\^A\9-
10
' ..^x A r\ ^po^
'••rK.
""•■•PO4.
AvQ. 7 8 9 10 II 12 14 15 Ife \1 i\ AvQ. 7 8 9 10 II U 14 15 \(> 17 il
Fig. 34.— Graphs of the daily studies of the water extract of soils 8A and SB, August, 1916.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 361
very dissimilar in their nature; the silty clay loams are representative
of decidedly complex soils with appreciable amounts of clay and other
colloids, while the fine sandy loams belong to the light open soils which
are common in the West. The extraction graphs yielded by both types
are so similar that it is believed that comparisons can safely be made
both within the types and between the two groups.
The extractions of the season of 191 6 are those to which the greatest
importance is attached and are those which will largely be considered
in this discussion. The previous year's work brought the soils into
comparable condition and also showed the probable limits of the agree-
ment between duplicates. It is believed that the results obtained in
this second season throw some light on several disputed points. The
method of constant comparison between the cropped and uncropped
soil is considered essential for the success of such a study. Only in this
manner can any conception of the inherent capacity of a soil be obtained.
When a large crop is growing on a soil, it is possible for it to affect the
soluble nutrients so that the extract given by it will be equal to a moder-
ately good or fairly poor soil.
Such a condition is seen with three fine sandy loams: No. 7, Hanford
fine sandy loam, a soil of moderate production; No. 8, Fresno fine sandy
loam, a very good soil; and No. 9, Kimball fine sandy loam, a soil which
has produced poor crops.
The nutrients extracted from each of these three soils when cropped
give practically duplicate graphs. But with the unplanted duplicate
striking differences are seen. The better soils all show greater differ-
ences between the cropped and the uncropped soils than do the three
poorest soils, No. 3, 9, 10, and 12. With these three it is possible that
plant food is the limiting factor, but with certain soils of intermediate
production, such as No. 4 and 10, it is less evident why these should
not fall in the most productive group. It is claimed, however, that
this method of study gives an expression of the inherent capacity of the
soil to produce water-soluble plant food.
Between these three poor soils and the group of highest production
there are large differences in the range of soluble nutrients. These
three soils would not be classed as unfertile soils in general farming.
The only one which had a small crop upon it when the original sample
was collected was soil No. 12, Arnold fine sandy loam. The oats grow-
ing on this body of soil were noticeably smaller than in other portions
of the field. Under the controlled conditions of the experiment the
differences between these soils and the more productive must be con-
sidered significant, since they have been among the poorest soils each
year. It may, therefore, be stated that among the 10 soils studied the
3 lowest in productivity also show the lowest inherent capacity to furnish
soluble nutrients.
27809°— 18 5
362 J our7ial of Agricultural Research voi. xii, no. 6
This conclusion is notably at variance with the earlier statements of
Whitney and Cameron (61). It is believed that the experimental method
which has been followed brings out the relationship between different
soils with unusual clarity and effectiveness.
The influence of the crop on the soluble nutrients is another point
which has been denied by the same authors. This is shown by the
following quotation from Bureau of Soils, Bulletin 22 :
At the same time we have detected no constant decrease in the amount of soluble
salts which could be easily detected with the methods used during the advance
stages of growth of the crop, notwithstanding the considerable withdrawal there must
be by the plants {61, p. 60).
The contrast between the curves of the planted and unplanted soils is
evident at the most cursory glance. It is also clearly shown that the
soil did not immediately recover and yield the same quantities of soluble
nutrients as the uncropped plot, even after the crop was recovered. At
the last observation recorded, on February 12, there was still an appre-
ciable difference between the duplicate containers. In this connection
it is pertinent to refer to the data obtained in a preliminary study in
which a cropped soil was extracted with a diluted solution from an
uncropped portion of it. In that study it was shown that this diluted
solution exerted a depressing effect on the nutrients extracted from the
cropped soil. This has an important application.
At the period of greatest growth of the crop, the water extract of the
cropped soil approached much more closely to distilled water in the case
of the planted soil than it did with the unplanted duplicate. In conse-
quence of this fact, a greater portion of the extract obtained from it was
from the solution of the actual soil minerals than was the case with the
uncropped soil. The real difference between the two duplicate con-
tainers was therefore greater than the graphs would indicate. They can
only be taken as the minimum difference which actually existed.
The data obtained by Hoagland, using the freezing-point method,
corroborate this difference between the planted and unplanted soils.
He has also observed that soils 9, Kimball fine sandy loam, and 12,
Arnold fine sandy loam, are notably lower than any of the other soils in
the concentration of their solutions. This coincided with the observa-
tions made from a study of the water extracts. Considering soil 3,
Yolo silty clay loam, the next lowest soil in water extract, he could not
draw any definite conclusion.
Jensen {32) in his studies on sugar beets has described a decrease in
water-soluble nutrients and has also found this less noticeable in the case
of phosphates. King's {35) figures also showed some decrease in soluble
compounds as the crop advanced in growth. Harris and Butt {24) have
observed differences in soluble salts and nitrates between cropped and
fallow soils. It is believed, however, that the use of the unplanted dupli-
cate soil and the periodic observations made upon it and the cropped
Feb. II. 1918 Effect of Season and Crop Growth on Soil Extract 363
container have given a better expression of the potential power of the
soil than has been obtained in the past.
The study of the unpTanted soil is especially valuable because of the
definite information. which it furnishes in regard to the effect of fallowing
and cultivation. From the results of the present investigation it can
be stated that the changes occurring in water-soluble compounds, both
organic and inorganic, as the result of these practices, are great and far-
reaching in effect. They entirely justify the importance attached to
such treatment in the past. In biennial cropping, alternating with
fallowing, as practiced in California, it is probable that fully as much is
gained from the increase of water-soluble nutrients as from the moisture
stored up in the soil.
In the preceding studies the data obtained have been considered solely
as a measure of the water-soluble nutrients obtained by a conventional
procedure of extraction. From the work performed by Hoagland in
collaboration the amounts extracted are definitely related to the actual
soil solution. Even though the figures so obtained do indicate a range
of concentrations higher than the actual truth, it has been previously
pointed out that this does not alter the relationships which have been
established. It is therefore believed, with this corroborative evidence,
that the changes observed in the water extract reflect actual changes in
the soil solution.
SUMMARY
(i) The water-soluble nutrients in 13 soils, of two different types,
have been periodically determined during two seasons.
(2) Throughout the second season comparisons were made between
the planted soil and its uncropped duplicate.
(3) Notable differences were observed between the nitrates, calcium,
potassium, and magnesium present in the water extracts from the
cropped and uncropped soils.
(4) The phosphates did not exhibit corresponding differences. Great
dissimilarities were observed in the phosphate content of different soils,
but in any one soil the amount was practically constant in both the
cropped and uncropped plot.
(5) Striking differences occurred between the soluble nutrients present
in the various uncropped soils.
(6) While the crops were growing, the concentrations of nutrients in
8 of the 13 planted soils were practically the same. These 8 included
both good and poor soils,
(7) The three poorest soils yielded the smallest amounts of water-
soluble nutrients and the smallest differences between the cropped and
uncropped duplicates.
2^^ A Journal of Agricultural Research voi. xii, no. 6
(8) The comparisons between the planted and unplanted duplicates
furnished valuable indexes of the inherent capacities of the soils to pro-
duce nutrients.
(9) The accuracy of the methods of analysis an,d of the extraction
procedure employed was determined, and the mean and maximum errors
involved were estimated.
(10) The amounts of the water-soluble nutrients obtained by varying
the ratio of soil to water were studied. The relationship of the com-
pounds extracted did not change essentially in the lower concentrations.
(11) By comparison with freezing-point determinations the concen-
tration of the soil solution calculated from the water extract was ^hown
to be from two to four or five times as great as the actual soil solution.
(12) Variations in the water extract were correlated with variations
in the freezing points of the same samples of soil.
(13) From the results of the freezing-point determinations it is con-
cluded that variations in the water extract reflect actual changes in the
soil solution.
(14) The results of the investigation show that large amounts of
water-soluble nutrients are developed by cultivation, fallowing, and bi-
ennial cropping, and demonstrate the soundness of these practices.
LITERATURE CITED
(i) American Public Health Association. Laboratory Section.
1912. STANDARD methods FOR THE EXAMINATION OP WATER AND SEWAGE.
ed. 2, 144 p. Bibliography, p. 137-140.
(2) BallEnEGGER, Robert.
1913. SOIL STUDIES BY THE AID OF THEIR WATER SOLUTIONS. (Abstract.) IN
Exp. Sta. Rec, v. 30, no. 6, p. 516. 1914. (Original article in
Foldtani Kozlony, v. 43, no. 7/9, p. 359-366. 1913. Not seen.)
(3) BoGUE, R. H.
1915. THE ADSORPTION OF POTASSIUM AND PHOSPHATE IONS BY T\TICAL SOILS OP
THE CONNECTICUT VALLEY. In Jour. Phys. Chem., v. 19, no. 8, p. 664-
695, 13 fig.
(4) BouYOucos, G. J.
191 7. MEASUREMENT OF THE INACTIVE, OR UNFREE, MOISTURE IN THE SOIL BY
MEANS OF THE dilatomETER METHOD. In Jour. Agr. Research, v. 8,
no. 6, p. 195-217, I fig. Literature cited, p. 217.
(5) and McCooL, M. M.
1916. THE FREEZING-POINT METHOD AS A NEW MEANS OF MEASURING THE CON-
CENTRATION OF THE SOIL SOLUTION DIRECTLY IN THE SOIL. In Mich.
Agr. Exp. sta. Tech. Bui. 24, p. 592H531, 2 fig.
(6) Briggs, L. J.
1902. FILTRATION OF SUSPENDED CLAY FROM SOIL SOLUTIONS. In U. S. Dept.
Agr. Bur. Soils Bui. 19, p. 30-40, fig. 5.
(7) and McCall, A. G.
1904. AN artificial ROOT FOR INDUCING CAPILLARY MOVEMENT OF SOIL MOIS-
TURE. In Science, n. s., v. 20, no. 513, p. 566-569.
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 365
(8) Briggs, L. J., and Shantz, H. L.
1912. THE WILTING COEFFICIENT FOR DIFFERENT PLANTS AND ITS INDIRECT
DETERMINATION. U. S. Dcpt. AgT. BuT. Plant Indus. Bui. 230, 83 p.,
9fig-
(p) Martin, F. O., and Pearce, J. R.
1904. THE centrifugal METHOD OP MECHANICAL SOIL ANALYSIS. U. S. Dept.
Agr. Bur. Soils Bui. 24, 38 p., 7 fig., 2 pi.
(10) Cameron, F. K.
1910. AN INTRODUCTION TO THE STUDY OF THE SOIL SOLUTION. In Jour. Phys,
Chem., V. 14, no. 4, p. 320-372, 3 fig.
(II)
(12)
191 1. THE SOIL SOLUTION, THE NUTRIENT MEDIUM FOR PLANT GROWTH. 136
p., 3 fig. Easton, Pa., London.
1912. THE r6lE op THE LYSIMETER IN SOIL SOLUTION STUDIES. In Orig.
* Commun. 8th Intemat. Cong. Appl. Chem., v. 15, p. 49-50.
(13) CossA, Alphonse.
1866. UEBER DIE BESTIMMUNG DER IM WASSER LOSLICHEN BESTANDTHEILE
DER ACKERERDEN. In Landw. Vers. Stat., Bd. 8, p. 54-59.
(14) Eichhorn, H.
1863. UEBER DIE AUFNAHME DER NAHRUNGSMITTEL AUS DEM BODEN DURCH
WASSER. (Abstract.) In Jahresber. Agr. Chem., Jahrg. 4, 1861/62,
p. 3-7. Original articles cited in footnotes.
(15) Engels, O.
1912. die wirkung einiger losungmittel aup die im boden enthaltenen
pflanzennahrstoffe; phosphorsaure, kali und kalk im ur-
SPRUNGLICHEN UND ABSORPTIV GEBUNDENEN ZUSTANDE. In Landw.
Vers. Stat., Bd. 77, Heft 3/4, p. 269-304.
(16) Gedroitz, K. K.
1906. variability op the concentration of the soil solution and op the
content of the soil op easily soluble compounds in dependence
ON EXTERNAL CONDITIONS. (Abstract.) In Exp. Sta. Rec, v. 19,
no. 4, p. 315. 1907. (Original article in Zhur. Opuitn. Agron. [Russ.
Jour. Expt. Landw.], v. 7, no. 5, p. 521-561. 1906. Not seen.)
(17) GiLTNER, Ward.
1913. REPORT OF THE BACTERIOLOGIST. In Mich. Agr. Exp. Sta., 52d Ann.
Rpt., 1912/13, p. 149-166.
Quotes Van Suchtelen, p. 149-155.
(18) Cola, Giuseppe.
1905. STUDI SUI RAPPORTI TRA LA DISTRIBUZIONE DELLE PIANTE E LA COSTI-
TuziONE Fisico-CHiNiCA DEL suoLO. In Ann. Bot. [Rome], v. 3, fasc.
3. P- 455-512, pi. 13.
(19)
(20)
1910. SAGGio Di UNA TEORiA OSMOTICA dell'Edafismo. In Ann. Bot. [Rome],
V. 8, fasc. 3, p. 275-548, I fig., pi. 13-14.
(21)
1911. OSSERVAZIONI SOPRA 1 LIQUIDI CIRCOLANTI NEL TERRENO AGRARIO. In
Gior. Geol. Prat., anno 9, fasc. 3/4, p. 131-159.
Also in Ann. R. Accad. Agr. Torino, v. 54, 1911, p. 33-67. 1912.
1912. LA vegetazione dell'appennino piemontese. In Ann. Bot. [Rome],
V. 10, fasc. 3, p. 189-338.
366 Journal of Agricultural Research voi. xii. no. 6
{22) Grouven, H.
1858. bEmERkungen zu uEbig's ansichten. (Abstract.) In Jahresber.
Agr. Chem., Jahrg. i, 1858/59, p. 13-14- i860. (Original article in
Agron. Ztg., Jahrg. 13, No. 30, p. 465-470. 1858. Not seen.)
(23) Hall, A. D., BrenchlEy, Winifred E., and Underwood, Lilian M.
1913. THE SOIL SOLUTION AND MINERAL CONSTITUENTS OP THE SOIL. In Phil.
Trans. Roy. Soc. London, ser. B, v. 204, no. 307, p. 179-200, 2 tig.
(24) Harris, F. S., and Butt, N. J.
1917. EFFECT OF IRRIGATION WATER AND MANURE ON THE NITRATES AND TOTAL
SOLUBLE SALTS OF THE SOIL. In Joiir. Agr. Research, v. 8, no. 9, p.
333-359. 18 fig. Literature cited, p. 359.
(25) HeydEN, Eduard.
1865. UEBER DEN PHOSPHORSAUREGEHALT IN WASSRIGEN BODENAUSZUGEN.
(Abstract.) In Jahresber. Agr. Chem., Jahrg. 8, 1865, p. 33-34.
1867. (Original article in Ann. Landw., Jahrg. 45, p. 189-194. 1865.
Not seen.)
(26) HiLGARD, E. W.
1903. METHODS OF PHYSICAL AND CHEMICAL SOIL ANALYSIS. Cal. Agr. Exp.
Sta Cir. 6, 23 p., i fig.
(27) HOAGLAND, D. R.
191 7. THE FREEZING-POINT METHOD AS AN INDEX OF VARIATIONS IN THE SOIL
SOLUTION DUE TO SEASON AND CROP GROWTH. In Jour. Agr. Research,
V. 12, no, 6, p. 369-395. 9 fig-
(28) Hoffmann, Robert.
1863. UEBER QUALITAT UND QUANTITAT DER AUS DER ACKERERDE DURCH EINES
WASSER AUFNEHMBAREN BODENBESTANDTHEiLE. In Landw. Vers.
Stat., Bd. 5, p. 193-197-
(29) ISCHEREKOV, V.
1907. THE OBTAINING OF THE SOIL SOLUTION IN UNALTERED CONDITION. (Abs-
tract.) In Exp. Sta. Rec, v. 19, no. 5, p. 409. 1908. (Original
article in Zhur. Opuitn. Agron. [Russ. Jour. Expt. Landw.], v. 8, no.
2, p. 147-166. 1907. Not seen.)
(30)
1911. SOIL SOLUTIONS. (Abstract.) /« Exp. Sta. Rec, v. 28, no. 4, p. 320-321.
19 13. (Original article in Zhur. Opuitn. Agron. [Russ. Jour. Expt.
Landw.], v. 12, no. 4, p. 558-560, 1911. Not seen.)
(31) Jarriges, V.
1861. ANALYSE VON kleEmuden bodenarten. (Abstract.) In Jahresber.
Agr. Chenu, Jahrg. 4, 1861/62, p. 36-37. 1863. (Original article in
Chem. Ackersmann, 1861, p. 83. 1861. Not seen.)
(32) Jensen, C. A.
1916. SOLUBILITY OF PLANT-FOOD ELEMENTS AS MODIFIED BY FERTILIZERS. In
Jour. Amer. Soc. Agron., v. 8, no. 2, p. 100-105.
(33) Jordan, W. H.
1916. MEASUREMENTS oP SOIL FERTILITY. N. Y. State Agr. Exp. Sta. Bui.
424, p. 389-412.
(34) King, F. H.
1904. investigations in soil management, being three of six papers on
the influence op soil management upon the water-soluble
SALTS IN SOILS AND THE YIELD OP CROPS. l68p.,2lfig. MadisoU, Wis.
(35)
1905. INVESTIGATIONS IN SOIL MANAGEMENT. U. S. Dept. Agr. Bur. Soils Bul.
26, 205 p., 9 fig., 4 pi-
Feb. II, 1918 Effect of Season and Crop Growth on Soil Extract 367
(36) King, F. H., and Jeffery, J. A.
1900. THE SOLUBLE SALTS OF CULTIVATED SOILS. In Wis. Agr. Exp. Sta., i6th
Ann. Rpt. [iSgSj/gg, p. 219-243, 3 fig.
(37) ^"<i Whitson, a. R.
1900. SOLUBLE SALTS OP CULTIVATED SOILS. In Wis. Agr. Exp. Sta., 17th Ann.
Rpt., [iSggj/igoo, p. 204-226, fig. 27-33.
(38)
1901. DEVELOPMENT AND DISTRIBUTION OK NITRATES IN CULTIVATED FIELD SOILS.
7n Wis. Agr. Exp. Sta., i8th Ann. Rpt., [igooj/oi, p. 210-231, fig. 48-52.
(39) LiEBiG, Justus von.
1858. UEBER EINIGE EIGENSCHAFTEN DER ACKERKRUME. In Ann. Chem. U.
Pharm., Bd. 105 (n. R., Bd. 29), Heft i, p. 109-144.
(40) Lyon, T. L., and Bizzell, J. A.
1913. THE PLANT AS AN INDICATOR OF THE RELATIVE DENSITY OF SOIL SOLU-
TIONS. In Proc. Amer. See. Agron., v. 4, 1912, p. 35-49, i fig.
(41) Maschhaupt, J. G., and Sinnige, L. R.
I912. DETERMINATIONS OF THE VALUE OF PLANT FOOD CONSTITUENTS IN SOILS
and FERTILIZERS AS DEPENDENT UPON SOLUBILITY. (Abstract.) In
Exp. Sta. Rec, v. 27, no. 6, p. 515. 1912. (Original article in
Verslag Landbouwk. Onderzoek. Rijkslandbouwproefstat. [Nether-
lands], 1912, no. II, p. 19-73. 1912. Not seen.)
(42) Mitscherlich, E. a.
1907. EINE chemische bodenanalyse Fur pflanzenphysiologische for-
SCHUNGEN. In Landw. Jahrb., Bd. 36, Heft 2, p. 309-369, 10 fig., i pi,
(43) Morgan, J. F.
1917. SOIL SOLUTION. In Science, n. s., v. 45, no. 1149, p. 19.
(44) Pantanelli, E.
1914. ricerche sulla concentrazione del liquido circolante nei terreni
LiBici. In Bul. Orto Bot. R. Univ. Napoli, t. 4, p. 371-383.
(45) Peters, E.
i860. studien tjber den boden, aus dem laboratorium zu tharand. in
Landw. Vers. Sta., Bd. 2, p. 113-151.
(46) ScHLCESiNG, Th6ophile.
1866. suR l'analyse des principes solubles de la terre veig^tale. In
Compt. Rend. Acad. Sci. [Paris], t. 63, no. 24, p. 1007-1012.
(47) .
1870. analyse des eaux contenues dans les terres arables. In Compt.
Rend. Acad. Sci. [Paris], t. 70, no. 2, p. 9S-102.
(48) Schlcesing, Thtophile, jr.
1901. l'acide phosphorique dissous par les eaux du sol, son utilisation
PAR LES PLANTES. In Ann. Sci. Agron., s. 2, ann. 7, t. 2, fasc. i,
p. 77-82.
(49) ■
1902. SUR LES PHOSPHATES DU SOL SOLUBLE A l'Eau. In Compt. Rend. Acad.
Sci. [Paris], t. 134, no. 23, p. 1383-1385.
(50) ScHREiNER, Oswald, and FailyER, G. H.
1906. COLORIMETRIC, TURBIDITY, AND TITRATION METHODS USED IN SOIL INVESTI-
GATIONS. U. S. Dept. Agr. Bur. Soils Bul. 31, 60 p., 5 fig., i pi.
(51) ScHULzE, Franz.
1864. UEBER DEN PHOSPHORSAURE-GEHALT DES WASSERAUSZUGS DER ACKE-
RERDE. In Landw. Vers. Stat., Bd. 6, p. 409-412.
(52) Schumacher, Wilhelm.
1863. UEBER DIE BODENLOSUNGEN UND SOLCHE BETREFFENDE VEGETATION-
SVERSUCHE. In Landw. Vers. Stat., Bd. 5, p. 210-222. '
368 Journal of A gricultural Research voi. xii, no. e
(53) Snyder, Harry.
1904. THE WATER-SOLUBLE PLANT FOOD OF SOILS. Ill Science, n. s., v. 19,
no. 491, p. 834-835.
(54) ^"d Hummel, J. A.
1905. SOIL INVESTIGATIONS. Minn. Agr. Exp. Sta. Bui. 89, p. 189-212, 2 pi.
(55) Thompson, H. S.
1850. ON THE ABSORBENT POWER OK SOILS. In Joxir. Roy. AgT. Soc. England,
V. II, p. 68-74.
(56) TOULAIKOV, N. M.
1913. THE INFLUENCE OF THE OSMOTIC PRESSURE OF THE SOIL SOLUTION ON THB
GROWTH OP SPRING WHEAT. (Abstract.) In Intemat. Inst. Agr.
[Rome] Mo. Bui. Agr. Intel!, and Plant Diseases, year 5, no. 11,
p. 1426-1429. 1914. (Original article in La P^dologie, aim. 15,
no. 4, p. 71-103. 1913. Not seen.)
(57)
1916. OSMOTIC PRESSURE OF THE SOIL SOLUTION AND THE VITREOUS STATE OP
BELOTOURKA GRAIN. (Abstract.) In Chem. Abstracts, v. 10, no. 23,
p. 3127-3128. 1916. (Original article in Zhur. Opuitn. Agron. [Russ.
Jour. Expt. Landw.], v. 17, p. 79-92. 1916. Not seen.)
(58) Ulbricht, R.
1863. EiN BEiTRAG zur METHODS DER BODEN-ANALYSE. In Landw. Vcrs.
Stat., Bd. 5, p. 200-209.
(59) Van Suchtelen, F. H. H.
I912. METHODE zur GEWINNUNG DER NATtJRLICHEN BODENLOSUNG. /»
Jour. Landw. Bd. 60, Heft 4, p. 369-370.
(60) Way, J. T.
1850-52. ON THE POWER OP SOILS TO ABSORB MANURE. In Jour. Roy. AgT.
Soc. England, v. 11, p. 319-380, 1850; v. 13, p. 123-143, 1852.
(61) Whitney, Milton, and Cameron, F. K.
1903. THE chemistry of THE SOIL AS RELATED TO CROP PRODUCTION. U. S.
Dept. Agr. Biu". Soils Bui. 22, 71 p.
(62) Wolff, Emil.
1864. ENTWURF ZUR BODENANALYSE. In Landw. Vers. Stat., Bd. 6, p. 141-
171.
(63) WundER, Gustav.
i860. uEber DIE IN DEM boden enthaltenden losungen. In Landw. Vers.
Stat., Bd. 2, p. 104-112,
PLATE 14
A. — General views of soil containers.
B. — Bins for storage of surplus soil.
Effect of Season and Crop Growth on Soil Extract
Plate 14
Journal of Agricultural Research
Vol. Xll, No.6
THE FREEZING-POINT METHOD AS AN INDEX OF
VARIATIONS IN THE SOIL SOLUTION DUE TO
SEASON AND CROP GROWTH ^
By D. R. HOAGLAND,
Assistant Professor of Agricultural Chemistry, College of Agriculture of the University
of California
INTRODUCTION
In the preceding article by Stewart (8)^ it has been conclusively shown
that water extracts of different soils may have widely varying concen-
trations of important nutrient elements, and that the water-soluble
substances in the 13 cropped soils were strikingly diminished in quantity
when the barley crop had reached its maximum power of absorption.
In the uncropped soils significant seasonal variations were also noted.
These studies suggested the importance of correlating the water extracts
with the actual soil solution, the immediate source of nutrients for the
plant.
Various attempts have been made to separate the soil solution, but in no
case has any appreciable quantity of solution been obtained when a soil
contained only the optimum percentage of moisture. Recently Bouyou-
cos and McCool (2) have proposed a method which seems capable of
giving direct experimental evidence concerning the concentration of the
soil solution. The procedure consists in determining the depression of
the freezing point in the soil itself under varying moisture conditions.
The soils used in the investigation described by Stewart seemed uniquely
adapted for further study by the freezing-point method. Accordingly
such determinations were made on the various soils under controlled
conditions, and the present paper deals with observations made on the
depressions of the freezing point in soils as affected by season, cropping,
moisture content, and other factors of significance in plant growth.
GENERAL METHOD OF PROCEDURE
The technic employed was essentially that described by Bouyoucos
and McCool (2) . The experience of the Agricultural Chemistry Laboratory
confirms their statements with regard to the possibility of obtaining
closely agreeing duplicates and general consistency of results. The
freezing-point blank with distilled water, however, was not found to
maintain a constant value from day to day, and to obviate any possible
1 Approved for publication in the Journal of Agricultural Research by the Director of the Agricultural
Experiment Station of the University of California.
2 Reference is made by number (italic) to " Literature dted," p. 394-395.
Journal of Agricultural Research, Vol. XII, No. 6
Washington, D. C Feb. 11, 1918
ly Key No. Cal.— 17
(369)
370
Journal of Agricultural Research
Vol. XII, No. 6
error from this source, determinations on pure water were checked
several times each day. The general magnitude of error in technic is
indicated by the following typical instances of duplicate experiments:
Soil No.
loA
loA
iB.
iB.
I5A
Freezing-
point de-
pression.
°c.
0.065
. 062
.066
. 069
.049
Soil No.
ISA
14A
14A
4B.
4B.
Freezing-
point de-
pression.
°c.
o. 052
• 043
.050
•157
•15s
DESCRIPTION OF SOILS USED
In the article by Stewart {8) the reader will find a detailed description
of the soils under the same laboratory numbers. For the sake of con-
venient reference, a list of the soils used is included here (Table I).
Each soil was investigated under three general conditions: (i) Soil from
tanks in which barley crop was grown for two seasons ; (2) corresponding
soils kept under identical conditions but with no crop the second season ;
(3) soils air-dried, sifted, and kept two years in closed bins. All tank
soils were kept constantly at nearly optimum moisture contents with
distilled water.
Table I. — Description of soils used in this investigation
Laboratory No.
Soil series and type.
Origin.
Yolo silty clay loam
Sacramento Valley.
do
Santa Clara Valley.
7
Hanford fine sandy loam
Southern California.
8 ...
Fresno fine sandy loam
San Joaquin Valley.
Kimball fine sandy loam
Southern California.
10
Te junga fine sandy loam
Madera fine sandy loam
Arnold fine sandy loam
Do.
II
San Joaquin Valley.
Do.
Standish fine sandy loam
Virgin desert soil.
UNFREE WATER OF SOILS
One of the most important factors affecting the concentration of the
soil solution is the moisture content. This has already been pointed
out by Bouyoucos and McCool, and the former has recently devised a
dilatometer method for more accurately measuring the unfree water
of a soil (7). In the present investigation an attempt was made to
approximate the unfree water in each soil by careful determinations of
the freezing-point depressions at different moisture contents. These
data could then be used in reducing the observed depressions to definite
and comparable moisture percentages, especially where the observed
Feb. II, 1918 Freezing-Point Method and Soil Solution
371
moisture contents differed only slightly from the standard. It has
been shown by Bouyoucos and McCool that freezing-point depressions
vary with the water content of the soil, not usually in direct ratio, but in
such manner as to necessitate the assumption that a certain fraction
of the total water present is so combined that it does not form an effective
part of the soil solution and is not subject to freezing. The percentage
of combined water would vary greatly with the type of soil, clay soils
having a large proportion of their water in the unfree state. The above
considerations, advanced by Bouyoucos and McCool, have been made
the basis for the determinations of unfree water in the soils used in this
investigation. Portions of each soil were divided into two samples the
moisture contents of which were so adjusted as to give a difference of
about 5 per cent. Each sample was thoroughly mixed and kept over-
night in a tight jar. Careful determinations of total moisture and freez-
ing-point depression were then made. From these data it is possible
to calculate what proportion of the moisture must be subtracted from
the total in order that the percentages of free water may give the same
ratios as the freezing-point depressions. Such estimates may not have
a high degree of accuracy, but it is probably sufficient for the purposes
in hand. Table II presents the results for the unfree water in each
soil.
Table II. — Estimation of unfree water in soils
Soil No.
iC
iC
2C
2C
3C
3C
4C
4C
sc
sc
6C
6C
7C
7C
Moisture.
Freezing-
point de-
pression.
Unfree
water.
Per cent.
*C.
Per cent.
20. 0
0.098
1 18.0
22. 4
•034
18.0
23.0
. 140
. 069
} 13-0
18.4
24. 0
•357
.149
> 14.0
19. 6
23.6
.362
. 210
} 13-0
18.8
23. 2
.089
.050
} 13-0
19. 6
24.4
.282
.079
} 17- 7
12.4
17.6
.284
.167
1 6.0
Soil No.
8C.
8C.
9C.
9C.
loC
loC
iiC
iiC
12C
12C
14C
14C
Freezing-
Moisture.
point de-
pression.
Per cent.
°C.
16.6
0-215
8.1
•655
12.8
.099
17.2
•059
12.8
•153
17.8
.074
13.2
'2,2,^
18.8
• 197
13.2
.130
18.0
. 089
14. 0
•254
18.4
.149
Unfree
water.
Per
cent.
4.0
6.0
8.0
6.0
4.0
7-S
SEASONAL VARIATIONS IN CROPPED AND IN UNCROPPED SOILS
One of the primary objects of this investigation was to follow through
the season the concentrations of the soil solution in each soil and to com-
pare at each period of sampling the cropped soils with their uncropped
duplicates. The first determinations of freezing-point depressions were
made early in July, 1916, and at intervals from that date until the fol-
lowing May. The barley crop was planted in May, 1916, and harvested
372
Journal of Agricultural Research
Vol. XII, No. 6
in August. The samples of soil used in this work were identical with
those employed by Stewart in making water extractions. In general,
the moisture contents of the uncropped and cropped soils were very
similar, but it was not always feasible to keep them in absolute agree-
ment. The data are therefore presented in two forms: First, as a table
showing the observed depressions and actual percentages of total moisture
present, together with the corresponding osmotic pressures and parts
per million of total solids in solution, as calculated by the methods of
(&ASIS or a2'/« noiSTURC)
time: from
plantiino.
(bASISaOF ZZ.'/' nOlSTURE)
I0WIC5 12WKS I6WKS
JULIO. JUL24 AUG 21
51WKS time: FRom
MAY. 1 PLANTING.
Fig. I. — Graphs of the depressions of the freezing point in soils i and 2, with and without crop.
Bouyoucos and McCool. In the second place, the observed depressions
have been calculated to uniform moisture contents, 17 per cent for all
of the fine sandy loams and 22 per cent for the silty clay loams. In mak-
ing these estimates the percentages of unfree water already presented
have been used. The corrected depressions have been plotted for each
pair of soils. In this way the cropped and the uncropped soils, as well
as the different soils in each group, may be compared readily (fig. 1-8).
(Cf. Table III.)
Feb. II. 1918 Freezing-Point Method and Soil Solution
373
5^s>ls or Z2.'A moisturl)
5IWKS TVHL FBOM
MAT\. PLANTING.
(E>A.51t> OF 22% mOlSTUREL)
lOWKS IZWK5 I6WKS Z4W1<:5
JULIT JUL3I AUGZe. OCT. 30
5EWK5 40WKS
DEC. 26 FtM9
5IWKS TIME. fROn
f-WYT PLANT IN O
Fig. 2.— Graphs of the depressions of the freezing point in soils 3 and 4, with and without crop.
374
Journal of Agricultural Research
Vol. XII. No. 6
fDAtj»5 OFZZ'/ MOISTURfL)
I0WK5 I2WK5 I6WK5 24.WK5
JULIl JUL3I AUGZ6 OCT. 50
5ZWK3. 40WKS
DE:C2.6 FE.e>19.
5IWK5
MAY 7.
Fig. 3.— Graphs of the depressions of the freezing point in soil 5, with and without crop.
Feb. II, 1918 Freezing-Point Method and Soil Solution
375
.055
-
.050
^/ \
/ \
.045
oy \
.040
-
^Diis--''^ (5A51SOF 2.Z°/. MOI
035
f\
/
.0^
-
"V^ N. ^^^^
025
\
f
.020
\
/
\ o7
.015
-
\ /
.010
_
.005
-
C[
1
1 1 t 1 1
I0WKS12WKSI6WKS
JULH JUL31 AU0Z8
OCT. 50
^2WK5
DLC 26
40WKS
FEb 19
.5IWK3 TlhELFROn
MAY 7. PLA\NTlN<j
Fig. 4. — Graphs of the depressioas of the freezing point in soil 6, with and without crop.
27809°— 18 6
376
Journal of A gricultural Research
Vol. XII, No. 6
T/4 MOISTURf:)
I0WK5I2WK5.I6\NK5 24WK,^ iZNNKS 4-OVVK.S
JUU7 JUL3I AU0 2B OCT 30 DE.C 26 FLb 13
5IWK5 TIME. TROM
HA^T PLANTING.
(C)A51S OF 11'/- nOISTUPLE-")
lOWKS. IZWIO.ieWKSi 24WKS 52WKS 40WKS
JULI0.JUL24AUGEI OCT 23 DCC.Ja FE.B.I2
■5IWK5 TIML FROM
nAY7 PLANT\NG.
FlG. 5. — Graphs of the depressions of the freezing point in soils 7 and 8, with and without crop.
Feb. M. i9i8 Freezing-Point Method and Soil Solution
377
(5A5ie> OF n'A noiSTUREi)
I0WK5 I2WKS I6WKS Z4WKS 52WK5 40WK.5>
OULI0JUL24AUCil OCT 22, DLCIS FE.BIZ
51\N/K5 T\ML FROM
MAY \ PLM^JTING
^
.025
-
D2Q
.0J5
SOIL \05_
^
NONE..
(5A515 Of 17"/- M
.010
soilM£^
c2^
-^^^4^y
.0^
^^^-^r'^"^^'^
c
II 1 1
1 1
I0WK5I2WIC5 16WK5 Z4-VVK5 5ZWK5 4.0WKS
JULIQ JUL24 AUG.2I OCT.23 DEIC.IO FL5I2
51WKS TIME FCOn
MAY I PLANTING.
Fig. 6.— Graphs of the depressions of the freezing point in soils 9 and 10, with and \nthout crop.
378
Journal of Agricultural Research
Vol. XII, No. 6
CgOPjVOlVC
(bA&IS OF \T/. nOISTURJL)
IOWKSI2WK5l6WKe> 24-WK.5 5ZWK5 4-OWKS
JULI0JULa4AUG.21 OCTZi DE_C IS FELb. 17
5IWK5 time: FROn
MA^^Y I PLANTING.
(e>At>(5 OF \iy- ri015TUR£.^
' I0WKSIZWKSI6WK5
JUU7JUL24^UG2I
Z4WK5 / 52LWK5 40W1C5
OCT 23 DELCia TELB 17
5IWIO TiMt rfion
^^Yl PLANTING
Fio. 7. — Graphs of the depressions of the freezing point in soils ii and 12, with and without crop.
OF IT/. MOISTURE.^
I0WKSI2WK5I6WK5 24.WKt> 5ZWK5. 40WX.5
JUL.ITJUL5\^UG^6 OCT 30 DE.C. iS FEL&O
51WK.5 TIME. FR.Or-1
MAY 7 PLANTING.
Pig. S. — Graphs of the depressions of the freezing point in soil 14, with and without crop.
Feb. II. 1918 Freezing-Point Method and Soil Solution
379
Table III. — Observed and calculated freezing-point depressions
Date.
Moisture.
Soil
lA.
Soil
iB.
Observed
freezing-point
depression.
Soil
lA.
Soil
iB.
Osmotic
pressure (atmos-
pheres).
Soil
lA.
Soil
iB.
Calculated total
solids in soil
solution .
Soil lA.
Soil iB.
Freezing-point
depression cal-
culated to uni-
form moisture.a
Soil
lA.
Soil
iB.
July 10,
July 24,
Aug. 21
Oct. 23.
Dec. 18
Feb. 12
May I .
July 10.
July 24.
Aug. 21
Oct. 23
Dec. 18
Feb. 12
May I .
July 10.
July 24,
Aug. 21
Oct. 23
Dec. 18
Feb. 12
May I .
July 17,
July 31-
Aug. 28
Oct. 30 .
Dec. 26
Feb. 19
May 7 .
P. ct.
23.2
22. 5
23-9
21.8
23-4
22. 7
21. 9
Soil
2A.
20. 4
19.7
23-7
18.3
20. 4
20. 2
17.4
Soil
3A.
18.6
19. o
22. O
20. 5
20. 9
20. 6
19. I
Soil
4A.
18.8
18.6
19. 2
17-3
19-3
17. 6
16.5
P. ct.
22. 4
22. 7
23.0
21. I
22.3
21. 7
20.3
°C.
0.030
. 029
. 019
• 050
•037
• 056
.082
" c.
o. 064
.065
. 062
.118
•095
. 108
.138
Soil
2B.
Soil
2A.
Soil
2B.
Soil
2A.
21. 4
20.8
21. 3
18.9
20. o
20. 2
17-3
o. 026
034
008
055
047
056
131
o. 044
.047
.049
. 092
.087
. lOI
. 201
0.32
.41
. 10
.66
.56
.68
1.47
Soil
3B.
Soil
3A.
Soil
3B.
Soil
3A.
22. 2
22. 2
22. 7
20. 9
20. 7
19.9
17-5
o. 130
.078
. 017
•075
.048
. 062
.068
0-055
.036
.034
. no
. 060
. 120
• ^2,S
I. 46
•94
. 20
.90
•58
•74
.82
Soil
4B.
Soil
4A.
Soil
4B.
Soil
4A.
18.5
19. 6
20. o
16. 7
18.0
17.2
I5^8
o. 048
.050
. 017
.068
•055
.098
. 106
0.047
.066
•057
.082
157
0.58
.60
. 20
.82
.66
I. 19
I. 29
0.77
•78
•74
1^43
I- IS
I- 31
1.56
P. p.m.
900
900
600
I, 600
1, 200
1,800
2, 600
P. p.m.
2,000
2,000
I, 900
3,700
3,000
3,400
4,300
' C.
039
033
028
047
050
066
080
Soil
zB.
Soil 2A.
0.53
•56
•59
I. II
1.05
1. 22
2. 42
800
1, 100
300
1,700
1,500
I, 800
4, 100
Soil 2B.
Soil
2A.
Soil
3B.
Soil 3 A .
66
44
41
72
45
52
4, 100
2, 400
500
2,300
1,500
1, 900
2, 100
Soil
4B.
S0U4A.
0.56
.80
.68
■99
I. 46
I. 179
1.8
1,500
1, 600
500
2, 100
I, 700
3, 100
3,300
1, 400
1,500
1,500
2, 900
2, 700
3,200
6,300
o. 021
025
010
032
039
045
064
SoilsB.
Soil
3A.
I, 700
I, 100
I, 100
3,400
I, 900
3,800
4, 200
0.075
049
017
061
041
051
043
Soil4B.
Soil
4A.
1,500
2, 100
1,800
2, 600
3,800
4, 600
4,900
0.031
031
012
033
038
050
041
" C.
3. 070
. 076
•77
. ogi
. 102
. 100
•079
Soil
2B.
o. 041
. 041
•045
. 060
.068
.081
. 096
Soil
3B.
D. 056
•037
•037
•095
. 050
.089
•059
SoU
4B.
o. 029
.048
.044
• 034
. 067
. 069
.049
a Soils I, 2, 3, 4, s, and 6 calculated to 22 per cent of moisture. Soils 7, 8, 9, 10, 11, la, and 14 calculated
to 17 per cent of moisture.
38o
Journal of Agricultural Research
Vol. XII. No. 6
Table III. — Observed and calculated freezing-point depressions — Continued
July 17.
July 31-
Aug. 28
Oct. 30 .
Dec. 26
Feb. 19
May 7. .
July 17.
July 31-
Aug. 28 .
Oct. 30.
Dec. 26.
Feb. 19.
May 7 . .
July 17.
July 31-
Aug. 28
Oct. 30
Dec, 26
Feb, 19
May 7.
July 10.
July 2r4.
Aug. 21
Oct, 23
Dec. 18
Feb. 12
May 7.
July 10.
July 24.
Aug. 21
Oct. 23
Dec. 18
Feb. 12
May I .
Moisture.
Soil
SA.
P. ct.
16. 2
18.4
21. 9
16.8
18.2
18. I
16,9
Soil
6A.
23-4
24. I
24- 5
24.8
23.0
Soil
7A.
12.5
Soil
8A.
13.2
14. I
15-9
9.9
12. 7
10. 6
Soil
sB.
p. ct.
18. 7
20. 6
21. I
17- 5
18.0
17. 2
19. 2
Soil
6B.
23.8
25.2
24.9
24.4
24.8
25. o
22. 6
0-399
Soil
7B.
15-9
16. 9
16.6
13-6
14.7
14. o
12.8
Soil
SB.
Soil
9A.
14-3
14. o
16. 9
10. 7
11. o
9.4
7-1
Soii
9B.
13-4
13.2
13- I
9-7
10. 6
9-5
6.9
observed
freezing-point
depression.
Soil
sA.
" C.
0.093
029
017
079
066
106
079
Soil
6A.
. 040
.068
.085
.066
. 104
Soil
7A.
0.028
. 026
. 007
.082
.038
•053
. 070
Soil
SA.
038
015
, 021
, 060
,049
,062
, 122
Soil
9A.
O. OI6
. 007
• 014
.028
.031
•057
.118
Soil
sB.
C.
o. 069
• 039
• 157
.088
. 106
.115
. lOI
Soil
6B.
097
086
072
lOI
103
126
153
Soil
7B.
o. 048
. 042
• 03s
• 151
. 108
• 093
. 116
Soil
SB.
o. 064
•053
.056
.138
. Ill
•135
. 214
Soil
9B.
0.034
. 026
. 042
. 092
.058
.087
. 200
Osmotic
pressure (atmos
pheres).
Soil
sA.
Soil
sB.
0.83
•4:7
Soil
6A.
4. til
.82
I. 02
.80
I. 26
Soil
7A.
0.34
32
Soil
6B.
I. 17
I- OS
.86
I. 22
1-25
1-52
1.85
Soil
7B.
0.58
■ 50
.42
1.82
1-31
I- 13
I. 41
Soil
8A.
o.'46
. 18
•25
•72
•59
•74
1.47
Soil
SB.
0.77
.64
.68
1.56
1-34
1-52
2.58
Soil
9A.
Soil
9B.
0. 41
• 32
• 50
1. II
.70
1. OS
2. 41
Calculated total
solids in soil
solution.
Soil 5A.
P. p.m.
2,900
900
500
2,500
2, 100
2,500
Soil
6A.
12, 000
1,300
2, 100
2, 700
2, 100
3,300
Soil
7A.
900
800
200
2, 600
I, 200
1, 700
2, 200
Soil 5B.
P. p.m.
2, 200
I, 200
4,900
2,800
3i30o
3,600
3,200
Soil
63.
3,000
2, 700
2, 300
3,200
3,200
3>900
4, 800
Soil
7B.
1,500
1,300
1, 100
4,700
3,400
2, 900
3,600
Soil
SA.
I, 200
500
700
I, 900
I, 500
1,900
3,800
Soil
9A.
600
200
400
900
I, 000
I, 800
3,700
Soil
SB.
2, 000
I, 700
I, 800
4,300
3,400
4, 200
6, 700
Soil
9B.
1, 100
800
1,300
2, 900
I, 800
2,700
6,300
Freezing-point
depression cal-
culated to uni-
form moisture.
Soil
sA.
"C.
033
017
017
033
038
060
034
Soil
5B.
Soil
6A.
Soil
7A.
o. 020
.018
. 009
.066
• 034
. 041
. 041
Soil
8A.
o. 027
. oia
. 019
. 027
•033
• 032
•039
Soil
9A.
014
005
014
012
014
018
012
'C.
o. 044
•033
. 141
.044
•059
•054
. 070
Soil
6B.
. 102
0.138
.150
• 053
. 121
. lOI
•157
• 135
. 170
. 109
.214
. 128
.174
Soil
7B.
0.043
. 042
•034
. 194
.085
.068
, 072
Soil
SB.
o. 051
.044
.051
. 064
•073
. 071
. 071
Soil
9B.
o. 023
. 017
. 027
.031
. 024
.028
. 016
Feb. II. 1918 Freezing-Point Method and Soil Solution
381
Table III. — Observed and calculated freezing-point depressions — Continued
Date.
Moisture.
Observed
freezing-point
depression.
Osmotic
pressure (atmos-
pheres).
Calculated total
solids in soil
solution.
Freezing-point
depression cal-
culated to uni-
form moisture.
Soil
loA.
Soil
loB.
Soil
loA-
Soil
loB.
Soil
loA.
Soil
loB.
Soil
loA.
Soil
loB.
Soil
loA.
Soil
loB.
July 10
July 24
Aug. 21
Oct. 23
Dec. 18
Feb. 12
May I
P. ct.
15-7
16. 0
19. 6
14.9
15-1
14. I
II- 5
P. ct.
16.6
16. I
17.0
13-6
14, 6
13-4
II. 0
" C.
0.037
.023
. 012
. 041
. 046
.066
.094
° C.
0.056
.051
•057
.099
.079
.125
.228
0.44
.28
•14
.49
•55
-79
I. 14
0. 67
.61
.68
1. 19
•95
I- 51
2-75
P. p.m.
I, 200
700
400
1,300
1, 400
2, 100
2, 900
P. p.m.
1,800
I, 600
1,800
3, 100
2,500
3,900
7,100
' C.
0. 028
. 021
•015
•033
.036
• 04S
•033
'C.
0.054
.047
•057
. 061
•058
•075
. 076
Soil
iiA.
Soil
iiB.
Soil
1 1 A.
Soil
iiB.
Soil
iiA.
Soil
iiB.
Soil
iiA.
Soil
iiB.
Soil
iiA.
Soil
iiB.
July 10
July 24
Aug. 21
Oct. 23
Dec. 18
Feb. 17
May I
14.9
12.7
16.6
13-9
14-5
II. 0
II. 2
15.8
16.3
13.2
13-6
13-5
10. 7
0.034
-035
.027
•037
.047
.065
. 121
0. 060
. 041
.052
. lOI
.134
• 133
.203
0. 41
.42
•32
.44
.56
.78
1.47
0. 72
•49
.62
1. 22
I- 51
1.50
2.45
I, 100
I, 100
800
1, 200
1,500
2, 000
3,800
I, 900
1,300
I, 600
3,200
4,200
4, 2GO
6,300
0. 028
. 021
. 026
.027
•036
-043
-055
0. 028
•037
■049
.066
•093
. 091
.087
Soil
12A.
Soil
12B.
Soil
12A.
Soil
12B.
Soil
12A.
Soil
12B.
Soil
1 2 A.
Soil
I2B.
Soil
12A.
Soil
12B.
July 17
July 24
Aug. 21
Oct. 23
Dec. 18
Feb. 17
May I
II. I
14.4
16.3
14.4
13-9
12.7
10. 0
12.3
13-5
14-5
II. 4
II. 7
II. 4
9.6
0.033
. 001
.003
. 061
.030
•043
.048
0.054
.036
. 026
.138
•059
.083
.077
0. 40
. 01
.04
•73
•36
•52
.58
0.65
•44
•32
1.56
•71
I. 01
.92
I, 000
50
100
I, 900
900
1,300
1,500
I, 700
I, 100
800
4,300
1, 800
2, 600
2,400
0. 018
. 001
•003
■049
.023
. 029
. 022
0.034
. 026
. 021
.079
•035
.047
•033
Soil
14A.
Soil
14B.
Soil
14A.
Soil
14B.
Soil
14A.
Soil
14B.
Soil
14A.
Soil
14B.
SoU
14A.
Soil
14B.
July 17
July 31
Aug. 28
Oct. 30
Dec. 26
Feb. 19
May 7
IS- 7
17-9
20. 4
17- S
17.0
16. I
15-7
18.4
19. I
19. I
16. 2
16. 5
15.0
20. 5
0. 041
.013
. QIC
•033
•039
.063
-043
0.045
. 027
.031
•09s
.068
.123
. 060
0.49
.16
. 12
.40
•47
.76
•52
c
I
I
54
32
37
IS
82
49
72
1,300
400
300
1,000
1, 200
2, 000
1,300
1,400
800
I, 000
3,000
•2, 100
3,800
1,900
0-035
. 014
. 014
-035
-039
•057
•037
0. 052
•033
.038
.087
• 065
•097
.082
382 Journal of Agricultural Research voi. xii. no. 6
It is strikingly evident from these data (Table III) that the freezing-
point depressions are not constant during the season, and that the con-
centrations of the soil solution are uniformly lower in the cropped soils
than in the same soils uncropped. This latter observation is especially
noteworthy, since comparisons of two samples of the same soil are pecu-
liarly applicable. Any errors due to unfree water, type of soil, optimum
moisture content, etc., would be practically constant. The same general
relations hold, whether the original data are considered, or the corrected
figures as used in the graphs. The only difference is that more exact
comparisons may be made when the moisture contents are reduced to a
uniform percentage.
The logical conclusion from the results given in Table III is that the
concentration of the soil solution may vary widely at different periods of
the year, and that the growth of the plant has a pronounced effect in
lowering the concentration. Furthermore, this effect is of long duration.
The soils which had been cropped are still decidedly lower in the concen-
tration of the soil solution, as compared with the fallowed soils, eight
months after harvesting the crop. It does not necessarily follow that
the differences between the cropped and the uncropped soils are to be
ascribed solely to the depletion of the soil solution by the plant. The
evidence presented by Burd (j) is confirmatory of the idea that certain
biological activities are more intense in the soil without crop, as shown by
the increased production of nitrates and greater solubility of calcium and
magnesium. It is nevertheless true that the lowest concentrations in
the cropped soils occur at about the time when the crop has completed
its maximum absorption of nutrients. In certain of the fine sandy loams
the concentrations have been reduced to a very low point, corresponding
to only a few hundred parts per million of total solids. This condition
must be the result of withdrawals by the plant. The magnitudes of
absorption, as given by Burd, are quite comparable with quantities
present in the soil solution. After the minimum has been reached at the
height of the growing season, the freezing-point depressions slowly- in-
crease up to the last date recorded in the graphs. May i. It is highly
probable that several weeks after cultivation a more marked increase in
the concentration of the soil solution would take place.
At this point it is desirable to compare the data presented by Stewart
{8) for the water extracts with the determinations made by the freezing-
point method. It will be noted that the same general relations obtain
in both cases. All the elements for which analysis was made are subject
to marked fluctuations, with the exception of phosphorus. This sub-
ject will be given further consideration later in the article.
Feb. II. r9i8 Freezing-Potfit Method and Soil Solution 383
VARIOUS FACTORS AFFECTING THE SOIL SOLUTIONS
As a result of the observations on the tank soils which have just been
described, it was decided to study various other factors possibly affecting
the depression of the freezing point. What conditions other than crop-
ping might increase or decrease concentration? The first step was to
study the soils which had been held in a nearly air-dry state in bins. The
results of these determinations have already been set forth in Table II,
and comparisons may now be made between tank soils and the correspond-
ing bin soils. It will at once be noted that a number of the latter show
very much higher concentrations of the soil solution than either the
cropped or the uncropped soils during most of the season. All three
portions of each soil were originally derived from the same sifted and
homogenous mass of soil. The existing dififerences must therefore be
due to subsequent treatments which are correlated with three levels of
concentration: (i) A very low concentration in the cropped soil, (2) a
higher one in the uncropped soil, and (3) highest of all in some of the
stored soils. The latter in a number of cases gave four or five times as
great a depression of the freezing point as the corresponding cropped soils.
The fact that the uncropped soils in the tanks had in some cases a less
concentration of the soil solution than the stored soils may be explained
by the treatment of the preceding year when both tank soils were cropped.
It may also be true that long storage of the bin soils under the special
conditions has brought about a decomposition of soil minerals as well as an
increase in nitrates, with the result that a more highly concentrated soil
solution is produced as soon as the soil is mixed with its optimum quan-
tity of water. The extractions of the bin soils confirmed the freezing-
point results. Water extracts showed a correspondingly high solubility
for nearly all constituents.
With the establishment of the relations of freezing-point lowerings of
soils under several conditions it became of interest to make a somewhat
more detailed study in an effort to determine what factors are especially
active in changing the concentration of the soil solution.
EFFECT OF INCUBATION ON CONCENTRATION OF SOIL SOLUTION
The first experiments dealt with the effect of long standing at laboratory
temperatures. Samples were obtained from the tank soils, and after the
determinations of freezing-point lowerings portions were kept in tight
Mason jars for from one to three months. At the end of the specified
periods the freezing points were again determined, with the results given
in Table IV.
384.
Journal of Agricultural Research
Vol. XII, No. 6
Table IV. — Effect of incubation at laboratory temperature on depressions of freezing
point
Soil
No
iMois-
ture.
Freezing-point
depression.
Soil No.
Mois-
ture.
Freezing- point
depression.
Soil No.
Mois-
ture.
Freezing-point
depression.
Dec. 27
Mar. 29
Dec. 19
Jan.
9-16
Oct. 23
Nov. 3
4A...
4B...
5A...
5B...
6A...
Per cent
19-3
18.0
18.2
18.0
24-5
24.8
15-9
14.7
13-9
II. 7
17.0
16.5
'C.
0.055
. 121
,066
. 106
,085
.103
.038
. 108
.030
•059
•039
.068
'C.
0.073
■n3
.127
•153
. 120
• 149
•052
.081
•013
. 046
. 091
. 096
lA. ..
iB. ..
2A. ..
2B. ..
3A...
3B...
8A...
8B. ..
9A...
9B...
loA. .
loB. .
Per cent
23-4
22.3
20. 4
20. 0
20. 9
20. 7
12.7
12.5
II. 0
10. 6
15- I
14. 6
"C.
0.037
•095
.047
.0S7
.048
. 060
•049
. Ill
•031
•058
TO46
•079
"C.
0. 050
.152
•050
. 100
•063
. no
. 062
. 142
.038
. 072
.058
• 105
loB. .
loB. .
loB. .
icB. .
loB. .
loB. .
loB. .
Per cent
13.6
12.8
14.4
14. 6
16. 0
17.0
17.8
'C.
0. 099
•c.
0.439
Dec. 5
Dec. 13
6B...
7A...
7B
12A. .
12B..
14A. .
14B..
0-153
.127
. 120
. 092
.080
.074
0. 180
. 169
. 148
.114
. 104
.097
It will be evident from Table IV that in the majority of cases there
has been a distinct increase in the depression of the freezing point under
the conditions of storage. It is significant that at the end of the storage
period the uncropped soils still maintain their superiority with regard to
the concentration of the soil solution. The time at which the sample
is taken may in some cases influence the further changes taking place at
laboratory temperatures. This suggestion is supported by the data for
soil loB. The samples taken in November gave a much greater increase
in concentration after standing than was the case with the sample taken
in December. Doubtless these differences may be associated \\-ith con-
ditions favorable for nitrate production, but the extraction experiments
have shown that other elements are increased in solubility simrltaneously
with the increased production of nitrates.
EFFECT OF CARBON-DIOXID GAS ON CONCENTRATION OF SOIL
SOLUTION
The increase in concentration of the soil solution noted above, as well
as the seasonal changes, may probably be related to the activities of
microorganisms and their production of carbon dioxid. Several experi-
ments were therefore undertaken to demonstrate the effect of carbon/
dioxid gas on freezing-point depressions. A sample of soil 5A containing
17.7 per cent of water was placed in a stoppered bottle, and a current of
carbon-dioxid gas was passed through the moist soil for about five
minutes. This treatment was repeated several times. The bottle was
then closed tightly and the soil permitted to stand overnight in contact
with the atmosphere of carbon dioxid, after which the depression of the
freezing point was determined and compared with another sample of
the soil exactly the same but not treated with carbon dioxid. The
control sample gave a depression of 0.083° C. and the treated soU 0.138°.
Feb. II. i9i8 Freezing-Point Method and Soil Solution
385
A similar experiment was performed on' the 5B soil, containing 16.7 per
cent of moisture. The results in this case were 0.156° depression for the
control and 0.218° for the sample treated with carbon dioxid. It is
evident from these data that the gas has a striking effect in increasing the
concentration of the soil solution. Here, again, it is interesting to
observe that the cropped and uncropped soils maintain the same general
relation to each other, even after the treatment described. Water
extractions made on the same soils showed that the total water-soluble
material was considerably increased by the carbon-dioxid treatment.
Calcium was particularly affected, but several other elements were also
made more soluble.
EFFECT OF DRYING ON CONCENTRATION OF SOIL SOLUTION
One of the important influences which may affect the chemical state
of the soil is that of drying. King (4) has shown that drying soils in the
oven causes a considerable increase in the quantity of water-soluble
material. Bouyoucos and McCool (2) have pointed out that the evidence
obtained by their freezing-point method tends to the same conclusion.
The writers have desired in the present research to apply this method to
the study of the influence of air-drying on the soils from the tanks.
Freezing-point depressions were first determined upon the moist soils,
and samples were then spread out in the laboratory and allowed to
become thoroughly air-dried. They were then mixed in a mortar with
suitable quantities of distilled water, allowed to remain in closed jars for
several hours, and the freezing points, as well as the percentages of
moisture again determined. The results are given in Table V.
Table V. — Effect of drying on freezing-point depressions
Soil No.
lA.
iB.
4A.
4B.
5A.
5B.
8A.
8B.
9A.
loA
loB
iiA
iiB
Moisture.
Per cent.
22. 7
20. I
17. 6
17. 2
17-7
16.7
10. 6
10. 8
9.4
9-5
14. I
13-4
13- 5
Freeziag-point depres-
sions.
Orisrinal
soil.
°C.
O. 056
108
oq8
148
,083
156
062
135
057
087
066
125
065
After dry-
ing and
retnoisten-
ing.a
0.050
. 191
.089
. 114
.068
. 100
.058
. 142
.083
. 126
.094
• 136
. 067
. 110
o Calculated to same moisture content as original sample.
386
Journal of Agricultural Research
Vol. XII, No. 6
It will be noted that there are no marked or consistent changes in the
freezing-point depressions as a result of the air-drying. In samples 5A
and 5B water extractions (i to 5) were made with the result that only
negligible increases occurred in the quantity of any of the constituents,
with the exception of the organic matter which was made somewhat
more soluble. It may be suggested that a further increase in the con-
centration of the soil solution might subsequently be caused through
stimulation of bacterial activity caused by the additional organic nutrj-
ment made available. In this case the solubility of various elements
would be increased indirectly by drying, but there is no evidence that
nutrient elements become more soluble by the simple process of drying
at ordinary temperatures.
EXPERIMENTS WITH LEACHED SOILS
All of the foregoing work has consistently indicated that a given soil
may yield very dififerent soil solutions under different conditions. In
order to establish the lower limits, it was decided to leach several soils
with distilled water and then determine the depression of the freezing
point. These were determined immediately after leaching and at inter-
vals over several months, during which time the soils were kept at
laboratory temperature in closed jars. Table VI shows the results ob-
tained.
Table VI. — Effect of leaching on freezing-point depressions
Soil No.
2B.
loB.
8C.
5C.
Treatment.
fSoo gtn. leached with 2 liters of
\ water.
.do.
1,000 gm. leached with 10 liters
of water.
.do.
Freezing-point depression.
Before leaching.
Deter-
mined.
O. 117
306
207
. 060
Water.
Per cent.
17. 2
14-3
21.5
After leaching.
Date.
Nov. 3
Nov. 8
Dec. 5
Feb. 26
May 2
Nov. 3
Nov. 8
<!Dec. 5
Feb. 26
May 2
Jan. 9
Feb. 6
iFeb. 26
[May 2
fFeb. 6
{Feb. 26
[May 2
Deter-
mined.
°C.
0.034
. 021
•033
•034
.030
. 027
. 020
.032
. 040
.051
. 004
. 006
. 012
•043
• 013
.049
. 064
Per cent.
20. 4
17. 2
17. 6
22. 4
Feb. II, 1918 Freezing-Point Method and Soil Solution 387
It seems quite clear from an inspection of these data that the concen-
tration of the soil solution may be reduced to an extremely low point
through leaching and that this condition is maintained over a considerable
period of time. Leaching has been shown by Lipman (5) to inhibit
bacterial action markedly, and in the absence of such activity, it seems
that the ability of the soil to recover its high level of concentration is
very limited.
RELATION OF WATER EXTRACTS TO SOIL SOLUTION
Before entering upon the final discussion it is desirable to consider at
this point certain more detailed relations of water extracts to the soil
solutions, as indicated by the freezing-point method. The work of
Bouyoucos and McCool enables us to obtain very useful and interesting
estimates of the concentration of the soil sola i ion, but obviously it is
only the total concentration which is measured. We are quite unable to
predict the nature of the substances whose resultant effect is expressed
by the lowering of the freezing point. Only by some method of water
extraction is it possible to gain any insight into the proportions of indi-
vidual elements. Various comparisons between osmotic pressures in
the soil solution and concentrations in water extracts may, however, be
made, which, in certain directions, make possible very interesting
deductions.
When different soils are to be compared, the first inquiry must be
concerned with the amount of lowering of the freezing point which
occurs for each 100 p. p. m. of total solids. This phase of the question
has already been investigated by Bouyoucos and McCool, who found
that extracts of varied types of soil gave very nearly constant results,
approximately 0.0032° C. depression for each 100 p. p. m. of total solids
in the solution. The writers have obtained closely similar figures for the
soils used in this investigation as is made evident by the following data :
Freezing-point depression per 100 p. p. m. of total solids
Soil No. "C.
2 o. 0032
4 0031
9 0032
10 0030
8 0034
5 0030
Average. 0032
By the use of this constant the approximate strength of the soil solution
in terms of parts per million of total solids may be calculated, and this
has been done in Table VII.
388
Journal of Agricultural Research
Vol. XII, No. 6
Table VII.
-Comparison of total solids in soil solution as calculated by freezing-point
and extraction methods
Date.
Nov. 15
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Dec. 21
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Dec. 27
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Feb. 12
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Soil
No.
4A
4B
5A
sB
6A
6B
7A
7B
12A
12B
14A
14B
lA
iB
2A
2B
3A
3B
8A
8B
9A
9B
loA
loB
iiA
iiB
4A
4B
SA
5B
6A
6B
7A
7B
12A
12B
14A
14B
lA
iB
2A
2B
3A
3B
8A
SB
9A
9B
loA
loB
iiA
iiB
Total
mois-
ture.
Per d.
17-3
16. 7
16.8
17-5
24
24
14
13
14
11. 4
17-5
16. 2
23-4
22.3
20. 4
20. o
20. 9
20. 7
12. 7
12. s
11. o
10. 6
15- I
14. 6
14-5
13.6
19-3
18.0
18.2
18.0
24- 5
24.8
15- 9
14.7
13-9
12. 7
17. o
16. 5
22. 7
21. 7
20. 2
20. 2
20. 6
19.9
10. 6
10.8
9.4
9-5
14. I
13-4
13-5
Free
water.
Per d.
4-3
3
3
4
6
6
8
7
10.
7
10.
8
5
4
7
7
Freezing-
point de-
pression.
C.
068
082
079
088
068
lOI
082
151
061
138
033
095
037
095
047
087
048
060
049
III
031
058
046
079
047
134
055
121
066
106
085
103
038
108
030
059
039
068
056
108
056
lOI
062
120
062
135
057
087
066
125
065
Concen-
tration of
total
dissolved
solids
calculated
from
freezing-
point de-
pression.
P.p.
100
600
500
800
100
200
600
700
900
300
000
000
200
000
500
700
500
900
500
500
000
800
400
500
500
200
700
800
100
300
700
200
200
400
900
800
200
100
700
400
700
200
900
800
900
200
800
700
100
900
000
200
Concen-
tration of
total
dissolved
solids
calculated
from
extracts
to free
water.
P. p. m.
10, 000
000
500
900
900
900
400
400
500
800
800
300
800
400
800
300
200
600
200
500
700
200
800
700
600
600
300
800
700
300
100
100
700
800
400
300
400
800
800
700
700
600
300
100
800
900
600
100
900
400
900
300
Total
dissolved
solids in
100 grams
of moist
soil cal-
cixlated
from
freezing-
point de-
pression.
Gm.
O. 009
. 010
. 010
•013
.013
. 021
.023
.036
. 020
.032
. 010
. 026
. 006
.013
. on
. 019
. 010
.013
.013
.030
• 005
.008
. 010
. 017
.013
.032
. on
. 019
. on
. 017
.018
.023
. 012
.030
. 009
. 014
. on
. 019
. 007
. on
. 012
.023
•013
. 026
■013
. 029
. 006
. 009
.013
. 021
.015
.032
Total
sissolved
solids in
100 grams
of moist
soil cal-
culated
from
extracts.
Gm.
0.043
•055
. 029
. 040
.031
. 046
.030
• 049
. 016
.036
.028
•055
. 026
•053
. 021
.044
. 029
.038
.028
. 064
. 019
. 029
. 020
.038
.031
• 050
• 052
.074
. 040
. 067
. 042
. 072
. 027
.068
. 024
■033
.032
•043
•045
. 069
.049
•055
•03s
. 042
• 032
•054
. 029
•039
.030
.051
.036
. 062
Feb. II, 1918 Freezing-Point Method and Soil Solution
389
Table VII.
-Comparison of total solids in soil solution as calculated by freezing-point
and extraction methods — Continued
Date.
Soil
No.
Total
mois-
ture.
Free
water.
Freezing-
point de-
pression.
Concen-
trated
total
dissolved
solids
calculated
from
freezing-
point de-
pression.
Concen-
trated
total
dissolved
solids
calculated
from
extracts
to free
water.
Total
dissolved
solids in
100 grams
of moist
soil cal-
culated
from
freezing-
point de-
pression.
Total
dissolved
solids in
100 grams
of moist
soil cal-
culated
from
extracts.
Feb, 19
Do
4A
4B
5A
ll
6B
7A
7B
12A
12B
14A
14B
Perd.
17.6
17.2
18. I
17.2
24.8
25.0
14. 6
14. 0
12.7
II. 4
16. I
15.0
Perd.
4.6
4.2
4.2
7-1
7-3
8.6
8.0
8.7
7-4
8.6
7-5
"C.
0. C598
.148
. 106
•115
.066
. 126
.053
•093
•043
.083
.063
.123
P. p. m.
3, ICO
4, 600
3>300
3, 600
2, 100
3,900
1, 700
2, 900
1,300
2, 600
2, 000
3>900
P. p.m.
8,800
16, 000
7, 100
II, 900
4,900
7,500
2,900
5,300
2,300
3,900
3,900
7, 200
Gm.
0. 014
. 019
.017
.015
.015
. 029
.015
.023
. Oil
. 019
.017
. 029
Gm.
. 041
. 067
•036
.050
•035
•055
.025
. 042
. 020
Do
Do
Do
Do
Do
Do
Do
Do
. 029
•034
•054
Do
Do
Upon several occasions the total solids were determined in the i-to-5
extracts of all soils. These were also calculated in terms of concentra-
tion of total soil moisture, as well as of free soil water. From the depres-
sions of the freezing point, calculations were made of concentrations
expressed as total solids in parts per million of the soil solution with the
constant just described (Table VII). The first and most striking observ-
ation based on these results is that the relation between the total solids
in the extracts of cropped and uncropped soils is also manifested by the
depressions of the freezing point. In fact, throughout the experiment
this correlation was noted. The fluctuations pf the important nutrient
elements in the water extracts already described by Stewart were in
general agreement with the changes in the concentration of the soil
solution. The low and high points came at about the same periods of
the growing season in both cases.
From these considerations it might seem justifiable to assume that the
material present in the actual soil solution forms at least an important
fraction of the total solids dissolved in the i-to-5 extract. It is, how-
ever, highly improbable that an extract should consist exclusively of the
material present in the soil solution. Necessarily an additional quota
would be derived from other more or less soluble substances, and the
quantity dissolved would depend upon the conditions of extraction. A
number of years ago Mitscherlich (6), in an important study on extracts
of soils prepared with carbonated water, emphasized this point of view.
His procedure consisted in making extracts with varying proportions of
carbonated water, ranging from 5 to i up to 30 to i . From the data so
390 Journal of Agricultural Research voi. xn. No. 6
obtained he constructed graphs and proved that with increasing quan-
tities of water more material was dissolved. He therefore divided the
total dissolved solids into two fractions : (a) That part soluble even with
low moisture content, and (b) the additional material dissolved by the
excess of solvent.
A number of difficulties arise in connection with this method of arriving
at the concentration of soil solution. In the first place, it is not safe to
assume that a curve based on one range of extractions can accurately
be extended to cover another range of extractions. Indeed, the experi-
mental data indicate that with the smaller proportions of water the
curves may change their direction very appreciably and it is unfor-
tunately impracticable to obtain extracts for analysis in those concen-
trations which correspond to optimum moisture conditions. Another
limiting factor, previously neglected, has been described by Stewart.
This concerns the differential effect of the solvent. The actual solvent
in any case is not pure water, but pure water plus the solids already
dissolved in the soil solution, and these vary enormously with changing
conditions, even in the same soil. It is quite obvious that this factor
would modify any calculations of the concentration of the soil solution
based on water extracts. Moreover, a question may arise with regard
to the relation of the soil extract to the residual solution. It is possible
that the equilibrium may not be the same for the low and high propor-
tions of water, especially in soils of colloidal character.
Notwithstanding these difficulties met with in attempting to predict
accurately the concentration of the soil solution from soil extracts, it is
still possible to make certain valuable comparisons of the two methods.
If we should contrast the concentration of the soil solution, calculated
from the extracts to the total moisture of the soil, with the concentra-
tions shown by the freezing-point method, there would be a general
similarity of magnitude. Logically, however, a comparison is much
more justifiable when the extracts are calculated not to the total soil
water, but to the free water, in the sense meant by Bouyoucos. It is
then apparent that the concentration of the soil solution calculated by
the extraction method is from two to five times that indicated by the
depressions of the freezing point (Table VII). In other words, there is
a considerably greater quantity of total solids dissolved in the i-to-5
extract than is actually present in the soil solution at any given moment.
The general order of magnitude of the two quantities is, however, evi-
dently not disproportionate, the material extracted from the sandy
loams averaging only about twice the quantity actually present in the
soil solution, as shown by the freezing-point method.
The total amounts of dissolved material in 100 gm. of soil have been
calculated by multiplying the percentages of free water by the concen-
trations in terms of parts per million of total solids. If we use the
Feb. II, i9i8 Freezing-Point Method and Soil Solution 391
freezing-point depressions as a basis, it appears that from 0.0 1 to 0.03
gm. of total solids is in solution for each 100 gm. of moist soil, while
the total solids obtained by i-to-5 extractions vary from 0.02 to 0.06
gm. per 100 gm. of soil.
In order to obtain approximate estimates of the material dissolved in
an extract additional to that present in the soil solution, two soils (No.
5 and 8) were subjected to the procedure of successive extraction. One
kgm. of each soil was extracted with a total of 10 liters of distilled water.
Ten extractions were made, with i liter of water for each extraction.
The results of the 10 extracts are plotted in figure 9. It will be noted
that the first extract contained a very much greater quantity of dissolved
solids than any subsequent extract. It was suggested by Stewart that
some idea of the extra dissolved material might be formed by using as a
basis of calculation the quantity dissolved in each liter after the extracts
had become relatively constant. Thus, for soil 5 roughly 150 p. p. m. of
total solids were found in each of the later extracts, and for soil 8, 100
p. p. m. By assuming that the rate of solubility is uniform, in a i-to-5
extract there would be 5 times 150, or 750, p. p. m. of extra material dis-
solved in the case of soil 5, and 500 p. p. m. for soil 8. The total solids
dissolved from i kgm. of soil by 5 liters of water aggregate 1,300 and
1,100 p. p. m. of total solids, respectively. This means that not more
than 50 per cent could have originally been present in the soil solution.
If this correction were applied to the extracts, the agreement with the
data obtained by the freezing-point method would in some cases be
fairly close, but usually the extracts would still give somewhat higher
results. This fact is reasonably explained on the assumption that the
first liter of solvent has dissolved out some material relatively soluble
although not actually present in the soil solution, while the later extracts
contain only difficultly soluble substances.
GENERAL DISCUSSION
It will now be well to correlate certain broad relations which may be
deduced from the data presented in this paper through some further
reference to the principal investigation of this series by Stewart. It
should be emphasized again that neither from the water extraction nor
freezing-point methods is there any evidence that the soil solution has a
constant composition. On the contrary, the soil solution even of the
same soil may vary greatly under different conditions. The water-
extraction method indicates that this variation may occur with all the
principal nutrient elements with the exception of phosphorus. These
data accord with the view expressed by Bouyoucos and McCool, that
the soil solution should not be considered as saturated. The opposite
conception as outlined in the earlier work of Whitney and Cameron (9)
is not upheld by the present investigation. If the great excess of nearly
insoluble minerals in a soil were the determining factor in the soil solu-
27809«— 18 7
>5
O
PHOSPHATE(P0,) SOILS
L
£34-567^9
NUMBLR. Of EXTRACT.
Fig. g.-Graphs showiag the results of successive extractions of soils s and 8. One liter of water to i
kgm. of soil for each extraction was used.
Feb. II, 1918 Freezing-Point Method and Soil Solution 393
tion, then its concentration in the cropped soil should not be strikingly
diminished over a long period of time after the crop is removed, which in
fact is the case. Also the soil solution in a leached soil has a greatly
diminished concentration which is not markedly increased by long
standing. In brief, the evidence is quite opposed to the theory that
there is an immediate restoration of equilibrium when the soil solution is
diminished in concentration by the plant or other agency.
One explanation of the fluctuations in the soil solution lies most prob-
ably in the varying nature of the solvent, especially in its content of
carbon dioxid, which may well be one of the preponderating factors.
The effect of carbon dioxid on the solution of soil minerals has been
demonstrated by Mitscherlich, and the present investigation has indi-
cated the effect of carbon dioxid in increasing the depression of the freez-
ing point. It is generally conceded that the partial pressure of carbon
dioxid in the soil atmosphere is greater than that of the outside air.
Russell (7) has shown that the content of carbon dioxid in the soil may
vary greatly at different times of the year, according to the intensity of
bacterial action. In addition to carbon dioxid, other products of bac-
terial activity may have a further influence, the extent of which is still
problematical.
It has been a general teaching of agricultural art that soil fertility is
increased by those operations which tend to bring about optimum
biological conditions in the soil. The experiments recorded in this
investigation seem to afford direct evidence that the soil solution may
be greatly affected by the activity of microorganisms and indirectly
therefore by cultivation, temperature, organic matter, etc. Finally, the
results of the experiments reported in this paper have considerable
significance to the plant physiologist, since comparisons may be made
between osmotic pressures in nutrient solutions and in the soil solution
as it actually exists in the soil under conditions favorable to crop
growth. It will be observed that in none of the soils did the plants
obtain their nutriment from a highly concentrated solution. The
general range of concentrations was from a maximum of 0.5 or i.o
atmosphere to a minimum of o.i or 0.2 atmosphere. Under the condi-
tions of approximately optimum moisture contents there was no sharp
distinction between the silty clay loams and the fine sandy loams. By
lowering the moisture content of the clay loams considerably, it is true
that a very high concentration of the soil solution results, but under
moisture conditions favorable to plant growth the solution is dilute.
From certain of the soils stored in the bins a different idea of the con-
centration of the soil solution might be obtained, since this treatment
has in some cases greatly increased the soluble material. The true
conception of the nature of the nutrient media in soils can only be
attained when soils are studied throughout the season under conditions
suitable for crop growth.
394 Journal of Agricultural Research Voi. xii. No. e
Preliminary experiments have been carried out with the use of sand
and water-culture methods with the nutrient solutions whose concentra-
tions are comparable with those of the soil solution. The results have
indicated that the osmotic pressures existing in the soil are also most
favorable to the growth of barley in culture solutions. When con-
siderably greater concentrations are maintained, decreased yields result,
while very much lower osmotic pressures are suboptimum. Further
experiments are now in progress.
It has not been possible in this investigation to make any general
correlation with crop yield, although it is worthy of note that the two
soils of lowest production. No. 9 and 12, show consistently low con-
centrations of the soil solution and also yield water extracts containing
exceptionally small quantities of nutrient elements. In the discussions
of Stewart and Burd the relations of chemical analysis of the medium
to crop production are discussed more completely.
SUMMARY
(i) Freezing-point depressions have been determined on 13 soils
under a variety of conditions.
(2) The concentration of the soil solution has been found to vary with
the season and also as a result of treatment with carbon dioxid, leaching,
incubation, etc.
(3) The growth of a crop markedly diminishes the concentration of
the soil solution. This effect is still evident at the beginning of the
following season.
(4) The soil solutions under conditions favorable to crop growth were
found to be very dilute, particularly at the height of the growing season.
(5) Certain general agreements between the extraction and freezing-
point methods are discussed.
LITERATURE CITED
(1) BOUYOUCOS, G. J.
I917. MEASUREMENT OP THE INACTIVE, OR UNFREE, MOISTURE IN THE SOIL
BY MEANS OF THE DiLATOMETER METHOD. In Jour. Agf. Research, v. 8,
no. 6, p. 195-217, I fig. Literature cited, p. 217.
(2) and McCooL, M. M.
1916. THE Freezing-point method as a new means of measuring the
CONCENTRATION OF THE SOIL SOLUTION DIRECTLY IN THE SOIL. Mich.
Agr. Exp. Sta. Tech. BuL 24, p. 592-631, 2 figs.
(3) BURD, J. S.
1917. WATER EXTRACTIONS OF SOILS AS CRITERIA OF THEIR CROP-PRODUCING
POWER. In Jour. Agr. Research, v. 12, no. 6, p. 297-309, i fig.
(4) King, F. H.
1905. INVESTIGATIONS IN SOIL MANAGEMENT. U. S. Dcpt. Agr. Bur. Soils Bul.
26, 205 p., 7 figs., 4 pis.
Feb. II, 1918 Freezing-Point Method and Soil Solution 395
(5) LiPMAN, C. B.
1916. PRELIMINARY EXPERIMENTS ON SOME EKECTS OP LEACHING ON THE SOIL
FLORA. In Soil Science, v. i, no. 3, p. 291-297. Literature cited,
p. 297.
(6) MiTSCHERLICH, E. A.
1907. EINE CHEMISCHE BODENANALYSE PUR PPLANZBNPHYSIOLOGISCHE POR-
SCHUNGEN. In Landw. Jahrb., Bd. 36, Heft 2, p. 309-369, 10 figs.,
ipl.
(7) Russell, E. J.
1915. THE ATMOSPHERE OP THE soil: ITS COMPOSITION ANDCAUSES OF VABUATION.
In Jour. Agr. Sci., v. 7, no. i, p. 1-45, 17 figs.
(8) Stewart, G. R.
191 7. effect of season and crop growth in modifying the soil extract.
In Jour. Agr. Research, v. 12, no. 6, pp. 311-368, 24 fig., i pi.
(9) Whitney, Milton, and Cameron, F. K.
1903. THE CHEMISTRY OF THE SOIL AS RELATED TO CROP PRODUCTION. U. S.
Dept. Agr. Bur. Soils, Bui. 22, 71 p.
ADDITIONAL COPIES
OP THIS PUBLICATION MAY BE PROCTJBED FROM
THE STTPEKrNTENDENT OF DOCUMENTS
GOVEENMENT FEINTING OFHCE
WASHINGTON, D. C.
AT
20 CENTS PER COPY
SuBscBiPTioN Price, $3.00 Pes Yeab
——'"'"'■ ~-~-~— — -— —TnmMi—
Vol. XII KEBRUARY 18, 1918 No. 7
JOURNAL OP
AGRICULTURAL
RESEARCH
CONTENXS
Page
Efficacy of Some Anthelmintics _ - _ - _ 397
MAURICE C. HALL and WINTHROP D. FOSTER
(Contribution from Bureau of Animal Indastiy)
Tobacco Wildfire -449
FREDERICK A. WOLF and A. C. FOSTER
(Contribution from North Carolina Agricultural Experiment Station)
Gipsy-Moth Larvae as Agents in the Dissemination of the
White-Pine Blister-Rust - - - - - - 459
G. FLIPPO GRAVATT and G. B. POSEY
( Contribution from Bureau of Plant Industry)
PDBUSHED BY AUTHOMTY OF THE SECRETARY OF AGRICULTURE,
WITH THE COOPERATION OF THE ASSOCUTION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINOTON, D. C.
WAtHINOTOM I OOVERNMENT PNINTINa OfTIOS 1 1818
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARL F. KELLERMAN, Chairman RAYMOND PEARL*
Physiologist and Associate Chief, Burtau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Entomologist and Assistant Chief, Burtau
of Enfomotogy
Biologist, Maine Agricuitural Experim$nl
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition, The
Pennsylvania Stale College
E. M. FREEMAN
Botanist, Plant Pathologist and Assistant
Dean, Agricultural EiperimenI Station of
the University of Minnesota
All correspondence regarding articles from the Department of Agricultiu'e should be
addressed to Karl F. Kellerman, Journal of Agricultiu^ Research, Washington, D. C.
*Dr. Pearl has undertaken special work in connection with the war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition;
State College, Pa.
JOINAL OF AGRIdMAL RESEARCH
Vol. XII
Washington, D. C, February i8, 1918
No. 7
EFFICACY OF SOME ANTHEIyMINTICS
By Maurice C. Kali, ^ formerly Assistant Zoologist, and Winthrop D. Foster, Junior
Zoologist, Zoological Division, Bureau of Animal Industry, United States Department
of Agriculture
INTRODUCTION
Although the use of anthelmintic treatment is an old practice in human
and veterinary medicine, the efficacy of the various substances employed
as anthelmintics is not well known. What information is available is
based largely on clinical observations, efficacy being judged to a large
extent on a consideration of the improvement or lack of improvement in
the patient's health after treatment. In other instances the efficacy of
the treatment has been checked by fecal examinations for worms passed
and for eggs persisting in the feces; but, while this method gives real and
valuable information, it is somewhat inexact. The methods employed
for examining feces for worms passed are often rather casual; and nega-
tive findings in examining feces for eggs, especially when a small number
of preparations are examined for only a few days after treatment, are
not always conclusive.
A more satisfactory method of investigating anthelmintic efficacy is to
administer treatment to animals, to collect all feces passed for a number
of days, to recover from them all worms present, and then to kill the
animals and collect all worms remaining. In this way it is possible to
arrive at a fair idea of the anthelmintic effect to be expected from a drug,
the correctness of the conclusions depending, of course, on the number of
experimental animals used and their degree of infestation.
While it is thus possible to express the efficacy of a drug in the form of
a mathematical ratio, the writers are fully aware that such ratios, except
when based on extensive data, can not be considered an accurate index
of the efficacy of the drug, since many factors, not entirely within con-
trol, such as the individual reaction of the animal, the amount of material
in the alimentary tract, and the potency of the drug, all enter into the
problem.
In carrying out this series of experiments the plan of the writers was
to test as many drugs as possible having a known or alleged anthelmintic
value, abandoning those which gave no results, and making further
Journal of Agricultural Research,
Washington, D. C.
Iz
' Resigned September 19, 1916.
(397)
Vol. xn. No. 7
Feb. 18, 1918
Key No. A— 34
398 Journal of Agricultural Research voi. xii. no. 7
experiments with the more promising. It is therefore possible that some
of the drugs tested only once and on a limited number of animals may
have more anthelmintic value than the tests indicate. On account of
the extent of the field to be covered, the writers did not feel justified in
devoting more effort to those drugs which gave small promise of success.
Some such method as the above has been employed by previous in-
vestigators. Hutcheon {iSgiy made numerous tests of anthelmintic
treatments for stomach worms in sheep and goats in South Africa,
and followed the treatments by post-mortem examinations to determine
the immediate effect on the worms. Stiles {1901, 1902) did similar work
in this country, and a number of veterinarians and stockmen made investi-
gations involving treatment, post-mortem examination, and clinical ob-
servation. But, so far as the writers are aware at present, a detailed
series of experiments covering the treatment of animals and the collection
of all worms from the feces for a number of days up to the time of making
a post-mortem examination in which all worms remaining were collected,
has not been reported.
In this work the following method was pursued: The animals were
given an appropriate dose of the anthelmintic to be tested, the method
of dosage varying with the purpose of the experiment and the substance
to be tested, preliminary purgation being undertaken or omitted as de-
sired. Treatment was usually administered in the morning, and all feces
were collected every morning thereafter until the animal was killed. The
feces were washed through a set of graded screens and the screens exam-
ined for worms. The animal was usually killed the morning of the fourth
day after the administration of the last dose of the anthelmintic, and all
parasites remaining were collected and counted.^ The percentage of effi-
cacy was then estimated from the number of worms found on post-
mortem examination and that number plus the number passed after
treatment. When preliminary purgation was resorted to, the feces were
collected on the following day, the day of administering the anthelmintic,
to ascertain what worms if any were removed by simple purgation.
For convenience we have arranged our experimental data in three
groups: (i) Simple purgatives, (2) a group including anthelmintic shaving
a mineral base and coal-tar products, and (3) a group covering the vege-
table anthelmintics. Tables I to V of the discussion of results summarize
the results of the experiments.
' Bibliographic citations in parentheses refer to " Literature cited." p. 446-447.
' In conducting these experiments the writers occasionally found on post-mortem examination dead
worms in the large intestine or rectum which normally are found in the small intestine. In such cases,
as the worms were evidently in the process of passing out, they are credited as being removed by the an-
thelmintic.
veh. i8, i9i8 Efficacy of Some Anthelmintics 399
EXPERIMENTS WITH SIMPLE PURGATIVES
CALOMEly
For worms in dogs. — Calomel in, fairly large single doses followed
by castor oil proved inefficacious as a remedy for ascarids in dogs.
Four infested pups weighing 3.6 to 4.5 kgm. were fasted from noon
of the day before treatment, given 4 to 5 gm. of calomel, and on the
day following treatment given 1 5 mils (milliliters) of castor oil. Three
of the dogs passed no worms after treatment ; one passed a single as-
carid (Belascaris marginata). Two of the dogs which passed no worms
when examined post-mortem showed 10 and 8 B. marginata, respec-
tively. The treatment was so evidently inefficacious that the other
two dogs were not sacrificed for post-mortem examination. The results
of this experiment do not accord with the claims occasionally made
as to the anthelmintic value of simple purgatives and cathartics.
CASTOR oily
For WORMS in dogs. — Although the efficacy of castor oil as an an-
thelmintic was not tested as a separate experiment, in several experi-
ments it was given as a preliminary purge. The feces passed follow-
ing its use were examined, and the worms passed compared with the
number found post-mortem. In these experiments only 24 hours
were allowed between the administration of the oil and the subsequent
anthelmintic. It is therefore possible that the oil alone might have
shown greater efficacy if the feces were collected for a longer time.
However, in our experience the greater number of worms are passed
within 24 hours after the administration of an anthelmintic; and in the
case of a drug like castor oil, which usually acts within a few hours after
its administration, it is not likely that the conclusions of the writers are
much in error.
In II experiments in which castor oil was used as a preliminary purge
on 50 dogs it removed 27 B. marginata out of a total of 351. Its effect
on hookworms (Ancylostoma caninum) and whipworms (Trichuris de-
pressiuscula) was practically nil. It was ineffective more frequently than
it was effective, and in no case did it remove all the ascarids from a
single dog. It has therefore little to recommend it as an anthelmintic.
It may, however, have some diagnostic value in veterinary practice to
confirm a clinical diagnosis of ascarid infestation in dogs when a micro-
scopic examination of the feces is impossible, though it is evident that
even for this purpose it would not always be reliable.
When worms are passed as the result of the increased peristalsis due to
a simple purgative, there is always the suspicion that the worms were dead
and would have passed out anyway and that the purge merely hastened
their removal.
400 Journal of Agricultural Research voi. xii, no. ^
EPSOM SALT
For worms in dogs. — Epsom salt was tested once as a preliminary
purge on three dogs which were given the salt in molasses at the rate of
4 gm. to each 5 kilos of weight. It had no effect on ascarids {Belascaris
marginata), hookworms (Ancylostoma caninum), whipworms {Trichuris
depressiuscula) , Dipyidium caninum, or Taenia pisiformis, all of which
were found present at post-mortem examination. Epsom salt is ex-
tremely distasteful to dogs and is likely to promote vomiting. The
salt is best given in a vehicle like molasses, which serves to disguise
the taste. Aside from this objectionable feature, the salt is less effective
in dogs than other purgatives, such as castor oil or calomel, and is not
particularly useful in canine practice.
For worms in hogs. — Campbell {19 17), in a discussion of the treat-
ment of swine for worms, mentions the use by a western veterinarian
of Epsom salt dissolved in the drinking water. He says that the hogs
are kept away from water for 12 to 24 hours, in warm weather not so
long, in cold weather even longer. One ounce (28.35 g^i-) of the salt is
allowed for pigs and up to eight ounces (226.79 g^i.) for hogs, then twice
this quantity is added to allow for waste. It is then given in the drinking
trough. The pigs being thirsty drink it. This completes the treatment.
A test of this treatment was made by the junior writer on two pigs weigh-
ing about 23 kgm. (50 pounds) each. The animals were kept in separate
pens on board floors and deprived of both food and water for 24 hours.
Each pig was then given 226.8 gm. (8 ounces) of Epsom salt dissolved in
about 3.8 liters (i gallon) of water and placed in the drinking troughs.
The pigs took a few swallows of the water and refused to taste more,
showing their resentment by overturning the troughs and spilling the
contents. The experiment was therefore repeated a few days later, the
troughs being nailed to the floor. After 24 hours without food or water
the pigs were given the salts as before and as in the previous experiment,
took only a few swallows, refusing to drink more. Although no other
water was given them for the next 24 hours, the troughs remained appar-
ently as full of the salt solution as at the beginning of the experiment.
As the pigs had now been without food and had drunk practically no
water for 48 hours, the salt water was removed and fresh water and food
given. The examination of the feces during the next three days revealed
no parasites excepting one nodular worm (Oesophagostomum dentatum).
The pigs were not killed, as the experiment was evidently a failure. A
microscopic examination of the last fecal samples collected showed
numerous ascarid eggs.
As it was evident that the Epsom salt in a solution of the strength
given was extremely distasteful, it was decided to repeat the experiment
with a weaker solution which the pigs might be induced to drink if suffi-
ciently thirsty. Accordingly, the pigs were again deprived of food and
Feb. i8, 1918 Efficacy of Some Anthelmintics 401
water for 28 hours and then given a solution made by dissolving 0.45
kgm. (i pound) of Epsom salt in 22.7 Hters (6 gallons) of water. As the
drinking troughs would hold only 3.8 liters (i gallon) each, each trough
when full contained 75.6 gm. of the salt in solution. It was intended
to refill the troughs as fast as they were emptied until each pig had taken
1 1.35 liters (3 gallons) of the solution, or 226.8 gm. of Epsom salt.
As in the previous experiment, the pigs merely tasted the solution -and
refused to drink more. No other water was given them for the next 28
hours during which time they had two meals of dry feed. As the pigs
had now been without a drink for 56 hours, the experiment was closed.
Another microscopic examination of the feces passed at the close of the
experiment revealed the presence of ascarids in both pigs.
It is evident from the above that a solution of Epsom salt, even when
less than 2 per cent in strength, is so distasteful to pigs that they will
not drink it even after a relatively long period of thirst. It may also be
noted that 11.4 liters (3 gallons) of water is much more than a pig of
22.67 kgm. (50 pounds) weight could consume in a day even if very
thirsty, so that even if the pigs had drunk the weaker solution the treat-
ment would have had to be prolonged for two or three days, greatly
, decreasing the purgative effect to be expected from a single dose of a
large quantity of Epsom salt.
In this connection it may be stated that in an experiment with oil of
chenopodium on hogs, conducted some years previously by the junior
writer, four hogs were given 113.4 gm. (4 ounces) of Epsom salt mixed
with a bran-mash feed, all the animals eating together. No difficulty
was experienced in getting the animals to eat the mixture, but the
amount of the salt allowed to each hog was only one-eighth of the dose
allowed in the present experiment.
EXPERIMENTS WITH ANTHELMINTICS OF A MINERAL NATURE AND
COAL-TAR PRODUCTS
TARTAR EMETIC
For worms in hogs. — Tartar emetic is commonly used as a remedy for
roundworms in horses, and this led the writers to test its effects upon
worms in hogs. Two small pigs were used, weighing 8.6 and 14.5 kgm.
They were each given 259 mgm. of tartar emetic dissolved in water,
followed after a short interval with 29.57 mils of castor oil. Winslow
(1913) gives the emetic dose of tartar emetic for hogs as 4 to 10 grains (259
to 648 mgm.). The drug in this case exerted no emetic effect. One of the
pigs passed 5 ascarids (Ascaris suum) after treatment and showed the
presence of 5 others on post-mortem; the other pig passed none and
showed 19 on post-mortem. Owing to the more or less common habit
which pigs have of devouring ascarids, it is possible that the second pig
also passed worms which were eaten; and the first may have passed
402 Journal of Agricultural Research voi.xu, No. 7
more than were recovered in the feces saved after treatment. A few
whipworms (Trichuris suis) and nodular worms (Oesophagostomum den-
iatum) were passed ; a few of the former were found in one of the pigs
post-mortem, not looked for in the other, and about 250 and 500 nodu-
lar worms were found, respectively, in the two pigs post-mortem.
This experiment demonstrated that tartar emetic has an anthelmintic
action on worms in swine, but further trials will be necessary before con-
clusions can be drawn as to its efficacy. The method of administration
used in the experiments is not likely to prove suitable in practice.
CHIvOROFORM
Chloroform has been more or less used for some time, either alone or
in combination with other substances, as an anthelmintic. Recently
Alessandrini {191 5) has commended it very highly for use against hook-
worms and other worms. He states that it has these advantages : Com-
bined with castor oil it requires no special preparation of the patient; it
can be administered in a single dose; it does not cause local or general
disturbances either immediately or subsequently; it is perfectly well tol-
erated and is not nauseous. It may be given in 3- to 4-gm. doses dis-
solved in olive oil or castor oil and is thoroughly efficacious against hook-
worms, whipworms, pinworms, and ascarids. Chloroform is a constitu-
ent of Hermann's mixtures, the formula of one of which is given on page
403. Schultz {1911) finds the chloroform to be the active anthel-
mintic ingredient of Hermann's white mixture. He found chloroform
very effective against hookworms in the dog and says of it :
It has proven rapid in its action and thus far not followed by any evil after effects.
Should the chloroform -castor oil mixture act as favorably in human beings as it has
for me in dogs, it will prove a universal worm remedy of great importance.
In view of the irremediable damage resulting from overdoses of chloro-
form, Schultz is inclined, however, to favor the use of the less effective
drug thymol. Billings and Hickey (1916) use a chloroform-castor-oil
mixture following the use of chenopodium in the treatment of hookworm
and other roundworm parasites of aliens detained at the immigrant sta-
tion at Angel Island, California. For an adult they administer 20 mils
of a mixture of chloroform and castor oil containing i .8 mils of chloroform.
The dose is graduated according to the apparent age.
The writers' experiments with chloroform as a drug for expelling hook-
worms from dogs inclined them to the belief that it was extremely
satisfactory compared with thymol, and on the suggestion of Dr. Charles
Wardell Stiles, of the United States Public Health Service, a note was
sent to Dr. Reid Hunt, of Harvard University, asking his opinion as to
the danger of the administration of chloroform. In his reply he states:
It has been shown .... that even one administration of chloroform causes distinct
changes in the liver, from which, however, animals almost always completely recover.
I presume that a single dose is efficient and ordinarily harmless ; but I think that if the
Feb. i8. 1918 Efficacy of Some Anthelmintics 403
liver or heart had already been injured or diseased, a single does might have serious
consequences. It would seem advisable not to repeat a dose for several days at
least.
The experiments of the writers with chloroform were as follows :
For worms in dogs. — Five dogs, weighing from 5.5 to 17.7 kgm., were
given, in the morning after fasting from the evening of the day before, 0.2
mil of chloroform per kilo of body weight mixed with 3 mils of castor oil
per kilo of body weight. All of the dogs were infested with hookworms
{Ancylostoma caninum), four very lightly, one heavily, three had a few
ascarids (Belascaris marginata),iour a few (i to 27) whipworms (Trichuris
depressiuscula) , two a few tapeworms (Dipylidium caninum), and one
numerous (250) tapeworms of the same species. The dog which had the
single hookworm failed to lose the parasite as a result of the treatment,
but none were left in the other lightly infested dogs, while more than
half the hookworms were passed by the heavily infested dog (457 passed
and 355 found post-mortem). One dog passed a single ascarid and showed
none post-mortem, two others showed i and 2 ascarids, respectively,
post-mortem, but had passed none after treatment. Two of the dogs out
of the four infested with whipworms passed a few of these worms, and all
four showed a few on post-mortem. After the treatment no tapeworms
were recovered from the feces of the dogs infested with D. caninum.
In this experiment chloroform and castor oil proved rather highly
efficacious in removing hookworms, removing all in three cases of light
infestation, failed to remove any in one case, and removed over half of
the parasites in a case of heavy infestation. The results in the case of
ascarids and whipworms are not striking. In the case of D. caninum
there is no evidence that chloroform is of value as an anthelmintic.
The failure of chloroform to remove all the hookworms from the heavily
infested dog indicates that repeated treatments may be necessary in
cases of heavy infestations, but in view of the dangers attending the use
of chloroform the advisability of repeating the dose is questionable.
Several days at least, perhaps a week or longer, should be allowed to
elapse to allow time for the animal to recover from the possible ill effects
of the first treatment. Further experiments are desirable.
A further test of the efficacy of chloroform in combination with other
drugs as an anthelmintic for hookworms in dogs was made with the
remedy in the form of Hermann's mixture. This preparation consists,
according to Railliet {19 15), of the following:
JOleoresin of male-fern 4 g™-* or)
|Oil of eucalyptus 2 gm. J
Chloroform 3 Z^-
Castor oil 4° g™-
He suggests its use as a substitute for thymol. Although the prepa-
ration is intended to include either male-fern (Dryopteris filix-mas) or
eucalyptus oil, through an error both drugs were included. In conduct-
ing this experiment two objects were aimed at : (i) To determine whether
404 Journal of Agricultural Research voi. xii, no. 7
or not the chloroform was the sole or principal cause of the efficacy of
the mixture as a remedy for hookworms {Ancylostoma spp.) ; (2) whether
the combination of castor oil and male-fern was likely to produce symp-
toms of male-fern intoxication, owing to its greater degree of absorption
when combined with an oil.
After a fast of 24 hours, two dogs weighing 9.5 and 16.5 kgm., re-
spectively, were given the full dose of Herman's mixture as detailed
above. Two other dogs weighing 9.6 and 12.4 kgm., respectively, were
given the mixture without chloroform. Nearly three-fourths of the
hookworms {Ancylostoma caninum) in the two dogs receiving the full
mixture were eliminated by the treatment, one dog passing 16 A.
caninum after the treatment and having only 3 left at the post-mortem
examination. On the other hand, out of 39 A. caninum in the two dogs
receiving the mixture without chloroform, only 3 were eliminated. The
mixture without chloroform showed some slight efficacy for whipworms,
but the figures are not striking. Two of the dogs were infested with
Dipylidium caninum, one passing a number of fragments with several
heads within 30 minutes after dosing. The other dog passed a few
segments. Both dogs were found free from D. caninum at post-mortem
examination. As these dogs had received the mixture without chloro-
form, it is evident that the taeniacidal value of the mixture was due to
the male fern or eucalyptol and net to chloroform. This may reasonably
be considered a demonstration of the efficacy cf male fern as a taeniacide.
To sum up, it appears from the above that Hermann's mixture (includ-
ing both eucalyptol and male-fern) is an efficient vermifuge for hook
worms, owing largely, if not entirely, to the chloroform content. It is
also efficacious for Dipylidium caninum, probably on account of the
male fern it contains. It seemed to have some slight effect on whip-
worms, but the evidence of this was by no means convincing. The
combination of male-fern and castor oil seemed to have no deleterious
effect on the experiment animals, in this respect supporting the opinion
of Lenhartz (1902) and Seifert {1908). Further experimentation, how-
ever, is desirable to elucidate this point.
For worms in sheep. — Chloroform was tested on two sheep weigh-
ing 25 and 28 kgm., respectively. The dose for each was mixed with
60 mils of castor oil, the smaller sheep receiving 5 mils of chloroform
(0.2 mil per kilo) and the other 10 mils (nearly 0.4 mil per kilo). Both
sheep died four days after dosing and showed, on post-mortem exami-
nation, lesions of gastro enteritis and pneumonia in the congestive stage.
The sheep were very lightly infested with parasites. None were recov-
ered from the feces of the sheep which received the smaller dose; three
stomach worms {Haemonchus contortus), five nodular worms (Oesopha-
gostomum columbianum) , and five hookworms {Bunostomum trigonoceph-
alum) were found in this sheep post-mortem. The sheep which received
the larger dose passed nine stomach worms and one nodular worm after
treatment, and two stomach worms were found post-mortem.
Feb. i8, 1918 Efficacy of Some Anthelmintics 405
The chloroform had an evident anthelmintic effect on the sheep which
received the larger dose. The fatal results of the treatment in both
cases, however, indicate that chloroform is not a promising anthelmintic
for use on sheep.
ETHKR
For worms in dogs. — Four dogs weighing from 2.38 to 4.89 kgm.
were given ether in a dose of 0.8 mil per kilo mixed with 15 mils of
castor oil for the two smaller dogs and 30 mils of castor oil for the two
larger dogs. The afternoon of the day before treatment the dogs were
given a preliminary dose of castor oil, and were not fed until several hours
after receiving the ether. No worms were recovered from the feces passed
between the administration of the preliminary dose of castor oil and the
administration of the ether and oil mixture. This mixture was much
resented by the dogs, and all of them were salivated by it and showed
more or less evidence of collapse. One of the dogs passed a few tape-
worm segments (Taenia pisiformis), but no other tapeworm material was
recovered from the feces ; nor were any tapeworms found post-mortem.
No hookworms were recovered from the feces, but all showed infesta-
tion on post-mortem examination, the number of worms being 7, 21,
233, and 242, respectively. One of the dogs passed i ascarid and one
passed 6. At the post-mortem examination no ascarids were found in
the former and 11 in the latter. The other two passed no ascarids;
4 were found post-mortem in one, but i of these was in the rectum,
evidently about to be passed; i ascarid was found post-mortem in the
other of these two dogs.
The conclusion from this experiment is that the ether exhibited a
rather slight anthelmintic action against ascarids, no evident action
against hookworms, and probably was instrumental in the removal of a
tapeworm which was presumably present in one of the dogs, in view of
the discovery of segments in the feces.
iodoform
For worms in dogs. — Among the remedies prescribed for ascarids in
human subjects iodoform has been occasionally recommended. Schid-
lowski {quoted by Seifert, 1885, p. g8) gives it in the form of a powder
mixed with sodium bicarbonate in doses of o.oi to 0.06 gm. three times a
day, followed by a dose of castor oil on the last day. In the present ex-
periment it was intended to give the maximum amount, 0.18 gm., in one
dose to each of four dogs ranging in weight from 11. 8 to 14 kgm. How-
ever, through an error in weighing the drug, the dogs were each given
0.018 gm. instead of 0.18 gm. This error was discovered afterward, and
the next day the full dose, 0.18 gm., was given. Thus, each dog received
a total of about 0.2 gm. of iodoform, given in capsule with sodium bicar-
bonate. The dogs were starved for 24 hours before treatment , and allowed
one meal between the first and second doses. After each dose 29.6 mils
of castor oil were administered. One Taenia segment was passed by one
4o6 Journal of Agricultural Research voi.xii, N0.7
of the dogs, no worms being recovered. As the treatment was obviously-
inefficacious, only two of the dogs were killed. The post-mortem exami-
nation showed these animals infested with various numbers of intestinal
nematodes and tapeworms. As far as can be judged from a single experi-
ment, iodoform is valueless as an anthelmintic for intestinal parasites
in dogs, even when given in doses in excess of that prescribed for human
subjects.
COPPER SUI.PHATE
For worms in sheep. — The use of copper sulphate as an anthelmintic
against stomach worms in lambs was developed by Hutcheon {1892, 1895),
who reported thousands of cases of its successful use in South Africa.
His favorable reports were based largely on clinical findings, but in a
number of cases he treated animals with the copper-sulphate solution
and killed them a short time afterward to determine whether the worms
in the stomach were dead or alive. The solution he used is approxi-
mately that which would be obtained from 0.45 kgm. of pure copper
sulphate, powdered fine and dissolved in 35.96 liters of warm water.
Only clear blue crystals are used, and it is best to powder these and
then to dissolve the powder in a small quantity of hot water and to add
cold water to make up the required amount. He gave the solution in
the following doses.
Lambs 3 months old ^ oimce (22. i7mils).
Lambs 6 months old iK ounces (44. 56 mils).
Sheep 12 months old 2K ounces (73. 9 mils).
Sheep 18 months old 3 oimces (88. 7 mils).
Sheep 24 months old 3K ounces (103. 5 mils).
Stiles and others have tested and recommended copper sulphate in
these doses.
It did not appear to the writers that such refinement in dosing was
called for in treating sheep; therefore, for the sake of simplicity a i per
cent solution was made up and administered in amounts of 100 mils to
sheep a year old or older, and in amounts of 50 mils to lambs under a
year old.
Five sheep less than a year old were dosed with copper sulphate, two
receiving 0.5 gm. each of powdered copper sulphate in capsule, and three
receiving 50 mils each of the i per cent solution.
The two sheep receiving powdered copper sulphate in a capsule passed
a few stomach worms, and on post-mortem showed over 6,000 in one
case and over 4,000 in the other. No nodular worms were recovered
from the feces, but the post-mortem examination showed over 100 in one
case and nearly 200 in the other. Both showed a few tapeworms
(Montezia sp.) and hookworms post-mortem, none having been recovered
from the feces after treatment. Several other species of nematodes were
also found in varying numbers at the post-mortem examination.
The three sheep which received the copper-sulphate solution passed,
respectively, 120, 240, and 314 stomach worms, and showed on post-
Feb. i8, 1918
Efficacy of Some Anthelmintics
407
mortem o, 49, and 3, respectively. There was no marked effect upon
the nodular worms, although one of the sheep passed 2 worms of this
species, 175 were found post-mortem, and 143 and 21, respectively, were
found in the two others, from whose feces none were recovered after
treatment. Two of the
sheep showed hook-
worms on post-mortem
examination, and, al-
though none were found
in the other one, it is
quite probable that none
were present when the
animal was treated, as
none were reco ve red
from the feces after
treatment.
Evidently the pow-
dered copper sulphate
in capsule exhibited no
anthelmintic action. On
the other hand, sup-
porting the experience
of Hutcheon, Stiles, and
others, a i per cent
copper - sulphate solu-
tion in 50 -mil doses
proved very efficacious
in the removal of stom-
ach worms. It had no
evident effect upon
other intestinal para-
sites.
In view of the efficacy
of the copper-sulphate
solution, a test was
made to determine how
readily and easily it
could be administered,
with a view to deter-
mining whether large
numbers of animals
could readily be treated with the solution. A dosing apparatus devised
by the senior writer was constructed and used as follows :
A I per cent solution of copper sulphate was made up and placed in
a small water-tight keg (fig. i). At the side of the keg, near the bottom,
Fio.
I. — Apparatus, with control, for administering copper-
sulphate solution to sheep.
4o8 Journal of Agricultural Research voi. xii, no. ?
a perforated cork with a glass tube through the perforation, was inserted
in an auger hole. A rubber tube was connected with the glass tube.
The keg was placed on a high shelf in the barn and the rubber tube from
the bottom of the keg connected with one of two glass tubes that per-
forated a rubber cork in the bottom of a graduated glass cylinder, the
top of the cylinder being on a level with the bottom of the keg. The
graduated glass cylinder was fastened by wires to two nails driven into
the wall in such a way as to maintain it in a vertical position.
A second rubber tube was connected with the second glass tube in
this cork, this tube terminating at the other end in a piece of metal
tubing. The glass cylinder was graduated at 50-mil intervals and had a
capacity of 150 mils. Close to this cylinder, pinch controls were fastened
on the rubber tubing leading into and out of the cylinder. By pinching
the control on the rubber tube connecting the keg with the glass cylinder,
the copper-sulphate solution was allowed to flow from the keg to any
desired graduation in the lower cyHnder. By letting this close and
pinching the other control, the solution was allowed to flow from the
cylinder to the metal tubing at the other end of the dosing tube, the size
of the dose delivered being noted on the cylinder. The metal tubing
was held in the sheep's mouth by one man, while another man controlled
the size of the dose.
In actual practice, with two men operating this apparatus, as noted, and
with a third holding the sheep and a fourth bringing them up, 25 sheep
were given 50-mil doses in 15 minutes, and 27 sheep were given loo-mil
doses in 25 minutes, a total of 40 minutes for 52 sheep. In the 50-mil dose,
which is perhaps enough for sheep of any age, since it combines safety
with efficacy, sheep can be dosed at the rate of i ^ a minute. This is 100
an hour, or 800 sheep for an eight-hour day.
It should be noted that of the sheep that received lOO-mil doses of the
solution, the equivalent of i gm. of copper sulphate, 2 sheep died in the
course of the next two days. These sheep were 10 months old, and these
d©ses would appear to be too large. Neither of the dead sheep showed
any indications of traumatic pneumonia, but the fourth stomach of one
of them was much congested.
On a subsequent occasion the 50 sheep surviving from this experiment
were dosed with 50-mil doses, using the apparatus described. No bad
results of any sort were experienced. The sheep have been similarly
treated subsequently, and occasional post-mortem examinations of the
sheep in this lot show almost complete freedom from stomach worms.
COPPER SUIvPHATE
For worms in dogs. — In view of the efficacy of copper sulphate against
stomach worms in sheep, an attempt was made to determine whether the
well-known emetic action of this drug would entirely prevent its use as
an anthelmintic for dogs. Four dogs were each given 0.5 gm. of copper
Feb. i8. 1918 Efficacy of Some Anthelmintics 409
sulphate dissolved in 10, 20, 30, and 40 mils, respectively, of water, and 2K
hours later they were given 14.79 to 29.57 mils of castor oil each. There
was prompt emesis in a few minutes after the administration of the cop-
per sulphate. Fecal examinations for the following four days showed only
I worm, a whipworm. Because of the evident defects of the treatment,
the dogs were not killed and examined. It is obvious that the emetic
action of copper sulphate precludes the use of this substance as an an-
thelmintic for dogs.
GASOUNE
For worms in sheep. — Gasoline has been extensively used as a treat-
ment for stomach worms, but some authorities consider that there are
dangers attending its use. Stiles (1901) says of it:
Gasoline has recently gained considerable reputation as a vermifuge. I have used
it in a number of cases and have found the claims made for it to be more or less justi-
fied. Three objections, however, arise to its use, and I can not, therefore, consider
it an ideal treatment. These objections are:
(i) Not less than three doses, and usually four to six, are required to expel the
worms.. Its use involves a great expenditure of labor, and it is, therefore, imprac-
ticable on the large ranches.
(2) While several doses are not necessarily injurious to the stock, still, if the doses
are large, repeated drenches cause a more or less severe congestion of the bowels.
Not only that, but repeated handling of range sheep, with the necessary preliminary
treatment of withholding food, is injurious to the animals.
(3) If used on animals suffering from plevirisy, it is likely to be fatal. I have had
several fatal cases of this kind.
Luckey {191 5) says:
Gasoline for acute cases is a specific. One dose is enough. * * * The average
man can not give gasoline without killing the animal. One can not give a sheep
with a little bit of pneumonia gasoline without killing it.
Arbuckle {1916) says of himself and Joe Wing:
We were the first men in the country to employ the gasoline treatment success-
fully. Wing had heard of it as a remedy used in France. We were also among the
first to discover that this was not a practicable treatment.
Gasoline is usually given in milk, linseed oil, or flaxseed tea, which
makes the treatment considerably more expensive, and, as these feeds
are not always available in large quantity, the treatment is not well
adapted to the needs of large flocks.
Stiles {1901), Ransom {1907), and others have recommended }{ ounce
(7.39 mils) of gasoline as the dose for lambs and }4 ounce (14.79 mils) as
the dose for sheep. Coffey {191 5) recommends larger doses; lambs at
weaning to get X ounce, X ounce, and ^ ounce, respectively, on each of
three successive days; sheep to receive >^ ounce, K ounce, and i ounce,
respectively, on each of three successive days, these doses to be given
thoroughly emulsified in 5 ounces (147.87 mils) of milk linseed oil, or flax-
seed tea.
4IO Journal of Agricultural Research voi. xii.no. 7
In the writers' first experiment four lambs were treated. Two of
them weighing 21.09 9-°^ 21.77 kgm., respectively, were given 7.39 mils
of gasoline in 148 mils of milk at each dose, the equivalent of the dose
commonly recommended, and the others, weighing 26.3 and 22.2 kgm.,
respectively, were given 14.8 mils of gasoline in 148 mils of milk at
each dose, the equivalent of Cofifey's doses of X» Ki ^^^ H ounce for
lambs. The dose named was given to the lambs on each of three suc-
cessive days.
The two lambs which received 7.39 mils of gasoline passed about
one-fourth of the total number of stomach worms (Haemonchus contortus)
present, no hookworms {Bunostomum irigonocephalum) , of which only a
very few were present, and several nodular worms (Oesophagostomum
columbianum) , of which a considerable number were found post-mortem.
There was no evident effect on various other species of nematodes and
tapeworms present in one or both of the lambs. The two lambs which
received 14.8-mil doses of gasoline were more lightly infested; one which
passed 58 H. contortus showed none on post-mortem ; the other passed 4
and showed 2 post-mortem. There was very little effect on hookworms
present in small numbers in one of the lambs.
All four of the lambs showed lesions of pneumonia which were sus-
pected to be of traumatic origin.
In the second experiment three sheep were treated. The larger dose of
gasoline, recommended by Coffey and found most efficacious by the
writers, was employed and was given on each of three successive days
by a stomach tube passed down the esophagus, but not far enough to
direct the fluid into the rumen. In this way the writers expected to
avoid the possibility of causing traumatic pneumonia. None of the
sheep passed any stomach worms, and none were found post-mortem,
so the experiment throws no light on this subject. The treatment
showed a slight efficacy for hookworms {Bunostomum trigonocephalum) .
It failed to remove any nodular worms {Oesophagostomum columbianum)
from one sheep, but removed all specimens of this species from another
lightly infested sheep. It apparently removed all specimens of Cooperia
sp. from the only animal infested with this parasite.
From this experiment it would appear that gasoline has some sUght
effect on intestinal worms in lightly-infested sheep, but the evidence is
not sufficient to justify its use for this purpose without further experi-
mentation. No lesions of pneumonia were observed in this experiment,
although the stomachs of all three sheep showed lesions suggesting healing
ulcers with traces of hemorrhage.
As the question of the efficacy of gasoline as a vermifuge for stomach
worms in sheep and the likelihood of this treatment's causing lesions of
pneumonia and gastritis did not seem to be settled, a further trial of
gasoline was carried out by the junior writer. In this experiment two
Feb. i8. 1918 Efficacy of Some Anthelmintics 411
sheep were used and treated by drenching with increasing doses of gaso-
line in milk as recommended by Coffey {19 13). The dosage given was
15, 22, and 30 mils in 236.6 mils of milk on the first, second, and third
days, respectively. The same dose was given to each animal.
The first sheep passed no stomach worms and had none on post-
mortem examination, so the efficacy of the treatment in this case is
undetermined. A few hookworms only were passed. The second
sheep, which was lightly infested with stomach worms, passed none, but
passed a few specimens of hookworms and nodular worms, forming a
small percentage of the total number present.
In this experiment, although the sheep received much larger doses
than in the previous experiments and the medicine was given in a drench,
there were no evidences of pneumonia or gastritis.
Considering the three experiments with gasoline as a whole, involving
the use of nine experiment animals, the writers find that this treatment
removed over one-fourth of the stomach worms present and had some
slight efficacy for hookworms and nodular worms. This compares very
unfavorably with the efficacy of copper sulphate for stomach worms
when given in solution. The gasoline treatment has also the further
disadvantages that it must be given three times, and in a vehicle such
as linseed tea or milk, which is an additional expense. There is also the
possibility of causing traumatic pneumonia, although the subsequent
experiments with this drug indicate that gasoline is not necessarily
more dangerous as a drench than is copper sulphate.
PETROLEUM BENZIN
For worms in sheep. — As already stated. Stiles (1901) was favorably '
impressed with the use of gasoline as a vermifuge in spite of certain objec-
tions that he notes. It was suggested by Dr. B. H. Ransom, Chief of the
Zoological Division, Bureau of Animal Industry, that our failure to get
more satisfactory results might be due to the difference in the commercial
gasoline of the present day and that used by Stiles. It was recalled that
in 1 90 1, when Stiles conducted his experiments, the automobile industry in
this country was in its infancy, and there was little demand for gasoline.
Consequently the petroleum distillers included only the most volatile
hydrocarbons in gasoline, reserving the heavier fluids for kerosene, their
principal product. At the present day, with conditions reversed, the
distillation temperatures of gasoline have been greatly extended, with a
consequent increase in the specific gravity and lessening of volatility.
In order to determine whether the lessened volatility of the present-
day gasoline was related to its inefficacy as a vermifuge, a test was made
by the junior writer of petroleum benzin, U. S. P. This product, dis-
tilled between 45° and 60° C, represents only the most volatile hydro-
412 Journal of Agricultural Research voi. xii. no. 7
carbons of petroleum and is probably more like the commercial gasoline
used by Stiles {1901) than the present-day commercial product.^
Two full-grown sheep were used for the experiment. The animals were
starved for 24 hours and then given 15 mils of petroleum benzin in 150
mils of milk, on three consecutive days. They were drenched through a
tube held in the mouth. In no case were any symptons of intoxication
from the fumes of the benzin observed during or after its administration,
and the post-mortem showed no lesions which could be attributed to the
action of the drug.
The chief points of interest in this experiment are :
(i) The apparent superiority of refined gasoline (petroleum benzin)
over commercial gasoline as an anthelmintic for stomach worms {Hae-
monchus contortus) and hookworms {Bunostomum trigonocephalum).
It proved 88 per cent efficacious for stomach worms while the latter was
only 35 per cent efficacious, as shown by the summary of three experi-
ments. Its efficacy against hookworms is still more marked. It was
73 per cent efficacious for this parasite, while gasoline proved only 5 per
cent efficacious.
(2) Its entire inefficacy against nodular worms (Oesophagostomum
columhianum) . No worms of this species were removed by petroleum
benzin, while the three experiments with gasoline showed an efficacy of
16 per cent for this parasite.
The post-mortem examination of the two sheep showed that they were
both very lightly infested with stomach worms {Haemonchus contortus),
most of which had been removed by the treatment. The remedy re-
moved all hookworms (Bunostomum trigonocephalum) from one sheep
and more than half the number present from the second sheep. It was
entirely inefficacious against other intestinal nematodes and tapeworms
(Moniezia spp.). Its failure to destroy any of the nodular worms
(Oesophagostomum columhianum) in the large intestine may be explained
on the grounds that it is perhaps more readily absorbed than commercial
gasoline, and, hence, was largely absorbed before reaching the colon.
It should be pointed out that in this experiment the animals were starved
for 24 hours before treatment, while in the experiments with gasoline they
were not starved. This may have had something to do with the results
notwithstanding the slowness with which the rumen becomes empty when
sheep are starved.
In summary it may be stated that petroleum benzin showed marked
superiority over commercial gasoline as an anthelmintic both for stomach
worms (H. contortus) and hookworms (B. trigonocephalum). While it
did not quite equal the treatment with the copper-sulphate drench as a
1 According to the issue of the Journal of the American Medical Association for October 14, 1916, the
present commercial gasolineincludeshydrocarbonsdistillingatatemperatureashiKh as 175° C. (Tydeman,
F. W. I,. NAPHTHALENE IN GASOLINE FOR AUTOMOBILES. In Jour. Amer. Med. Assoc, v. 67, no. 16,
p. II75-)
Feb. i8, i9i8 Efficacy of Some Anthelmintics 413
remedy for stomach worms, it was far superior to anything else tried as a
remedy for hookworms {B. irigonocephalum) and is worthy of further
experimentation to test its efficacy on more heavily infested animals.
For the treatment of stomach worms copper sulphate must still be con-
sidered superior, not only because it is somewhat more efficacious but on
account of its lesser cost and the fact that only one dose of copper sulphate
is required, whereas three doses of petroleum benzin are advisable.
PHENOLS
The phenols a re a group of organic compounds composed of hydroxy
derivatives of the benzene series, the hydroxyl radical being linked
directly to the nucleus. The refined phenols include phenol (CgHgOH),
commonly called carbolic acid; cresol (CgH^CHgOH), commonly called
cresylic acid or kresol; and the higher phenols. The term "crude
phenols" is used in general to designate those unrefined mixtures of the
phenols proper with certain hydrocarbon oils and other impurities with
which they become associated in the course of their preparation, whether
from coal-tar, wood-tar, or blast-furnace gases. There are on the market
numbers of trade preparations in the form of soaps, powders, ointments,
or liquids which contain refined or crude phenols as essential constituents.
A number of these liquid phenol preparations have been used and recom-
mended as anthelmintics. Some of these liquids are insoluble in water,
in which case they may or may not be capable of being emulsified, and
others are soluble in water.
For worms in sheep. — Three preparations were tested, which may
be referred to as A, B, and C. A and C are advertised as remedies for
stomach worms, while B, though not advertised on the container for the
treatment of stomach worms {Haemonchus contorius), has since been
recommended by the New Zealand Department of Agriculture for that
purpose. Two sheep were treated with A and one sheep each with B
and C. One of the sheep treated with A passed four nodular worms
(Pesophagostomum columhianum) . This sheep and the one treated with
B died the day after treatment. The other sheep passed no worms.
Of the two sheep receiving A, the one which died had 422 stomach
worms (Haemonchntortus) us coand numerous intestinal nematodes. The
other sheep killed four days after drenching had no stomach worms post-
mortem and few intestinal nematodes. As this sheep was evidently unin-
fested with stomach w;orms (H. contorius), it must be left out of consider-
ation. The sheep treated with C passed no worms and had 8 stomach
worms (//. contortus) post-mortem.
A second trial was made of A on two sheep, the dose given being the
same as in the previous experiment — i tablespoonful of the product (14.8
mils) to I pint (473 mils) of milk, as advertised on the label. The sheep
passed no worms, and at post-mortem examination were found lightly
infested with stomach worms and intestinal nematodes.
27810°— 18 2
414 Journal of Agricultural Research voi.xii.no. 7
While the experiment is inconclusive in regard to the efficacy of coal-
tar phenols for stomach worms in sheep, since the experiment animals
were so lightly infested, it failed completely as a remedy for lightly
infested animals, and there seems to be no reason for considering that it
would be more successful in heavily infested sheep.
There can be little doubt that the treatment is dangerous. Two out
of four sheep in the first experiment died, the post-mortem showing
lesions of pneumonia, pleurisy, and gastritis. In the second experiment
both sheep collapsed after drenching, but seemed to recover. The pneu-
monic condition observed in the dead sheep may be attributed to getting
some of the fluid in the lungs, an error in drenching, to be sure, but one
almost impossible to avoid when giving as much as 473 mils of drench.
In the writers' experience sheep will usually take quietly 118 to 177
mils of fluid by drench; but, after more is given, they begin to struggle,
making the drenching increasingly difficult. Furthermore, the hemor-
rhagic lesions in the stomach of one of the dead sheep would seem to indi-
cate the absorption of phenols through the gastric mucosa.
For worms in dogs. — A further test of the anthelmintic efficacy of
phenols was made, using another preparation, which is recommended as
an anthelmintic for worms in dogs, the dose recommended being 5 to 8
drops (0.3 to 0.48 mil) in a tablespoonful (14.79 mils) of castor oil. The
writers used 0.48-mil doses, administering this dose to each of two dogs,
weighing, respectively, 8.6 and 11.3 kgm. On the fourth day after
treatment one dog passed two ascarids and the other passed a headless
chain of segments of Taenia hydatigena. This latter dog showed
an infestation with ascarids {Belascaris marginata), hookworms
{Ancylostoma caninum), and tapeworms {Taenia sp.), and, as the
treatment had been unsuccessful except for the partial removal of a
tapeworm, the animal was not killed. The other dog was killed
and found to have three B. marginata and three A. caninum.
The treatment, therefore, was somewhat efficacious against ascarids
in one case, but entirely inefficacious in the other. It was effective
against A. caninum in both dogs. It also appears to be ineffective
against Taenia, since it did not bring away the head, but this is a point
that should have been confirmed post-mortem.
We may conclude from the experiments on dogs and sheep, above
recorded, that the phenols in the form of commercial disinfectants and
dips are likely to be of little value and dangerous as anthelmintics.
EXPERIMENTS WITH VEGETABLE ANTHELMINTICS
OLEORESIN ASPIDII
For worms in dogs. — Oleoresin aspidii is the classic remedy for
use against tapeworm. In the discussion of chloroform as an
anthelmintic for hookworms (page 403) the writers have already
Feb. i8, 1918 Efficacy of Some Anthelmintics 415
shown that male-fern in a combination known as Hermann's mix-
ture is efficacious against Dipylidium caninum and may have
some slight value against whipworms (Trichuris depressiuscula)
and hookworms (Ancylostoma caninum), although the efficacy of
the mixture for hookworms is largely due to the chloroform contained
in it. The following experiments in which male-fern alone was used
corroborate these conclusions.
In the first experimental test of the drug five dogs were used ranging
in weight from 6.4 to 15.9 kilos. The treatment was preceded by
calomel (65 to 194 mgm., according to the weight of the animals) the
afternoon of the preceding day and followed about 45 minutes after
treatment by Epsom salt in molasses. One to three mils of male-fern
were administered. Within an hour after treatment one dog had passed
a mass of fragments of Dipylidium caninum,, including at least four
heads, and another some chains of segments of Taenia sp. When the
dogs were killed, no specimens of Taenia sp. were found, and only two
specimens of D. caninum, one of which was in the colon, evidently
about to pass out. It is evident that the remedy was entirely efficacious
against Taenia sp., since the dog which was observed to pass Taenia seg-
ments after dosing with male-fern was found uninfested on post-mortem
examination. It was slightly less efficacious for D. caninum, but appar-
ently removed all but one specimen. It is possible that the male-fern
removed all individuals of D. caninum not attached by burrowing into
the intestinal mucosa, leaving embedded heads to renew strobila. This
would account for the failure to find more than four heads in the rela-
tively large mass of segments passed.
Male-fern had but little effect on ascarids (Belascaris marginata) and
removed only one-fourMi of the hookworms {Ancylostoma caninum)
present, in spite of the fact that at one time it was regarded as an appro-
priate remedy for hookworms {A . duodenale) in man and much used for
the purpose. It should be stated, however, that, in view of the small
number of hookworms involved in this experiment, only four being
present, the conclusion that male-fern is inefficacious against hookworms
is hardly warranted. None of the three whipworms (Trichuris depressi-
uscula) present were removed by this drug, although Miller (1904)
reported the successful removal of whipworms from dogs with it.
A further test of the efficacy of oleoresin of aspidium against tapeworms
in dogs was conducted by the junior writer. After the usual 24-hour
fast, two dogs weighing 20.4 and 11.34 kgm. were given 2.7 and 1.8
mils, respectively, of male-fern, followed by 162 mgm. of calomel.
Prior to treatment the smaller dog had been repeatedly seen to pass
chains of proglottides identified as Taenia pisiformis, while only a few
Taenia eggs had been seen in the feces of the larger dog. The presence
of these eggs may have resulted from contamination of the specimen.
41 6 Journal of Agricultural Research voi.xii. No. 7
Following the anthelmintic the smaller dog passed a few chains of
proglottides the first day after treatment, and on the second day 17
fragments and 4 heads. No tapeworms were passed by the larger dog
during the four days following treatment, and no nematodes were passed
by either dog. The feces of the larger dog were examined for Taenia
eggs four days after administering the vermifuge, and were found nega-
tive. As this dog had passed no Taenia, and as the original presence
of tapeworms was somewhat doubtful, it was dropped from the experi-
ment. The post-mortem examination of the smaller dog showed no
Taenia, 15 hookworms, and 13 whipworms. The remedy was therefore
entirely efficacious for species of Taenia and entirely inefficacious for
hookworms (Ancylostoma caninum) and whipworms {Trichuris depres-
siuscula), thus confirming the opinion derived from the previous experi-
ment, that male-fern is very efficacious for tapeworms in dogs and
inefficacious for nematodes.
For worms in cats. — Six cats which had been fed with Cysticercus
fasciolaris fonned the subject of this experiment. Each cat was given
0.8 mil of oleoresin of aspidium, followed by 130 mgm. of calomel for the
four largest cats and 97 mgm. of calomel in the case of the two smaller
cats. Within half an hour one of the cats, which was weak from con-
finement in a cage and which was suffering from coryza, had died. The
post-mortem examination revealed an intense congestion of the gastric
mucosa. The characteristic odor of male-fern was noticeable in the
stomach, but not in the intestines. A half -grown individual of Taenia
taeniaeformis and several ascarids (Belascaris cati) were found in the
small intestine, while three dead B. cati were in the colon. Another cat,
which was weak from a previous infestation with mange, passed several
chains of T. taeniaeformis shortly after the administration of the male-
fern and died three days later. A third cat vomited one T. taeniaeformis
and one B. cati and still had one B. cati post-mortem. Two of the six
cats were uninfested with Taenia spp., since they passed no tapeworms
and none were found post-mortem. Of the four infested animals two
died from the treatment, but three of the four were entirely freed from
their Taenia infestation, the fourth dying before the remedy had an
opportunity to reach the parasite.
It would appear from the above that male-fern is efficacious in removing
tapeworms from cats, but that it is apparently more toxic in the case of
cats than with dogs, and should only be prescribed for healthy animals.
Probably some other taeniacide could be used which might prove as
effective and less dangerous. It should be noted that the dose of male-
fern given was less than the minimum dose 15 minims (0.9 mil), recom-
mended by Winslow {igis). With the exception of the one ascarid,
which was vomited, the treatment had no effect on the few ascarids {Be-
lascaris cati) and hookworms {Ancylostoma caninum) which were present.
Feb. i8, 1918 Efficacy of Some Anthelmintics 417
PELLETIERINE TANNATE
For worms in cats. — Pelletierine tannate has long been recognized
as an effective taeniacide. According to the U. S. Dispensatory — ^
The efficacy of pelletierine as a taeniacide has been abundantly confirmed, and
it appears to be established that the tannate is the most effective and the least dan-
gerous form of the remedy, — probably because its insolubility prevents its rapid ab-
sorption and enables it to come in prolonged contact with the worm.
A test of this drug was made by the junior writer on two cats weighing
approximately 3 and 4 kilos each. Sixty-five mgm. of pelletierine tannate
per kilo were administered in capsules to each animal and were followed
one hour later by 25 mils of castor oil. Prior to the administration of
the anthelmintic the cats were starved for 24 hours.
The animals were fed shortly after the administration of the castor oil.
All feces passed during the four days following the administration of
the drug were negative for parasites. Cat 2 vomited immediately after
feeding, or a little more than an hour after taking the pelletierine. On
post-mortem examination cat i had 29 Belascaris cati and 4 Dipylidiuw,
caninum; cat 2 had i Taenia taeniae formis.
In this experiment pelletierine tannate proved unsuccessful as an
anthelmintic for species of Taenia, Dipylidium, or ascarids in cats. The
complete failure of the drug is not easy to understand. In the case of
cat 2 the drug might have been expelled in the vomitus before it had
acted on the Taenia, but this is unlikely, since the vomiting did not occur
until more than an hour after the ingestion of the pelletierine.
As the pelletierine tannate used had been in the laboratory for at least
six years it was submitted for examination to the Biochemic Division of
the Bureau of Animal Industry. The report was returned that the sam-
ple responded to all the tests for pelletierine taimate given in the Pharma-
copeia. The Pharmacopeia specifies that the drug shall be kept in small
well-stoppered bottles away from the light, a condition complied with in
this case. There seems to be no reason to suppose that the pelletierine
tannate that was used had undergone deterioration, and its failure to
give results remains unexplained.
For worms in dogs. — As pelletierine tannate had shown a surprising
inefficacy in regard to tapeworms in cats, a further test of this drug was
made by the junior writer, using dogs as experiment animals.
The drug was aarainistered to three dogs weighing from 11.8 to 14
kilos, at the rate of 16 mgm. per kilo and was followed one hour later by
castor oil. Previous to the treatment 35 mils of castor oil were given
to each dog as a preliminary purgative, and the animals were then fasted
for 24 hours.
' Wood, G. B., and Bache, Franklin, thb disebnsatory op thb united states op America, ed. 19.
p. 6°o. Philadelphia and London, 1907.
41 8 Journal of Agricultural Research voi.xii, no. 7
Following the administration of the preliminary purgative, two dogs
each passed two Taenia proglottides. Following the dose of pelletierine
tannate and castor oil, one dog passed five Taenia proglottides, and an-
other dog passed two proglottides. No tapeworm heads or nematodes
were found in the feces. The post-mortem examination revealed a few
nematodes and Dipylidium caninum, but no Taenia sp.
It is evident that all three dogs were infested with species of Taenia,
since proglottides were recovered in the feces of all animals. The fact
that all dogs were free from Taenia on post-mortem examination is indi-
cative of the efficacy of the drug as a taeniacide. The failure to recover
the worms in the feces probably resulted from neglect on the part of the
attendant collecting the feces.
Pelletierine tannate apparently proved entirely inefficacious against
Dipylidium caninum or intestinal nematodes in dogs, since none of
these parasites was found in the feces, but some were present post-
mortem. As there is a strong probability that all feces were not col-
lected, the writers can not be certain with regard to the total inefficacy
of the drug for these parasites.
The sample of pelletierine tannate used in this experiment was pur-
chased from the same manufacturer and at the same time as the sample
used in the previous experiment on cats. It may therefore be assumed
to have the same potency as the sample used in the previous experi-
ment, which was tested for purity by the Biochemic Division.
ARECA NUT
For worms in dogs. — Areca nut (Areca catechu) is not infrequently
prescribed as an anthelmintic for ascarids in dogs. Railliet {1915) gives
the dose of areca nut as 2 to 4 gm. combined with 10 to 20 gm. of soluble
cream of tartar,^ the latter presumably being used merely as a vehicle.
In this experiment 6 gm. of areca nut and 36 gm. of soluble cream of tartar
were formed into 12 pills, 4 of which were given to each of three puppies
weighing between 1.8 and 2.3 kgm. Previous to the experiment the dogs
were each given 1 4.79 mils of castor oil, which failed to remove any worms,
and were then starved for 24 hours. All the dogs were infested with
ascarids (Belascaris marginata) and hookworms {Ancylo stoma caninum),
as shown by previous fecal examinations. The day following the ad-
ministration of the vermifuge one of the dogs passed 4 B. marginata
and was found dead the following morning. The other dogs passed no
worms. The post-mortem of the dog that died showed hemorrhagic
areas in the colon and feces stained with blood. The other organs were
• Soluble cream of tartar, a preparation seldom used by American veterinarians, is boro-tartrate of potas-
sium, made by boiling together four parts of cream of tartar with one part of boric acid in a large amoimt of
water. When most of the water has evaporated, the process of evaporation is completed in a drying oven
and the resultant salt is pulverized and stored in well-stoppered bottles. Soluble cream of tartar is very-
deliquescent, and hence well adapted for making a piUmass.
Feb. 18. 1918 Efficacy of Some Anthelmintics 419
entirely normal in appearance. This puppy had 23 ascarids {B. mar-
ginata) and a few hookworms (^4. caninum) and whipworms (Trichuris
depressttiscula). The other dogs had varying numbers of ascarids,
hookworms, and whipworms. The treatment removed only 4 out of 67
ascarids and had no effect on the hookworms and whipworms present.
It should be stated in comment that the areca nut used was ground 14
months previous to the experiment and hence may have lost some of its
potency. This leaves open the question as to the efficacy of freshly
ground areca nut, but it appears from the experiment that areca nut
ground one year previously is not a very efficacious anthelmintic for
dogs, and may possibly cause serious digestive disturbances. While it is
not possible to determine whether the colitis noted in the dead dog
resulted from the drug, it should be stated that this puppy had been
weak and emaciated for some time previously and was in poor con-
dition to undergo anthelmintic treatment.
For worms in poultry. — Areca nut is quite commonly prescribed
as a remedy for tapeworms in poultry. The following experiment was
made to test its efficacy against tapeworms and incidental nematode
parasites of the intestinal tract of chickens :
One gm. of powdered areca nut was given to each of six chickens,
weighing about 453 gm. each, the dose being given in 4 mils of olive
oil. Prior to the experiment the birds were fasted for 24 hours.
The total number of worms passed following the vermifuge were 20
Heterakis papulosa, which were passed by three of the six birds on the
first day, and some tapeworm segments, including at least one head
with these. On post-mortem examination large numbers of tapeworms
and H. papulosa were found, besides other nematodes, which, from
their location in the esophagus and proventriculus, would not be likely
to be affected by anthelmintics. Probably more tapeworms were
passed than were counted, since, when a small tapeworm head is
passed and then given an opportunity to dry, it is very difficult to
detect even by a careful examination of the feces. But at best the
drug seems to have had very little efficacy against H. papulosa or tape-
worms.
The areca nut used in this experiment was ground at least four years
previously. This leaves open the question as to the efficacy of the
freshly ground product. If it is true that the drug loses its potency
after grinding, this constitutes a serious objection to its use in commercial
products, since it is unlikely that it will be freshly ground when purchased.
santonin
Santonin is the classic remedy for ascarids and forms the basis of
most of the worm remedies for children. It has also been widely used
and recommended in canine practice. As this drug is largely of Euro-
420 Journal of Agricultural Research voi.xii.no. 7
pean or Asiatic origin, the present price of santonin is almost prohibitive
in veterinary medicine, and its place is being taken by the native and, in
the writers' experience, more effective drug chenopodium. Santonin is
usually combined with or followed by a purgative to promote elimina-
tion of the parasite. For this purpose calomel or areca nut is much
used.
Santonin and calomel for worms in dogs. — The dose of santonin
for dogs recommended by Winslow {19 13) is from i to 3 grains (65 to
195 mgm.). Taking 130 mgm. as the dose for an average dog weighing
iokgm.,the writers gave seven dogs, ranging in weight from 3.8 to 9.5
kgm., doses graded from 32 to 130 mgm., accompanied by the same
amount of calomel. The treatment was preceded by the administration
of castor oil, 29.57 mils to dogs weighing 4.5 to 9.5 kgm., and 7.39 mils
to pups under 4.5 kgm. Food was withheld the previous day. Seven
ascarids (Belascaris marginata) were passed following the administration
of castor oil and 13 B. marginata and two whipworms (Trichuris depres-
siuscula) after santonin and calomel. At the post-mortem examination
it was found that about one-fourth of the ascarids remaining after the
action of the castor oil later had been removed by santonin and calomel.
The treatment was very inefficacious for whipworms, removing 2 out of
72, and entirely inefficacious for Dipylidium caninum and Taenia.
It would appear from the foregoing that santonin and calomel, the
remedy usually prescribed for ascarids, is not very efficacious for dogs
in single doses.
As the efficacy of the santonin as shown by the above experiment
was considerably less than had been expected, considering the well-
established reputation that santonin has as an ascaricide, a second experi-
ment was undertaken to determine the efficacy of santonin and calomel
in repeated doses. In this experiment four dogs, weighing from 1.8 to
9 kilos, were given graded doses of equal quantities of santonin and calo-
mel, the dose varying from 32 to 130 mgm., according to weight. The
first dose was given after a preliminary fast, and the second dose was given
two days after the first, food being withheld the evening before the
second treatment also, the same dose being given at each treatment.
The first treatment was preceded by castor oil, which failed to eliminate
any worms. One of the dogs passed no ascarids and had none post-
mortem, so that it was evidently not infested and must be left out of
consideration. From the three remaining dogs it eliminated 7 out of
10 ascarids, a distinct gain in efficacy compared with the previous
experiment, in which a single dose was given. In this case the drug
also showed a fair degree of efficacy against whipworms {Trichuris
depressiuscula), parasites which in the writers' experience are difficult to
remove with any degree of certainty with any of the anthelmintics tested.
The drug was entirely ineffective against hookworms and Dipylidium
caninum.
Feb. i8. 1918 Efficacy of Some Anthelmintics 421
Santonin and calomel for worms in hogs. — Before the present
European war, when the price of santonin justified its use for live
stock, it was much recommended for roundworms (Ascaris suum)
in swine. Since the separate dosing of a herd of swine requires
considerable time and labor and the handling of the animals is likely
to excite them and may lead to their being injured, it i§ usually the
custom to mix the medicine with the feed, allowing a few hogs to eat
from the same trough at one time. If the drug is well mixed with the
feed and the animals are of about the same size, it is assumed that they
will get approximately the same dose. This method was tested by the
junior writer for a number of drugs. In this series of experiments the
worms passed were not counted; nor were the animals killed. The
feces were previously examined for ascarid eggs, and the efficacy of the
drug was judged solely by the presence or absence of eggs in the feces
three days after the administration of the anthelmintic, care being taken
to verify all negative findings by one or more repeated fecal examinations
at 3-day intervals.
In the first experiment, after a preliminary fast of 24 hours, three
hogs weighing 15.4 to 18 kgm. were given santonin and calomel allowing
130 mgm. of each per hog. The drug was shaken up in 473 mils of water,
poured over a bran-mash feed and thoroughly stirred in. The feces
of all the pigs were positive for ascarids {Ascaris suum or suilla) at the
beginning of the experiment. Three days after the treatment the feces
of all the pigs were still positive. As the treatment failed to free any
of the pigs from ascarids {A. suum), it was repeated, giving 194 mgm.
of santonin and calomel instead of 130 mgm. The second treatment was
equally without results. A third trial was made in which the pigs were
dosed individually after preliminary starvation, each pig receiving 194
mgm. each of santonin and calomel. This treatment resulted in the
apparent elimination of all ascarids from one of the pigs, the other two
remaining infested. These hogs were then dosed individually with
259 mgm, each of santonin and calomel, all subsequent fecal examinations
being negative.
It appears from the above that santonin and calomel should be given
in repeated and separate doses to be effective. Probably three doses of
santonin at intervals of three days would be effective. No symptoms
of poisoning were observed either in dogs or pigs, although the pigs were
given a dose more than twice as large as is usually prescribed for animals
of their size. According to the Veterinary College at Ames, lowa,^ not
over 4 grains (259 mgm.) of santonin should be given to the largest hog.
It would seem that the treatment of swine with santonin requiring
apparently three separate doses to be effective, is far too laborious and
costly for use, especially as we have a more reliable and less expensive
drug in oil of chenopodium.
• Treating pigs for worms. In Breeder's Gaz., v. 64, no. 8, p. 315. 1913.
422 Journal of Agricultural Research voi. xii.no. 7
Santonin and ar^ca nut for worms in dogs. — Areca nut is
frequently prescribed with santonin, since it furnishes the necessary
laxative, and having anthelmintic properties of its own is supposed
to be an adjuvant to the santonin. In this experiment four dogs
weighing from 2.7 to 6.8 kgm. were given 33 to 130 mgm. of santonin
combined with 0.78 to 1.94 gm. of areca nut. The dogs were given a
preliminary dose of castor oil which removed 7 ascarids (Belascaris
marginata). Following the administration of the santonin and areca
nut 3 B. marginata were passed. At the post-mortem examination 30
ascarids {B. marginata) were found besides several hookworms (Ancy-
lostoma caninum), whipworms (Trichuris depressiuscula), and i Dipy-
lidium caninum. One of the dogs vomited the santonin, which may
account for its failure to act in this case. In this case santonin proved
less efficacious for ascarids than castor oil, and had no effect on the
other intestinal worms present.
Santonin and areca nut for worms in hogs. — Combinations of
santonin and areca nut are frequently prescribed for pigs, to be mixed
with the feed. The following, copied from the Breeder's Gazette, is
typical :
Santonin 2}4 grains.
Areca nut i dram.
Calomel K grain.
Sodium bicarbonate K dram.
This quantity is considered sufficient for a 100- pound hog. As the pigs
experimented with weighed only 24.5 kgm. each, two-thirds of the amount
prescribed was allowed for each pig and mixed with the feed, after
starving the animals 24 hours. Three days after the experiment the
feces were still positive for Ascaris suum, and a second trial was made
using the full amount prescribed in the formula. This experiment was
also unsuccessful in ridding any of the hogs of ascarids. A third trial
of the formula was made on a pig weighing 24.5 kgm. with the object of
testing the efficacy of repeated doses. Five times the amount pre-
scribed was made up and divided into seven powders, one powder being
mixed with the feed every morning for seven consecutive days. At the
end of the period no diminution in the number of ascarid eggs in the
feces was discernible.
As far as these experiments can be analyzed, it appears that santonin
and areca nut offer no advantages over santonin and calomel, and that
either combination must be repeated and given in individual doses on
an empty stomach to be efficacious. The last experiment illustrates
the inadvisability of giving drugs to hogs, mixed with the feed, since in
this case, a drug already shown to be efficacious in repeated doses was
apparently a complete failure when given in this manner.
Feb. 18. 1918 Efficacy of Some Anthelmintics 423
THYMOL
For worms in dogs. — Prior to the beginning of the great European
war, thymol was the classic remedy for hookworm in man, and it was
only after the war had made this drug practically unobtainable that
serious efforts were made to find a substitute for it. In a general way
it was regarded as a satisfactory drug, and we expected to find a fairly
high coefficient of efficacy for it in our experiments.
In our first experiment thymol was administered at the rate of 130
mgm. per kilo of body weight, giving the thymol in capsules to 2 dogs
and in drench to 2 dogs ; in the latter case the thymol was dissolved in a
small amount of alcohol and this added to water to make a fine suspension.
After an interval of a half hour to an hour, calomel in doses of about 65
mgm. per 2.5 kilos of weight of dog was given. All dogs were fasted from
the previous day.
As regards hookworms, thymol removed 15 out of a total of 114 worms
present, being entirely ineffective in the case of three dogs, two of which
received the thymol in capsule and one of which received it in aqueous
suspension. In one dog, however, it removed 15 out of 16 hookworms.
In regard to ascarids, thymol removed nearly three-fourths of these
worms present, a very satisfactory showing considering that thymol is
not usually considered especially valuable as an ascaricide. It was not
efficacious when administered in aqueous suspension. Thymol also
exhibited some efficacy against whipworms and was very slightly ef-
fective for Dipylidium caninum.
In the above experiment thymol showed a low degree of efficacy
against hookworms, the parasite for which it is usually prescribed, and
was more efficacious against whipworms and still more efficacious against
ascarids. In the experience of the writers a number of drugs were
found to be occasionally efficacious against whipworms, while at other
times they were decidedly inefficacious. This variation may perhaps
be explained as a matter of accident, depending on whether or not the
drug succeeds in penetrating to the cecum in its passage through the
alimentary tract.
An objection to thymol at present is its cost and scarcity.
In order to determine whether the low degree of efficacy displayed
against hookworms by thymol was due to the mode of administration,
the experiment was repeated. This time the animals were subjected to
preliminary purgation with Epsom salt, and the administration of
calomel after treatment was delayed for five to six hours. The thymol
was given in the same dosage, 130 mgm. per kilo of body weight, and was
given in aqueous suspension.
Three dogs weighing 4 to 9.3 kilos were used for the experiment. They
were given Epsom salt at the rate of 3.7 mils for each 5 kilos of weight,
the day prior to the administration of the anthelmintic, and five to six
424 Journal of Agricultural Research voi. xii, no. 7
hours after giving thymol they were given calomel at the rate of 65 mgm.
for each 2.5 kilos of body weight.
One of the dogs died immediately after receiving the thymol, and the
remaining two dogs each passed one ascarid. The post-mortem exam-
ination showed that one or both dogs were infested with hookworms
(Ancylostoma caninum), whipworms (Trichuris depyessiuscula), asca-
rids {Belascaris marginata), and Dipylidium caninum. The treatment,
therefore, proved an entire failure for all worms present except ascarids,
for which it showed a rather low degree of efficacy in this experiment.
As the thymol used had not proved as efficacious against hookworms
as might be expected, a sample was sent to the Biochemic Division of
the Bureau for analysis. An analysis was made by Dr. Custis, who
reported that it responded to all tests for thymol and showed no impuri-
ties. Thymol is apparently a stable phenol; the writers are unaware
of any evidence indicating that it deteriorates.
Since thymol in a single dose proved so ineffective in the foregoing
experiments, an experiment was undertaken to determine its efficacy
in repeated doses. Four dogs weighing 2.3 to 3.9 kilos were given a
preHminary purgation with 20 mils of castor oil on the day previous
to treatment and were then fasted for 24 hours. Thymol was given
at the rate of 130 mgm. per kilo of body weight, and this dose repeated
three and five days later for a total of three doses. After the second and
third doses of thymol, 97 to 130 mgm. of calomel were given. The first
dose of thymol was given dissolved in alcohol to one dog and resulted
in the immediate death of the animal. The other dogs were given the
thymol in aqueous suspension.
The thymol removed 8 out of the 16 hookworms {Ancylostoma
caninum) present in the dogs, a rather poor showing in view of the pre-
liminary fasting, purgation, and repeated doses, but decidedly better
than in the previous experiments. It removed the only ascarid {Belas-
caris marginata) present and showed some slight degree of efficacy against
whipworms {Trichuris depressiuscida) .
It appears from these three experiments that thymol to be at all suc-
cessful as an anthelmintic for hookworms {Ancylostoma caninum) must
be preceded by purgation and fasting and given in repeated doses. At
least two such courses of treatment should be given to remove the
greater number of the worms present, and further treatment combined
with prophylaxis is desirable. As a matter of fact, this is the usual
procedure in administering thymol in human practice, and it is usually
reaUzed by physicians that it is very difficult to remove all the hook-
worms present, as it is observed that hookworm eggs often persist in the
feces after repeated thymol treatments.
In the three experiments considered together, thymol removed over
half thje ascarids present, but showed very little efficacy for whipworms or
Feb. i8, i9i8 Efficacy of Some Anthelmintics 425
Dipylidium caninum. That the administration of thymol to dogs is
not without danger is shown by the death of 2 out of the 1 1 experiment
animals.
The low efficacy and the danger in the use of thymol do not compare well
with the comparatively high efficacy and safety in the use of chloroform,
so far as the experiments with dogs are concerned. While the writers
are impressed with the dangers in connection with the administration of
most anthelmintics, and these dangers are quite impressive for chloro-
form as well as thymol, nevertheless, it seems that, with the exceptions
already noted in the discussion of chloroform for cases where there is heart
trouble, or lesions of the parenchymatous viscera, chloroform is not only
much more effective than thymol, but, in therapeutic doses, is safer.
Turpentine
Turpentine is a remedy very commonly advocated for use against
nematode parasites, especially those in chickens, horses, and swine. The
obvious objection to its use is its well-known injurious effect on the
kidneys.
For worms in poultry. — Since the treatment of chickens for worms
is for obvious reasons so seldom undertaken by veterinarians and so
commonly by owners of poultry who are not especially trained in medi-
cal lines, it is not surprising that the dose advocated for use in this con-
nection is very variable. Some writers recommend a half teaspoonful of
turpentine in an equal amount of olive oil ; others recommend i to 3 tea-
spoonfuls of turpentine undiluted.
An experimental test of the efficacy of the lighter dose of turpentine
against worms in poultry was made as follows: Six chickens weighing
between 0.45 and 0.9 kgm. were given 2 mils of turpentine mixed with
2 mils of olive oil, the birds being fasted from the previous day and the
dose being followed at once with 8 mils of castor oil. About five hours
after treatment all the birds had passed some feces, the feces having an
odor of turpentine.
The treatment appeared to be fairly satisfactory for the large round-
worm (Ascaridia perspicillum) in the small intestine of chickens, since it
removed more than three-fourths of the worms present, as shown by
post-morten examination. It had little effect in cases of infestation with
large numbers of cecum worms (Heterakis papulosa) with which chicks
are frequently infested. According to the experience of the writers, this
worm is difficult to remove with any anthelmintic, since its location
protects it to a greater or less extent from contact with the drug.
Turpentine was equally inefficacious as a remedy for tapeworms in
fowls, removing only 8 out of 444. It should be stated, however, that it
is very difficult to count the tapeworm heads which may be present in the
426 Journal of Agricultural Research voi. xii, No. 7
feces. On account of their minute size they dry quickly, and unless the
feces are examined very soon after they are passed, many specimens will
be unrecognized. The remedy, therefore, may have been more effective
for tapeworms than the figures indicate. Its principal advantage, how-
ever, seems to be as a remedy for roundworms (Ascaridia perspicillum).
For worms in dogs. — Four dogs weighing 2.5 to y.j kilos were given
preliminary purgation with castor oil, and dosed with turpentine at the
rate of i mil per kilo of weight. The turpentine was given mixed with
castor oil, 15 to 30 mils, according to the weight of the animals.
The preliminary purgation with castor oil did not bring away any
worms. One of the dogs died on the fifth day after treatment.
Only one worm, an ascarid, was passed; to this should be added one
ascarid found in the rectum of the dog that died, making two ascarids
removed by the turpentine.
In view of the lack of results from the treatment, the dogs in this
experiment were not killed. Just previous to the experiment, micro-
scopic examination of the feces for eggs had shown that all of the dogs
were infested with either hookworms or ascarids or both. On the last
day of the experiment the feces were again examined for eggs, and eggs
of the same worms were found to be still present.
All the dogs showed symptoms of distress immediately after the
treatment with turpentine, the principal feature being a temporary
paralysis of the hind quarters.
It may be concluded that turpentine in doses of i mil per kilo of
weight is not very efficacious against ascarids in dogs, is entirely ineffi-
cacious against hookworms, gives rise to pronounced suffering and tem-
porary paralysis of the hindquarters, and may kill the dog.
For worms in hogs. — Turpentine is frequently prescribed as a
remedy for roundworms {Ascaris suum) in swine. It is often given in
repeated doses mixed with the feed. A better way is to make an emul-
sion of the turpentine with equal parts of flaxseed decoction. This is
more easily miscible with the feed, and avoids the burning caused by
the ingestion of pure turpentine.
An emulsion of turpentine and flaxseed decoction, made by boiling
85 gm. of flaxseed in 296 mils (10 ounces) of water, straining, and adding
an equal amount of turpentine, was fed to three hogs by the junior writer.
The hogs weighed from 45.36 to 68.04 kgm. The equivalent of 7.4 mils
of turpentine per hog or 44.36 mils of the mixture, was mixed daily in
the feed of the three hogs for seven days. At the end of seven days all
the hogs were listless, refused to eat, and were constantly voiding small
amounts of urine, the nephritic symptoms continuing for a week after
the treatment stopped. All the feces of the hogs contained ascarid
(Ascaris suum) eggs after the experiment was concluded.
Feb. i8, 1918 Efficacy of Some Anthelmintics 427
The treatment was decidedly inefficaceous and highly dangerous.
The animals never gained as rapidly as the other hogs kept under the
same conditions but not dosed with turpentine. While separate dosing
might have shown some efficacy for the treatment, the injury to the
hogs precludes its recommendation as an anthelmintic.
FICUS LAURIFOLIA
For worms in dogs. — The latex of Ficus laurifolia has been highly
recommended by Berrio (1911) and by Mouat-Biggs (1915) for use in
expelling whipworms from man, and is also said to have been adopted by
the Venezuela State Board of Health as the official remedy for use against
hookworms. Berrio gives doses of 25 to 30 gm., followed by castor oil.
Mouat-Biggs gives the latex morning, noon, and night, mixed with water,
which may be sweetened, or with milk, in doses of 10 to 40 gm., accord-
ing to the age of the patient.
A sample of the expressed juice of the latex kindly supplied to the
writers by Dr. Gonzalez-Rincones of Caracas, Venezuela, was tested on
three dogs weighing from 4 to 29 kilos. Fifteen to 30 mils of the
juice were given, preceded by a purge of 30 mils of castor oil, and
followed three or four hours after the administration of the latex by
15 to 30 mils of castor oil. Although all the dogs were infested with
hookworms, whipworms, or ascarids, as shown by a previous fecal
examination, no parasites were passed. Since the remedy was entirely
inefficacious, the dogs were not killed.
The sample forwarded to the writers did not conform to the published
accounts of the latex, which describe it as of thick, sirupy consistency
and milky white in color, but was instead a watery fluid with something
resembling curds floating in it.
A second sample conforming in all respects to the descriptions of the
latex was subsequently received through the courtesy of Dr. Rincones.
This drug was tested by the junior writer on three dogs weighing 4.5
to 18.2 kilos. The latex was given at the rate of 2.6 mils per kilo of
weight and followed by castor oil at the rate of 3 mils per kilo of weight.
The treatment was preceded by a dose of castor oil the day before, and
the animals were starved for 24 hours. The drug removed 9 out of a total
of 11 ascarids (Belascaris marginata and Toxascaris limhata), i out of
33 hookworms {Ancylostoma caninum), and i out of 50 whipworms
(Trichuris depressiuseula) , thus showing but little anthelmintic value for
the two worms (hookworms and whipworms) for which it is particularly
recommended, and a very satisfactory anthelmintic efficacy for ascarids.
It was ineffective against Dipylidium caninum, no specimens being passed
and several being found at autopsy. Of the three nematodes on which
the latex had some effect, ascarids are, according to the experience of the
writers, most easily removed — probably because they do not attach
428 Journal of Agricultural Research voI.xii.no.?
themselves to the mucosa like hookworms and whipworms. The writers
have usually found that any drug which is at all efficacious against hook-
worms is equally so against ascarids.
Evidently the latex has defitiite anthelmintic properties, and although
its efficacy against hookworms {Ancylostoma caninum) and whipworms
(Trichuris depressiuscula) was very slight in this test, it is worthy of
further experimentation in repeated dosage. As the liquid expressed
from the latex appears to have no value as a vermifuge, the active prin-
ciple may be considered to reside in the solid portion sustained in the
latex in the form of an emulsion.
SPIGEUA
For worms in dogs. — Spigelia, or pinkroot, is the dried rhizome and
roots of Spigelia marilandica, a plant native to the Southern and South-
western States. Its virtues as an anthelmintic are said to have been
known to the Cherokee Indians. According to the U. S. Dispensatory,
19th edition,* it is generally considered among the most powerful
anthelmintics, and is used especially for roundworms. The fluid extract
is official.
A test of the fluid extract of Spigelia was made by the junior writer.
In this experiment four dogs weighing from 4.8 to 18 kgm., all of which
were infested with hookworms, and one with whipworms, were treated
with Spigelia in doses graduated from 1.5 to 6 mils, according to the
weight of the dogs. The drug was followed by 194 to 324 mgm. of calo-
mel. The dogs were previously dosed with 29.57 mils of castor oil and
then starved for 24 hours.
One dog was not killed, as it passed no worms. As the previous fecal
examination for this dog showed hookworms, it is evident that the drug
was ineffective in this case for hookworms. The three dogs remaining
passed altogether 6 hookworms, including 3 found in the rectum post-
mortem in the process of elimination. There were 199 hookworms left
unaffected by the treatment. Spigelia removed i out of 18 ascarids and
2 out of 29 whipworms. A few segments of Dipylidium caninum were
passed, but no heads.
The remedy therefore appears to have but little efficacy as an anthel-
mintic for any of the common parasites of dogs, but further trial is advisa-
ble to determine this point.
tobacco
For worms in chickens. — Herms and Beach {1916) have devised a
method of treating poultry for worms, consisting in the administration
of chopped tobacco stems and the liquid in which they are steeped.
Finely chopped tobacco stems, 453 gm. or i pound, enough for 100
birds, are soaked for two hours in enough water to keep them covered.
1 Wood, G. B., and Bache, Franklin. Op. cit., p. 1162.
Feb. i8. 1918 Efficacy of Some Anthelmintics 429
Both the stems and water are mixed with half the usual ration of ground
feed and given to the fowls. Two hours later one-fourth the usual ration
is given mixed with Epsom salt at the rate of 312 gm. for each 100 fowls.
The treatment is to be repeated one week later. The cost is said to be
only 10 cents for 100 fowls.
In the present experiment six chickens were deprived of food for 24
hours and then fed the tobacco mash; two hours later they were fed
the Epsom salt mash in the proportions recommended by Herms and
Beach. The next day they were given what was left of the tobacco
mash, since they had refused to clean it up the first time. A mixture of
bran and tobacco was used, and the birds were not very eager for it,
even after the preliminary fasting.
The remedy removed 30 out of 162 Heterakis papulosa, 3 out of 39
tapeworms, and had no effect on Tetrameres sp., a parasite which, from
its location in the mucous glands of the proventriculus, would not be
likely to be affected. This treatment is apparently intended especially
for Ascaridia perspicillwm, since the "roundworm" figured in the paper
by Herms and Beach {1916) is evidently of this species. Unfortu-
nately the efficacy of the treatment for this parasite could not be deter-
mined, since no species of Ascaridia were present in the birds used in
the experiment.
While the tobacco treatment failed to free any bird from all of the
worms of any given species that might be present, nevertheless the
treatment seems to have been more successful against Heterakis papulosa
or against tapeworms than the other treatments tried.
In view of the difficulty of removing H. papulosa from the ceca, it
may be assumed on the showing here that this treatment would be
effective in removing Ascaridia perspicUhim from the small intestine,
especially if the dose is repeated, as recommended by Herms and Beach.
OIL OF CHENOPODIUM
Oil of chenopodium is derived from the distillation of the seeds or of the
entire leafy part of Chenopodium anthelminticiim L-., sometimes referred
to under the synonym " CJienopodium a^nbrosioides anihelminticum
A. Gray," and commonly called "chenopodium," "American wonn-
seed," or "Jerusalem oak." According to Henkel (1913) —
Wormseed has been naturalized in this coiintry from tropical America and occurs
in waste places from New England to Florida and westward to California.
Infusions made from chenopodium were used in the United States
by the early settlers as a treatment for infestation with ascarids in man,
and its anthelmintic properties are said to have been known by the
Indians. Oil of chenopodium has only recently come into prominence
as a result of the shortage of thymol and santonin, for which it has
proved an effective substitute.
27810°— 18 3
430 Journal of Agricultural Research voi. xii. No. 7
The chemistry of the oil of chenopodium has been studied by Nelson
(jpii, 19 1 3), who was unable to ascertain its exact chemical nature,
but concludes that it is an unstable dioxid. Its physiological action
and toxicity have been the object of numerous experiments by Salant
et al. (1915) and Nelson (1911, 1913). They note that it is a respira-
tory depressant and that it decreases vagus instability and diminishes
frequency of heart action.
In regard to its toxicity the minimum lethal dose is twice as great
when given by the mouth as when given hypodermatically. Repeated
doses have a cumulative effect, and the toxicity is enhanced when the
experiment animals are starved. It is less toxic when combined with
nonessential oils, such as olive, cottonseed, or coconut oil. In cats 0.6
mil per kilo by mouth was invariably fatal, but in dogs 0.5 mil caused
only vomiting. Adrenalin and digitalis were found to be antagonistic
to oil of chenopodium.
Heiser {19 15) notes that over 100,000 cases of hookworm infestation
involving both Ancylostoma duodetmle and Necator americanus have been
treated in the Orient with chenopodium.
For worms in dogs. — In view of the present great interest in cheno-
podium, the writers have made a considerable number of tests of this
substance with a view to determining its efficacy against worms in dogs
and other domestic animals, and in nearly all cases they have found it
extremely efficacious, especially for ascarids. The following experiment
was conducted to determine the efficacy of oil of chenopodium when
administered in one dose at the rate of 0.3 mil per kilo.
Eight dogs, weighing from 2 to 10.3 kilos, were given a preliminary
purge of castor oil, which resulted in the removal of 9 ascarids. After
starving for 24 hours, they were given chenopodium in the dosage indi-
cated, mixed with 2.5 to 15.5 mils of castor oil, according to weight.
The treatment was found extremely efficacious for ascarids, removing
160 worms and leaving only 2. From all but one of the dogs it removed
all ascarids, and removed 10 out of 12 from that one. Its effect on hook-
worms and whipworms was less striking, about one-fourth of the total
number present being removed. It had very little efficacy against
Dipylidium caninum. It was evident that the dose given was uimeces-
sarily large, since the feces smelled strongly of chenopodium, and six of
the eight dogs vomited from two to four hours after treatment. An
objectionable feature of the chenopodium treatment, as given, was the
excessive ptyalism that it caused and the fact that the dogs objected to
the taste of chenopodium in the castor oil and resisted its administration.
In the second experiment with chenopodium an attempt was made to
overcome the ptyalism by giving the chenopodium in a capsule. Eight
dogs, weighing from 5.5 to 9 kilos, were given chenopodium in capsules
at the rate of 0.2 mil per kilo on each of three successive days, the drug
being followed with five times its volume of olive oil the first two doses,
Feb. 18, 1918 Efficacy of Some Anthelmintics 431
and five times its volume of castor oil the third dose. While this treat-
ment removed 15 out of 17 ascarids (Belascaris marginata), it was
ineffective for hookworms (Ancylostoma caninum), whipworms {Tri-
churis depressiuscula) , and Dipylidium caninum. This method was less
satisfactory than that employed in the previous experiment. The castor
oil given was insufficient to overcome the constipating effect of the
chenopodium; the chenopodium, being promptly released from the cap-
sules, was brought undiluted against the gastric mucosa, causing con-
siderable irritation, as shown by the fact that all dogs vomited promptly
after the administration of the drug. Two of the dogs died from the
treatment, one from traumatic pneumonia, due to the fact that the cap-
sule opened in the lar3mx, allowing oil to penetrate the lungs. The other
dog which died was a bitch, containing 11 well-developed fetuses, and
it is likely that the pregnancy in this case was a condition contributing
to the animal's inability to withstand the large amount of chenopodium
given.
A third experiment was undertaken to test the efficacy of oil of cheno-
poditmi when given in doses of o.i mil per kilo, and the dose repeated
daily for a total of six doses. The chenopodium was given mixed with
10 times its volume of olive oil, and was preceded each day by castor oil
in amount equal to the olive oil ; i minim of chloroform per kilo of weight
was added to the mixture of chenopodium and olive oil the first day and
to the castor oil on the following days. The animals were kept on half
feed during the time the treatment was being given.
In this experiment all of the 8 ascarids present and 94 out of 133
hookworms were removed. This increased efficacy against hookworms
is in the writers' opinion due largely to the chloroform administered
with the chenopodium, since in none of their numerous experiments with
chenopodium have they ever found it so efficacious for hookworms when
given with castor oil only. The treatment was entirely ineffective for
whipworms (Trichuris depressiuscula) or for Dipylidium caninum.
This method of administration proved generally satisfactory. The
individual dose was sufficiently diluted to prevent any undue ptyalism
when administered, and there were no symptoms of acute distress, which
occurred when the oil was given in a capsule.
The results of this experiment led the writers to make another test of
chenopodium and chloroform as a remedy for mixed infestation, cheno-
podium being in their experience the most successful remedy for ascarids
and chloroform the best remedy for hookworms. This combination
seems to be entirely compatible both physiologically and pharmaceu-
tically. These drugs when combined with castor oil form a homogeneous
and fairly stable mixture, and the action of each drug is not inhibited by
the other.
Four dogs having been given a preliminary purgative of castor oil,
followed by a 24-hour fast, were given chenopodium and chloroform, both
432 Journal of Agricultural Research voi. xii. no. 7
at the rate of o.i mil per kilo. The drugs were given in castor oil, vary-
ing in amount from 12 to 40 mils, according to the weight of the animals.
In this case, as usual, the chenopodium proved efficacious against ascarids
(Belascaris marginata) , removing 7 out of 8. The remedy showed an
appreciable effect on the whipworms (Trichuris depressiuscula) , removing
about one-fifth of the total number. In regard to hookworms, however,
only 7 out of 61 were removed.
In order to ascertain whether the efficacy of the remedy for hookworms
would be enhanced by increasing the dose, a second trial was made in
which 0.2 mil per kilo, or double the amount of both chenopodium and
chloroform, were administered. As in the previous case, the treatment
was preceded by the administration of castor oil 24 hours before dosing,
,and the dogs allowed no food until after the administration of the anthel-
mintics. The chenopodium and chloroform were given in castor oil,
allowing 29.57 ^^^ for each dog. The preliminary purgative removed
one ascarid (B. marginata) and some segments of Dipylidium caninum.
The combination of chenopodium and chloroform in the dosage given
proved highly successful. All ascarids and over half the hookworms
present were removed. As usual, it was entirely ineffective for Dipy-
lidium caninum,. Undoubtedly the dosage of chenopodium and chloro-
form at the rate of o. i mil per kilo is too small for the best results, and
the efficacy of both drugs is enhanced by doubling the amount. At the
same time this increased dose seems to be well within the limits of safety
and has been tested several times by the junior writer, always with
satisfactory results.
In view of the fact that castor oil is more or less objectionable to many
persons and is not well tolerated by others, the writers undertook an
experiment in which liquid petrolatum was substituted for castor oil in
connection with oil of chenopodium. Four dogs weighing 4.5 to 15 kilos
were given chenopodium in doses of 0.2 mil per kilo mixed with 10 mils
of liquid petrolatum, the drug being followed at once by 20 mils of
liquid petrolatum. The animals were fasted from noon of the previous
day. One of the dogs died on the third day following the administration
of chenopodium. The treatment was entirely inefficacious for ascarids,
hookworms, Dipylidium sp., or Taenia sp., and removed only 2 of 28
whipworms. While the experiment indicates that the liquid petrolatum
diminishes the efficacy of the chenopodium, it must be noted that there
was only one ascarid present, and while the treatment should have removed
the worm, this is a rather small basis on which to judge the performance
of the anthelmintic. On the other hand, the liquid petrolatum seems to
be unequal to the task of overcoming the constipation and toxic effects
from the chenopodium, as the treatment resulted in the death of one dog.
It is the opinion of the writers that the mechanical lubrication resulting
from the use of liquid petrolatum is not sufficient to overcome the con-
stipation and that an active purgative, preferably castor oil, is indicated.
Feb. i8. 1918 Efficacy of Some Anthelmintics 433
The foregoing experiments with chenopodium and chloroform, in
comparison with other standard anthelmintics used for the removal of
worms from dogs, indicate that oil of chenopodium is the best anthel-
mintic of those tried for use against ascarids {Belascaris marginata and
Toxascaris limhata). In the 6 experiments, involving the use of 34
experiment animals, it removed 194 out of 200 ascarids, an efficacy of
97 per cent. It is at the same time probably a little more efficacious
against whipworms (Trichuris depressiuscula) than any other anthel-
mintic tested. In the entire six experiments its average percentage of
efficacy was only 12, a figure which illustrates the difficulty experienced
in dislodging this parasite. Moreover, it is probable that the chloroform
used in three of the experiments may be responsible for the removal
of some of these worms, since both chloroform and chenopodium were
found to have a limited efficacy for whipworms when given separately.
As regards hookworms (Ancylostoma caninum), the experiments of the
writers with chenopodium do not show as high an average efficacy
for this drug as they had been led to expect, considering the warm
indorsement chenopodium has frequently received as an anthelmintic
for the hookworm of man. Undoubtedly chenopodium as well as thymol
would have to be given in repeated doses and the treatment renewed
in order to secure a high percentage of efficacy. The writers have found
chloroform much more efficacious than chenopodium for hookworms in
dogs, and, when given at the rate of 0.2 mil per kilo, is apparently free
from danger.
For worms in hogs. — A series of experiments to test oil of chenopo-
dium as an anthelmintic for ascarids in swine were carried out by the
junior writer and seemed to indicate the great value of this drug when
given to hogs individually after a period of starvation. In these experi-
ments the hogs were not killed, the effect of the anthelmintic being
judged by the presence or absence of ascarid eggs in the feces passed
subsequent to treatment.
In the first experiment with chenopodium an effort was made to
determine the minimum effective dose necessary to free a hog from
ascarids. In this case the dose of oil was combined with 3.7 mils of
areca nut and mixed with the feed, after starving the pigs. Each pig
was fed separately, thus assuring full dosage to each animal. Two pigs,
weighing 11.32 and 14.06 kgm., respectively, were given 0.2 mil of
chenopodium, while a third pig weighing 37.9 kgm. received 0.4 mil.
This latter pig was apparently free from ascarids after a single dose,
while the two other pigs each received three additional doses of 0.27 mil,
0.36 mil, and 0.54 mil before their feces were negative for ascarid eggs.
It is evident from the above that, even when given mixed with feed,
a method later shown to be unsuitable, a relatively small dose of oil of
chenopodium may prove efficacious. It should be stated in comment
that at the time these experiments were conducted there was little
434 Journal of Agricultural Research voi. xri. No. 7
modern literature in regard to the use of oil of chenopodium, and, so far
as the writer is aware, no use had been made of the oil as an anthelmintic
for swine. Hence, the drug was administered cautiously. Subsequent
experiments have shown that doses of chenopodium relatively many
times larger are perfectly well tolerated by swine and are more certain
in action.
In a second experiment with swine chenopodium was administered
in castor oil, and was not given mixed with feed but was fed with a spoon
to each hog individually. Four of the five hogs used were each given
28.35 gn^- of Epsom salt mixed with their feed, while the fifth hog was
given none. All animals were then starved for 24 hours before the
administration of the chenopodium. The smallest pig, weighing 12.25
kgm. was given 0.8 mil; the others, weighing 12.7 to 28.12 kgm. were
given I mil. The dose of chenopodium in each case was given with
29.57 niils of castor oil. This treatment was entirely successful. All
five hogs had shown numbers of ascarid eggs in their feces prior to the
treatment, and all were found free from ascarid eggs a week later. The
feces of the hogs were again examined three days later and found negative.
Two of the five hogs received a third fecal examination, which was
consistently negative.
It appears from the above that chenopodium is an excellent anthel-
mintic when given in individual doses after preliminary starvation, and
in most cases may be relied on to remove all the ascarid worms present.
This opinion was confirmed by further experiments. The preliminary
dosing with Epsom salt did not seem to be of any benefit, since the cheno-
podium was as successful in the case of the hog which received no salt
as in the case of those that did. Subsequent experience has shown that
in the chenopodium treatment for hogs preUminary purgation is an un-
necessary expense.
As there is always a demand among stockmen for something that can
be mixed with the feed to rid their hogs of worms, an experiment was
undertaken to determine the possibilities of this form of medication.
As hogs, especially if large, are very difficult to handle and as much time
and eff'ort would be required to dose separately each animal of a large
herd, the advantage of mixing the medicine in the feed is obvious. Unfor-
tunately this method proved impracticable, as shown by the following
experiment.
One mil of chenopodium, 29.57 mils of castor oil, and 237 mils of lin-
seed decoction made by boiUng 113 gm. of linseed in 296 mils of water,
were thoroughly shaken together into a homogeneous emulsion. This
amount was allowed for each of foyr hogs weighing 20.4 to 32.2 kgm. The
emulsion was distributed evenly through a bran mash which all four
hogs ate together. The animals had previously been fed a mash con-
taining Epsom salt at the rate of 28.35 gm. for each animal, then starved
for 24 hours before treatment. The fecal examination a few days after
Feb. i8. 1918 Efficacy of Some Anthelmintics 435
the treatment showed that one hog was freed from ascarids, the three
others still remaining infested.
It is evident that the administration of drugs by mixing them with
the feed is a method which gives little promise of success, and the far
greater efficacy achieved when the drug is given individually after a
fast more than compensates for the extra labor involved.
In order to be certain that the comparative failure in this treatment
resulted from the method of administration and not from an insufficient
dose of chenopodium, the experiment was repeated, allowing 2.5 mils
of oil to each of the three hogs remaining infested. This experiment
resulted in a total failure, all the hogs remaining infested after the
experiment.
The above experiments were concerned solely with the action of cheno-
podium against ascarids, no attention being paid to its action on other
intestinal parasites. In order to test its action on nodular worms
(Oesophagostomum dentatum) three doses of oil of chenopodium were given
to one hog at intervals of 10 days, two fecal examinations being made
after each dose. The chenopodium was given with 3.7 mils of areca
nut after preliminary starvation, and, so far as could be judged from
the fecal examinations, there was no diminution in the number of nodular
worms. On account of their location in the cecum and colon, where
they are protected by a large mass of fecal material, it is not surprising
that a vermifuge given by the mouth has little or no effect upon nodular
worms. The extensive colon of the pig is rarely, if ever, approximately
empty, even after prolonged starvation. Probably the best way to
reach worms in this location is by the use of diluted anthelmintics in the
form of enemata, as suggested by Railliet {1915).
As Chenopodium anthelminticum grows abundantly in the vicinity of
Washington, D. C, a test was made to determine the possibilities of
feeding the entire plants to the hogs. The test was made when the plants
were in full seed and therefore contained the maximum amount of oil.
A large armful of chenopodium plants was placed daily in a hogpen con-
taining three hogs. The amount given represented about all the hogs
would clean up in a day, and the experiment continued for 19 days.
Although the hogs received their usual daily ration, they ate the chenopo-
dium plants readily. At the end of 19 days one hog was apparently
free from ascarids (A scans suum).
While this treatment has the obvious disadvantage of inaccuracy in
dosage and is applicable only in those regions where the chenopodium
plant grows abundantly, it has the great advantage of costing nothing
but the labor of gathering the weeds. At least it would seem worthy of
further trial.
A test of the dried seeds of chenopodium was made in the following
experiment: Each of three infested hogs was given daily 3 gm. of
chenopodium seed and 3 gm. of areca nut mixed with the feed. The
436 Journal of Agricultural Research voi. xn, no. ?
treatment was continued for 1 1 days, at which time each hog had received
33 gm. of chenopodium seed, or the equivalent of about i mil of the oil.
One of the three hogs was apparently freed from ascarids.
The treatment, besides being of uncertain efficacy, cost more than a
single treatment with chenopodium oil, since a large amount of both
chenopodium seed and areca nut must be consumed. Its efficacy would
doubtless be increased if given after a period of starvation, but this is
obviously impossible when the treatment must be continued for several
days.
It appears from the above experiments that oil of chenopodium is an
excellent anthelmintic for ascarids in hogs when given in suitable dosage
after a preliminary period of starvation. The best purgative with which
to combine it is castor oil. This has the advantage of relative cheapness,
certainty of its action, and of being easily miscible with oil of chenopo-
dium. The chenopodium may be given first and followed immediately
by castor oil, or the two may be given together. The efficacy of oil of
chenopodium is greatly decreased or entirely lost if the drug is mixed
with the feed and several animals allowed to dose themselves while
eating together. Chenopodium seed and areca nut have but little
efficacy when given daily, mixed Avith the feed, and the expense of this
treatment, on account of the number of doses required, is actually
greater than a single dose of the relatively high-priced oil of chenopodium.
Chenopodium plants seem to have some value as an anthelmintic and,
when these are available, the cost of anthelmintic treatment is greatly
reduced. However, further experimentation is desirable along this line.
The following experiments with chenopodium, in which the hogs were
subsequently killed and all worms counted, confirmed the writers'
opinion regarding the efficacy of oil of chenopodium when given under
suitable conditions :
Two pigs weighing 8.16 and 14.06 kgm., respectively, were given i mil
of oil of chenopodium, while a third, weighing 9.98 kgm., was given 2
mils. The drug was followed by 59 mils of castor oil in the case of one
pig and given in 59 mils of castor oil in the case of the second pig. Eleven
ascarids and six nodular worms were passed by the pigs. Post-mortem
examination showed that all animals were freed from ascarids {Ascaris
suum), while numerous nodular worms {Oesophagostomum dentatum)
were found in each animal. The experiment adds further proof of the
efficacy of oil of chenopodium as an ascaricide, and demonstrates that it
also has some efficacy for nodular worms, although probably only a few
were removed. As already stated, these worms are so protected by the
large amount of fecal material in which they lie that it is unlikely that
any anthelmintic given by the mouth will have much effect upon them.
As already noted in the previous experiments, oil of chenopodium,
when mixed with the feed of a number of hogs feeding together, had
shown little or no efficacy as an anthelmintic. Since, however, in these
Feb. i8, 1918 Efficacy of Some Anthelmintics 437
experiments the pigs were not killed and the results were based solely
on the fecal examinations, it was decided to repeat the experiment,
killing the pigs and counting the worms and thus determining definitely
whether this method could be recommended.
A sow which had been starved for 24 hours was fed a mash of corn
meal and bran into which had been stirred an emulsion consisting of
3.5 mils of oil of chenopodium, 60 mils of castor oil, and 473 mils (i pint)
of milk. Following the treatment the sow passed 19 stomach worms
{Arduenna strongylina) and 13 nodular worms (Oesophagostomwm den-
tatum). A post-mortem examination revealed the presence of 19 ascarids
(Ascaris suum), 13 stomach worms (A. strongylina), and numerous
nodular worms (O. dentatum). The treatment proved a complete failure
in eliminating ascarids, thus confirming the opinion already expressed
regarding the inefficacy of chenopodium when given mixed with the feed.
The treatment removed over half the stomach worms present and
demonstrates the possibilities of the drug for this species of parasite.
Although it also had some efficacy against nodular worms, as usual the
percentage removed was very small.
In cooperation with Dr. Ernest, of the Tuberculosis Eradication
Division, of the Bureau of Animal Industry, the junior writer was recently
enabled to make a practical test of the chenopodium treatment of swine
on a large scale. The test was carried out at the Green Berry Point Farm
of the United States Naval Academy, where about 176 swine ranging
from young pigs weighing 15.88 kgm. or less to large boars and brood
sows, were given the treatment.
The pigs were starved for 24 hours before treatment and confined in
relatively small pens so that they could be easily caught. The brood
sows and boars were kept separate. The pigs were caught in the small
pen and lifted one at a time over the fence into a larger inclosure, where
they were treated. Four laborers were employed in this work, two
capturing the pigs and two holding the animals while they were dosed.
Restraint was made fairly easy by backing the pig into a comer of the
inclosure. While one attendant held the pig in this position the other
kept the animal's jaws apart, using two loops made of harness straps;
one loop, passing under the upper jaw, was pulled upward while the other
loop, passing over the under jaw, was pulled downward. The head was
kept tilted upward and the pressure on the straps, besides keeping the
mouth wide open, served to hold down the tongue and prevent the hog
from shaking its head.
A table to hold the bottles of chenopodium and castor oil and the
measuring glasses was placed conveniently near. Four mils of oil of cheno-
podium and I ounce (29.57 mils) of castor oil was allowed each hog under
100 pounds (45.36 kgm.), while those over that weight were given double
the dose. No attempt was made to weigh the animals, but the dose for
each hog was decided upon as it was brought up for treatment. The
438 Journal of Agricultural Research voi. xii, No. 7
two oils were measured separately, and were then poured together into
a large iron kitchen spoon, which was placed as far back as possible in
the animal's mouth. Subsequently, however, the iron spoon was dis-
pensed with, the oils being poured directly into the hog's mouth, care
being taken to hold the graduate glasses out of reach of the hog's teeth.
When the medicine had been given the pressure on the straps was
relaxed to enable the hog to swallow, the head still being held high. If
the hog refused to swallow, it could always be induced to do so by
plugging the nostrils with the finger tips. This forces the animal to breath
through its mouth, and to do so it must first swallow.
Wefound this method fairly rapid for hogs weighing not over 100 pounds
(45.36 kgm.), 65 animals being dosed in one hour. It took eight hours
to dose the entire herd of 176 hogs, the labor of handling the heavy
brood sows making the operation much slower than when pigs of medium
size were being treated. In treating the brood sows it was necessary
to throw them on their backs and to hold them in this position while
dosing. As this caused considerable excitement to the sows, those
which from their appearance were soon to farrow were left untreated.
Only one of the 176 hogs treated was injured. This animal, which
was accidentally dropped, injuring the spinal cord, was killed and exam-
ined for worms. One hundred ascarids (Ascaris suum) were taken from
a piece of intestine not over 30 cm. long, and many more remained
uncounted. Two days later the manager of the farm reported that a
great many ascarids were seen among the feces on the place, and about
two weeks later one of the pigs which was killed for food was found to
be entirely free from ascarids. The treatment therefore appears to have
been very successful. The treatment required 1.13 kgm. of oil of che-
nopodium and 7.5 liters of castor oil. With chenopodium at $11 per
kilo ($5 a pound) and castor oil at 66 cents per liter, the treatment in
this instance cost a trifle under 10 cents per hog, exclusive of the labor.
Even at the present high price of drugs, the cost is trifling compared
with the increased profit which may be expected to be derived from
healthy animals.
For WORMS in sheep. — Four lambs weighing 16.6 to 26.1 kilos were
dosed with oil of chenopodium at the rate of 0.2 mil per kilo, the dose
which was found most effective for dogs. The medicine was given as a
drench emulsified with 147.9 mils of milk. Following the treatment,
one of the lambs contracted pneumonia, probably as the result of
some of the drench's entering the lungs. The treatment succeeded
in removing all the stomach worms from three lightly infested lambs,
but failed completely in the case of one heavily infested lamb, suggest-
ing that in this case the drench did not reach the fourth stomach di-
rectly, but was modified or absorbed in the rumen.
The treatment was fairly efficacious for hookworms (Bunostomum iri-
gonocephalum) , removing two-thirds of those present, in this respect
Feb. i8. 1918 Efficacy of Some Anthelmintics 439
being considerably more efficacious than it was found to be against
hookworms in dogs. It was inefficacious for other intestinal worms.
On the whole, the use of chenopodium for stomach worms and hook-
worms in sheep seems to promise considerable success when properly
administered, and is at least worthy of further trial.
For worms in poultry. — ^To test the efficacy of oil of chenopodium
against worms in poultry, six chickens were dosed at a rate of about
0.4 mil per kilo. Each bird weighed 0.5 kgm. Each bird received 2
mils of castor oil followed at once by 0.2 mil of oil of chenopodium
mixed with 2 mils of castor oil, the birds being kept without feed the
previous day. The treatment was fairly satisfactory for Ascaridia
perspicillum, removing 9 out of 13 worms. In this connection it will
be recalled that the experiment in feeding tobacco stems to chickens
was carried out with birds not infested with A . perspicillum, so that the
writers have no data of their own to compare with the showing made
by oil of chenopodium. As already stated, Herms and Beach (1916)
found the treatment very efficacious for worms which were evidently
Ascaridia perspicillum, to judge from their illustration of the para-
sites.
Chenopodium removed only 2 out of 349 cecum worms (Heterakis
papulosa) and was entirely inefficacious against other nematodes. No
tapeworms were passed in the feces, but 22 were found post-mortem,
2 of which were in the large intestine. Even though these 2 are con-
sidered as having been removed by the anthelmintic, the showing is
not very creditable. It may be stated, however, that none of the sub-
stances tested by the writers for tapeworms on poultry have proved
very satisfactory.
DISCUSSION OF RESULTS
In order to present the foregoing data in a condensed comprehensive
summary, the various experiments have been tabulated by hosts
(Tables I-V). These tables show, for each host, the efficacy of the
different drugs tested against the more important parasites, as indicated
by the percentage of worms removed compared with the total number
present. Three columns of figures are given for each parasite, in order
that the reader may see at a glance not only the percentage of efficacy
but the data from which this percentage is derived, and thus be able
to judge how conclusive or otherwise the figures presented may be.
The tables also show the number of host animals used for each drua:
tested, and the size of the dose, usually based on the weight of the
animal. A reference column gives the pagination of the experiments
described in detail in the text. Where several experiments have been
conducted in testing one drug, the results are combined in the tables
into a single set of figures.
440 ' Journal of Agricultural Research voI.xii.No. 7
While it is realized that in most instances these data are insufficient
to express results in percentages, and for this reason reference to the
percentage of efficacy of a drug has been avoided in the text, it was
thought that comprehensive tables like the following would prove
convenient in summarizing the results and would at least serve to indi-
cate what drugs offer promise of success. Too much emphasis, however,
should not be placed on negative results based in most cases on insuffi-
cient data.
The tables do not include some of the earlier experiments carried
out on hogs, in which the efficacy of the drug tested was determined
solely by the results of fecal examination without killing and examining
the animals post-mortem.
Feb. 18, 1918
Efficacy of Some Anthelmintics
441
S <u VI
S OJ ui
SB c
030
^u'2
h' c-a
5 C osg o
(^ 8 a " 3
e-^
5Bc
o u o
H 5^
<Lt G tt f-H U
Hfcl
3Bc
031;
s ci"s art--
1 ■* "*2^
sa
V ■*
to ro
O O
O •»
O O -O
O O r^ Ov ■ O O\oo
o >o o o
O ■-I -o o o
^^ fO >^j o o\
O f o
^ tn o
O^n •* O
•^-•3 a
C e M
8ga
aeag
d d o d 6
a
•» o
l?^a
O -w O O
00 00O0'-«
■^ O 1^00 o
t 1 1 ■<»• -a-
3 S
u g
•a 3
I - "
Si - ^- -S
« >, "S a tr "
^•3 g las 21-3
j a - a a v^J^-o 3 g ~ ^ g
^"•i1=3ia-P^sN .aSl-o-Slgl
gs
a-o
■u a
a -a
•a §
« o
V 01
•rt c.
aSi
.3 a
V
•d
a
a
o
"o
0
•0
•2
♦J
L^
1
.a
a
o
■a
a
l-n«
ap.«
1 I'lp
i IB
a 5^yg
- a ojfc CI
. g « a i'*j
Sit: a a a oj
o 2 a in o
3 1: a O' a V o o
ig S o oj o o^
' a; £ ci< a, 55 M ID
^B-S
f til
" o S
^t: a.
442
Journal of Agricultural Research
Vol. XII, No. 7
Table II. — Percentage of efficacy of various anthelmintics for swine
Drug tested.
Epsom salt in solution b
Tartar emetic and cas-
tor oil.
Chenopodium and cas-
tor oil.
Emulsion of chenopo-
dium and castor oil
mixed in feed.
a .
It
Page.
400
401
436
437
Dose.a
75.6 to 227
gm. in water.
260 mgm. in
water.
I to 2 mils. . . .
3.5 mils.
Ascarids.
^•0
3 Q
I'm- O
Whip-
worms.
3a
2
Is
t-r QJ
Nodular
worms.
3 a
gse
stomach
worms.
sag?
o The dose indicated is for the anthelmintic, not for the laxative with which it is combined.
6 Entirely ineffective. Pigs would not drink either the stronger or the weaker solution.
c Present,
d Numerous.
Table III. — Percentage of efficacy of various anthelmintics for sheep
Drug tested.
Chloroform and castor oil.
Copper sulphate:
In capsule
In drench
Gasoline in milk. (Sum
mary of 3 experiments.)
Petrolemn benzinin milk
Coal-tar phenols in milk.
Chenopodium
0*3
Page
404
406
406
411
413
438
Dose.o
5 to 10 mils.
o.s gm
(50 mils of I per
cent solution.)
7.5 to 30 mils ,
15 mils
2.8 to 15 mils
3.3 to 5.2 mils in
147.8 mils milk.
Stomach worms.
10, S4I
726
978
3 o,
64.0
0.3
93- o
Hookworms.
.fl-O
3 a
25
S!fi
Nodular
worms.
•^-g
■3 y
0.6
16.0
a The dose indicated is for the anthelmintic, not for the laxative with which it is combined.
6 Both sheep died following the treatment.
e Two sheep died; the others passed no worms.
Table IV. — Percentage of efficacy of various anthelmintics for poultry
Drug tested.
Areca nut with olive oil
Turpentine and castor oil
Tobacco stems and Epsom salt.
Chenoiwdiimi and castor oil. . .
Page.
419
42 s
428
439
Dose.o
1 gm
2 mils
454 gni. per 100
birds.
0.2 mil
Ascaridia
perspicillum.
3 a.
oThe dose indicated is for the anthelmintic, not for the laxative with which it is coi|fbined.
Feb. i8, 1918
Efficacy of Some Anthelmintics
443
Table V. — Percentage of efficacy of various anthelmintics for cats
Drug tested.
Oleoresin of aspidium and calomel.
PeUetierine tannate and castor oil <
^0
Page,
416
417
66
Dose.o
0.8 mil
64 mgm. per
kilo.
Ascarids.
S il
Taenia.
|&
2: Ph°
Hookworms.
3 a
2:
a The dose indicated is for the anthelmintic, not for the laxative with which it is combined.
b Two cats died shortly after treatment. The third cat vomited one Taenia.
c Treatment was entirely ineffective. No tapeworms or nematodes removed.
CONCLUSIONS
Making due allowance for the paucity of data in regard to certain
drugs, the writers consider that the following may be reasonably advanced
as the result of their investigations.
Simple purgatives, calomel and castor oil, may have some slight value
as anthelmintics, but it is hardly sufficient to justify their use for this
purpose. Ascarids in dogs are sometimes removed by castor oil given as
a preliminary purge, and this fact may prove of benefit in veterinary
practice as a diagnostic measure when the more accurate method of
microscopic fecal examination can not be carried out. However, castor
oil failed to remove ascarids more frequently than it succeeded, and in no
case were all the ascarids removed from any one animal. As many of
the experiments on dogs were preceded by a dose of castor oil, the writers
have fairly extensive data on this subject.
The most reliable vermifuge for ascarids, whether in dogs or swine, is
oil of chenopodium. This drug, which was tested out on 34 dogs in six
experiments, showed an efficacy for the entire series of 97 per cent.
It rarely fails to remove all the ascarids present in a dog if given at the
rate of 0.2 mil per kilo, preceded by a dose of castor oil and the animal
starved for 24 hours before treatment.
The chenopodium treatment is also very efficacious for ascarids in
swine, and when properly administered may be expected to remove most,
if not all, of the worms present. It would seem, however, that neither
chenopodium nor any other drug tested will give satisfactory results if
mixed with the daily ration and the animals allowed to dose themselves;
it is best given to each pig individually in suitable dosage, preceded by
a fast. While this method necessarily involves considerable labor when
treating animals as unruly as swine, the labor can be reduced by sort-
ing the hogs roughly into classes according to size and confining them
in inclosures which will permit them to be caught with a minimum
amount of struggUng. The treatment has proved practical on a large
444 Journal of Agricultural Research voi. xii. No. 7
scale and the results, as far as they could be determined, were entirely
satisfactory.
Oil of chenopodium appeared to be effective for stomach worms in
sheep, although the data on this subject are not sufficient to warrant
its recommendation. It is also of some efficacy for hookworms in sheep
and in dogs, though in the latter case chloroform was found more reliable.
Other remedies which seem to have more or less merit as anthelmintics
against ascarids are the latex of Ficus laurifolia, santonin in repeated
doses, and thymol. Although thymol in repeated doses is fairly effica-
cious against hookworms, it was inferior to chloroform for this purpose,
causing more distress. An excellent preparation for mixed infestation
in dogs consists of equal parts of oil of chenopodium and chloroform, given
at the rate of 0.2 mil per kilo, combined with 30 mils of castor oil. This
preparation may be expected to remove all the ascarids present, a large
proportion of hookworms, and possibly a certain percentage of whip-
worms. This latter parasite seems to be very difficult to eliminate, and
nothing tried by the writers proved very efficacious, almost any anthel-
mintic occasionally proving successful. This experience may perhaps be
explained by an intermittent peristalsis of the cecum, which occasionally
allows the anthelmintic to enter, but which usually excludes it. Although
chloroform was fairly successful in removing stomach worms from sheep,
both animals upon which it was tried subsequently died from its effects,
and it would seem to be too dangerous for use on sheep.
In the case of stomach worms in sheep, copper sulphate was found to be
the most satisfactory remedy, the experiments confirming the findings of
Hutcheon. A simple apparatus (fig. i) devised by the senior writer
reduces the labor involved in drenching a flock of sheep and insures
accurate dosage. Petroleum benzin also proved satisfactory and was
more efficacious for hookworms than copper sulphate. However, it is
much more expensive than copper-sulphate solution, must be given three
times, and in a vehicle like milk, which adds greatly to the expense. The
fact that petroleum benzin (refined gasoline) proved efficacious, while
commercial gasoline was considerably less so, is perhaps related to the
differences in specific gravity and consequent volatility of the refined
product compared with the commercial product.
Among anthelmintics intended for use against tapeworms, male-fern
proved efficacious when tested on dogs. In the case of cats it removed
all tapeworms from 75 per cent of the animals tested, though it proved
fatal to two out of six animals which were somewhat enfeebledf rom disease.
Apparently it is more toxic to cats than dogs and should be prescribed
with caution and only given to healthy subjects. So far as can be
judged from a single experiment with dogs, there seems to be no danger
in combining male-fern with castor oil, as is done in the so-called Her-
mann's mixture. In fact, the writers are inclined to agree with Seifert
Feb. 18, 1918 Efficacy of Some Anthelmintics 445
{1908) that the administration of castor oil after male-fern will avoid
the toxic effects of the latter by causing its rapid and thorough elimina-
tion, and that for this purpose no other purgative is quite so effective.
This subject, however, should receive more study before conclusions are
drawn.
Pelletierine tannate was a failure in the one experiment in which it
was tested on cats, but was efficacious on dogs. No remedy was effica-
cious against tapeworms in poultry. Of the four drugs tested, chenopo-
dium gave the best results for this purpose, but its efficacy for tapeworms
is very slight.
Turpentine proved the most efficacious of the remedies tested on
poultry for the removal of Ascaridia perspicillum, while chenopodium
was nearly as good. When tested on dogs and pigs, turpentine was not
very efficacious; and, as it caused grave symptoms of nephritis in pigs
and caused the death of some of the experiment dogs, its use upon these
animals is inadvisable.
The treatment with chopped tobacco stems recommended by Herms
and Beach for ascarids in poultry proved fairly efficacious for Heterakis
papulosa and would presumably be at least as efficacious for Ascaridia
perspicillum, since this latter worm is more easily reached by anthel-
mintics than is H. papulosa.
There are a large number of drugs showing a greater or less degree of
efficacy for the various intestinal parasites of domestic animals. Usually
their action is selective — that is, they show a pronounced efficacy for
certain species of intestinal worms, while they are decidedly less efficacious
or entirely inefficacious against other intestinal parasites. If we consider
the ideal anthelmintic one which will remove all worms of a given class
or species, and do this every time in a single dose, we find that very few
drugs approach this ideal.
Among the drugs which have given the best results under experimental
conditions for the purposes intended and concerning which the writers
have sufficient data to warrant positive conclusions may be mentioned
the following :
(i) Copper sulphate in drench for stomach worms in sheep.
(2) Oil of chenopodium for ascarids in pigs and dogs.
(3) Oleoresin of male-fern for tapeworms in dogs.
(4) Turpentine for Ascaridia perspicillum in fowls.
(5) Chopped tobacco stems for Heterakis papulosa in fowls.
27810°— 18 i
446 Journal of Agricultural Research voi.xii, No. 7
LITERATURE CITED
Alessandrini, Giulio.
191 5. LE MALATTIE DE PARASSITI ANIMALI NEGLI ESERCITI COMBATTENTI. Itt
Policlinico, sez. prat., v. 22, no. 25, p. 822-827.
ArbucklE.
1916. ROTATE YOUR PASTURE AND AVOID DISASTER. In Amcf. Sheep Breeder,
V. 36, no. 7, p. 402-403.
Berrio, Posada.
1911. traitement de la trichoc^phalose par le latex d'higueron. In Rev.
M6d. et Hyg. Trop., t. 8, no. 3, p. 191-192. Discussion, p. 192-193.
Billings, W. C, and Hickey, J. P.
1916. SOME points about hookworm disease, its diagnosis and treatment.
/nJour.Amer. Med. Assoc, V. 67,no. 26, p. 1908-1912,4 fig.
Campbell, D. M.
1917. EPSOM salt an effective vermifuge. In Amer. Jour. Vet. Med., v. 12,
no. 5, p. 314.
Coffey, W. C.
1915. checking stomach worms in sheep. In Breeder's Gaz., v. 68, no. 6, p. 192.
Heiser, V. G.
1915. RECENT EXPERIENCES IN THE ORIENT WITH CHENOPODIUM AS A REMEDY
AGAINST HOOKWORM AND OTHER INTESTINAL PARASITES. In JoUT. Amer.
Med. Assoc, v. 65, no. 6, p. 526-527.
Henkel. Alice.
1913. AMERICAN MEDICINAL FLOWERS, FRUITS, AND SEEDS. U. S. Dept. AgT.
Bul. 26, 16 p., 12 fig.
Herms, W. B., and Beach, J. R.
1916. ROUND WORMS IN POULTRY— LIFE HISTORY AND CONTROL. Cal. AgT. Exp.
Sta. Circ. 150, 7 p., 3 fig.
HUTCHEON, D.
x89if . THE TREATMENT OF WIRE- WORMS IN SHEEP. In Agt. JouT. Cape Good Hope,
V. 4, no. I, p. 8-9.
189IC. WIRE-WORMS IN SHEEP AND GOATS, AND THEIR TREATMENT WITH SXn<PHATB
OP COPPER. In AgT. Jour. Cape Good Hope, v. 3, no. 19, p. 179-181.
1892. BLUESTONE FOR WIRE-WORMS. In Agr. JouT. Cape Good Hope, v. 4, no.
20, p. 24c.
1895. WIRE-WORMS. In Agr. Jour. Cape Good Hope, v. 8, no. i, p. 18-19.
Lenhartz, Hermann.
1902. BEHANDLUNG DER DURCH DARMSCHMAROTZER HERVORGERUFENEN
ERKRANKUNGEN. In Handbuch der speciellen Therapie innerer Krank-
heiten. Aufl. 3, Bd. 4, p. 604-631. Jena.
LUCKEY, D. F.
1915. HOW TO TREAT A FLOCK OP 300 LAMBS INFESTED WITH STOMACH WORMS
(STRONGYLUS CONTORTUS)? In Proc 24th Ann. Meeting Mo. Vet. Med.
Assoc, p. 68-69.
Miller, F. H.
1904. haemorrhagic colitis of the dog due to infection with the tricho-
CEPHALUS DEPRESSIUSCULUS. (True whipworm). In Amer. Vet. Rev.,
V. 28, no. 8, p. 722-729, I pi.
MOUAT-BIGGS, C. E. F.
I915. THE TREATMENT OF ANKYLOSTOMIASIS IN VENEZUELA. In TranS. SoC Trop.
Med. and Hyg., v. 8, no. 7, p. 216. Discussion, p. 217-218,
Feb. i8, 1918 Efficacy of Some Anthelmintics 447
Nelson, E. K.
1911. a chemical investigation of the oil op chenopodium. u. s. dept.
Agr. Bur. Chem. Circ. 73, 10 p.
19 13. A CHEMICAL INVESTIGATION OP THE COMPOSITION OF THE OH, OF CHENOPO-
DIUM. U. S. Dept. Agr. Biir. Chem. Circ. 109, 8 p.
Railliet, a.
191 5. l'EMPLOI DES ME;dICAMENTS dans LE TRAITEMENT DES MALADIES causiSes
PAR DES Ni&MATODES. In Rcc. Med. Vet., t. 91, no. 15, p. 490-513.
Ransom, B. H.
1907. STOMACH worms (hAEMONCHUS CONTORTUS) IN SHEEP. U. S. Dept. Agr.
Bur. Anim. Indus. Circ. 102, 7 p.
Salant, William, and Livingston, A. E.
1915. the influence op the oh, op CHENOPODIUM ON THE CIRCULATION AND RES-
PIRATION. In Amer. Jour. Physiol., v. 38, no. i, p. 67-92, 14 fig.
and Mitchell, C. W.
I915. THE INFLUENCE OP OIL OF CHENOPODIUM ON INTESTINAL CONTRACTILITY.
In Amer. Jour. Physiol., v. 39, no. i, p. 37-53, 9 fig.
and Nelson, E. K.
1915. THE TOXICITY OF OIL OF CHENOPODIUM. In Amer. Jour. Physiol., v. 36,
no. 4, p. 440-463-
SCHULTZ, W. H.
191 1. REMEDIES FOR ANIMAL PARASITES. A STUDY OP THE RELATIVE EFFIQENCY
AND DANGER OF THYMOL AS COMPARED WITH CERTAIN OTHER REMEDIES
PROPOSED FOR HOOKWORM DISEASE. In JouT. Amer. Med. Assoc, v.
57, no. 14, p. 1102-1106.
Seifert, otto.
1885. DIE DARMPARASITEN DES MENSCHEN. In Dcut. Med. Ztg., Bd. 6, No. 8,
p. 85-87; No. 9, p. 97-98; No. 10, p. 109-111.
Cites Schidlowski, p. 98.
1908. KLINISCH-THERAPEUTISCHER TEIL. In Braun, Max. Die tierischen Paia-
siten des Menschen. Aiifl. 4, p. 477-623. Wurzburg.
Stiles, C. W.
1901. treatment for round worms in sheep, goats, and cattle. u. s. dept.
Agr. Bur. Anim. Indus. Circ. 35, 8 p.
1902. FURTHER INVESTIGATIONS ON VERMINOUS DISEASES OF CATTLE, SHEEP, AND
GOATS IN TEXAS. U. S. Dept. Agr. Bur. Anim. Indus. i8th Ann. Rpt.,
1901, p. 223-229.
WiNSLOW, Kenelm.
I913. VETERINARY MATERIA MEDICA AND THERAPEUTICS, ed. 7, 781 p. New York,
TOBACCO WILDFIRE^
By Frederick A. Wolf, Plant Pathologist, and A. C. Foster, Assistant Plant Path-
ologist North Carolina Agricultural Experiment Station
INTRODUCTION
During the past season (191 7) a leaf disease of tobacco {Nicotiana
tahacum) has been the cause of much concern to tobacco growers because
of its destructiveness. Attention was first directed to it during June
when tobacco was being transplanted. Subsequently during the entire
growing season numerous complaints of this disease were received by
members of the staff of the North Carolina Agricultural Experiment
Station and of the State Department of Agriculture. Because of the
severity of the epidemic and the insistence by growers that this disease
was manifestly different from any they had ever seen, an investigation
was begun.
A preliminary survey of the literature on diseases of tobacco revealed
the fact that this disease was clearly unlike any which had previously
been described. Furthermore, the apparently water-soaked margin of
the diseased areas, the tissues of which, upon microscopic examination,
were found to be teeming with bacteria, suggested that the disease was
probably of bacterial origin. Accordingly, attention in the first studies
was centered upon the etiology of the disorder. A brief statement (7) ^
concerning this work, in which attention was directed to the presence of
the disease, was duly prepared.
It was also pointed out that the causal organism was undescribed,
and the name "Bacterium tahacum Wolf and Foster" was suggested.
Promise was made in that report of a description of the morphological and
cultural studies upon Bact. tahacum. Besides making this description, it
is the present purpose to adequately describe the disease and to indicate
our present knowledge of its economic importance, distribution, and
dissemination.
HISTORY AND DISTRIBUTION OF THE DISEASE
While the disease was first definitely recognized in June, 191 7, near
Wendell, N. C, it is impossible to determine with certainty for how many
seasons prior to the present one the disease has existed. It is quite
probable, to judge from the testimony of several reliable informants, that
the disease caused the loss of practically the entire crop in one field, near
Wendell, in 191 6. Mr. E. G. Moss, Assistant Director, in Charge of the
Granville Branch Tobacco Station, Oxford, N. C, is convinced, as is also
' Approved tor publication by B. W. Kilgore, Director, North Carolina Agricultural Experiment Station.
* Reference is made by nimiber (italic) to "Literature cited," p. 458.
Journal of Agricultural Research, Vol. XII, No. 7
Washington, D. C. ^ Feb. 18, 1918
ma ' KeyNo. N.C— 9
(449)
450 Journal of Agricultural Research voi. xii. No. ^
the senior writer, that the malady was observed by him during 191 6 at
Creedmoor, N. C. To judge from the additional fact that the disease has
been collected during the past season in 19 counties within North Caro-
lina (Surry, Stokes, Forsythe, Guilford, Rockingham, Caswell, Alamance,
Orange, Person, Durham, Chatham, Moore, Hoke, Wake, Johnston,
Franklin, Granville, Vance, and Warren) and in 3 within Virginia (Pitt-
sylvania, Halifax, and Mecklenburg) it is highly probable that the disease
existed prior to the present year. Whether or not this disease occurs in
other of the States in which tobacco is grown is not known, except in the
case of Wisconsin. A letter from Prof. James Johnson, Department of
Horticulture, University of Wisconsin, who has had occasion to observe
the disease in North Carolina, states that a similar bacterial spot, although
not nearly so destructive as in North Carolina, has been observed by him
in Wisconsin.
ECONOMIC IMPORTANCE
The disease is universally conceded by growers of tobacco to be the
most destructive one which attacks this crop. Losses, ranging from
those which were inappreciable to those in which almost the entire crop
was destroyed, were sustained in every locality where the disease was
present. In some instances fields upon the same farm were observed
to be badly affected, while others had little or none of the disease. In
some sections, too, the disease occurred upon every farm within a radius
of several miles, while in others it was present only in an occasional
field. It is not possible, therefore, because the disease was not uniformly
destructive over the entire area in which it is known to occur and because
time has not been afforded to make a careful survey, to obtain a reasonable
estimate of the damage wrought. Some idea of the losses, however, can
be gained from statements taken from reports kindly furnished by a
number of growers, who compared their leaf-tobacco sales with neighbors,
whose crops were free from wildfire or were at most only slightly affected.
Some of these reported losses averaging $100 per acre for their entire
crop. One correspondent estimated his total loss at $5,000 and said
that hundreds of farmers in his section suffered an equal acreage loss.
APPEARANCE OF THE DISEASE
The disease was first noted early in June during a period of rainy
weather accompanied by nights which were so cold as to retard the
growth of tobacco. The affected plants in many fields perished, necessi-
tating replanting a second or a third time. A period of relatively dry,
warm weather of about a month's duration followed, during which time,
the crop made an extremely rapid growth, as shown by the fact that the
plants were sufficiently mature to be topped. At this stage of their
development another rainy season of about a week's duration occurred
and was followed by another epidemic of the disease. The disease
Feb. i8, 1918 Tobacco Wildfire 451
appeared so quickly, spread so rapidly, and affected the leaves so seri-
ously that it was commonly given the appropriate designation "wildfire."
The foliage alone seems to be subject to attack. The first evidence
of disease is the appearance of circular, chlorotic areas varying from 0.5
to I cm. in diameter. Within 24 hours after this chlorosis is first noted
minute brown areas will have formed at the centers of the spots (PI.
15, A). Within another day these spots will have enlarged greatly
PI. 15, B), and a border of water-soaked appearance marks the margin
of the necrotic tissues. Within a few more days the diseased areas are
2 to 3 cm. in diameter and are often strikingly concentric with shades
of tan to dark brown, the centers being lightest in color (PI. 16, A).
Such spots have a broad translucent border, which is in turn surrounded
by a chlorotic halo that pales out into adjacent tissues (PI, 15, B; i6, A).
When the spots are numerous, they fuse, causing large, irregular areas
of leaf tissue to become dry. These dead areas remain intact in case
there is no precipitation. When dewy nights and intermittent showers
occur, however, the dead areas rot out so that the leaves present a ragged
appearance (PI. 16, C) which is especially manifest when large numbers
of infections occur upon a single leaf.
Not uncommonly the leaves on one side of the plant are more seriously
diseased than those on the opposite side and there may even be a uni-
lateral destruction of these leaves resulting in distortion, as shown in
Plate 16, B. The vascular tissues seem not to be invaded, but the
organism confines its attack to parenchymatous tissues.
Two other leaf spot diseases of tobacco, frogeye and speck, are present
within North Carolina, from both of which wildfire is easily distinguish-
able. Frogeye appears as circular, brown spots, with a darker border
and with grayish centers. Upon this gray center may be seen the
fructifications of Cercospora nicotianae E. and E., or other fungi associated
with the disease. No chlorosis accompanies these spots. Speck, which
results from a deficiency of potash, appears as tan-colored, irregular
areas which are first present at a distance from the principal veins.
When this disease is accompanied by chlorosis, there is no definite halo
around the lesions. In the case of neither of these diseases is the margin
of the affected areas water-soaked in appearance, and in neither of them
do the affected tissues disintegrate and fall out.
ISOLATIONS AND INOCULATIONS
On June 13, fresh material of tobacco wildfire was collected and
isolations were made by planting on poured plates of nutrient agar frag-
ments of tissue from the margin of affected areas. Contamination was
avoided by washing the leaves prior to making the planting in mercuric
chlorid and then rinsing them in sterile water. Several types of colonies
developed along the margins of these plantings, the most common of
which was Bad. tabacum, which appeared as glistening, grayish white
452 Journal of Agricultural Research voi.xii. No. 7
colonies. It was possible, in some cases, to make transfers directly from
these colonies to tubes of agar and secure pure cultures. In others,
dilution poured plates were first made, and the organism was transferred
from certain of the colonies which developed to tubes of agar. Diseased
material was collected several times subsequently, and numerous speci-
mens were received by mail, so that opportunity was afforded during the
season to isolate the organism from several sources.
Some preliminary inoculations were made on June 13 in which the
inoculum consisted of macerated, diseased leaves upon which a quantity
of water was poured. About 100 young tobacco plants growing in a
flat in the greenhouse were then sprinkled with this water. Four days
later infections were evident by the appearance of numerous yellow spots
with pin-point-like centers.
On the evening of June 28 two potted tobacco plants, about i8 inches
in height, were inoculated with pure cultures of Bad. tabacum. A
watery suspension from agar cultures was sprinkled upon these plants,
after which they remained covered with a bell jar for 36 hours. On the
morning of July 2 numerous chlorotic areas had formed, which by July 5
had changed to large dry spots, typical of wildfire. No difficulty was
experienced in reisolating the organism from these lesions.
On July 10, 16 plants which had been transplanted in the field on
June 14 were inoculated by sprinkling them with suspensions made from
bouillon cultures. These plants were not seen again until July 15, when
large brown areas had formed abundantly, whereas adjacent uninocu-
lated plants remained healthy.
Another series of inoculations, involving 18 plants growing in pots
placed outside of the greenhouse, was made on July 25. In this case
the leaves were immersed in a bacterial suspension. Seventy-two hours
later the first evidence of infection was observ^ed. Here again the
organism was recovered from mature lesions.
Another set of inoculations, involving 12 potted tobacco plants, was
made with what proved to be Boat, tabacum isolated from spots on
cowpeas (Vigna sinensis) which had been planted between the hills in
a badly diseased tobacco field. The spots on cowpeas, from which these
isolations were made, were very similar in appearance to 3- or 4-day-old
lesions on tobacco. Eight tobacco plants were inoculated on July 27
and four on August 2. Inoculation was effected by sprinkling the plants
with a bacterial suspension. By August i in the first case and August 9
in the second there was no doubt that the diseased areas, which had
formed upon all of the inoculated plants, were typical of wildfire. The
organism was reisolated from these spots and, together with transfers
from the original cultures from cowpeas, was used in inoculating cowpeas.
Here again the same method as before was used in making inoculations.
Only a few spots developed upon the several plants employed in two sets
of inoculations. These spots were similar to those on cowpeas growing
Feb. i8, 1918 Tobacco Wildfire 453
in the field of diseased tobacco. Microscopic examination, furthermore,
showed that the dead tissues were filled with bacterial organisms.
The lesions, both naturally and artificially produced, are believed to
have originated around punctures made by leaf-hoppers, which were
abundantly present on these plants throughout the season. The wild-
fire organism is capable of multiplying within the cells weakened as a
result of the withdrawal of their contents by the feeding of these insects,
but is not able to parasitize normal cells. Drops of moisture laden with
bacteria certainly dripped from the diseased tobacco plants to the cowpeas
beneath them, and could thus have supplied the inoculum which caused
the cowpea foliage to become spotted. This explanation is supported
by the observation that the lesions on cowpeas did not increase in size
beyond pinpoint-like dead areas, indicating that Bact. tabacum can not
adapt itself to invade healthy tissues, and by the further fact that no
new spots developed subsequently on the naturally and artificially
inoculated plants. Furthermore, spots never developed on cowpeas
growing at a distance from diseased tobacco plants — that is, where they
could not be infected through the agency of water dripping from diseased
tobacco plants. Bact. tabacum, therefore, is not parasitic upon cowpeas,
and its chance occurrence upon this crop indicates that conclusions as to
the pathogenicity of bacteria when judged from inoculation experiments
in which the inoculum is introduced through wounds are not entirely
convincing. In view, therefore, of the fact that tobacco is not grown in
the vicinity of West Raleigh, where the inoculation experiments were
conducted, and that all of the uninoculated plants within the greenhouse
grew to maturity without any manifestation of wildfire, there is no doubt
that all of the infections which were secured resulted from inoculations
with the organism in hand. When judged by the readiness with which
infection occurs, Bact. tabacum is to be regarded as a very vigorous
pathogene.
Aside from the inoculations upon cowpeas, only two other host species,
bell peppers (Capsicum, annuum,) and Jimson weed {Datura tatula), were
employed, with negative results.
PATHOLOGICAL ANATOMY
Affected tissues were fixed in 95 per cent alcohol, embedded in paraffin,
sectioned, and stained with carbol-fuchsin. The presence of a crystal-
line substance whose nature is described in a recent paper by Ridgway (5)
interfered seriously with the cutting of suitable sections. In tissues in
which the cells had not yet become dry and collapsed, bacteria are abund-
antly present within the intercellular spaces (fig. i). In mature lesions,
however, they occur also within the cells. The contents of such cells
appear to have been completely destroyed, whereas the walls have
undergone little disintegration. The complete disintegration of diseased
tissues, which occurs in the presence of excessive moisture, results, it
454
Journal of Agricultural Research
Vol. XII, No. 7
is believed, from the activity of other species of bacteria which enter
the cells following invasion by Baci. tabacum. These species appear
always to be present in old lesions as judged by the isolation studies.
DESCRIPTION OF BACTERIUM TABACUM
Bacterium tabacum, emend.
The primary cause of tobacco wildfire is a grayish white, rod-shaped
organism with rounded ends. It is motile by means of a single polar
Fig. 1. — Parenchyma cells from the margin of a lesion showing Bacierium tabacum in the intercellular
spaces and within the cells.
flagellum which is about twice as long as the body of the bacterium
(fig. 2, a). Motility can be observed when fresh material is examined in
a drop of water or when prepara-
y ^ |k ( •tW tions are made from i8-hour-old
A V V A f ^^N bouillon or agar cultures. Flagella
m { (2. 1 M ■ ^^^ f ^^^ easily demonstrated when the
/ / / m Aw ^^^^3 organism from such cultures is sub-
's I ^ / JF m ^H^^ jected to the staining method out-
U ^ ^^ ^ fl lined by Morrey (2). The bacterium
^^^^f ''^^--^ ^W usually occurs singly within the host
^^ ^'^"yy^^ tissues, but in culture chains of as
^^Btb^ima^^i^mmm^ .^mmmm,u,^^B^ many as five elements have been
noted (fig. 2, fc). The limits of size
vary from 2.4 to 5 by 0.9 to 1.5 \x,
the most common size being 3.3 by
1.2 /x. No involution forms have
been observed; neither have endospores been demonstrated.
The organism stains readily with aqueous- and carbol-fuchsin, anilin
gentian-violet, and methylene blue. It is Gram-negative, however,
and is not acid-fast. Neither is a capsule demonstrable by the methods
of Rabiger or Welch.
The organism has been cultured mainly upon glycerin agar and potato
agar. Colonies appear in these media on the second day in poured plates
Fig. 2. — a, Flagella of Bacierium tabacum stained
by Morrey 's method; b. Bad. tabacum, from
bouillon stained with carbol-fuchsin, showing
arrangement of the elements.
Feb. i8, 1918 Tobacco Wildfire 455
kept at a temperature of 20° to 25° C. By the fourth day the surface
colonies have attained a diameter of 2 to 3 mm. They are grayish white
in color, circular in outline, are appreciably raised, and have a smooth
margin and a smooth, wet -shining surface. Buried colonies are biconvex.
In stroke cultures a filiform growth which widens at the base of the
slant is formed. Growth is moderate and does not give rise to the
production of odor. In stab cultures growth is best at the surface of
the agar, and the line of puncture is filiform. In stab cultures on nutri-
ent gelatin a filiform growth also appears along the line of puncture,
with the greatest growth at the surface of the medium. No evidence of
liquefaction occurs until the tenth day, when it becomes crateriform,
and is complete within 30 days.
Growth on potato cylinders is nontypical in appearance, and there is
no evidence of diastatic activity.
With a 2 per cent solution of Difco peptone as a basal solution, five
solutions were prepared by adding i per cent of one of the following
carbon compounds : Dextrose, saccharose, lactose, glycerin, and dextrin.
No gas formed in fermentation tubes containing any of these media.
A vigorous growth with strong clouding and a surface pellicle occurred
in all in the open arm. In the presence of dextrose and saccharose a
distinctly visible clouding gradually extended upward into the closed
arm, while in the case of the other carbon compounds the closed arms
remained clear. The organism is therefore regarded as aerobic in general.
Acid formation in stab cultures on litmus-glycerin, litmus-dextrose,
litmus-lactose, and litmus-saccharose agar begins within four to six
days, but no gas formation occurs on any of these media.
Growth on litmus milk presents a very characteristic appearance.
During the first three days following inoculation there is a deepening
of the blue color. Two days later coloration begins to become strati-
form, and by the seventh or eighth day four distinct layers are evident.
The upper one is between plumbeus and violaceous in color (5), the
next is lilacinus, the third, violaceous, and the lowermost approximates
caesius. These colors lose their intensity after a few days and become
more or less blended, and by the tenth day there is evident reduction
of the litmus and precipitation of the casein. Reduction proceeds
rather slowly and is complete by the twenty-fourth day.
There is no reduction of nitrates in nitrate-peptone broth, although
a conspicuous clouding occurs. Furthermore, no gas is formed; and
the tests for ammonia, indol, and skatol were negative.
The thermal death point of this organism, as determined by exposing
newly inoculated tubes of bouillon in the customary manner, was found
to be about 65° C. It is manifestly quite sensitive to desiccation,
since no growth appeared after six days when bouillon cultures were
placed on sterile slides in sterile petri dishes.
The group number of Baclerhim iabactim according to the numerical
system of the Society of American Bacteriologists is 2>2i-2222032.
456 Journal of Agricultural Research voi. xii, No. 7
DISSEMINATION OF WILDFIRE
The fact that the disease appeared in epidemic form twice in one
season and that each epidemic followed a rainy period, with little or
no new infection in the interim suggests that dissemination is primarily
influenced by moisture. This is in accord with observations on the
dissemination of certain other plant diseases of bacterial origin, as
angular leafspot of cotton (4) and Citrus canker (6).
Very striking evidence was found in two instances that wind is also
a potent factor in the spread of wildfire. In one instance a field was
observed before and several days following a rain which was accom-
panied by a high wind. The disease had advanced in consequence
over a distance involving 1 6 to 18 rows lying parallel to diseased tobacco.
The disease terminated rather abruptly beyond this distance. In
another locality no disease occurred, except in the case of a field of
approximately an acre in area, the plants for which had been brought
for a distance of several miles. Here, again, the disease spread, fol-
lowing the same storm into an adjacent field to the leeward, was most
abundant near the field of diseased plants, and gradually became less
in the direction away from the diseased field. The organism had very
evidently been spread by wind-blown rain, a phenomenon in accord
with observations by Faulwetter (j) upon the angular leafspot of cotton.
When the first epidemic was prevalent in the vicinity of Oxford, N. C,
thrips were abundantly present upon tobacco and were popularly sus-
pected of being responsible for the spread of the disease. Accordingly,
diseased leaves bearing numbers of these insects were collected, and
the thrips were liberated upon healthy plants in the greenhouse at
West Raleigh. Careful watch was kept, but no evidence of wildfire
developed upon any of the plants. In the same season only a few
thrips were found in diseased fields near Wendell, N. C, during an
entire afternoon's search. For these reasons it is improbable that
thrips are to be regarded as agents of dissemination.
Following the first outbreak of wildfire, opportunity was afforded to
make numerous observations upon the origin of the disease. In every
instance where the disease occurred in the field it has been possible to find
that plants in the seed beds or "plant beds" were also affected. It was
adjudged, therefore, that the disease must have been introduced into the
plant beds either through the use of infected seed or through the agency
of fertilizers. One large seed farm upon which diseased plants occurred
was visited in searching for wildfire lesions upon seed pods, and, further-
more, affected pods were carefully sought lor in many other diseased
fields with negative results.
Since the disease is already so widely spread, which suggests that it
must have had some common agency of dissemination, and since tobacco
stems were incorporated in certain fertilizers as the source of potash,
inquiry was directed to determine the possibility of the introduction and
Feb. i8, 1918 Tobacco Wildfire 457
dissemination of wildfire by fertilizer materials. Attempts to isolate
Bad. tahacum from diseased leaves which had passed through the curing
process gave negative results in the case of three samples tested. In
view of the fact that no growth occurred in bouillon cultures exposed to
temperatures above 65° C. for 10 minutes, as has previously been reported,
it is highly improbable that the organism could survive for several hours
temperatures of 180° F. and above, as are maintained for several hours
in the last part of the curing process. Furthermore, in the preparation
of tobacco stems for incorporation with fertilizer materials they are sub-
jected to a sufl&cient degree of heat to insure complete sterilization.
SUMMARY
(i) A leafspot disease of tobacco called "wildfire," which is more
destructive than any other malady affecting this crop, has appeared
within North Carolina and Virginia.
(2) It has been collected during the past season in 19 counties within
North Carolina, 3 within Virginia, and occurs also in Wisconsin.
(3) Wildfire first attracted attention at time of transplanting tobacco
and appeared again in epidemic form at time of topping the crop.
(4) The disease originated in the seed bed or plant bed, but only nega-
tive evidence had been secured that infection comes from the seed.
(5) The leaves alone are attacked, and the symptoms are entirely
unlike those of other foliage disease of tobacco.
(6) The primary cause of wildfire has been found to be a wet-shining
grayish white, i -flagellate organism, which is herein described as Bac-
terium tahacum. Its period of incubation is about 72 hours, and large
lesions are formed within a week.
(7) The disease is of the necrotic type, involving parenchyma tissues.
(8) Moisture is of prime importance in the spread of wildfire. When
rains are accompanied by wind, dissemination is especially rapid.
458 Journal of Agricultural Research voi. xii, No. ^
LITERATURE CITED
(i) Faulwetter, R. C.
1917. DISSEMINATION OF THE ANGULAR LEAFSPOT OP COTTON. In JoUf. AgT.
Research, v. 8, no. 12, p. 457-475, 2 fig. Literature cited, p. 473-475.
(2) MORREY, C. B.
1917. THE FUNDAMENTALS OP BACTERIOLOGY. 289 p., 165 fig. Philadelphia
and New York.
(3) RiDGWAY, C. S.
1916. GRAIN OP THE TOBACCO LEAP. In Jour. Agr. Research, v. 7, no. 6, p. 269-
288, 2 fig., pL 15-17. Literature cited, p. 287.
(4) Rolfs, F. M.
1915. ANGULAR LEAF spot OF COTTON. S. C. Agr. Exp. Sta. BuL 184, 8 fig.,
30 p., 9 pL Literature cited, p. 30.
(5) Saccardo, p. a.
1894. chromotaxia. Ed. 2, 22 p., 2 pL Patavii.
(6) Wolf, F. A.
1916. citrus canker. In Jour. Agr. Research, v. 6, no. 2, p. 69-100, 8 fig.,
pL 8-11. Literature cited, p. 98-99.
(7) and Foster, A. C.
1917. bacterial leap spot of tobacco. In Science, n. s., v. 46, no. 1189,
p. 361-362.
PLATE 15:
Bacterium tabacum:
A. — Tobacco leaf , fotir days after artificial inoculation, showing chlorosis and lesions.
B. — Natural infection with brown lesions bordered by tissues of a water-soaked
appearance.
Tobacco Wildfire
Plate 15
Journal of Agricultural Research
_M.---y
Vol. XII, No. 7
Tobacco Wildfire
PLATE 16
Journal of Agricultural Research
Vol. XII, No. 7
PLATE 16:
Bacterium tabacum:
A. — Natural infection. Lesions are large and concentrically zonate.
B. — Numerous confluent lesions on one side of the midrib have resulted in distortion
of the leaf.
C. — Almost the entire leaf is involved and a portion of the rotted tissues have fallen
out. Natiu"al infection.
27810°— 18 5
GIPSY-MOTH LARV^ AS AGENTS IN THE DISSEMINA-
TION OF THE WHITE-PINE BUSTER-RUST^
By G. Flippo GravatT, Assistant Pathologist, and G. B. PosEy, Scientific Assistant,
Investigations in Forest Pathology, Bureau of Plant Industry, United States Depart-
ment of Agriculture
INTRODUCTION
Very little has been done to correlate the widespread distribution of
the white-pine blister- rust, caused by Cronartium rihicola Fischer, with
factors governing the dissemination of the spores of the causal organism.
The early occurrence of telia on leaves of currant and gooseberry plants
(Ribes spp.) in localities distant from known infections on pines (Pinus
spp.), together with the absence of definite knowledge of instances of
overwintering on the former hosts, is suggestive of distant seasonal
spread of the disease by aeciospores from pines.
Larvae of the gipsy moth (Porthetria dispar L.) feed on the Peridermium
stage of Cronartium rihicola and carry thousands of . aeciospores on their
bodies. As Collins^ found that larvae of the gipsy moth are blown as
far as 20 miles, these insects are a potential agent in distant spread of
the blister-rust. The gipsy moth is distributed over a large portion of
the white-pine region of New England.
GIPSY-MOTH INFESTATION ON DISEASED PINE
In the fall of 191 6 a stand of white pine covering an area of from 5 to
7 acres at Kittery Point, Maine, was found to be severely infected with
Cronartium rihicola. This growth ranged from young seedlings to mature
trees 80 feet tall and random ]4-a.cre plots in this area showed 65 to
100 per cent of the trees to be diseased. The number of infections on
individual trees ranged from i to more than 300, and it was estimated
that there were 75,000 to 100,000 separate infections in trees on this
area.
In the infected plot gipsy-moth-egg clusters were found in varying
abundance on limbs and stems of pines of all sizes, and were located
from near the ground to the tops of the largest trees. In a number of
cases egg clusters were present on the diseased bark, and in one instance
four were located on a single canker.
« The writers are indebted to Mr. A. F. Burgess, of the 0£Gce of Gipsy Moth and Brown-tail Moth In-
vestigations, Bureau of Entomology, for helpful suggestions. Further work is being carried on in co-
operation with the aboTe office.
' Collins. C. W. dispersion of gipsy moth larv^ by the wind. U. S. Dept. Agr. Bui. 273, 23 p.,
7 pi. 1915. Bibhography, p. 22-23.
METHODS USED IN DETERMINING WIND DISPERSION OF THE GIPSY MOTH AND SOME OTHER
INSECTS. /« Jour. Econ. Ent.. v. 10. no. i. p. 170-176, 2 pi. 1917-
Journal of Agricultural Research. Vol. XII, No. 7
Washington, D. C. Feb. 18, 1918
mb Key No. G 135
(459)
460 Journal of Agricultural Research voi. xii, no. 7
On May 25, 1917, large numbers of gipsy-moth larvae were found in
and around ruptured blisters, and several days later some of the bUsters
where the presence of larvae had been previously noted were empty, and
spore production was apparently arrested. Subsequent observation on
these blisters showed that no further spore production took place, while
on the same cankers blisters which were artificially protected from larvae
continued to produce spores in abundance until June 25. To determine
the rate of blister destruction on June 9 a number of larvae were placed
on a twig infection which had 38 sporulating pustules. Many of the
larvae crawled away or dropped off, but a sufficient number remained to
destroy the fruiting layer in practically every blister by noon of the
following day, with the result that no subsequent spore production took
place.
Cessation of spore production in injured blisters was caused by the
destruction of fruiting hyphae. Usually the spores and hyphae were eaten
away first and then the larvae very often ate through the base of the
fruiting layer to a depth of several millimeters. Apparently, after blis-
ters no longer furnished suitable food for the larvae, they began feeding
on the areas of the yellowish, discolored, infected bark outside the fruit-
ing region and in some cases a large per cent of the outer bark of next
year's sporulating zone was destroyed. Careful observations on many
larva-infested cankers showed that spore production was prematurely
arrested in 25 to 100 per cent of the pustules, the percentage usually aver-
aging highest on small twigs.
SPORES ON LARViE
Larvae working in blisters collected so many aeciospores on their hairy
bodies that they appeared nearly the color of spores in mass. On differ-
ent dates larvae were taken from blisters and placed in separate capsules,
precautions being taken against including spores not on the bodies.
These were taken into the temporary laboratory and spore counts made
of the bodies and the alimentary tracts. Spores for counting were re-
moved from larvae by washing the bodies in series of water, and alcohol
mounts on slides followed by final examinations to assure thoroughness
in the method used. This procedure proved quite effective, and, where
carried through 5 to 10 washings, practically all spores were removed
from the outside of the bodies. Counts were made on the spores adher-
ing to the inside of the capsule and added to the total found in the wash-
ings. After bodies of the larvae had been thoroughly washed, they were
dissected and counts made of spores in the alimentary tracts. On May 26,
June 4, and June 1 1 fifteen small larvae were collected. Spore counts on
the bodies of these 15 gave a minimum of 1,120, a maximum of 28,320,
and an average of 18,100. Counts of spores in the alimentary tract gave
a minimum of 1,740, a maximum of 48,570, and an average of 26,022
Feb. i8, i9i8 Gipsy-Moth Larv(E and Blister-Rust 461
To determine the approximate amount of spore material passed through
the alimentary tract, 20 larvae were placed on fresh cankers in a feeding
tray. After they were settled and had fed normally for several hours, a
sheet of paper was placed under the cankers for the collection of pellets.
A total of 423 pellets were dropped within a period of 13 hours. Counts
of the spores in these pellets gave from 3,960 to 12,450, with an average
of 8,160, which is at the rate of 318,616 spores excreted per day per larv^a.
Germination tests made of the spores on the bodies of larvae collected
on cankers gave positive results, and approximately the same percentage
of germination was observed as on spores taken directly from these
cankers. Germination tests of spores in pellets have given very poor
results; in only one case did several spores germinate. In many cases
spores taken directly from these cankers also failed to germinate in
laboratory tests.
WIND DISPERSION OF LARV^ AS A FACTOR IN BLISTER-RUST SPREAD
At Kittery Point, Me., aeciospores were produced from April 29 to
July I, with maximum spore production from May 10 to 25. Collins^
gives the hatching period for gipsy moth larvae in this section for 191 2,
1913, and 1914 as May i to 23, April 29 to May 14, and May 11 to 28,
respectively. The season of 191 7 was approximately one week later than
usual. The period of wind dispersion of larv^se is given as ranging from
18 to 30 days, starting one to two weeks after the first caterpillars hatched.
Observations by the writers showed varjang numbers of larvae feeding on
blisters from May 25 to June 25.
Collins^ working with wind dispersion of larvae of the gipsy moth over
a series of several years showed that they were carried in wind currents
to distances as great as 20 miles. The same author states that approxi-
mately 50 per cent of the lar\'ae caught at distances of 6 miles or less had
fed previously.
The writers, using fly-paper traps, placed 10 to 30 feet from the nearest
pine infection, and so arranged as to exclude larvae that may have reached
the trap by crawling, caught 75 small larvae. Four of these caterpillars
had, respectively, 35, 105, 185, and 2,180 aeciospores on their bodies,
which establishes the fact of local wind dispersion of aeciospore-bearing
gipsy-moth larvae. That spores carried on bodies of larvae may remain
viable for a considerable length of time is borne out in viability tests
under laboratory conditions, wherein aeciospores germinated after
remaining in vials for a period of two months.
Examination of wild and cultivated species of Ribes at various points
throughout Kittery Township showed an abundance of gipsy-moth
larvae feeding on the foliage, and in many cases they were observ^ed
crawling on the under surfaces of leaves. Quite a number of the larva-
' CoiLiNS. C. W. 191S. Op. cit. « Collins, C. W. 1917. Op. cit.
462 Journal of Agricultural Research voi. xii, no. 7
infested plants showed areas producing uredospores, and in four instances
the only leaves showing blister-rust infections were those which had been
injured by insects.
Sixty larvae collected on species of Ribes were examined for aeciospores.
Of these one larva collected on June 14 on the und^er surface of a wild
gooseberry plant showed 280 aeciospores and 520 uredospores on its
body. The gooseberry plant was heavily infected with blister-rust,
being located only 20 feet from pine infections. Germination tests of
the spores from this larva gave two germinating aeciospores and many
germinating uredospores, thus bearing out the fact that gipsy-moth
larvae do carry viable spores to Rihes spp. and also showing the part
which insects may play in local distribution of the disease by uredospores.
PRACTICAL importance:
The facts given in regard to the gipsy-moth larvae show that these
insects are certainly a factor in the spread of the blister-rust locally
from pines to Rihes spp. Their habit of feeding and crawling over the
lower leaf surface, where the stomata are located, gives the spores borne
on their bodies a good opportunity for causing infection. The prob-
ability of the spread of blister-rust from pines to distant Rihes spp. is
undoubted, since Collins' work shows that the gipsy-moth larvae are
blown by winds of varying intensity for distances of 20 miles. Though
wind is considered to be the most important factor in aeciospore dissemina-
tion, gipsy-moth larvae undoubtedly play an important part. Other
insects have been collected from infected pines with thousands of aecio-
spores on their bodies, but these insects were not present in sufficient
numbers to make them of importance in comparison with the number of
gipsy-moth larvae present.
SUMMARY
(i) The period of hatching and of wind dissemination of gipsy-moth
larvae came within the period of spore production of the blister-rust on
pines.
(2) Larvae fed abundantly on spores and injured the fruiting layer of
the pustules so that further spore production was arrested.
(3) Larvae from blister-rust cankers had thousands of viable spores on
their bodies. A small percentage of the larvae collected from fly paper
and from species of Ribes near infected pines showed aeciospores on their
bodies.
(4) Gipsy-moth larvae were found feeding on leaves of Rihes spp.,
and in some cases the only infected leaves on plants of this genus were
those showing insect injury.
(5) The Bureau of Entomology has shown that these larvae are
blown by the wind up to a distance of 20 miles. Within this distance
the larvae are potential agents in the spread of the white-pine blister-
rust (within the area infested by the gipsy moth).
ADDITIONAL COPIES
or THIS PUBLICATION MAY BE PROCXTRED FEOM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
20 CENTS PER COPY
Subscription Price, 83.00 Per Year
Vol. XII KBBRUARY 25, 1918 No. 8
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTKNTS
Page
Influence of Carbonates of Magnesiimi and Calciiun on
Bacteria of Certain Wisconsin Soils - - - ► 463
H. L. FULMER
( CffifilribxiUon trom Wisconsin Agricultural Experiment Station )
Humus in Mulched Basins, Relation of Humus Content to
Orange Production, and Effect of Mulches on Orange
Production - -- - - - - - - 505
CHARLES A. JENSEN
(Ccutribtttion from Bureau ol Plant Industry )
Relation of Kinds and Varieties of Grain to Hessian^pFly
Injury - - - - - - - - - . 519
JJOSES W. McCOLLOCH and S. C. SALMON
( Contrlbntioa from Kansas Agricultural Experiment Station)
PUBUSHED BY ADTHOMTY OF THE SECRETARY OF AGRICULTURE,
WITH THE COOPERATION OF THE ASS0CL4TI0xN OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
WASMINGXON, t). C.
wASMiNOToit : aovenNMCNT pmnTwa ornce : i«ib
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Pkyskiogist and A ssociale Chief, Buretnt
of PI an I Indu-stry
EDWIN W.- ALLEN
Ckie.f, Office of Experiment Slaitons
CHARLES L. MARLATT
EntomoloQist and Assistant Chief, Bureau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agriculfural Experimtnt
Station
H. P. ARMSBY
Director, Institute of Animal Nutrition, The
Pennsylvania State College
E. M. FREEMAN
Botanist, Plant Pathologist and Assistant
Dean, Agricttlturcl Experiment Station of
the University of Minnesoia
All correspondence regarding articles from the Department of Agriculture should be
addressed to Karl F. Kellermajti, Journal of Agricultural Research, Washington, D. C.
*Dr. Pearl has undertaken special work in connection with the war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College* Pa.
JOMALOFAGMQUIAIISEARCH
Vol. XII Washington, D. C, February 25, 1918 No. 8
INFLUENCE OF CARBONATES OF MAGNESIUM AND
CALCIUM ON BACTERIA OF CERTAIN WISCONSIN
SOILS ^
By H. ly. FuLMER,^
Assistant Agricultural Bacteriologist, Wisconsin Agricultural Experiment Station
THE PROBLEM
In the course of a study of the biology of certain acid soils it was
tound that magnesium carbonate causes a great increase in the reproduc-
tion of soil bacteria. Just what effect this great increase in number of
bacteria has on the fertility of the soil is a problem beyond the scope of
this paper. However, it is a well-established fact that the majority of
the changes of soil constituents are brought about by microorganisms.
These substances, which are constantly undergoing changes, are influ-
enced by the number and kinds of bacteria.
The beneficial effect of lime, calcium carbonate, and magnesium car-
bonate on the growth of higher plants is generally preceded by an increase
in the number of bacteria. Evidence is sufficient to warrant the con-
clusion that an increase in the number of soil microorganisms is usually
reflected in a more rapid decomposition of organic matter and a greater
liberation of the inorganic soil constituents, chiefly potassium and phos-
phorous.
Although many important data have been obtained in relation to the
effect of liming on the bacteria in soil, there still remain problems which
deserve careful investigation — for example, the amount and kind of lime
and calcium and magnesium carbonates that can be applied most eco-
nomically and yet give the best results.
The investigations of this paper were undertaken with the hope of
throwing some light jpon the problem of soil bacteria and their response
to applications of magnesium carbonate, calcium carbonate, and limestone.
The effect of these substances on pure cultures of bacteria and on the
ordinary soil flora was observed in —
(a) Acid Colby silt loam soil;
(6) Acid Plainfield sand;
(c) Neutral Miami silt loam.
' This paper is submitted in partial fulfillment of the requirements for the degree of doctor of philosophy
in bacteriology in th^ Graduate School of the University of Wisconsin, July, 1917.
' The writer is indebted to Dr. E. B. Fred, of the Wisconsin Experiment Station, for supervision and
suggestions in the experimental work and preparation of the manuscript.
Journal of Agricultural Research, Vol. XII, ^fo. 8
Washington, D. C. Feb. as, 191S
me Key No. Wis.— ir
(463)
464 Journal of Agricultural Research volxii.no.s
REVIEW OF LITERATURE
A complete review of the literature, showing the effect of lime (CaO)..
calcium carbonate (CaCOg), and magnesium carbonate (MgCOg) on soil
microorganisms, is not attempted in this paper. Only a brief resume of
certain of the more important papers is given.
Although this paper is primarily concerned with the relation of soil
microorganisms to calcium and magnesium carbonates, it was deemed
worth while to include a brief review of results obtained from the use of
lime and limestone. Under ordinary soil conditions, calcium oxid, or
lime, is soon converted into calcium carbonate. Because of the frequent
occurrence of magnesium in limestone, this substance was included in
the discussion.
NUMBKR OF ORGANISMS
In 1 90 1 Chester (7)^ made plate counts of an acid soil which had re-
ceived lime at the rate of i ,000 to 4,000 pounds per acre. In every case
the lime increased the total number of bacteria; the greatest gain was
noted where the largest amount of lime was applied.
Fischer {18) treated soil with lime and calcium carbonate at the rate
of 0.1 to 3.0 per cent by weight of soil. The calcium carbonate and lime
were added in gram-molecular equivalents. His results showed that
0.1 and 0.2 per cent of lime after three days gave an increase in the
total number of bacteria. Treatments amounting to 0.5 per cent and
more were harmful at first, but later gave an increase beyond that of the
control. The data showed that a slight increase in the number of bac-
teria occurred when calcium carbonate was added to the soil.
Several years later, Hutchinson (25) made somewhat similar experi-
ments, using lime and calcium carbonate. He also found that lime at
first exerted a depressing effect but later gave rise to an increase in the
number of bacteria. He concluded that the reduction in the number of
bacteria immediately after treatment with lime was due to the caustic
effect of the oxid, since no injury was noted after the oxid changed to
carbonate. He believed the benefit from liming was due in part to the
gain in soluble organic matter, to the improvement in the physical con-
dition, and to the correction of the acid reaction of a soil, all of which
tended to bring about a better environment for the development of bac-
teria. However, this investigator held that these changes brought
about by the action of liming did not seem sufficient to account for the
enormous increase in plant growth. He assumed, therefore, that the
action of lime was much the same as an antiseptic — that is, it caused a
partial sterilization.
In a later publication, Hutchinson and MacLennan (24) reported the
results of experiments with lime and calcium carbonate on five different •
soils. The range of reaction in these soils varied from neutral to strongly
1 Reference is made by number (italic) to " Literature cited," pp. 500-504.
Feb. 25. 1918 Influence of Carbonates on Soil Bacteria 465
acid. Lime was added in amounts varying from o.i to i.o per cent,
while calcium carbonate was added in amounts of i.o per cent only.
Besides the total number of bacteria, these investigators determined the
nitrate and ammonia nitrogen of the soils subjected to this treatment.
The results of the experiments showed that in all of the soils, liming
increased the number of bacteria. At first the heavier applications of
lime retarded and later stimulated the reproduction of microorganisms.
The acid soil required more lime to retard the growth of bacteria than
the nonacid soils. As a result of liming, ammonification and nitrifica-
tion in all of the soils was increased. The authors concluded that differ-
ent soils varied in relation to the amount of lime needed to effect partial
sterilization. They assumed that each soil absorbed a certain amount
before the antiseptic action began.
Miller (57) carried out a rather intensive study of the effect of lime
and calcium carbonate in both calcium-rich and calcium-poor soils, and
also in soils low and high in organic matter. His results were in accord-
ance with those obtained by many other investigators — namely, that
lime in small amounts increased the number of bacteria, while larger
application, decreased at first and later increased the number. To this
investigator it appeared that lime was a direct stimulant to the soil
flora. The same general increase was obtained with soil low and soil
high in calcium content, and also in soils low and high in organic matter.
Unlike lime, calcium carbonate brought about only a slight increase in
the number of bacteria.
Beckwith, Vass, and Robinson (5) applied lime at the rate of 2 tons
per acre to both acid and nonacid soils. Only the soils with an acid
reaction or where large amounts of organic matter were added showed
an increase in the number of bacteria from the lime treatment. Under
the same conditions ammonification and nitrification responded in a
like manner.
Soil was treated with lime, magnesium oxid, and magnesium car-
bonate by Lemmermann and Fischer (34). They found that mag-
nesium oxid caused a greater increase in the number of bacteria than
did either lime or magnesium carbonate.
Engberding (13) treated clay soil with o.i per cent of lime and with
0.5 per cent of magnesium oxid. Both treatments increased the number
of bacteria, although the lime seemed to give better results.
Jenkins and Britton {26) showed that by using heavy applications of
lime with raw-bone meal, the latter was decomposed more rapidly.
Fabricius and Van Feilitzen (14) noted an increase in the number of
bacteria in moor soils treated with lime.
Lemmermann, Fischer, Kappen, and Blanck (55) reported a gain in
the number of bacteria in cultivated and noncultivated moor soils and
in clay soils when lime or calcium carbonate was added. However, an
increase in the number of bacteria was not shown in a sandy clay soil
or in a sandy soil.
466 Journal of Agricultural Research voi.xii.no. s
AMMONIA AND NITRATES
Withers and Fraps (62, 63) added calcium carbonate to soil treated
with nitrogenous material and observed a gain in nitrates which was
greater in the calcium-carbonate soil than in the control. Koch {31)
reported similar results by applying lime. Fraps (jp) treated acid and
nonacid soils with calcium and magnesium carbonates and noted an
increase in nitrates in all cases, especially in the acid soil. However,
calcium carbonate gave better results than were obtained with mag-
nesium carbonate.
By applying calcium carbonate to a soil treated with ammonium
sulphate, Lemmermann, Blanck, Heinitz, and Von Wlodeck {36) no-
ticed a loss in ammonia. Lipman, Brown, and their associates (4, 5,
42, 43) studied the effect of calcium carbonate on the actitivies of soil
microorganisms. Lipman and Brown {44) showed an increase in am-
monification with monocalcium phosphate [CaH4(P04)2] and dicalcium
phosphate [Ca2H2(P04)2], but a decrease with tricalcium phosphate
[Ca3(P04)2] when applied to soil.
Wohltmann, Fischer, and Schneider {64) inoculated liquid media
with soil treated with magnesium oxid and lime and observed an increase
in both ammonification and nitrification. Hutchinson {22) observed
that in liquid cultures magnesium carbonate retarded nitrification in
practically every case, while calcium carbonate had but little influence
on this process.
Vogel {57) inoculated a nitrogenous solution containing calcium
carbonate with a soil suspension and obtained an increase in ammonia.
A similar test for nitrification was made, but an increase in nitrates was
not obtained. Somewhat similar results were obtained by Paterson
and Scott (55). Hutchinson and McLennan (25) reported that calcium
carbonate treatments caused a slight increase in nitrification in soil.
Greaves {21), working with a Utah soil high in lime and magnesium
oxid, found that calcium carbonate in all concentrations increased the
formation of ammonia, while magnesium carbonate retarded ammonifi-
cation except when applied in very small amounts. The chlorids of
magnesium and calcium appeared to be very toxic. With a Japanese
soil Machida {48), of the Japanese Experiment Station at Tokio, noted
that calcium chlorid retarded ammonification, while magnesium chlorid
increased it. He found that nitrification was favored more by the use
of magnesium carbonate than by calcium carbonate.
The work of McBeth and Wright (49) showed that the chlorids, the
sulphates, and especially the carbonates inhibited nitrification. Lyon
and Bizzell (47) reported that 10 days after treatment with lime the
number of bacteria was doubled.
Fred and Graul (20) concluded that the accumulation of nitrates from
casein or gelatin in acid soils was not materially benefited by calcium
carbonate. In many cases in acid soil calcium carbonate increased and
Feb. 25, 1918 Influence of Carbonates on Soil Bacteria 467
later decreased the nitrate content, and it was assumed that a loss of
nitrate nitrogen occurred because of the increase in the total number of
microorganisms.
Allen and Bonazzi (j) showed that the addition of ground limestone
to a noncalcareous soil brought about a more rapid nitrification. Kelley
{28) studied the effects of calcium and magnesium carbonates on ammoni-
fication and nitrification in a California soil high in basic substances.
His results showed that calcium carbonate benefited ammonification
slightly and nitrification to a great extent, while magnesium carbonate
was toxic to both processes. This investigator failed to show any def-
inite ratio of calcium and magnesium which favored the nitrifying and
ammonifying power of this soil. With Hawaiian soils, high in lime and
magnesium oxid, Kelley (29) obtained similar results. Dolomitic and
calcareous limestones gave results similar to those obtained with calcium
carbonate.
Kellerman and Robinson (27) obtained nitrification of ammonium
sulphate in a soil with a high magnesium content. When calcium car-
bonate was added, an increase in nitrification was noted; with magnesium
carbonate a decrease, except when the latter substance was added to the
soil in very small amounts. Owen (52) reported that magnesium car-
bonate favored nitrification more than calcium carbonate.
White (6j) and Voorhees and Lipman {58) treated soil with pure lime
and with a lime containing magnesium. Better nitrification and in most
cases better ammonification was obtained from the magnesium than
from the nonmagnesium limed soil. Soil treated with magnesium car-
bonate and inoculated into a liquid medium was shown by Lipman and
Brown {41) to retard nitrification.
It was shown by Ehrenberg (12), Lemmermann, Aso, Fischer, and Fres-
enius {37), and Wheeler, Sargent, and Hartwell {60), that when lime or
calcium carbonate was applied to soil, the decomposition of organic
matter was accelerated.
NITROGEN FIXATION
Fischer {15, 16) demonstrated the fact that both lime and magnesium
oxid increased the reproduction of Bacillus azoiobacter in soil. The oxid
of magnesium seemed to give better results than lime.
So much is Bacillus azotohacter influenced by lime that Christensen and
Larsen (9) suggested the use of this organism to measure the reaction of
soil. In a later publication Christensen (8) showed that the growth of
this organism in solution took place only when inoculated with a basic
soil. In their work with Danish soils, Weis and Bornebusch (^g) confirmed
Christensen's results. Loew (45) showed that lime added to a soil in-
creased the growth of the film of B. azotohacter formed in liquid cultures.
Results similar to Loew's were obtained by Cauda (6) when calcium car-
bonate was used.
468
Journal of Agricultural Research
Vol. XII, No. 8
Fischer (/d) obtained a better growth of Bacillus azotobacter from limed
than from unlimed clay soil. On the other hand, Koch, Litzendorf , KruU,
and Alves (32) reported that lime retarded free-nitrogen fixation. Kriiger
(5j) treated soil with lime and obtained an increase in nitrogen fixation.
Purer and better film growth of Bacillus azotobacter was obtained by
Ashby (2) with magnesium carbonate than with calcium carbonate.
The effect of calcium carbonate and magnesium carbonate on the
fixation of nitrogen by Bacillus azotobacter, both in liquid cultures and in
soil, was studied by Lipman and Burgess (40). In every case magnesium
carbonate alone proved toxic. These authors observed that when 15
parts of calcium carbonate were mixed with i part of magnesium car-
bonate, the latter was no longer toxic to this organism.
SUMMARY OF LITERATURE
From the citations just given it appears that the addition of calcium
and magnesium, either in the form of oxid or carbonate, to soil, and
especially to acid soil, brings about conditions favorable to the growth
of certain groups of microorganisms. There are many factors which have
been given little or no consideration — for instance, what relationship
exists between the total number of bacteria in soil and the quantity of
soil acid neutraHzed? With few exceptions, little attention has been
directed toward the relative effect of calcium and magnesium carbonates
on the soil flora. There exists a diversity of opinion with regard to the
relation of bacteria to these two compounds. This lack of harmony may
be due to the difference in the soil types which have been studied.
Again, it seems that no one has tried to measure the effect of calcium
and magnesium carbonates on pure cultures of bacteria in sterihzed
acid soil. To obtain information with regard to these points, a series of
experiments was planned.
EXPERIMENTAL WORK
For this study three Wisconsin soils, acid Colby silt loam, acid Plain-
field sand, and neutral Miami silt loam, were used. The Colby silt loam
was collected near Marshfield, the Plainiield sand from Hancock, and
the Miami silt loam from Madison. At the laboratory each soil was
passed through a 4-mm. sieve and thoroughly mixed.
The percentage composition of these soils is given below :
Constituent.
Colby silt
loam.
Miami silt
loam.
Plainfield
sand.
Potassium
Nitrogen
Phosphorous. ..
Calcium oxid . .
Organic matter
I- 51
. 198
. 072
. 0907
3-91
2. 16
•IS
•IS
.185
2.74
0-93
.09
.032
. 0023
I. 41
Feb. as; 19x8 Influence of Carbonates on Soil Bacteria 469
The calcium-carbonate requirement of the acid soils was determined
according to the Truog barium-hydroxid method. In calculating the
amount of acid in each soil only the active acidity was considered. For
every 100 gm. of Colby silt loam on the dry basis 1.05 gm. of calcium
carbonate were required to correct the active acidity, and for 100 gm. of
Plainfield sand 0.21 gm. Three different bases were used to neutralize
the acidity in these soils — namely, pure precipitated calcium carbonate,
pure precipitated magnesium carbonate, and commercial ground lime-
stone. The limestone, the analysis of which showed 53 per cent of calcium
oxid and 43 per cent of magnesium oxid, was ground to pass through a
loo-mesh sieve.
Aside from the compounds just named, monocalcium phosphate was
used in certain experiments. The phosphate was employed to find out
whether or not the calcium of this phosphate salt would serve in a like
capacity as that of calcium carbonate. Accordingly the monocalcium
phosphate was added to the soil alone and in various mixtures with cal-
cium carbonate.
The calcium carbonate, magnesium carbonate, and limestone were
added in amounts sufficient to satisfy one-fourth, one-half, and full cal-
cium-carbonate requirement — that is, to neutralize one-fourth, one-half,
and the total active acidity. The phosphate was added in varying
amounts. After the bases and phosphate were thoroughly mixed with
the soil, the latter was then poured into earthenware jars and the moisture
content raised to one-half saturation with distilled water. At definite
intervals samples were drawn and plate counts made. The soil of each
jar, after the sample had been drawn, was poured on sterile paper, thor-
oughly mixed, and returned to the original jar. In order to reduce evapo-
ration and to prevent outside contamination, the jars were covered with
cheesecloth. The entire series of jars was incubated in the greenhouse
at approximately 22° C.
At regular intervals the effect of these compounds on the total number
of bacteria in the soil, on ammonification, and on nitrification was
studied.
INFLUENCE OF CALCIUM CARBONATE, LIMESTONE, AND MONOCALCIUM
PHOSPHATE ON THE NUMBER OF BACTERIA IN SOIL
Colby silt loam. — Two-kgm. portions of soil were treated as outlined
in Table I and incubated for a period of five months. During this time
eight plate counts were made with Heyden-Nahrstoflf agar. These plates
were incubated at 27° C. for 10 days. The influence of monocalcium
phosphate on the number of bacteria in the soil was tested simultaneously
with that of the carbonates. The data of this experiment are given in
Table I.
470
Journal of Agricultural Research
Vol. XII, No. 8
Table I. — Influence of calcium carbonate, limestone, and monocalcium phosphate on the
number of bacteria in Colby silt loam
Treatment.
None
One-fourth calcium
carbonate
One-half calcium
carbonate
Full calcium car-
bonate
One - fourth lime-
stone
One-half limestone. .
Full limestone. . . . . .
o.s sm. monocalcitmi
phosphate
9 gm. monocalcium
phosphate
0.5 gm. monocalcium
phosphate + one-
fourth calcium car-
bonate
o.s gm. monocalcium
phosphate + full
calcium carbonate .
s gm. monocalcium
phosphate + one-
f ointh calciimi car-
bonate
3 gm. monocalcium
phosphate -f full
calcium carbonate .
Number of bacteria in i gm. of dry soil.
After I
week.
iS, 600, 000
21,000,000
20, 500, 000
18, 100, 000
21,000,000
25,300,000
22,300,000
14, 300, 000
22,300,000
16,000,000
21,300,000
14, 300, 000
16, 100,000
Rela-
tive.
After 2
weeks.
100 xg,
113
III
97
120 18.
200,000
400,000
000,000
200,000
500,000
500,000
100,000
15,500,000
21,000,000
15,000,000
31,700,000
21,100,000
20, 900, 000
Rela-
tive.
100
116
I3S
110
127
lOI
94
80
109
78
109
108
After 3
weeks.
Rela-
tive.
12,800,000
21,300,000
14,000,000
13,100,000
15,000,000
15,500,000
20,700,000,
18,000,000
20, 200, 000
18,300,000
33,500,000
26,200,000
35,600,000
204
278
After 8
weeks.
000,000
000,000
000,000
000,000
,000,000
,900,000
, 100,000
,500,000
,600,000
17,300,000
15,400,000
15,300,000
12, 500,000
Rela-
tive.
100
138
100
138
123
114
131
X03
104
132
119
117
97
After 20
weeks.
Rela-
tive.
, 700, 000
,400,000
, 700, 000
,600,000
,500,000
,500,000]
, 700,000
, 400, 000
,300,000'
100
»4S
230
203
20t
173
189
95
134
17,400,000 365
i4f3°o>o<'<' 2*S
11,600,000
11,000,000
173
164
It will be seen from the figures in Table I that in practically every case
calcium carbonate, either pure or in the form of limestone (dolomitic),
increased the growth of bacteria to a considerable extent. As compared
with the untreated soil, the favorable influence of the calcium compounds
on the number of bacteria was greatest 3, 8, and 20 weeks after treat-
ment. Apparently these compounds of calcium, especially the carbonate,
have little influence on the soil flora for the first week. This is to be
expected, since calcium carbonate is very slowly soluble.
The most striking fact noted from the results of this experiment is
the marked stimulation of the microorganisms following small applica-
tions of calcium carbonate. Figure i is a diagram showing the effect
of calcium carbonate and limestone on the total number of bacteria.
One-fourth enough calcium carbonate to neutralize the entire soil acidity,
with only one exception, showed the greatest increase in the number of
bacteria. If grouped according to their effect on the number of soil
organisms, one-fourth calcium carbonate gave the greatest gain in the
number of bacteria, one-half, the next greatest, and full, the least. The
beneficial effect of calcium carbonate extended over the entire period of
five months — that is, the treated soil gave a decided increase in the total
number of bacteria as compared with the untreated soil. In general,
pure calcium carbonate proved superior to limestone in its effect on the
bacteria of Colby silt loam soil. This superiority of calcium carbonate
Feb. 2s. 1918 Influence of Carbonates on Soil Bacteria
471
as compared with limestone was due probably to the difference in solu-
bility of the two compounds; pure calcium carbonate is more soluble
than the dolomitic limestone.
The monocalcium phosphate in small amounts apparently did not in-
crease the total number of bacteria, whereas in larger amounts, applied
alone, it was slightly beneficial. In two instances a combination of
calcium carbonate and phosphate showed an increase in the total num-
ber of microorganisms. However, in most cases the increase was no
greater than that obtained with calcium carbonate alone.
I 2 i 4 5 e 7
I U/eeA
/ 2 3 4 5 6 7
ZWeehs
I 2 i t 5 b 1
3 U/ee/(i
I z i4 5 b r
t 2 3 ■^ 5 b 7
FiO. I. — Diagram showing the influence of calcium carbonate and limestone on the number of bacteria in
Colby silt loam.
i=- no treatment.
2"=one-fotirth cacium carbonate.
3= one-half calcium carbonate.
4= full calcium carbonate.
5= one-fourth limestone.
6= one-half limestone.
7= full limestone.
A comparison of the influence of calcium from calcium phosphate
with that from calcium carbonate on the number of bacteria in Colby
silt loam is shown in figure 2. Although there are many fluctuations,
the results indicate that calcium phosphate alone or with calcium car-
bonate increases the number of bacteria much sooner than does the
carbonate alone. Here the maximum gain with the phosphate was
noted 3 weeks after treatment instead of 20 weeks, as in the case of cal-
cium carbonate.
PLAiNFiEiyD SAND. — ^The preceding experiment was repeated with a
Plainfield sand, a soil very low in organic matter. The results obtained
in this test are presented in Table II.
472
Journal of AgrictUtural Research
Vol. XII, No. 8
Tabi,E II. — Influence of calcium carbonate, limestone, and monocalcium phosphate on the
number of bacteria in Plainfield sand
Treatment.
Number of bacteria in i gm. of dry soil.
After 1
week.
Rela-
tive.
After 2
weeks.
Rela-
tive.
Afters
weeks.
Rela-
tive.
Afters
weeks.
Rela-
tive.
After 2o
weeks.
Rela-
tive.
None
One-fourth calcium
carbonate
One-half calcium car-
bonate
Full calcium carbonate.
One-fourth limestone. . .
One-half limestone
Full limestone
o.s gm. monocalcium
phosphate
a gm. monocalcium
phosphate
0.5 gm. monocalcium
phosphate -f one-
fourth calcium car-
bonate
0.5 gm. monocalcium
phosphate -f- full
calcium carbonate . . .
a gm. monocalcium
phosphate + one-
fourth calcium car-
bonate
2 gm. monocalciimi
phosphate + full
calciiun carbonate
10, 300, 000
10, 000, 000
6, 600, 000
6, 000, 000
5, 800, 000
8, 000, 000
9, 700,000
8, soo, 000
S, 000, 000
16,000,000
12,000,000
11,200,000
97
64
3,500,000
3,600,000
4, 660, 000
60-5,520,000
56|4, 660,000
77 3,800, 000
94'4, 200, 000
84
4,200,000
4,200,000
100
102
130
158
130
109
120
120
120
120
80
5,500,000
6, 000, 000
11,200,000
18,300,000
11,000,000
10,000,000
9, 700,000
5,700,000
8,900,000
14,500,000
9,400,000
9,300,000
150 10,300, 000
100 4
10916
203 7,
3327.
200 8,
180! 7,
17617,
700,000
600,000
700,000
300, 000
500, 000
700, 000
200, 000
263
4, 100, 000
3,200,000
8, 000, 000
6, 800, 000
6,600,000
164
155
181
164
154
87
68
170
144
140
172
2,450,000
4, 850, 000
4 330,000
5, 800, 000
4, 000, 000
3,000,000
4,530,000
2, 650, 000
2,000,000
3,120,000
7,770,000
3,410,000
3,660,000
197
IJ2
236
i6i
12a
184
loS
83
86
317
139
148
12 3 4 5 6 7
I z i ^ s b t
Z U/eeHi
I Z i 4 5 6 7
3 WeeAs
12 3 4 5 6 7
6 U/eehi
ZOU/eehs
Flo. a. — Diagram showing the influence of calcium carbonate and monocalcium phosphate on the number
of bacteria in Colby silt loam.
i=no treatment.
2=0.5 gm. monocalcium phosphate.
3=2 gm. monocalcium phosphate.
4=0.5 gm. monocalcitmi phosphate -f-one-fourth calduza carbonate.
5=0.5 gm. monocalcium phosphate-1-fulI calcium carbonate.
6="2 gm. monocalcium phosphate -f one-fourth calcium carbonate.
7— a gm. monocalcium phospbate-l-full calcium carbonate.
Feb. 35, 1918
Infltience of Carbonates on Soil Bacteria
473
The results reported in Table II differ somewhat from those obtained
with Colby silt loam. One week after the treatment there was a decrease
in the number of bacteria in the soil treated with both the calcium car-
bonate and limestone. However, after the first week the treated soil
showed an increase in the number of bacteria.
In contrast with the results of the Colby silt-loam experiment, the
Plainfield sand, to which one-fourth calcium carbonate or limestone was
added, did not give any marked gain in the total number of bacteria
after i and 2 weeks. After 3, 8, and 20 weeks, one-fourth calcium car-
bonate caused a slight increase in the number of soil organisms. It is
evident that one-half or full neutralization of the soil acids by the calcium
carbonate was required to give the greatest increase in the number of
bacteria (fig. 3).
12 3-^567
I Week
I Z i ^ 5 6 7
2 U/ee/(s
12^^567
3 U/eeki
12 3-9 567
d UreeJts
I Z 3 i 5 6 7
20 WeeAi
Fio. 3.— Diagram showing the influence of calcium carbonate and limestone on the number of bacteria in
Plainfield sand.
I— no treatment.
3= one-fourth calcium carbonate.
3— one-half calcium carbonate.
4= full calcium carbonate.
S= one-fourth limestone.
6= one-half limestone.
7=full limestone.
In general, light applications of limestone gave a greater increase in
the number of microorganisms than did calcium carbonate. A difference
in the nature of the soil acid in Colby and Plainfield sand probably ac-
counts for the difference in quantity of calcium carbonate required to
stimulate the reproduction of bacteria.
Where monocalcium phosphate was added alone to the sandy soil,
practically no increase in the number of bacteria was obtained. A com-
bination of the phosphate with calcium carbonate apparently did not
stimulate the multiplication of bacteria any better than did calcium
carbonate alone except where a combination of 0.5 gm. of monocalcium
474
Journal of Agricultural Research
Vol. XII, No. 8
phosphate and one-fourth calcium carbonate requirement was used. It
is surprising that this combination should favor the development of bac-
teria more than a heavier application of the same combination (fig. 4).
INFI^UENCE OF MAGNESIUM CARBONATE ON THE NUMBER OP BACTERIA
IN SOIIv
Colby silt loam. — Since pure calcium carbonate or dolomitic lime-
stone failed to give a large increase in the total number of bacteria, an
attempt was made to determine what effect magnesium carbonate would
have on the soil flora. Accordingly an experiment was planned in which
pure magnesium carbonate was added to the soil. The magnesium car-
12 3 4 5 6 7
i Weeh
12 3 4 5 6 7
2U/eeki
I 2 3 4 5 6 r
3 ureehs
1^34567
<3 Uree/o
I 2 3 4 5 b 7
20 UreeAa
Fig. 4. — Diagram showing the influence of calcium carbonate and monocalcium phosphate on the number
of bacteria in Plainfield sand.
x«>no treatment.
9=0.5 gm. monocalcium phosphate.
3= 1 gm. monocalcium phosphate.
4=>o.s gm. monocalcium phosphate+one-fourth calcium carbonate.
5—0.5 gm. monocalcium phosphate+full calcium carbonate.
6= a gm. monocalcium phosphate+one-fourth calcium carbonate.
fee 2 gm. monocalcium phosphate+full calcium carbonate.
bonate was applied to the soil in the gram-molecular equivalent of cal-
cium carbonate. The procedure in this experiment was similar to that
just described. The results of this test are presented in Table III.
TablS III. — Influence of magnesium carbonate on the number of bacteria in Colby silt
loam
Treatment.
None ,
One-fourth mag-
nesiiun carbon-
ate
One-half magne-
sium carbonate.
Full magnesium
carbonate
Number of bacteria in i gm. of dry soil.
After
I week.
Rela-
tive.
25, 200, 00c
36, 600, 000
44, 800, 000
156,000,000
14s
178
61S
After
2 weeks.
21,000,000
29, 700, 000
41,000,000
125,000,000
Rela-
tive.
After
3 weeks.
19,300,000
30, 500, 000
45, 200,000
74, 500, 000
Rela-
tive.
157
234
386
After
8 weeks.
Rela-
tive.
14, 000, 000
17. 200.000
19.400.0001 138
59,100,000 422
After
20 weeks.
8, 600, 000
II, 700,000
12,200,000
26,000,000
Rela-
tive.
136
141
30a
Feb. 25, 191S
lyifluence of Carbonates on Soil Bacteria
475
Unlike calcium carbonate and limestone, full applications of magne-
sium carbonate increased the number of bacteria far beyond the increase
obtained with the one-half and one-fourth treatments. From the
figures in this table it will be seen that the effect of the application of
magnesium carbonate to Colby soil invariably increased the reproduction
of the soil bacteria, especially during the first, second, and third week.
i Z d 4
iUTeek
/ £ 3 4
ZU/eehs
I Z 3 4
ZWeeh
6U/eeh
I z 3 4
"PJO, 5. — ^Diagram showing the influence of magnesium carbonate on the number of bacteria in Colby silt
loam.
i=no treatment.
a = one-fourth magnesium carbonate.
3 = one-half magnesium carbonate.
4= full magnesium carbonate.
Figure 5 shows very conclusively the marked effect of magnesium car-
bonate on the number of soil bacteria. For instance, the increase with
full magnesium-carbonate treatment was more than six times as great
as that of the control. The absolute numbers varied with the time of
the count, but the ratio of numbers between the different quantities of
magnesium carbonate remained almost the same throughout the entire
476
Journal of Agricultural Research
Vol. XII. No. 8
period of 20 weeks. The increase was maintained for a period of 20
weeks, but the total number of organisms decreased greatly during this
time.
Plainfield sand. — It was next arranged to determine the effect of
magnesium carbonate on the number of microorganisms in Plainfield
sand. The results are summarized in Table IV.
Here again, magnesium carbonate caused a striking increase in the
number of soil microorganisms. The results are very similar to those
obtained with magnesium carbonate in Colby soil. An enormous
increase in the number of soil organisms was noted after i and 3 weeks.
The sudden drop in numbers after the 3-week period was due probably
to a mistake in weighing which resulted in a low moisture content.
After the 8- and 20-week periods, the gradual decrease in the number of
bacteria was observed, which agrees with the results obtained in Colby
silt loam (fig. 6).
Table IV. — Influence of magnesium carbonate on the number of bacteria in Plainfield
sand
Treatment.
Number of bacteria in i gm. of dry soil.
After I
week.
Rela-
tive.
A f ter 2
weeks.
Rela-
tive.
Afters
weeks.
Rela-
tive.
After 8
weeks.
Rela-
tive.
After 20
weeks.
Rela-
tive.
None
One-fourth mag-
nesium carbonate . .
One-half magnesium
carbonate
Full magnesium car-
bonate
10, 300, 000
21, 100,000
13, 500,000
73,300,000
100
203
131
7"
3,500,000
S, 100,000
5,300,000
8,100,000
100 5,500,000
145 10,400,000
15119, 300, 000
23o\sS, 300,000
100
190
350
1,060
4, 700,000
5,800,000
10,300,006
16,000,000
100
123
219
340
2,450,000
3,560,000
5,620,000
6,530,000
100
I4S
228
266
The results secured with these two soil types show clearly that mag-
nesium carbonate in certain soils is a potent factor in the reproduction
of soil bacteria.
iNIfLUENCE OF A MIXTURE OF CALCIUM AND MAGNESIUM CARBONATE ON
THE NUMBER OF BACTERIA IN SOIL
In every case magnesium carbonate gave a much greater increase in
the number of bacteria than did calcium carbonate. Therefore the
question which suggests itself is, What effect will a mixture of calcium
and magnesium carbonates have on the soil flora?
It was shown by many investigators, principally Loew (46) and his
associates, that a soil should contain calcium and magnesium in a certain
ratio in order to secure the best plant growth. On the other hand,
Lipman (38) made the following statement :
In their behavior toward salts, bacteria differ in some respects from both plants
and animals and occupy a position by themselves.
Feb. as. 1918
Infltcence of Carbonates on SoU Bacteria
477
From his study of B. subtilis, he concluded that no antagonism exists
between calcium and magnesium. In a later publication Lipman (39)
gave an extensive review of the literature bearing on the subject of the
/ 2 3 -f
; 2 3 4
12 3 4
I z 5 A
i 2 3 4
\U/eek
2 Wee hi
3 Wee As
a U/eeAs
^0 Wee/is
Fig. 6. — Diagram showing the influence of magnesium carbonate on the number of bacteria in Plainfield
sand.
i=no treatment.
2= one-fourth magnesium carbonate.
3-= one-half magnesium carbonate.
4= full magnesium carbonate.
proper lime-magnesia ratio in soils. He concluded that there is little
evidence to support this hypothesis of the lime-magnesia ratio. In accord
with Lipman's results, certain investigators {28 ^ 29) showed that various
478
Journal of Agricultural Research
Vol. XII, No. 8
mixtures of calcium and magnesium carbonates are not favorable to the
development of soil bacteria.
To study the effect of a mixture of calcium and magnesium carbonates
on the total number of bacteria in both Colby silt loam and Plainfield
sand, a series of tests was made. This mixture was employed in amounts
equivalent to the neutralizing power of calcium carbonate — that is,
enough of the mixture was added to neutralize one-fourth, one-half, and
all of the soil acids. Each mixture was made by adding che carbonates
in gram-molecular equivalent amounts. The results are presented in
Table V.
Table V. — Influence of a mixture of calcium and m,agnesiiim, carbonates on the number
of bacteria in Colby silt loam and Plainfield sand
COLBY SILT LOAM
Number of bacteria in i gm. of dry soil.
Treatment.
After
Rela-
After
Rela-
After
Rela-
After
Rela-
After
Rela-
I week.
tive.
2 weeks.
tive.
3 weeks.
tive.
8 weeks.
tive.
20 weeks.
tive.
ig, 200, ooo
100
27,600,000
100
27,300,000
100
8, 700, 000
100
One-fourth:
One-half caldumj
carbonate 1
One-half magne-j
slum carbonate. J
One-half:
One-half calcium]
36, 500, 000
190
29, 000, 000
105
27,500,000
100
20, 700,000
I4S
15,100,000
173
carbonate 1
One-half magne-|
slum carbonate. J
Full:
One-half calciumi
14, 700,000
76
32,500,000
117
38, 600, 000
104
17,000,000
120
12, 200,000
140
carbonate 1
One-half magne-|
slum carbonate. J
31,800,000
165
4S, 300, 000
163
44,500,000
162
20, 000, 000
140
I3j300,ooo
153
PLAINFIELD SAND
10,300,000
100
3,500,000
100
S, 000, 000
100
4, 700, 000
100
2,450,000
100
One-fourth:
One-half calciumi
carbonate 1
One-half magne-|
slum carbonate. J
One-half:
One-half calciumi
11,300,000
109
3,200,000
91
11,700,000
234
5,500,000
117
4,250,000
173
carbonate 1
One-half magne-|
sivun carbonate. J
Full:
One-half calciumi
11,800,000
"S
3,800,000
109
I3>300,ooo
266
8, 400, 000
180
3,500,000
143
carbonate 1
One-half magne-|
sium carbonate. J
13', 100, 000
127
5,000,000
143 18,400, 000
370
8, 600, 000
184
2,630,000
107
It will be seen from the figures of the table that a mixture of calcium
and magnesium carbonates increased the number of bacteria in CoBby
silt loam and Plainfield sand. The most interesting fact ascertained
from this test is that magnesium carbonate plus calcium carbonate is
less efficient in its efifect on the reproduction of bacteria than the equiva-
lent weight of magnesium alone.
Feb. 35, 1918 Influence of Carbonates on Soil Bacteria
479
INFLUENCE OF CALCIUM CARBONATE, CALCIUM CHLORID, MAGNESIUM
CARBONATE, MAGNESIUM CHLORID, DIBASIC MAGNESIUM PHOSPHATE,
AND MONOCALCIUM PHOSPHATE ON THE BACTERIA IN SOIL
Colby silt loam. — Since the increase in number of bacteria in soil
treated with magnesium carbonate seemed far too great to be accounted
for by the correction of the soil acidity alone, an experiment was made
to study the efifect of a neutral salt of magnesium, and a magnesium
phosphate, on the number of bacteria in Colby silt loam soil. Two
points were considered in planning this test : First, the action of the mag-
nesium and calcium ions on the bacteria; and second, the possibility
of the combining of the calcium or magnesium, especially the latter, with
the phosphate of the soil, thus liberating the phosphate in a more avail-
able form.
It was reported by Truog (56) that magnesium phosphate favors the
phosphorus assimilation by plants more than any other phosphate. If
this be true, then the bacteria (lower plant life) should be favored by a
phosphate when in this form. Therefore the question arises, Does the
magnesium carbonate when added to soil react with the phosphates to
form magnesium phosphate?
In order to make these tests, magnesium carbonate, magnesium chlo-
rid, dibasic magnesium phosphate, also calcium carbonate, calcium chlo-
rid, and monocalcium phosphate were added to the soil alone and in
various mixtures.
TabIvE VI. — Influence of calcium carbonate, calcium chlorid, magnesium carbonate,
magnesium chlorid, dibasic tnagnesium phosphate, and m.onocalcium phosphate on the
bacteria of Colby silt loam
Treatment.
Number of bacteria in i gm . of dry soil.
After I
week.
Rela-
tive.
After 2
Rela-
Afters
Rela-
After 8
weeks.
tive.
weeks.
tive.
weeks.
19, 000, 000
100
23,200,000
100
12,200,000
22, 100,000
116
14,000,000
60
17,700,000
17,600,000
92
27,500,000
118
10,500,000
37,600,000
197
43,300,000
186
26, 200,000
121, 100,000
637
200,000,000
862
58, 700, 000
13, 100,000
67
14,000,000
60
15,600,000
9,300,000
47
24,000,000
103
11,600,000
66, 100, coo
347
33,100,000
141
26,000,000
47,300>o°o
249
46,200,000
199
31,500,000
190,000,000
1,000
106, 000, 000
456
85,500,000
144,000,000
75S
197,000,000
849
78,000,000
Rela-
tive.
None
0.1 per cent magnesitun chlo-
rid
0.1 per cent calcium chlorid. . .
Full calcium carbonate
Full magnesium carbonate . . .
0.1 per cent dibasic magne-
sium phosphate
0.1 per cent monocalcium
phosphate
0.1 per cent dibasic magne-
sium phosphate-f-fuU cal-
cium carbonate
0.1 per cent monocalcium
phosphate-Ffull calcium
carbonate
0.1 per cent dibasic magne-
sium phosphate-l-full mag-
nesium carbonate
O.I per cent monocalcium
phosphate + full magne-
sium carbonate
28,000,000
16,000,000
18, 200,000
S9, 100, oco
165,000,000
30, 500, 000
38,000,000
74, 100, 000
38,000,000
191,000,000
223,000,000
57
65
31S
590
264
I3S
6S2
800
I4S
86
314
481
a 13
as8
700
639
The effect of magnesium phosphate on the number of bacteria was com-
pared with that of magnesium carbonate, calcium phosphate, and cal-
cium carbonate. If the action of the magnesium and calcium carbonates
27811°— 18 2
48o
Journal of Agricultural Research
Vol. XII, No. 8
on the bacteria is derived from the basic part of these compounds, then
the salts of these substances — namely, magnesium and calcium chlorid —
should increase the number of bacteria in a similar manner. These salts
were added in quantities equivalent to o.i per cent of the dry weight of
the soil. The data for this experiment are presented in Table VI.
The chlorids of calcium and magnesium failed to increase the number
of bacteria in Colby silt loam. Instead, the majority of cases showed a
loss, especially with calcium chlorid. Calcium and magnesium carbo-
nates, in agreement with a previous study, increased the number of
microorganisms. Magnesium carbonate stimulated the development of
the bacteria far greater than did calcium carbonate.
When the dibasic magnesium phosphate and monocalcium phosphate
were applied alone there did not seem to be any gain in the number of
bacteria. The mixtures of monocalcium phosphate and dibasic mag-
nesium phosphate with magnesium carbonate resulted in all except one
period in an increase in the number of bacteria beyond that obtained by
the use of magnesium carbonate alone. The mixtures made by adding
these same phosphates with calcium carbonate did not result in an
increase in the number of bacteria beyond that noted where calcium car-
bonate was used alone.
From the data of this experiment, which are diagramed in figures 7
and 8, it seems safe to conclude that neither the phosphates nor the
chlorids of magnesium and calcium alone caused a marked multiplica-
tion of the bacteria in Colby silt loam.
Miami silt loam. — A similar test of these different compounds was
made with Miami silt loam, a neutral soil. The chlorids of magnesium
and calcium were eliminated in this experiment. Since the effect of a
carbonate on the number of bacteria in Miami silt loam soil was not
known, the carbonates were added in amounts equal to the one-half and
full calcium carbonate requirement of Colby silt loam. The data for
this experiment are presented in Table VII.
Table VII. — Influence of calcium carbonate, magneshim carbonate, dibasic magnesium
phosphate, and monocalcium phosphate on the number of bacteria in Miami silt loam
Treatment.
None
One-half calcium carbonate
Full calcium carbonate
One-half magnesium carbonate
Full magnesium carbonate
0.1 per cent monocalcium phosphate
0.1 per cent dibasic magnesium phosphate. .
0.1 per cent monocalcium phosphate -h full
calcium carbonate
0.1 per cent dibasic magnesium phosphate
-t-full calcium carbonate
0.1 per cent monocalcium phosphate + full
magnesium carbonate
0.1 per cent dibasic magnesium phosphate
+full magnesium carbonate
Number of bacteria in i gm. of dry soil.
After 1 week,
10, 000, 000
10,300,000
13,800,000
17,000,000
27,000,000
10,600,000
15, 700,000
12,400,000
II, 100,000
20, 000, 000
21,600,000
Rela-
100
103
13S
170
270
106
157
200
216
After 2
weeks.
18,500,000
18,500,000
23 , 700, 000
25,400,000
52,400,000
20. 400, 000
27,000,000
24, 000, 000
20, 000, 000
35,000,000
43 , Soo, 000
Rela-
tive.
100
100
128
137
283
no
146
130
108
1S9
236
After 3 Rela-
weeks. tive.
9, 600, 000
13,800,000
15, 100,000
15, 500,000
21,600,000
12,300,000
14,300,000
13 , 500, 000
14, 600, 000
19, 000, 000
II, 700,000
100
143
IS7
161
22s
12S
ISO
140
isa
198
Feb. 25. 1918 Influence of Carbonates on Soil Bacteria
481
i 2 5 4 5
t U/eeA
/ i 3 4 5
12 5 4 5
3 Wee As
i Z i 4 5
6 U/eeA&
Fig. 7.— Diagram showing the influence of the carbonates and chlorids of magnesium and calcium on the
number of bacteria in Colby silt loam soil.
i=no treatment.
2= O.I per cent magnesium chlorid.
3=0.1 per cent calcium chlorid.
4=full calcium carbonate.
S=full magnesium carbonate.
482
Journal of Agricultural Research
Vol. XII, No. 8
12 3^667
12 5^367
I 2 3 t 5 6 7
I 2 5 -i 5 6 7
eOTee/ci
FiQ. 8.— Diagram showing the influence of dibasic magnesitun phosphate, monocalcium phosphate, calcium
carbonate, and magnesium carbonate on the number of bacteria in Colby silt loam.
1= no treatment.
a=o.i per cent dibasic magnesium phosphate.
3=0.1 per cent monocalcium phosphate.
4=0.1 per cent dibasic magnesium phosphate+
full calcium carbonate.
S=o.i per cent monocalcium phosphate+fuU
calcium carbonate.
6=o.i per cent dibasic magnesium phosphate+fuU
magnesium carbonate .
7=o.i per cent monocalcium phosphate+full mag-
nesium carbonate.
Feb. 25, 1918 Influence of Carbonates on Soil Bacteria
483
From the results which are given in Table VII and from figure 9 it
will be seen that both magnesium and calcium carbonate increased the
number of bacteria in Miami silt loam soil. Here, again, the increase
was much greater with magnesium carbonate than with calcium carbo-
nate. The greatest increase was obtained two weeks after treatment.
In all cases, the heavier applications increased the number of bacteria
more than did the smaller ones.
^ 3 4 5 6 7
tu/ceh
Z i ^ 5 6 7 Q 9 /O II
2 U/eeAi
2 i 4 5 b 7 & 1 10 ti
3 U/iseAs
Fig. 9. — Diagram showing the influence of calcium carbonate, magnesium carbonate, dibasic magnesium
phosphate, and monocalcium phosphate on the number of bacteria in Miami silt loam.
1= no treatment.
2= one-half calcium carbonate.
3= full calcium carbonate.
4= one-half magnesium carbonate.
S=full magnesimn carbonate.
6=0.1 per cent monocalcium phosphate.
7=0.1 per cent dibasic magnesium phosphate.
8=0.1 per cent monocalcimn phosphate+fuU calcium carbonate.
9=0.1 per cent dibasic magnesium phosphate-hfuU calcium carbonate.
10= O.I per cent monocalcium phosphate -l-full magnesium carbonate.
11=0.1 per cent dibasic magnesimn phosphate+full magnesium carbonate.
The results differentiate themselves from those obtained with acid
Colby silt loam soil since the phosphates increased the number of bacteria
in the neutral Miami silt loam. It was expected that these phosphates
would have a more beneficial effect in neutral than in acid soil, since it
is probable that the phosphate was hydrolyzed, thus adding more acid
to the soil. In every test magnesium phosphate gave a larger increase
in number of bacteria than was obtained by the use of calcium phos-
484 Journal of Agricultural Research voi. xii.no. s
phate. The dibasic magnesium phosphate increased the number of
bacteria to about the same degree as an application of one-half magne-
sium carbonate. A combination of each phosphate with each carbonate
did not prove beneficial in augmenting the number of bacteria in Miami
silt loam beyond that caused by each carbonate when used alone. It is
evident from the foregoing data, as shown in figure 9, that the relative
increase of the number of bacteria from the use of carbonates of calcium
and magnesium in Miami silt loam was not so great as when Colby silt
loam was treated with the same compounds. However, the phosphate
gave a greater relative increase in the neutral Miami silt loam than in
the acid Colby silt-loam soil.
INFlvUENCS 0]P MAGNESIUM CARBONATE, CAI^CIUM CARBONATE, LIMESTONE,
AND MONOCALCIUM PHOSPHATE ON THE BACTERIA IN SOIL,
Another series of jars was filled with Colby silt loam treated with
magnesium carbonate; a second series with the same soil treated with
calcium carbonate; a third series with limestone, and a fourth with
monocalcium phosphate. All jars were placed in the greenhouse and
the soil moisture maintained at half saturation. After one, two, and
three months samples of this soil were drawn and tested for ammonifying
power, for nitrate content, and for the number of bacteria.
The ammonifying power was determined by mixing 2 per cent of
dried blood, which contained 13.4 per cent of nitrogen, with 100 gm. of
soil. After adding the dried blood meal, the soil was placed in tumblers,
the proper amount of water added, the tumblers covered with petri-dish
covers, and incubated at 27° C. for six days. The ammonia was deter-
mined by the steam-distillation method.
In order to measure the nitrification in the treated and untreated soil,
samples were taken from the jars and the nitrate content determined
immediately. This was simply a study of nitrate accumulation in
the soil; no nitrogenous substance was added. Nitrates were deter-
mined by the phenol-disulphonic acid method. The data for the experi-
ments with Colby silt loam and Plainfield sand are shown in Tables
VIII to XI, inclusive.
From the data in Table VIII it will be seen that after one month the
increase in the number of bacteria in Colby soil, in conformity with
previous tests, was greatest with the magnesium-carbonate treatment.
The increase was very marked, about five times greater than that de-
rived from the use of calcium carbonate or limestone. The favorable
effect of the magnesium carbonate was noted after both the 2- and
3-month periods. In most cases the calcium carbonate and limestone
increased the number of bacteria, notably after two months. The
monocalcium phosphate apparently exerted no effect in increasing the
Feb. as, 1918
Influence of Carbonates on Soil Bacteria
485
number of bacteria, since it gave no increase except when used in combi-
nation with calcium carbonate, and then the increase was no greater
than that obtained with calcium carbonate alone.
TablB VIII. — Influence of calcium carbonate, magnesium carbonate, limestone, and
monocakium phosphate on the number of bacteria in Colby silt loam
Treatment.
Number of bacteria in i gm. of dry soil.
After
Rela-
I month.
tive.
14, 000, 000
100
14,370,000
102
12, 200,000
87
14,370,000
102
12,200,000
«7
14,370,000
102
13.340,000
95
15, 180,000
108
26,000,000
185
101,400,000
724
15,000,000
107
18,100,000
130
11,380,000
80
After
2 months.
Rela-
tive.
After
3 months.
Rela-
tive.
None
One-fourth calcium carbonate
One-half calcium carbonate
Full calcium carbonate
One-fourth limestone
One-half limestone
Full limestone
One-fourth magnesium carbonate
One-half magnesium carbonate
Full magnesium carbonate
0.1 per cent monocalcium phosphate
0.1 per cent monocalcium phosphate -1- one-
fourth calcium carbonate
0.1 per cent monocalcium phosphate -1- full
calcium carbonate
10, 700, 000
13,400,000
17,000,000
14, 800, 000
17,600,000
II, 200,000
14, 300, 000
11,900,000
25, 700,000
43 > 700, 000
10, 700, 000
10,800,000
13,600,000
100
158
138
164
104
133
III
240
lOI
127
8, 000, 000
8,800,000
7, 100,000
8, 700,000
10, 300, 000
8, 700, 000
10, 700,000
10, 900, 000
13, 200,000
33,500,000
8,300,000
7, 600, 000
10.300,000
100
no
83
108
128
loS
133
136
i6s
418
103
95
128
The data in Table IX are in agreement with those obtained with Colby
silt loam — that is, magnesium carbonate increased the number of bac-
teria in Plainfield sand to a considerable extent beyond that produced
by calcium carbonate or Umestone. The effect of the magnesium car-
bonate on the number of bacteria was most noticeable one month after
the treatment was applied. In general, limestone proved inferior to
calcium carbonate in stimulating the number of bacteria in Plainfield
sand, except that in one case the monocalcium phosphate when applied
alone did not increase the number of bacteria.
Table IX. — Influence of calcium carbonate, magnesium carbonate, limestone, and
monocalcium phosphate on the number of bacteria in Plainfield sand
Treatment.
Number of bacteria in i gm. of dry soil.
After
Rela-
I month.
tive.
6, 800, 000
100
6, 220, 000
91
6, 100,000
90
8, 770,000
128
S, 600, 000
83
9, 180,000
I3S
7, 700,000
"3
7, 240, 000
164
12,640,000
185
44,800,000
658
4,380,000
64
7,550,000
III
6,620,000
97
After
Rela-
3 months.
tive.
6,550,000
100
5,440,000
83
7,800,000
119
8, 440, 000
128
5, 900, 000
90
4, 900, 000
74
7, 660, 000
116
S, 400, 000
82
7,300,000
III
28, 500, 000
435
3,000,000
45
4, 670, 000
71
4, 500, 000
68
After
3 months.
Rela-
tive.
None
One-fourth calcium carbonate
One-half calcium carbonate
Full calcium carbonate
One-fourth limestone
One-half limestone
Full limestone
One-fourth magnesium carbonate
One-hall magnesium carbonate
Full magnesium carbonate
0.1 per cent monocalcium phosphate
0.1 per cent monocalcium phosphate-l-one-fourth
calcium carbonate
0.1 per cent monocalcium phosphate-1-full cal-
cium carbonate
2, 100,000
2,450,000
3, 200,000
2,900,000
I, 780, 000
2,550.000
2,550,000
3,000,000
5,000,000
4, 800, 000
2, 340, 000
3,230,000
2, 700,000
100
116
152
138
84
121
121
133
238
238
III
154
128
486
Journal of Agricultural Research
Vol. XII, No. 8
FORMATION OP AMMONIA
The ammonia determinations were made at the end of two and three
months. It will be seen from the data in Table X that all treatments
accelerated the formation of ammonia from dried blood. After two
months monocalcium phosphate plus calcium carbonate showed the
greatest effect, although the phosphate-alone treatment gave nearly as
good results. Of the carbonates, calcium carbonate appeared to have
the greatest effect in increasing the ammonifying power, an effect nearly
as great as that obtained from the phosphate applications. Magnesium
carbonate exerted a less effect than did calcium carbonate, and lime-
stone had even a less effect than was obtained with magnesium car-
bonate. After three months the phosphate still retained a lead in
stimulating ammonification. Limestone caused a greater increase in
ammonia-producing power than did calcium carbonate, while calcium
carbonate gave a greater increase than was obtained with magnesium
carbonate. From the data on Colby soil it is shown that the increase
in ammonifying power is not always parallel with the increase in the
total number of organisms.
Table X. — Influence of cakium carbonate, magnesium carbonate, limestone, and thouo-
calcium phosphate on the ammonification of dried blood in Colby silt loam and Plain-
field sand
Ammonia nitrogen in loo gm. of dry soil.
Treatment.
Colby silt loam.
Plainfield sand.
After 2
months.
In-
crease.
After 3
months.
In-
crease.
After 2
months.
In-
crease.
After 3
months.
In-
crease.
None
Mgm.
21- 1
39- o
Mgm,.
Mgm.
51-7
S5-5
61.6
62.3
61.6
64-5
69.7
S8.i
53-7
52- S
63.1
78.0
80.6
Mgm.
Mgm.
49.0
47-7
40. 1
40. 1
51.8
38.7
37-7
40-3
39-9
35-2
S6.7
SO. 5
48-5
Mgm..
Mgm.
SS-i
S2.2
47.0
S3- 1
Sl-o
51-3
73-8
46.7
47- S
47.0
75-3
73-3
65.7
Mgm.
One-fourth calcimn carbonate
+ 1-3
+ 3-8
-1- 9-9
-I-I0.6
+ 9-9
-fl2.8
-hiS.o
+ 6.4
+ 2.0
+ 1.8
-i-ll.4
+26.3
+ 28.9
- 1-3
- 8.9
- 8.9
-f 2.8
-10.3
-1 1. 3
- 8.7
- 9.1
-13-8
+ 7-7
-t-i-s
- o-S
— 2.9
Full calcium carbonate
One-fourth limestone
S3-0
45-7
40.6
44.2
49.0
51.6
57-1
6i. 0
57-5
72- S
+ 15-3
-f 8.0
-1-2.9
+ 6.S
+ 11. 3
+ 13-9
+ 19-4
+23-3
+ 19-8
+34-8
- 2.0
- 4-1
- 3-8
+18.7
- 8.4
- 7-6
- 8.0
One-half limestone
Full limestone
One-fourth magnesium car-
bonate
One-half magnesium carbon-
ate
Full magnesium carbonate . . .
O.I per cent monocalciiun
phosphate ....
O.I per cent monocalciimi
phosphate-)-one-fourth cal-
cium carbonate
-1-18. a
O.I per cent monocalcium
phosphate-(-full calcium car-
bonate
-f-io.6
The data in Table X for Plainfield sand showed very different results
from those obtained with Colby silt loam. After two months a decrease
in ammonifying power was noted, except in the case of the one-fourth
limestone treatment. The phosphate when applied alone gave the
greatest increase, but where calcium carbonate was added in combina-
tion with the phosphate, the formation of ammonia was not so great.
Feb. as, 1918
Influence of Carbonates on Soil Bacteria
487
With an increase of calcium carbonate there was a decrease in ammonia
production. This decrease was shown with all carbonates, magnesitun
carbonate causing the greatest decrease.
After three months the same order was held by these compounds in
stimulating ammonia formation. At this time the full limestone had a
greater effect than the smaller limestone applications; the effect was
almost as great as that due to the phosphate treatment. It is evident
from the results obtained in both Colby silt loam and Plainfield sand,
that the greatest accumulation of ammonia (six days' incubation) does
not occur where the largest increase in the number of bacteria was
obtained. Where the greatest number of bacteria developed, it seems
probable that the greatest amount of ammonia should be formed. Since
the substances which gave the highest number of bacteria also neutral-
ized the soil acids, it is probable that the ammonia partly escaped through
volatilization. Because of the open texture of the Plainfield sand, more
ammonia escaped from this soil than from the Colby silt loam.
Table XI. — Influence of calcium carbonate, magnesium carbonate, limestone, andm,ono-
calcium phosphate on nitrate accu7nulation in Colby silt loam and Plainfield sand
Treatment.
None
One-fourth calcium car-
bonate
One-half calcium carbo-
nate ,
Full calcium carbonate. . ,
One-fourth limestone
One-half limestone
Full limestone
One-fourth magnesium
carbonate
One-half magnesium car-
bonate
Full magnesiinn carbon-
ate
0.1 per cent monocal-
cium phosphate
0.1 per cent monocal-
cium phosphate -t- one-
fourth calcium carbon-
ate
0.1 per cent monocal-
cium phosphate 4- full
calcium carbonate
Nitrate nitrogen accumulated in 100 gm. of dry soil.
Colby silt loam.
Mgm.
S-86
7-23
6. 10
6.25
5- S3
S-86
6.51
6.58
8.13
13-28
S-63
6. 10
6.73
Mgm.
•24
•37
— -33
.00
•6S
•72
2. 27
7.42
— -23
.24
.87
•^ o
Mgm.
6.05
6-94
7-s6
8.18
6.0s
6. 40
6. 42
6.75
8.44
13.00
6.05
6. 42
Mgm.
0.89
I- SI
2. 13
.00
•35
•37
.70
2-39
6.9s
.00
•37
2-13
Mgm.
6.62
8.36
9^37
12.00
7.2s
7-25
8.56
9-43
12.00
IS- 00
7-2S
Mgm.
1.94
2^7S
5^38
•63
•63
•94
2.81
S-38
8.38
.63
2-75
4-63
Plainfield sand.
a 5
Mgm.
1.74
2.42
2.3s
2. 20
a. 20
1.87
1.97
a. 10
2. 20
2^34
2. 12
Mgm..
.61
.46
.46
•13
•23
•36
.46
.60
•38
.61
.06
^a
Mgm.
1.S8
i.8s
2.31
2. 22
1.80
1.67
1.87
1.24
2.78
1.85
Mgm..
•73
.64
. 22
.09
•29
■ -34
.64
I. 20
•27
^a
Mgm
^•33
1.67
3-55
2.92
1-73
2. 00
3.80
1. SO
2. 70
4.62
3-oS
Mgm.
0.34
1.23
i^S9
.40
.67
1.47
•17
^•37
3- 29
I- 75
•87
I-7S
ACCUMULATION OP NITRATES
From the data presented in Table XI it will be seen that nitrates
accumulated faster in both the treated Colby silt loam and Plainfield
sand than in the untreated. In both soils the magnesium carbonate
benefited nitrification more than did the other carbonates. Next in
order to magnesium carbonate was calcium carbonate, and lastly, lime-
488
Journal of Agricultural Research
Vol. XII. No. 8
stone, in increasing the nitrate content of Colby silt loam and Plainfield
sand; the heavier applications gave the highest nitrate accumulation.
In the case of the magnesium carbonate with Colby silt loam soil, the
increased accumulation reached as high as 8 mgm. The increase in
nitrate in the Plainfield sand ran parallel with that in the Colby silt loam
where similarly treated, though the accumulation in the sand was much
smaller, since this soil is low in organic matter.
t :^ 3 -^ S 6 7 Q 9 /O imJ3
I month
I £ i 'f 5 b 7 6 9
2mor,ih%
3 fVoni/is
Fio. lo. — Diagram showing the influence of calcium carbonate, magnesium carbonate, limestone, and
monocalcium phosphate on nitrate accumulation of Colby silt loam.
i=no treatment.
3= one-fourth calcium carbonate.
3= one-half calcium carbonate.
4= full calcium carbonate.
5= one-fourth limestone.
6= one-half limestone.
7= full limestone.
8= one-fourth magnesium carbonate.
9= one-half magnesium carbonate.
io= full magnesium carbonate.
11= O.I per cent monocalcium phosphate.
1 2= O.I per cent monocalcium phosphate+one-fourth calcium carbonate.
13=0.1 per cent monocaldum phosphate-1-full calcium carbonate.
The monocalcium phosphate apparently did not favor nitrification in
the Colby soil, while in the sand this substance proved beneficial. A
combination of the phosphate and calcium carbonate did not increase
nitrification beyond that obtained from the use of calcium carbonate
alone. The effect of these substances on the reproduction of soil
organisms and on the formation of nitrates was similar. A review of
the entire data is shown graphically in figure 10.
Feb. 25, 1918
Influence of Carbonates on Soil Bacteria
489
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, LIME-
STONE, AND MONOCALCIUM PHOSPHATE ON THE NITRIFICATION OF
GELATIN IN SOIL
In order to study the effect of calcium carbonate, magnesium car-
bonate, limestone, and monocalcium phosphate on nitrification in Colby
and in Plainfield sand to which nitrogenous material was added, a series
of tests was made. After the soils had been treated w^th the forenamed
substances separately and allowed to incubate for three months, loo-gm.
samples were drawn and to each portion 23.6 mgm. of gelatin were added.
The soil was placed in tumblers, which were covered with glass, and
incubated for six weeks at 27° C. At the end of this time the nitrate
content was determined. The results of this experiment are given in
Table XII.
Table XII. — Influence of calcium carbonate, magnesium carbonate, limestone, and mono-
calcium, phosphate on the nitrification of gelatin in Colby silt loam and in Plainfield
sand
Treatment.
Nitrate nitrogen in loo gm. of dry soil.
Colby silt loam.
After 6
weeks.
Increase.
Plainfield sand.
After 6
weeks.
Increase.
None ,
One-fourth calcium carbonate
One-half calcium carbonate
Full calcium carbonate
One-fourth limestone
One-half limestone
Full limestone
One-fourth magnesium carbonate
One-half magnesiiun carbonate
Full magnesium carbonate
0.1 per cent monocalcium phosphate
0.1 per cent monocalcium phosphateH-one-fourth calcium car-
bonate
0.1 per cent monocalcium phosphate-t-full calcium carbonate. .
Mgm.
20.87
22.31
25.00
28.43
20.87
23.12
25.00
23-12
26. 12
25.09
22.31
23-12
28.43
Mgm.
1-44
4-13
7.56
.00
8. 25
4-13
3.25
S-25
4-13
1.44
3.25
7-56
Mgm..
9.06
11.70
10.57
zi. 12
II. 70
II. 70
II. 12
II. 12
13.09
12.24
9.06
13.09
12.24
Mgm.
2.64
1.51
2.06
3.64
3.64
3.06
3. 06
4-03
3- 18
.00
4-03
3- 18
In reviewing Table XII it will be seen that with the exception of one-
fourth limestone treatment in Colby loam, and phosphate alone in
Plainfield sand, there was an increase in nitrification in the treated
soils. The increase in most cases was very slight, especially in the sand,
where only about half of the nitrogen of gelatin apparently was nitrified,
while in the Colby loam the greater part of the nitrogen of gelatin was
nitrified. A comparison of the amount of nitrate formed in the soil to
which magnesium or calcium carbonate was added, both with and without
the addition of gelatin (Table XI), showed a relatively greater nitrate
formation in the soils to which no organic nitrogen was added. These
results were to be expected, since it is likely that some of the ammonia
formed from the gelatin in the neutral or partly neutral soil escaped.
On the other hand, the combined effects of the gelatin and the carbonates
increased the multiplication of bacteria beyond that of the soil treated
with gelatin alone. The great gain in the number of bacteria is no doubt
490
Journal of Agricultural Research
Vol. XII, No. 8
followed by an increase in assimilation of nitrates. These results agree
with those obtained by Fred and Graul (20).
In the Colby silt loam the full calcium carbonate gave the best results
while magnesium carbonate and limestone rank in the order named.
The magnesium carbonate when added to the soil with gelatin did not
give as good results as calcium carbonate. This difference may be due to
the fact that the large increase in the number of bacteria caused by the
magnesium carbonate treatment favored a greater assimilation of nitrate
by the microorganisms. Plainfield sand apparently gave better results
where the one-fourth or one-half neutralization was obtained. Very
probably the sand releases more ammonia when neutral than does the
Colby silt loam. In both soils the phosphate benefited nitrification but
slightly.
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, LIMESTONE,
AND MONOCALCIUM PHOSPHATE ON NITROGEN FIXATION IN SOIL
An effort was made to study the influence which carbonate and phos-
phate treatments would have on the independent nitrogen-fixing organ-
ism, Bacillus azotobacter chroococcum. Three months after treatment,
Colby and Plainfield soils were sampled and the soil placed in large soup
plates, I per cent of mannit was added to each plate of soil, which was
then inoculated with a culture of B. azotobacter. After incubating for
four weeks in the greenhouse, the soils were dried and ground to a very
fine powder in a ball mill. Total nitrogen determinations (Kjeldahl
method modified to include nitrates) were then made. Duplicate deter-
minations were made from each plate and an average of these taken.
The results for this experiment are given in Table XIII.
Table XIII. — Influence of calcium carbonate, magnesium carbonate, limestone, and
monocakium phosphate on nitrogen fixation in Colby silt loam and Plainfield sand
treated with mannit and inoculated with Bacillus azotobacter
Treatment.
None
One-fourth calcium carbonate
One-half calcium carbonate
Full calcium carbonate
One-fourth limestone
One-half limestone
FuU limestone
One-fourth magnesium carbonate
' One-half magnesium carbonate
Full magnesium carbonate
0.1 per cent monocalcium phosphate.
0.1 per cent monocalcium phos-
phate + one-fourth calcium car-
bonate
0.1 per cent monocalcium phos-
phate + full calcium carbonate
Nitrogen in loo gm. of dry soil.
Colby silt loam.
After 4
weeks.
Mgm.
273.0
277. o
273.0
276.0
27S.O
273.0
278.0
277.0
278.0
275.0
274.0
2 79- o
283.0
Mgm.
275.0
276. o
275.0
276.0
275-0
27S-0
279. o
276.0
276.0
27S-0
276.0
282. o
277. o
Aver-
age.
Mom.
275.0
276.5
274. o
276.0
275- o
274.0
278.5
276.5
277. o
275- o
275-0
280.5
280. o
In-
crease.
Mgm..
+ 5-5
+ S-0
Plainfield sand.
After 4
weeks.
Mgm.
63.0
59- o
6i. o
64.0
6i.o
61. o
62. o
59- o
62. o
63.0
61. o
61. o
65.0
Mgm.
61.
59- o
65.0
Aver-
age.
Mgm..
0
0
59
0
61
0
63
0
60
0
62
0
63
0
59
.0
62
.0
63
0
61
60. o
65. o
In-
crease.
Mgm.
—2.0
-I-3-0
Feb. as. 1918 Influence of Carbonates on Soil Bacteria 491
From the data presented in Table XIII it appears that there was no
decided gain in total nitrogen in the Plainfield sand, and only a small gain
in Colby silt loam. It is difficult to explain why no gain, and in some
cases a loss, of nitrogen was found. The loss may have been due to a
disturbance in the balance of the flora of these soils. Since the mannit
which was added to the soil, and the neutralization of some of the soil
acids by the basic substances caused a great increase in the number of
bacteria, it is possible that a part of the nitrate nitrogen was set free as
elemental nitrogen. However, a gain in total nitrogen is shown in the
Colby silt loam soil due to the carbonate and limestone treatment.
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, AND
LIMESTONE ON ORGANISMS IN PURE CULTURE
An effort was made to determine the influence of calcium carbonate,
magnesium carbonate, and limestone on pure cultures of bacteria in steril-
ized soil. In order to simplify the vv^ork, only Colby soil was employed.
It was obser\^ed that when this soil is heated for a long period at a
high temperature, there is a reduction in soil acidity. Other investi-
gators have reported similar results in acid soils. Conner (10) showed a
decrease in acidity when soil was heated to complete dryness. He sug-
gested that possibly the acid silicates were hydrolyzed with the formation
of a base. On heating soil to 140° C. for one hour. Sharp and Hoagland
(55) noted a decrease in acidity. On the other hand, Schreiner and
Lathrop (54) increased the acidity in an acid-reacting soil by heating the
soil for three hours at 30 pounds' pressure, which in all probability
changed the organic matter to such an extent as to increase the acidity.
The work of Kelley and McGeorge (jo) and Darbishire and Russell (//)
showed a change in the soil consituents on heating, especially in the solid
inorganic constituents, which were made more soluble.
In order to overcome as much as possible this reduction in acidity,
various methods of sterilization were tried. Both dry and moist soils
were sterilized in Erlenmeyer flasks, which were heated in steam under a
pressure of 15 pounds for three hours. On testing this heated soil for
its degree of acidity by the Truog zinc-sulphid test, the dry soil showed
practically no change in its acid content, while the moist soil, one-half
and full water-saturated, showed a considerable decrease. The very wet
soil after sterilization contained less acid than that which received one
half as much water. From the evidence it seems safe to conclude that
the silicates of this soil are partially hydrolyzed with the formation of
bases.
Further tests were conducted with dry soil. It was dried for 24
hours at about 45° C. and placed in flasks; these were then plugged and
sterihzed for three hours at 15 pounds' pressure. After sterilization, the
soil received water sufficient to bring it to half saturation. This was
done to determine if hydrolysis took place by adding water after the soil
492
Journal of Agricultural Research
Vol. XII. No. 8
had been sterilized. At the end of four days tests were made on the
dry and on the watered soil. The acidity of both appeared unchanged.
From the results of the foregoing experiments it appears that dry
Colby soil can be sterilized and sterile water added until the soil is half
saturated, without interfering seriously with the original reaction of
the soil. Therefore this method of sterilizing soil was adopted for all
of the pure-culture work.
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, AND LIME-
STONE ON THE AMMONIFICATION OF BLOOD MEAL BY PURE CULTURES
OF BACTERIA IN SOIL
Colby silt loam. — One-hundred-gm. portions of dry soil were each
treated with calcium carbonate, with magnesium carbonate, and with
limestone separately. To the soil of the entire series 2 per cent of dried-
blood meal was applied. The treated soil was placed in 300-c. c. Erlen-
meyer flasks and sterilized as previously stated. Sterilized water was
then added to bring the soil to half saturation. The sterilized soil was
then inoculated with water suspensions of Bacillus tumescens and B.
suhtilis. The flasks were incubated at 27° C. for seven days and at the
end of this period the ammonia in each flask was determined. In Table
XIV are recorded the complete data for this experiment.
Table XIV. — Influence of calcium carbonate, magnesium carbonate, and limestone on
the am.monification of dried blood with pure cultures in Colby silt loam
Treatment.
Ammonia nitrogen in loo gm. of dry soil.
Bacillus tumescens.
After
7 days.
Increase.
Bacillus svbiilis.
After
7 days.
Increase.
None ,
One-fourth calcium carbonate. . . .
One-half calcium carbonate
Full calcitmi carbonate
One-fourth limestone
One-half limestone
Full limestone
One-fourth magiiesium carbonate
One-half magnesium carbonate. . .
Full magnesium carbonate
Mgm.
Mgni.
•3
15- S
16.4
16.4
II. 2
14. 6
14.8
15-7
16. s
17. 1
Mgm.
II. o
Mgm.
4.2
5-1
S-i
-o. I
3-3
3-5
4-4
S-2
S-8
16.0
16.0
II. 7
13- o
13-9
12- 5
17.6
21.0
50
S-O
0.7
2.0
2.9
1-5
6.6
10. o
The data in Table XIV show clearly that the ammonification of dried
blood by Bacillus tumescens or B. suhtilis was increased when the
sterilized soil was treated with calcium carbonate or with magnesium
carbonate or with limestone. Magnesium carbonate gave better results
than did calcium carbonate, while the latter gave better results than did
the limestone. In every case full treatment with any of the compounds
gave the largest increase of ammonia. This is in agreement with results
obtained in unsterilized Colby silt loam two months after treatment with
the limestone or calcium carbonate.
Feb. 2$, 1918
Influence of Carbonates on Soil Bacteria
493
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, AND LIME-
STONE ON BACILLUS AZOTOBACTER IN STERILIZED SOIL
Colby silt loam. — Dry soil in 200-gm. portions was treated with cal-
cium carbonate, with magnesium carbonate, and limestone, and placed
in 500-c. c. Erlenmeyer flasks. To each flask was added i per cent of
mannit by weight of dry soil. The flasks were plugged, sterilized, and
sterile water was added to bring the soil to the proper moisture content.
Inoculations of Bacillus azotobacier in pure culture were then made by
adding a i-c. c. suspension of the organisms to each flask. The entire
set of flasks was incubated at 27° C, and after one, two, and three weeks
plate counts, using mannit agar, were made. The results for this experi-
ment are recorded in Table XV.
Table XV. — Influence of calcium carbonate, magnesium carbonate, and limestone on
Bacillus azotobacter in Colby silt loam treated with mannit
Treatment.
Number of bacteria in i gm. of dry soil.
After I week.
After 2 weeks.
After 3 weeks.
None
One-fourth calcium carbonate
One-half calcium carbonate
Full calcium carbonate
One-fourth limestone
One-half limestone
Full limestone
One-fourth magnesium carbonate
One-half magnesium carbonate. .
Full magnesiiun carbonate
602,
200,
400,
<i.
<i,
<i,
200,
150,
800,
<i,ooo
4, 170,000
12,500,000
11,350,000
<I,000
<i,ooo
850,000
3, 620, 000
47, 000, 000
477,000,000
<I,000
2,440,000
32,300,000
45,500,000
<i,ooo
<i,ooo
26, 000, 000
10,350,000
268, 000, 000
355,000,000
The results of this experiment showed the striking effect of magnesium
carbonate on the multiplication of Bacillus azotobacter cells in Colby silt
loam soil. Calcium carbonate caused an increase in the number of these
organisms, but the increase was not as great as that obtained with mag-
nesium carbonate. With limestone the increase was very small. This
great gain with magnesium carbonate was obtained where amounts
equal to full neutralization were applied. After the third week the one-
half neutralization by magnesium carbonate also gave a large increase,
while two weeks after treatment the greatest gain was shown.
INFLUENCE of large APPLICATIONS OF MAGNESIUM CARBONATE ON
BACILLUS AZOTOBACTER IN STERILE SOIL
Colby silt loam.— Because magnesium carbonate when applied to give
neutralization, increased the number of Bacillus azotobacter in sterilized
Colby silt loam soil to a great extent, a further test was made to deter-
mine if heavier applications, enough added to give a distinct alkaline
reaction, would continue to increase the reproduction of Bacilhis azoto-
bacter. These applications were made by adding magnesium carbonate
suflBcient to make iXi iX» ^^^ double neutralization.
494
Journal of Agricultural Research
Vol. XII. No. 8
The data for this test, presented in Table XVI, showed that i}4 mag-
nesium— carbonate treatment after one week gave a greater increase in
the number of Bacil-
2,500,000
lus azotobacter cells
than was obtained by
the iX application;
after two and three
weeks, the iX treat-
ment gave a greater
increase than was ob-
tained by the i}4
treatment. In every
case the heaviest ap-
plication of magne-
sium carbonate caused
less increase in the
number of these or-
ganisms than did the
lighter treatments.
However, as compared
with the control, the
heavy application also
gave a great increase.
In comparing the
data of Table XVI
with those of Table
XV, it will be seen
that iX ^^^ ill most
cases i}4 magnesium-
carbonate treatment
caused a greater in-
crease in the number
of Bacillus azotobacter
organisms in sterilized
Colby silt loam than
the full treatment. It
is difficult to explain
why such an enor-
mous increase in the
number of these or-
ganisms was obtained
when more than
enough magnesium
carbonate was added
to neutralize the soil acids. From the data in Table XVI, one is led to
believe that this great multiplication in Bacillus azotobacter cells was due
12 3 4
lUTeek
t Z 3 -t-
BU/eeh
Fig. II.— Diagram showing the influence of large applications of mag-
nesium carbonate on Bacillus azotobacter in sterile Colby silt loam.
I— no treatment.
9=» il4 magnesium carbonate.
3= 1^2 magnesium carbonate.
4= double magnesium carbonate.
Feb. as. 1918
Influence of Carbonates on Soil Bacteria
495
to something besides the correction of the soil acidity,
in Table XVI are presented graphically in figure 1 1 .
The entire data
Table XVI. — Influence of large applications of magnesium carbonate on Bacillus azoto-
bacter in Colby silt loam treated with fnannit
Treatment.
Number of bacteria in i gm. of dry soil.
After I week. After 2 weeks.
After 3 weeks.
None
<i,ooo
13. 500,000
147. 500,000
10, 300, 000
<I.OOO
1,030,000,000
357,000,000
338,000,000
One and one-fourth magnesium carbonate
481,000,000
INFLUENCE OF MAGNESIUM CARBONATE ON BACILLUS AZOTOBACTER IN
STERILE NEUTRAL SOIL
Miami silt loam. — Since magnesium carbonate increased the number
of Bacillus azoiobacter cells in an acid soil when more than enough of the
carbonate was added to neutralize the acidity, an experiment was planned
to determine the effect this compound would have on Bacillus azoiobacter
in a neutral soil. The soil selected for this work was Miami silt loam,
previously described. Magnesium carbonate, along with i per cent
mannit, was added in one-fourth, one-half, and full neutralization based
on the acidity of Colby silt loam. The data for this test are tabulated in
Table XVII.
Table XVII.-
-Influence of magnesium carbonate on Bacillus azoiobacter in sterile
Miami silt loam treated with m,annit
Treatment.
Number of bacteria in i gm. of dry soil.
After I week.
Relative.
After 2 v/ceks
Relative.
After 3 weeks.
Relative.
None
3,020,000
2,870,000
44, 000, 000
67, 700,000
100
95
1,456
2,241
37,500,000
52,500,000
II, 200,000
26, 200, 000
J 00
140
30
70
51,000,000
127,000,000
137,000,000
392,000,000
One-fourth magnesium car-
bonate
249
One-half magnesium car-
bonate
Full magnesium carbonate
768
In studying the data in this table, it will be seen that in Miami silt
loam, a neutral soil, where magnesium carbonate was applied in an
amount great enough to neutralize all the active acidity in Colby soil,
there was a great increase in the number of Bacillus azoiobacter cells.
At this time no explanation can be offered to account for the drop in
number after two weeks where one-half and full magnesium carbonate
were applied, since the other counts, made after one and three weeks
with the one-half and full application gave a decided increase. The
count made after the 3-week period showed the greatest increase. From
the results of Table XVII it is clear that magnesium carbonate plaj's some
role in stimulating the B. azoiobacter organism other than that of
neutralization.
27811°— 18 3 '
496
Journal of Agricultural Research
Vol. XII, No. 8
INFLUENCE OF CALCIUM CARBONATE, MAGNESIUM CARBONATE, AND
LIMESTONE ON BACILLUS RADICICOLA (ALFALFa) IN STERILE SOIL
Colby silt loam. — This experiment was planned to measure the effect
of calcium carbonate, magnesium carbonate, and limestone on the
reproduction of Bacillus radicicola in sterilized Colby silt loam soil.
The alfalfa strain, which is supposed to be sensitive to acidity, was selected
for this work. The soil was treated with i per cent of mannit, and the
magnesium carbonate was added. Two-hundred-gm. portions of the
treated soil were placed in 500-c.c. Erlenmeyer flasks. The flasks were
then plugged and sterilized according to the method used in the pre-
ceding experiments. Inoculations were made by introducing a i-c.c.
suspension of the organisms.
From the data presented in Table XVIII it will be seen that calcium
and magnesium carbonates increased greatly the number of alfalfa
organisms in this soil. The magnesium carbonate after one week did
not increase the number of these organisms to any greater extent than
did calcium carbonate. The smaller amounts were surprisingly efifective,
though the increase was not as great as that obtained with larger
amounts. Limestone increased the number of these organisms to a
very slight extent.
TablB XVIII. — Influence of calcium carbonate, magnesium carbonate, and limestone
on Bacillus radicicola {alfalfa) in sterile Colby silt loam treated with mannit
Treatment.
Number of bacteria in i gm. of dry soil.
After I week.
After 2 weeks.
After 3 weeks.
None
12, 500
340, 000, 000
1,310,000,000
J, 750,000,000
101,000
227,000
2,450, 000
87, 000, 000
4, 100,000,000
4, 900, 000, 000
1,200
1,350,000,000
1,650,000,000
1,620,000,000
100,000
5,000
4,170,000
1,212,000,000
1, 112,000,000
600, 000, 000
1,170,000,000
1,870,000,000
50,000
240, 000, 000
1,540,000, 000
1,770,000,000
925,000,000
INFLUENCE OF CALCIUM AND MAGNESIUM CARBONATES ON BACILLUS
RADICICOLA (lupine) IN STERILIZED SOIL
Colby silt loam. — Since the alfalfa strain of Bacillus radicicola was
greatly benefited by the carbonates of magnesium and calcium, it seemed
desirable to see what effect a similar treatment would have on the lupine
strain of B. radicicola, which is frequently termed an acid-resistant
organism. Limestone was eliminated in this experiment; otherwise the
procedure for this test was similar to that in the preceding experiment.
The results of this test are tabulated in Table XIX.
Feb. 2Si 1918
Influence of Carbonates on Soil Bacteria
497
Table XIX. — Influence oj calcium carbonate and magnesium carbonate on Bacillus
radicicola {lupine) in sterile Colby silt loam treated with mannit
Treatment.
Number of bacteria in i gfm. of dry soil.
After I week.
After 3 weeks.
After 3 weeks.
None
112,000
875,000,000
1,300,000,000
1,837,000,000
886, 000, 000
9SO, 000, 000
1, 750, 000, 000
30,000
3, 770,000,000
3, 630, 000, 000
a, 760, 00c, 000
4,025,000, 000
4,837,000,000
8, 850, 000, 000
One-fourth calcium carbonate
5,530,000,000
3,820,000,000
4, 200, 000, 00c
125,000,000
5,000,000,000
9,170,000,000
One-half calcium carbonate
FtlJJ TTifi£:T7'";'1ir" c^rlffmatr-. , , ,
From the data given in Table XIX it is evident that the lupine strain
of Bacillus radicicola was benefited to as great an extent as the alfalfa
strain when Colby soil was treated with magnesium or calcium carbo-
nate. After the second and third weeks the counts showed that mag-
nesium carbonate gave better results than did calcium carbonate. The
untreated soil gave a slightly higher number than was obtained with the
alfalfa strain under similar conditions. This difference may be due to
the fact that the lupine strain is slightly tolerant to an acid reaction.
Here, as in the tests made with unsterilized soil, the smaller applications
stimulated multiplication to a greater extent, in proportion to the
amounts applied, than did the larger treatments. However, the maxi-
mum gain in number of bacteria was obtained with full magnesium-
carbonate treatment. From the data obtained in this and the preceding
experiment it is clearly evident that either magnesium or calcium car-
bonate (the magnesium to a somewhat greater degree than the calcium)
greatly increased the number of legume bacteria, alfalfa and lupine
strains, in an acid soil.
CONCLUSIONS
From a general review of the results of the experiments just described
It is clearly shown that carbonates of calcium and magnesium when
applied to acid Colby silt loam, acid Plainfield sand, and neutral Miami
silt loam increase the number of bacteria. A greater increase was
obtained in the acid soils than in the neutral Miami silt loam. These
carbonates increased the number of bacteria in both sterilized and un-
sterilized soil. The sterilized soil was inoculated with an organism in
pure culture. Not only was the number of bacteria in the soils increased
by applications of magnesium and calcium carbonates, but an increase
in activity of the various groups of bacteria was shown.
The form of carbonate which gave the best results was magnesium
carbonate. There was an enormous increase in the number of bacteria
produced by the magnesium carbonate over that of calcium carbonate.
Ground dolomitic limestone did not prove quite as efficient as calcium
498 Journal of Agricultural Research voi. xir, no. s
carbonate in increasing the number of soil bacteria. This variation can
possibly be explained by the difference in solubility of these compounds.
It was shown (30) that magnesium carbonate is more soluble in carbo-
nated water than is calcium carbonate, and that calcium carbonate is
more soluble than limestone. Dolomitic limestone, however, is less
soluble than the nondolomitic. This order of solubility is in agreement
with the order in which the bacteria of the soil responded to treatment
with these different compounds.
It does not seem correct to say that the entire influence which these
compounds had on the soil bacteria was due to the neutralization of the
soil acids. When neutral soil was treated with magnesium carbonate
or with calcium carbonate, the number of bacteria was increased, espe-
cially in the case of magnesium carbonate. Both in neutral and in acid
soil, the latter doubly neutralized, magnesium carbonate increased the
number of Bacillus azotobacter. All this would indicate that carbonates
either serve in part as a stimulant or effect an indirect action on other
compounds which are in turn rendered more soluble.
Since the magnesium and calcium carbonates increased the number
of bacteria in acid soil when added in small amounts, and since appli-
cation in acid soil of these compounds gave better results than when
applied to neutral soil, it appears that the greater part of this influence
on the bacteria was due to neutralization.
It may have been that the magnesium carbonate when added to the
soil was partially converted into magnesium phosphate. Truog (36)
pointed out the fact that for the amount of phosphate used, magnesium
phosphate increased the phosphorus content of plants more than did
other forms of phosphates. If this be the case, and the formation of
magnesium phosphate takes place in soil when magnesium carbonate is
applied, then magnesium chlorid and magnesium phosphate should give
an increase in the number of bacteria. The results from the magnesium-
chlorid treatment did not prove beneficial, however, even in so small
amounts; the chlorin radical may have been toxic. Magnesium phos-
phate did not prove favorable to the reproduction of bacteria in acid soil
but did in neutral soil. In the latter soil, the effect of magnesium phos-
phate was more beneficial than the effect of calcium phosphate. From
this evidence it appears that magnesium phosphate in a soil favorable
for the development of bacteria is a stimulant to the growth of bacteria.
This action of magnesium phosphate on bacteria in soil may in part
account for the enormous influence which magnesium carbonate exerts
in increasing the number of bacteria in both neutral and acid soils.
From all the data obtained in the various experiments performed,
magnesium carbonate appears to play an important part in the devel-
opment of soil bacteria, much more so than does calcium carbonate.
Magnesium phosphate, when applied to neutral soil, caused an increase
Feb. 25. 1918 Influence of Carbonates on Soil Bacteria 499
in the number of bacteria. In the neutral soil magnesium phosphate
proved superior to calcium phosphate in stimulating the bacteria; in
acid soil neither compound appeared to benefit the soil flora.
SUMMARY
(i) The number of bacteria in acid Colby silt loam and acid Plain-
field sand is increased by the applications of calcium carbonate, mag-
nesium carbonate, or limestone.
(2) Magnesium carbonate increases the number to a much greater
extent than does either calcium carbonate or limestone.
(3) Monocalcium phosphate and dibasic magnesium phosphate slightly
increase the total number of bacteria in neutral soil.
(4) Nitrification is benefited by limestone, calcium-carbonate, and
magnesium-carbonate treatment. Magnesium carbonate in soil to
which no nitrogenous substance was added favors nitrate accumulation
more than does either calcium carbonate or limestone. The phosphates
increase the accumulation of nitrate nitrogen to a very small extent.
When gelatin was applied to the soil, magnesium carbonate did not
benefit nitrification any more than calcium carbonate or limestone.
(5) Ammonification in Colby soil is benefited by all three forms of
the carbonates, while in Plainfield sand a decrease in ammonia is shown.
Monocalcium phosphate increases ammonification in both soils.
(6) Pure cultures of Bacillus tuinescens and B. suhtilis ammonify
blood meal better when sterile Colby soil is treated with any one of the
three forms of carbonates.
(7) A culture of B. azotobacter in the two soils treated with the car-
bonates and mannit fails to show an increase in total nitrogen in the
sand and only a slight gain in the Colby soil.
(8) Pure cultures of B. radicicola, of both alfalfa and lupine strains,
and B. azotobacter are greatly benefited when inoculated into sterile
Colby soil previously treated with magnesium or calcium carbonate.
Limestone barely increases the number of B. azotobacter in Colby soil.
In neutral and acid soils treated with magnesium carbonate until the
soils were strongly alkaline, B. azotobacter is greatly increased in munber
over that of the untreated.
(9) From the data considered as a whole, magnesium carbonate is
superior to calcium carbonate or limestone in stimulating the reproduc-
tion of bacteria in Colby gilt loam and Plainfield soils. In general, the
smaller applications of either compound give better results than do the
heavier applications.
500 Journal of Agricultural Research voi. xii, no. s
LITERATURE CITED
(i) Allen, E. R., and Bonazzi, A.
1915. ON nitrification: preliminary observations. Ohio Agr. Exp. Sta.
Tech. Bui. 7, 42 p., 6 fig. Bibliography, p. 41-42.
(2) ASHBY, S. F.
1907. SOME OBSERVATIONS ON THE ASSIMILATION OF ATMOSPHERIC NITROGEN
BY A FREE LIVING SOIL ORGANISM — AZOTOBACTER CHROOCOCCUM
BEiJERiNCK. In Jour. Agr. Sci., v. 2, no. i, p. 35-51.
(3) Beckwith, T. D., Vass, a. F., and Robinson, R. H.
1914. AMMONIFICATION AND NITRIFICATION STUDIES OF CERTAIN TYPES OF OREGON
SOILS. Ore. Agr. Exp. Sta. Bui. 118, 40 p., 30 fig. References, p. 40.
(4) Brown, P. E.
1911. SOME BACTERIOLOGICAL EFFECTS OF LI^^NG. lowa Agr. Exp. Sta. Re-
search Bui. 2, p. 49-107, 9 fig.
(5) —
1912. BACTERIOLOGICAL STUDIES OF FIELD SOILS. I. THE EFFECTS OF LIMING.
In Centbl. Bakt. [etc.], Abt. 2, Bd. 35, No. 11/13, p. 234-24S.
(6) Cauda, A.
1916. EXPERIMENTS ON THE GROWTH OF AZOTOBACTER. (Abstract.) In In-
ternat. Inst. Agr. [Rome] Mo. Bui. Agr. Intel, and Plant Diseases,
year 7, no. 6, p. 801-802. 19 16. (Original article in Staz. Sper. Agr.
Ital., V. 49, pt. 2, p. 125-131. 1916. Not seen.)
(7) Chester, F. D.
1901. STUDIES IN SOIL BACTERIOLOGY. Del. Agr. Exp. Sta. 13th Ann. Rpt.,
[i90o]/oi, p. 50-73.
(8) ChristensEN, H. R.
1915. STUDIEN UBER DEN EINFLUSS DER BODENBESCHAFFENHEIT AUF DAS BAK-
TERIENLE3EN UND DEN STOFFUMSATZ IM erdbodEN. In Centbl. Bakt.
[etc.], Abt. 2, Bd. 43, No. 1/7, p. 1-166,20 fig., 2 pi. Literaturver.
zeichnis, p. 160-166.
(9) ^rid Larsen, O. H.
1911. UNTERSUCHUNGEN UBER METHODEN ZUR BESTIMMUNG DES KALKBEDURF-
NissES DES BODENS. In Centbl. Bakt. [etc.], Abt. 2, Bd. 29, No. 9/1 1,
p. 347-380.
(10) Conner, S. D.
1916. acid soils and the effect of acid phosphate and other fertilizers
UPON THEM. In Jour. Indus, and Engin. Chem., v. 8, no. i, p. 35-40,
2 fig.
(11) Darbishire, F. v., and Russell, E. J.
1907. OXIDATION IN SOILS AND ITS RELATION TO PRODUCTIVENESS. THE INFLU-
ENCE OF PARTIAL STERILIZATION. In Jour. Agr. Sci., V. 2, no. 3, p.
305-326, 3 fig.
(12) EhrEnbErg, Paul.
1904. DIE BAKTERIELLE BODENUNTERSUCHUNG IN IHRER BEDEUTUNG PUR DIE
FESTSTELLUNG DER bodEnFruchtbarkeit. In Landw. Jahrb., Bd.
33, Heft I, p. 1-139, 2 pi.
(13) EngbErding, Diedrich.
1909. VERGLEICHENDE UNTERSUCHUNGEN UBER DIE BAKTERIENZAHL IM
ACKERBODEN IN IHRER ABHANGIGKEIT VON AUSSEREN EINFLUSSEN.
In Centbl. Bakt. [etc.], Abt. 2, Bd. 23, No. 21/25, p. 569-642, 4 fig.
Feb. 2s. i9i8 Influence of Carbonates on Soil Bacteria 501
(14) Fabricius, Otto, and Feilitzen, Hjalmar von.
1905. UEBER DEN GEHALT AN BAKTERIEN IN JUNGFRAULICHEM UND KULTIVIER-
TEM HOCHMOORBODEN AUF DIE VERSUCHSFELDE DEM SCHWEDISCHEN
MOORKULTURVEREiNS BEi FLAHULT. In Centbl. Bakt. [etc.], Abt. 2,
Bd. 14, No. 6/7, p. 161-168, 2 fig.
(15) Fischer, Hugo.
1905. EIN BEITRAG ZUR KENNTNIS DER LEBENSBEDINGUNGEN VON STICK-
STOFFSAMMELNDEN BAKTERIEN. In Jour. Landw., Bd. 53, Heft i, p.
61-66.
(16)
1905. EiN BEITRAG ZUR KENNTNIS DER LEBENSBEDINGUNGEN VON STICK-
STOFFSAMMELNDEN BAKTERIEN. In Centbl. Bakt. [etc.], Abt. 2, Bd.
14, No. 2, p. 33-34-
(17)
1905. ZWEITER BEITRAG ZUR KENNTNIS DER LEBENSBEDINGUNGEN VON STICK-
STOFFSAMMELNDEN BAKTERIEN. In JouT. L,andw., Bd. 53, Heft 3,
p. 289-297.
(18)
1909. UBER DEN EINFLUSS DES KALKES AUF DIE BAKTERIEN EINES BODENS.
In Landw. Vers. Stat., Bd. 70, Heft 5/6, p. 335-342.
(19) FRAPS, G. S.
1908. THE PRODUCTION OF ACTIVE NITROGEN IN THE SOIL. Texas Agr. Exp.
Sta. Bui. 106, 31 p., 5 fig.
(20) Fred, E. B., and Graul, E. J.
1916. SOME FACTORS THAT INFLUENCE NITRATE FORMATION IN ACID SOILS. In
Soil Sci., V. I, no. 4, p. 317-338. Literature cited, p. 337-338.
(21) Greaves, J. E.
1916. THE INFLUENCE OP SALTS ON THE BACTERIAL ACTIVITIES OF THE SOIL.
In Soil Sci., v. 2, no. 5, p. 443-480, 4 fig. Literature cited, p. 476-480.
(22) Hutchinson, C. M.
1912. studies in bacteriological analysis of indian soils. in mem.
Dept. Agr. India, Bact. Ser., v. i, no. i, p. 1-65, 6 pi.
(23) Hutchinson, H. B.
1913. the partial STERILIZATION OP THE SOIL BY MEANS OP CAUSTIC LIME.
In Jour. Agr. Sci., v. 5, pt. 3, p. 320-330, pi. 14.
(24) and MacLennan, Kenneth.
1914. THE RELATIVE EFFECT OF LIME AS OXIDE AND CARBONATE ON CERTAIN
SOILS. In Jour. Agr. Sci., v. 6, pt. 3, p. 302-322, 2 fig., pi. 9-10. Bib-
liography, p. 322.
(25)
1915. STUDIES ON THE LIME REQUIREMENTS OF CERTAIN SOILS. In JoUT. Agr.
Sci., V. 7, pt. I, p. 75-105, 4 fig., I pi. References, p. 105.
(26) Jenkins, E. H., and Britton, W. E.
1900. ON THE availability OF THE NITROGEN OF HARD RAW BONE AS AFFECTED
BY APPLICATIONS OF SLAKED LIME. In Conn. Agr. Exp. Sta., 23d
Ann. Rpt., 1899, p. 211-216, i pi.
(27) Kellerman, K. F., and Robinson, T. R.
1910. LIME AND LEGUME INOCULATION. In Scicnce, n. s., V. 32, no. 813, p.
159-160.
(28) Kelley, W. p.
1912. the effects of calcium and magnesium carbonates on some bio-
LOGICAL TRANSFORMATIONS OF NITROGEN IN SOILS. In Univ. Cal. Pub.
Agr. Sci., v. I, p. 39-49.
502 Journal of Agricultural Research voi.xii, No. 8
(29) KELtEY, W. P.
I915. AMMONIFICATION AND NITRIFICATION IN HAWAHAN SOILS. Hawaii AgT.
Exp. Sta. Bui. 37, 52 p.
(30) and McGeorge, William.
1913. THE EFFECT OF HEAT ON HAWAHAN SOILS. Hawaii Agr. Exp. Sta. Bui.
30. 38 P-
(31) KocH, Alfred.
19U. VERSUCHE t>BER DIE SALPETERBILDUNG IM ACKERBODEN. In Jour.
Landw., Bd. 59, Heft 3, p. 293-315.
(32) and others.
1907. DIE STICKSTOFFANREICHERUNG DES BODENS DURCH FREILEBENDE
BAKTERIEN UND IHRE BEDEUTUNG FUR DIE PFLANZENNAHRUNG. In
Jour. Landw., Bd. 55, Heft 4, p. 355-416, pi. 11-13.
(33) KRtJGER, W.
1909. EIN BEITRAG ZUR UNTERSUCHUNG DER STICKSTOFPUMSETZUNGEN IM
BODEN. Inaugural dissertation, Konigsberg, 1908. (Abstract.) In
Centbl. Bakt. [etc.], Abt. 2, Bd. 23, No. 6/9, p. 236.
(34) Lemmermann, O., and Fischer, Hugo.
I911. DIE WIRKUNG VON MAGNESIA UND VON KALK-MAGNESIAMISCHUNGEN
AUF DIE BAKTERIELLE TATIGKEIT IM BODEN. In Landw. Jahrb., Bd.
40, Heft 1/2, p. 244-254.
(35) et al.
1909. BAKTERIOLOGISCH-CHEMISCHE UNTERSUCHUNGEN. In Landw. Jahrb.,
Bd. 38, Heft 2, p. 319-364.
(36)
(37)
19II. UNTERSUCHUNGEN tVBER DAS VERHALTEN DES AMMONIAKSTICKSTOFFS IN
GEKALKTEN UND UNGEKALKTEN BODEN. In Landw. Jahrb., Bd. 41,
Heft 2, p. 163-216.
1911. UNTERSUCHUNGEN X)BER DIE ZERSETZUNG DER KOHLENSTOFFVERBIN-
DUNGEN VERSCHEIDENER ORGANISCHER SUBSTANZEN IM BODEN,
SPEziELL UNTER DEM EINFLUSS VON KALK. In Landw. Jahrb., Bd.
41, Heft 2, p. 217-256.
(38) LiPMAN, C. B.
1910. ON THE LACK OF ANTAGONISM BETWEEN CALCIUM VERSUS MAGNESIUM
AND ALSO BETWEEN CALCIUM VERSUS SODIUM. In Bot. GaZ., V. 49,
no. I, p. 41-50, 2 fig.
(39)
1916. A CRITIQUE OP THE HYPOTHESIS OF THE LIME-MAGNESIA RATIO. In
Plant World, v. 19, no. 4, p. 83-105; no. 5, p. 119-135.
(40) and Burgess, P. S.
I914. THE PROTECTIVE ACTION, AGAINST MgCOj, OF CaCOs FOR A. CHROOCOCCUM.
In Jour. Agr. Sci., v. 6, pt. 4, p. 484-498.
(41) LiPMAN, J. G., and Brown, P. E.
1907. LOSSES OF AMMONIA PROM CULTURE SOLUTIONS. In Jour. Amer. Chem.
Soc, V. 29, no. 9, p. 1358-1362.
(42)
1907. REPORT OF THE SOIL CHEMIST AND BACTERIOLOGIST. In N. J. Agr. Exp.
Sta. 20th Ann. Rpt., [i9o6]/o7, p. 139-204, 3 pi.
(43) arid Owen, I. L.
19IO. RELATION OF LIME TO AMMONIFICATION IN SOILS. N. J. Agr. Exp. Sta.
31st Ann. Rpt., [1909J/10, p. 114-119.
(44)
191 1. EXPERIMENTS ON AMMONIA AND NITRATE FORMATION IN SOILS. II. 1%
Centbl. Bakt. [etc.], Abt. 2, Bd. 30, No. 7/12, p. 156-181.
Feb. 25. 1918 Influence of Carbonates on Soil Bacteria 503
(45) LoEW, Oscar.
1913. STUDIES ON ACID SOILS OP PORTO RICO. Porto Rico AgT. Exp. Sta. Btll.
13, 23 p., I fig.
(46) and May, D. W.
I901. THE RELATION OP LIME AND MAGNESIA TO PLANT GROWTH. U. S. Dept.
AgT. Bur. Plant Indus. Bui. i, 53 p. 3 pi.
(47) IvYON, T. L., and BizzELL, J. A.
1909. SOME CONDITIONS FAVORING NITRIFICATION IN SOILS. In Science, n. s.
V. 30, no. 778, p. 773-774.
(48) Machida, S.
1905. ON THE influence OF CALCIUM AND MAGNESIUM SALTS ON CERTAIN
bacterial ACTIONS. In Japan Imp. Cent. Exp. Sta. Bui., v. i, no. i,
p. 1-16, I pi.
(49) McBeth, I. G., and Wright, R. C.
1912. CERTAIN FACTORS LIMITING NITRIFICATION. (Abstract.) In Science,
n. s., V. 35, no. 897, p. 392.
(50) MacIntire, W. H., Willis, L. G., and Hardy, J. I.
1914. THE NON-EXISTENCE OF MAGNESIUM CARBONATE IN HUMID SOILS
Tenn. Agr. Exp. Sta. Bui. 107, p. 151-202, 4 pi. References, p. 202,
(51) Miller, F.
1914. tjber den einfluss des kalkes auf die bodenbakterien. in
Ztschr. Garungsphysiol., Bd. 4, Heft 2, p. 194-206.
(52) Owen, W. L.
1908. THE EFFECT OF CARBONATES UPON NITRIFICATION. Ga. AgT.Exp.Sta.
Bul. 81, 42 p., 4 fig., 4 pi.
(53) Paterson, J. W., and Scott, P. R.
1912. INFLUENCE of certain SOIL CONSTITUENTS UPON NITRIFICATION. In
Jour. Dept. Agr. Victoria, v. 10, pt. 7, p. 393-400, 2 fig.
(54) Schreiner, Oswald, and Lathrop, E. C.
1912. the chemistry of steam heated soils, /n Jour. Amer. Chem. Soc,
V. 34, no. 9, p. 1242-1259, I fig.
(55) Sharp, L. T., and Hoagland, D. R.
I916. acidity and adsorption in soils as MEASURED BY THE HYDROGEN
ELECTRODE. In Jour. Agr. Research, v. 7, no. 3, p. 123-145, i fig.
Literature cited, p. 143-145.
(56) Truog, E.
1916. the utilization of phosphates by agricultural crops, including
a new theory regarding the feeding power of plants. wis.
Agr. Exp. Sta. Research Bul. 41, 50 p., 12 fig., 2 diagr.
(57) VOGEL, J.
I911. AMMONIAK- UND SALPETERASSIMILATION DURCH MIKROORGANISMEN
DES BODENS. In Centbl. Bakt. [etc.], Abt. 2, Bd. 32, No. 6/12, p.
169-179.
(58) VoorhEES, E. B., Lipman, J. G., and Brown, P. E.
1907. SOME chemical and bacteriological EFFECTS OF LIMING. N. J. Agr.
Exp. Sta. Bul. 210, 79 p.
(59) Weis, F., and Bornebusch, C. H.
1914. on the PRESENCE OF AZOTOBACTER IN DANISH WOODS AND ON THE
VALUE OP AZOTOBACTER CLOTURES FOR THE DETERMINATION OP
THE LIME REQUIREMENTS IN WOODLAND. (Abstract.) In Exp.
Sta. Rec, V. 34, no. 9, p. 814-815. 1916. (Original article in Forstl.
Forsogsv. Danmark, Bd. 4, No. 4, p. 814-815. 1914. Not seen.)
504 Journal of Agricultural Research voi. xii, no. s
(60) Wheeler, H. J., Sargent, C. L., and Hartwell, B. L.
1899. THE AMOUNT OP HUMUS IN SOILS AND THE PERCENTAGE OP NITROGEN
IN THE HUMUS, AS AFFECTED BY APPLICATIONS OP AIR-SLAKED
LIME AND CERTAIN OTHER SUBSTANCES. In Jour. Amef. Chem. Soc,
V. 21, no. II, p. 1032-1037.
(61) White, J. W.
1915. nitrification in relation to THE REACTION OP THE SOIL. In Penn.
Agr. Exp. Sta. Ann. Rpt., 1913/14, p. 70-80, 4 pi. References, p. 80.
(62) Withers, W. A., and Fraps, G. S.
1900. THE RELATIVE VALUE OF SOME NITROGENOUS FERTILIZERS. N. C. Agr.
Exp. Sta. Bui. 176, p. 15-22.
(63)
1902. NITRIFICATION IN DIFFERENT SOILS. In Jour. Amer. Chem. Soc, v. 24,
no. 6, p. 528-534.
(64) WoHLTMANN, F., FiscHER, Hugo, and Schneider, Philipp.
1904. BODENBAKTERIOLOGISCHE UND BODENCHEMISCHE STUDIEN AUS DEM
VERSUCHSFELDE. In Jour. Landw., Bd. 52, Heft 1/2, p. 97-126, i pi.
HUMUS IN MULCHED BASINS, RELATION OF HUMUS
CONTENT TO ORANGE PRODUCTION, AND EFFECT
OF MULCHES ON ORANGE PRODUCTION
By Charles A Jensen,
Assistant in Plant Malnutrition, Office of Biophysical Investigations, Bureau of Plant
Industry, United States Department of Agriculture
INTRODUCTION
The formation of humus in a soil and its consen^ation are points that
are usually given much weight in discussions of soil fertility. The free
formation of humus in a soil from organic matter added to it is usiially
held to indicate that the soil its biologically active, and proper biological
activity in a soil is conceded to be necessary under the usual agricul-
tural practices. Hence, whatever the absolute value of soil humus may
be as a factor in crop production, it is ordinarily conceded, other fac-
tors being equal, that a soil which readily forms humus when organic
matter is added is superior to one that does not.
Previous work (j)^ by this Office in southern California has shown
that the mottling of Citrus leaves varied inversely with the humus con-
tent in the soil. In other work by the same writers (2) it was shown
that on certain soil types in the general region of Riverside, California,
the use of the mulched-basin system in Citrus groves improved the tree
conditions more than the usual system of cultivation and furrow irriga-
tion, and that different green-manure substances when used as a mulch
affected tree growth and fruit setting in varying degrees.
It is therefore of practical value to obtain information about the rate
of humus formation from various organic substances when employed
either as a mulch or when worked into the soil, and to ascertain, if pos-
sible, whether the increase in humus is correlated with increase in tree
growth and fruit setting.
This work reports the study of humus formation in mulched basins in
Citrus groves, and the effect of different mulching materials on fruit
production and tree growth. The work was done at Riverside.
The term "humus" as used in this paper is the brown-colored organic
matter extracted from a soil by boiling it for two minutes in a 7.5 per cent
sodium-hydrate solution, after the removal of calcium from the soil with
I per cent hydrochloric acid.
The percentage of humus was determined colorimetrically by com-
paring the intensity of the color of the soil extract thus obtained with the
intensity of a standard humus solution prepared from the humus ex-
' Reference is made by number (italic) to " Literature cited, " pp. sij~si&
Journal of Agricultural Research, Vol. XII, No. 8
Washington, D. C Feb. 75, 191S
md Key No. G136
(S05)
5o6 Journal of A gricultural Research voi. xii. no. s
tracted from peat. In the great number of humus determinations made
by the Office of Biophysical Investigations in the Citrus areas studied in
California not a single humus extract obtained has been off-color, when
compared with the standard solutions prepared for comparison.
Gortner (4) obtained a black soil pigment from the soils he studied by
extracting them successively with 4 per cent sodium hydroxid, i per cent
hydrochloric acid, 4 per cent sodium hydroxid followed by water. The
soil residue obtained from the last sodium-hydroxid extraction, when
shaken with water in quantity, yielded the black pigment.
Soils from various Citrus areas studied in this work did not produce
black pigment in appreciable amounts when they had been extracted
with hydrochloric acid to the absence of calcium, boiled for two minutes
in a 7.5 per cent sodium-hydroxid solution, and washed with hot dilute
sodium-hydroxid solution on the filter till the "humus color" had been
removed. It might be noted that a boiling 5 per cent solution of sodium
hydroxid did not remove all the color from the soils, and that a 10 per
cent solution removed no more than a 7.5 per cent solution.
Following Gortner's procedure, black pigments were obtained from
some of the soils investigated in the Citrus areas of southern California.
It should be noted that the colorimetric reading should be made as
soon as the humus extract is obtained, as the color partially fades out
on standing.
As no evidence was obtained to indicate that any coloring matter was
left in the soil after the extraction with boiling sodium hydroxid of the
strength noted, and as the color tint was always the same as that of the
standard used, it seems that this rapid method for humus determination
is reliable for comparative studies, especially when used with soils in
the same general area.
Another objection to the colorimetric method is that undecomposed
organic matter, like dry alfalfa, sweet clover, etc., produces a color when
treated in the manner just described for mating humus determinations.
This objection is also brought out by Gortner (4). This matter was
looked into when the method was worked out. When dry hay susbtances
were extracted \\dth hydrochloric acid to the absence of calcium, the
chlorin washed out, and the residue boiled for several minutes with a
7.5 per cent solution of sodium hydroxid, the organic materials yielded
a yellow-colored solution entirely different in color from the brown to
black humus color.
A number of readings of these organic extracts gave a density of color
corresponding to about 0.0012 per cent humus in soil, if it is assumed
that I per cent of the substances was to be added to the soil. This is
about the working error in making humus determinations by the method
used, as will be seen from Table II, and such errors could have no influ-
ence on comparative results in the kind of study here reported. Indeed
there could never have been much of this off-color in the extracts, as a
Feb. 25, 1918 Humus in Mulched Basins 507
uniform tint was always obtained from the soils studied in making humus
extracts from them. The straw color produced by the extraction of soil
with I to 2 per cent of undecomposed organic matter added was found
to be so thoroughly obscured by the humus color of the soil that its pres-
ence did not interfere with the colorimetric reading.
It is well recognized that when hay material of many kinds is digested
with hot hydrochloric acid and extracted with ammonia, a solution is
obtained which in appearance and color can not be distinguished from
a humus extract from a soil.
Much of the literature relating to humus is discussed by Schreiner and
Shorey (/o).
HUMUS FORMATION IN MULCHED BASINS
Experiment I. — Mulched basins, 15 feet in diameter, were installed
in an orange grove near Riverside, Cal., in March, 191 5. The soil is a
red clay loam, derived from granite, which underlies the soil at a depth
of 3 to 4 feet. One half of the basins were mulched with about 180
pounds of alfalfa (Medicago saliva) each and the other half with about
20 cubic feet of good cow manure each. One half of the basins in each
mulching series received 200 pounds of lime dust each. This lime was a
by-product from the flues of a neighboring cement plant. It was com-
posed of about 45 per cent of calcium carbonate, about 20 per cent of
calcium hydrate, about 10 per cent of calcium sulphate, about i per cent
of potash, and a little phosphoric acid. The balance was mostly silicious
material.
During Augus^. of the same year another row of orange trees in the same
grove was basined and mulched with alfalfa. These basins were smaller
than those just referred to, and less mulch was used. Part of these
basins received about 100 pounds each of ground hme rock analyzing
about 90 per cent calcium carbonate. The rest of the basins in the row
were unlimed.
All surface organic matter was carefully removed, and soil samples
were taken to a depth of 3 feet in these basins, usually from three basins
in each experiment, the samples being separated into foot sections.
Corresponding foot samples from the three basins were composited, and
each composite sample was analyzed in duplicate. The percentage
reported in the following tables are the averages to a depth of 3 feet.
Usually the samples were collected just before irrigation.
Table I shows the average percentage of humus from time to time to a
depth of 3 feet of soil in the basins under the various treatments during a
period of from 12 to 17 months.
The low humus percentage is due in part to the removal of several
inches of the surface soil in constructing the basins, which suggests inci-
dentally that in making basins as little surface soil as possible should
be removed.
5o8
Journal of Agricultural Research
Vol. XII, No. 8
Table I. — Percentage of humus to a depth of j feet in mulched basins in orange groves
on clay loam, soil. Experiment I
Date.
1915-
Mar. 31
Aug. 13
Sept. 13
Oct. 12
Nov. II
Nov. 20
Dec. II
Average for 191 s.*
1916.
Mar. 17
Apr. 17
May 15
June 16
July 13
Ai»g. 7
Average for 1916. .
Ratio 1916 to 1915
Basin treatment and percentage of htunus.
Large basins installed in March.
Alfalfa
alone.
3-075
•073
•075
.084
.098
.187
. iq8
"3
•113
. Ill
. 076
. 112
.081
.185
113
Manure
alone.
118
130
136
153
186
188
180
.156
201
246
133
231
295
292
233
1.49
Alfalfa
and lime.
D. 113
. 106
.147
•155
.185
.150
.188
149
180
287
,181
288
272
215
237
1-59
Manure
and lime.
O. 129
. 112
•139
. 160
•151
•151
.156
143
179
213
141
220
207
227
1.38
Small basins in-
stalled in August.
Alfalfa
alone.
O. 166
. 114
. 118
130
171
257
^33
165
117
212
176
1-35
Alfalfa
and lime.
0.16s
. 092
. 108
096
"S
259
107
139
III
175
151
I- 31
It will be observed that the determinations show a fluctuation in the
percentage of humus from time to time. This is doubtless due partly
to the difficulty of getting uniform soil samples. This factor, however,
does not afford a complete explanation, because the same kind of fluc-
tuation was noted when determinations were made on soils kept in pots
in the laboratory, where better control conditions obtained and where
leaching was avoided. Neither can the fluctuations be due entirely to
the working error in making humus determinations, since these errors
are of smaller magnitude than the variations in the humus content of
the soils.
The following duplicate determinations made on March 31, 1915, the
averages of which are given on the first line in Table I, illustrate the
working error in making humus determinations by the method employed
(Table II).
Heinze (6) states that after humus has been formed other bacteria,
such as Azotobacter, commence to decompose it. McBeth (9) lays stress
on the fact that the cellulose destroying organisms in the soil break down
the organic matter in the soils with the formation of humus, and that the
nitrifying organisms break the humus up into still simpler compounds
through their nitrifying activities.
Feb. t5, X918
Humus in Mulched Basins
509
Table II. — Working error in making humus determ,inations
Soil No.
677
677
678
678
679
679
670
670
671
671
672
672
Per cent.
0.358
•358
. 041
•039
•045
. 046
• 250
• 250
.030
.028
.048
. 046
vSoil No.
676
676
677
677
678
678
682
682
683
683
684
684
Per cent.
o. 227
. 227
.051
. 050
.023
. 022
. 192
. 200
.056
.051
. 022
. 021
Soil No.
685
685
686
686
687
687
688
688
689
689
690
690
Humus.
Per cent,
o. 284
.284
.081
.081
.050
.050
.284
.384
.074
. 070
•043
. 04a
From these considerations it might therefore be expected that the
amount of humus in the soil in the basins would be subject to fluctua-
tion, the amount present at any one time depending upon the ascendency
of one or the other of these important groups of soil organisms.
The fluctuation in the percentage of humus makes it impracticable to
attempt to determine the rate at which humus was formed between
specific dates. It will be seen in Table I that there is frequently a de-
crease in the percentage of humus from one period to another, especially
during the second year, a point that might possibly apply to determina-
tions made by Gortner (5). He added silk, wool, flour, and alfalfa meal
to soils, kept the materials in earthenware jars, and at the end of a year
made humus determinations. In most instances he found less humus
at the end of the year than he found in the treated soils when the experi-
ment was started. It would have been of much interest if these deter-
minations had been made frequently during the year. None of the
losses of humus reported by him are greater than was sometimes found
in a month's time in humification studies carried out under laboratory
control in this Office, to be reported later.
A comparison of the amount of humus in the basins from year to year
is made by averaging the percentage found during each of the years
1915 and 1916, as shown in Table I. At the bottom of the table will be
found the ratio of the average percentage of humus in 1916 to that in
1915-
Experiment II. — In April, 191 5, a block of orange trees in another
grove on sandy loarri soil was basined. The basins were about 12 feet
in diameter, and the treatments were the same as in Experiment I,
except that less lime was used.
The results of the humus determinations are given in Table III, which
shows the fluctuation in the amount of humus from time to time. The
ratio of the average percentage of humus in the basins in 191 6 to that
in 1 91 5 is given in the bottom line in the table.
5IO
Journal of Agricultural Research
Vol. Xn, No. 8
Table III. — Percentage of humus to depth of ;j feet in mulched basins in orange groves on
sandy loam soil. Experiment II
Date.
1915
May 3
Aug. 9
Sept. 4
Oct. 4
Oct. 30
Nov. 30
Average for 191 5. ...
1916.
Mar. 17
Apr. 17
May 15
June 16
July 13
Aug. 7
Average for 1916
Ratio, 1916 to 1915. .
Basin treatment and jjercentage of humus.
Alfalfa Manure Alfalfa Manure
alone. alone. and lime, and lime.
O. 169
.128
. 129
• 230
.203
.185
174
251
227
129
216
195
175
199
I. 14
o. 159
. 121
. 146
.218
. 209
•159
169
241
220
180
213
167
296
1.30
D. 214
. 126
.187
.283
•255
.181
208
.301
. 296
. 163
•309
.244
•243
259
1.24
o. 196
. 164
. 189
. 260
•307
. 213
221
286
338
. 180
248
309
272
23
Experiment III. — In October, 191 5, a block of 60 trees in another
orange grove of heavy sandy loam soil was basined and mulched. The
basins were about 6 by 20 feet. The mulching materials used, the
quantity of each, and the percentage of humus in the soil are given in
Table IV. The percentages are averages to a depth of 3 feet. The
amount of bean straw used was not enough to cover the soil, and so did
not make an effective mulch. Some of these basins received in addition
to the mulching material, 100 pounds ground lime rock, analyzing about
90 per cent calcium carbonate.
Table IV. — Average percentage of humus to a depth of 3 feet in mulched basins in an
orange grove on heavy sandy loam,. Experiment III
Percentage of humus.
Humus
ratio —
1916 to
1915-
Quantity of mulch and lime
Oct. I J,
1915.
June 27,
1916.
per basin.
Alfalfa alone
0.223
. 160
.243
.241
. 264
.244
•319
•235
.254
•237
.279
.266
.265
0.316
.185
. 220
. 240
. 226
.163
.247
.194
.161
.215
.203
.279
. 206
I. 41
I. 16
.91
I. 00
.86
.67
.78
.83
•63
.91
•73
1.05
.78
125 pounds.
Alfalfa and lime
100 pounds of lime.
Bean straw alone
70 pounds.
100 pounds of lime.
Bean straw and lime
Manure alone
20 cubic feet.
Manure and lime
100 pounds of lime.
Barley hay alone
125 pounds.
Barley liay and lime
100 pounds of lime.
Sweet clover alone
125 pounds.
Sweet clover and lime
100 potmds of lime.
Bur clover alone .
125 pounds.
125 pounds.
100 poimds of lime.
Pine shavings alone
Pine shavings and lime
Feb. 25, 1918
Humus in Mulched Basins
511
The basins with alfalfa as the mulch were the only ones which increased
in percentage of humus during the interval given. Other experiments
have shown that bean straw readily humifies. It is possible that if this
had been applied in as large a quantity as the other substances, these
basins would have increased in humus also. The lime was not consistent
in its effect on the amount of humus formed.
It is to be noted, however, that the percentage of humus as found in
the summer, after the application of the mulches in the preceding fall,
does not necessarily indicate that in most of these basins the average
amount of humus had actually decreased. As shown in Tables I and III,
there are periodical fluctuations in the humus content, and the second set
of determinations shown in Table IV may have been made at a time when
the amount of humus was comparatively small. The figures in Table IV
probably give an indication of the relative humifying activity in the
basins. Each percentage given is an average of about six determinations.
Experiment IV. — A series of basins was installed in a lemon grove
near Corona, Cal., in October, 1915. The soil is sandy loam in texture
interspersed freely with gravel so that samples below 3 feet can not be
taken with the ordinary soil tube.
The mulching materials used were alfalfa and manure and, in addition
to the organic mulch, some of the basins received other artificial ma-
terials— ^viz, blood, phosphate, tankage, sulphur, bone meal, and lime.
In addition to the basined rows, several control rows were retained, and
were manured, irrigated, and cultivated in the usual way. The amount
of manure applied to the trees in the control rows was the same as appUed
in the basins.
Soil samples were collected after the basins were made, on October 20,
1915, and another set was taken on June 29, 1916. The detailed humus
determinations made on the latter date did not indicate that any of the
artificial substances added to the mulches produced any definite influence
on the humus content. The average percentage of humus to a depth of
3 feet in the alfalfa basins, manure basins, and in the control rows is
given in Table V. Each percentage figure given in the table is made up
of an average of about 21 determinations, each determination repre-
senting a composite of at least three samples.
Table V. — Average percentage of humus to a depth of 3 feet in unmulched soil and in
mulched basins in a lemon grove on light sandy loatn soil. Experiment IV
Percentage of humus.
Ratio
Soil treatment.
Oct. 20, June 29,
1915. 1916.
1916 to
1915-
Basins with alfalfa mulch
0. 360
•345
•551
0.306
.361
.380
0.85
1.05
.60
Basins with manure mulch
Furrow irrigation and siu-face cultivation ; manured
27811°— 18-
^12 Journal of Agricultural Research voi. xii, No. 8
The low initial percentage of humus in the basins is partly due to the
fact that 5 or 6 inches of the surface soil were removed in making the
basins, owing to sloping ground.
The percentage of humus in the manured basins in 191 6 was about the
same as in 191 5, while in the alfalfa basins it was less. The decrease in
the humus content in the manured soil furrow-irrigated and surface-
cultivated is very appreciable, indicating that this system did not con-
serve the humus as well as the basin system.
The grove in which this experiment was conducted was deteriorating
along with other groves in the neighborhood when the basins were in-
stalled. The whole grove was in better condition in October, 191 6, but
there was no apparent difference in the appearance of or the amount of
fruit on the trees basined as compared with trees not basined. The
mulch-basin system in experiments conducted by this office has always
produced a quicker response on trees on clay loam soil than on trees on
sandy loam or light sandy loam soil.
The deterioration of the grove above mentioned was evidently due to
bad soil conditions, which quite likely affected the bacterial flora in such
a way as to make it less efficient in converting the alfalfa into humus.
In most cases presented above the percentage increase in humus during
a period of from 9 to 1 7 months was greater in the basins on the clay loam
soil than in those on the sandy loam soil, with both manure and alfalfa
as the mulch. On the whole, the manured basins gained slightly more
in humus content on both types of soil than the alfalfa-mulched basins.
No correction has been made for the amount of humus added with the
manure. This material contained about 5 per cent of humus when ap-
plied, part of which undoubtedly found its way into the surface soil when
the basins were irrigated. The percentage increase of humus in the
manured basins therefore does not necessarily indicate that the manure
had been humified to a greater extent than alfalfa. Data to be presented
in another paper would indicate that the increase is likely due to the
humus added with the manure.
In all cases where lime was added to manure in basins the increase in
humus was somewhat less than when manure alone was used. In most
cases the addition of lime to alfalfa in basins produced slightly more
humus than when alfalfa was added alone. Frear and Hess (j) found
that field plots receiving manure and lime contained less active humus
than unmanured plots.
The amount of alfalfa or manure necessary to form a unit amount of
humus can not be stated with exactness from the data available. In
Experiments I and II the average increase in humus in the basins in
1 91 6 over the average amount present in 191 5 would indicate that i part
of humus was formed from about 10 to 12 parts of alfalfa, and i part of
humus from about 26 parts of manure. These calculations agree fairly
Feb. 2s, 1918
Humus in Mulched Basins
513
well with results stated by others (7, p. 128-129), but are necessarily
only estimates.
The data obtained in the humus studies in mulched basins did not
indicate that there was any very appreciable accumulation of humus in
the lower soil depths due to leaching from the surface foot of soil. The
detailed data showed that most of the change in humus content took
place in the surface foot.
RELATION OF HUMUS CONTENT OF THE SOIL TO ORANGE
PRODUCTION
Experiment I. — Picking records of the earlier basined trees in Experi-
ment I were obtained from the company on whose grove the experiment
was conducted, and are given in Table VI.
Table VI. — Effect of different mulching materials on orange production. Experiment I.
Picked May, igij
Basin treatment.
Average number of boxes of oianges
per tree.
191 7 yield
corrected on
basis of pre-
vious yields
(boxes per
tree).
Row
No.
Tree nimiben
1914
I9I5 ! 1917
in row.
Alfalfa alone
I. 49
I. 19
1.78
1.32
0.71
I. 00
I. 02
3-31
3.61
2. 16
I. 81
4-97
6.69
2. 56
2-55
3
3
3
3
I to 8
Alfalfa and lime
Manure alone
26 to 11
9 to 15
16 to 25
Manure and lime
This appreciable difference in fruit production of trees mulched with
alfalfa and manure, respectively, does not correlate with the humus con-
tent in the soil under the respective mulches, as may be seen from Table I.
The manured basins average a slightly higher percentage of humus than
the basins mulched with alfalfa.
It will also be noticed in Table VI that the trees basined and mulched
with manure in March, 191 5, had previously produced slightly more
fruit than the trees mulched with alfalfa.
When the oranges were picked in the spring of 1917, the color of the
oranges on the trees mulched with alfalfa was discinctly more golden
than the color of the oranges on the trees mulched with manure.
Experiment II. — Individual tree picking records had not been kept
of the trees in the grove in which this experiment was conducted. Indi-
vidual tree picking records were obtained in the spring of 191 7 from the
company on whose grove the experiment was conducted. In this experi-
ment a row of trees was left unbasined and was manured, furrow-
irrigated, and cultivated in the usual manner. The amount of manure
per tree was the same as that put into the basins.
The fruit production of the trees is given in Table VII. The number
of trees used in each set of mulched trees was from 13 to 15.
514
Journal of Agricultural Research
Vol. XII, No. 8
Table VII. — Influence on orange production of different mulches in basins and by furrow
irrigation and cultivation. Experiment II. Picked JuTie, 1917
Soil treatment.
Average for
each or;ianic
substance
with and
without
lime.
Basins with alfalfa mulch alone
Basins with alfalfa mulch and lime. . .
Basins with manure mulch alone
Basins with manure mulch and lime. .
Cultivated with manure alone, disked
Cultivated with manure and lime
4.8
4.0
2-5
It is not known that the trees used in this experiment differed in
yields previous to the installation of this experiment. However, as no
individual tree picking records had been kept previous to the experi-
ment, the comparative yields shown in Table VII do not carry the force
they would if they could have been compared with the previous perform-
ance of the same trees. In comparing the fruit production in experi-
ments I and III, it seems safe to infer that the results obtained in Ex-
periment II were indicative of the effect of the different organic mate-
rials used.
It will be noticed in Table III that the percentage of humus in the
manured basins was slightly higher than that in the basins mulched
with alfalfa, a result agreeing with that obtained in Experiment I.
No appreciable difference in the color of the fruit from the trees dif-
ferently treated was discernible.
Experiment III. — The block of 60 orange trees used in this experi-
ment had been used in a study of individual tree performance by the
Office of Horticultural and Pomological Investigations, Bureau of Plant
Industry. Individual tree records for the previous six years had been
obtained, and these were kindly furnished by the Riverside officials of
that Office. The picking records of the 191 6-1 91 7 crop are given in
Table VIII, together with the basin treatments. The basins were in-
stalled and mulched in October, 191 5.
No effect of the lime on fruit production could be definitely deter-
mined; hence, the yields of all the trees mulched with the same organic
matter have been averaged, including both limed and unlimed trees.
The alfalfa-mulched trees produced more fruit than the manure-
mulched trees, which result agrees with the results obtained in Ex-
periments I and II. It does not appear from Table VIII that any
legume whatever is superior to a nonlegume or to manure. The basined
trees mulched with sweet clover and bur clover did not produce as much
fruit as those mulched with alfalfa, bean straw, manure, or barley hay.
Feb. 2$. 1918
Humus in Mulched Basins
515
The color of the oranges on the trees mulched with alfalfa and bean
straw was distinctly more golden than the color of the oranges on the
trees mulched with the other organic substances. The foliage on the
bean-straw and alfalfa-mulched trees was considerably denser and
greener than that on the other trees.
Table VIII. — Effect of different mulching materials in basins on orange production.
Experiment III. Picked May, igij
Basin treatment.
Yield of oranges corrected on ba-
sis of previous performance of
trees.
Pounds
per tree.
Oranges
per tree.
Average
weight per
orange.
Alfalfa hay
Bean straw
Manure
Barley hay
Sweet clover hay
Bur clover hay. .
Pine shavings. . . .
308
289
261
202
217
216
163
881
744
611
497
584
530
402
Pounds.
350
389
427
407
372
408
406
On comparing the results of the humus determinations shown in
Table IV with the fruit production shown in Table VIII it is evident
that the percentage of humus in the soil does not correlate with the fruit
production. There is, however, a correlation between the fruit produc-
tion and the ratio of the humus content in 191 6 to that in 191 5, but such
correlation is not evident in Experiments I and II.
Closer examination of the picking records in Experiment III showed
that the bean-mulched trees produced more first-class fruit of the more
desirable sizes than the trees mulched with any of the other substances;
the alfalfa-mulched trees were second in this respect; manure third;
and barley fourth; while the trees mulched with sweet clover and pine
shavings produced the least number of fruits of the first quality.
It may be that some toxic substance is formed in the decomposition
of the pine shavings which might account for the poor condition of the
trees and for the small yield obtained from them; also the shavings prob-
ably had a deleterious influence on nitrification. Redwood boxes for
use in germinating Citrus seedlings proved unsuitable, because, when
the root tips came in contact with the wood, they promptly died.
Oranges from the trees in Experiments I, II, and III were analyzed,
the analyses being made by the branch office of the Citrus B3--products
Laboratory in Los Angeles. The analyses did not bring out any con-
sistent differences in the fruits from trees receiving different fertilizer
treatments.
2i6 Journal of Agricultural Research voi. xii.no. s
There was a consistent difference in the specific gravity of the oranges
from the Umed trees and from the unUmed, but the difference was small.
The average specific gravity of the oranges from the mulched and limed
trees was 1.0012, and of those from the mulched trees, but not limed,
was 1.0307. The specific gravity was calculated from the grading
results, and not by individual determinations of separate fruits. The
error in sizing by the grading machine would probably be compensated
by the number of fruits included, as the entire crop was used in the
calculation.
On the whole, the evidence obtained from these experiments does not
show that the humus content of the soil correlates with orange pro-
duction. Neither does the information obtained justify the statement
that the humifying activity in the soil correlates with fruit production;
but the results obtained would indicate that this point might be worthy
of further study.
It appears from the results here presented that the most important
function of organic matter as influencing orange production is not that
of merely furnishing humus; that humus in itself is not the most impor-
tant product of organic degradation for orange production. It appears
from other work (8) that a more important function of organic matter
is to make the plant food in the soil minerals more readily available.
It was found that organic substances, especially in a freshly decom-
posing condition, dissolved plant food elements in appreciable amounts,
even when the organic solvents contained practically no electrolytes.
This function and that of promoting the biological activities of the soil
seem to be more important rdles of organic matter in the soil than merely
to furnish humus.
SUMMARY
This report presents a study on (a) the changes in humus content in
soils in basins mulched with different organic substances, (b) the effect
of lime on humus content in soils in mulched basins, (c) the relation of
humus content in the soil to orange production.
By "humus" is meant the brown- to black-colored organic extract
obtained from soil leached with i per cent hydrochloric acid to the
absence of calcium and the soil residue boiled for two minutes in a 7.5
per cent sodium-hydrate solution.
Humus determinations in mulched basins in citrus groves showed a
fluctuation in the percentage of humus from time to time.
The average percentage of humus increased more in basins on clay
loam soil than in basins on lighter soil tubes, with manure and alfalfa
as mulching m^aterials.
Usually the percentage of humus in basins increased more when manure
was used as mulch than when alfalfa was used as mulch. This seemed,
Feb. J5, 1918 Humus in Mulched Basins 517
however, to be due more to the humus added with the manure, than
to the greater " humification " of the manure over the alfalfa.
When manure alone was used as mulch in basins the increase in hu-
mus was greater than when lime was added with the manure.
In most cases when lime was added to alfalfa in basins greater increase
in the humus content occurred than when alfalfa alone was used.
Blood, acid phosphate, bone meal, tankage, or sulphur did not show
any appreciable influence on the changes of humus content in mulched
basins.
It was not evident that there was any appreciable accumulation of
humus in the lower depths of soil due to the leaching of humus from the
surface foot of soil.
Thsre was no evident correlation between the amount of humus in the
soil in mulched basins and the amount of fruit on the trees.
There was no evident effect of lime on orange production in these
experiments.
Alfalfa and bean-straw mulch in basins on the heavier soil types
produced from 30 to 100 per cent more oranges per tree than manure
mulch. Manure mulch produced more oranges per tree than either
barley hay, sweet clover, bur clover, or pine shavings. These differ-
ences were obtained in the summer following the application of the
mulches in the preceding fall.
Alfalfa mulch and manure mulch in basins on the lighter types of
soil produced no observable differences on fruit production of lemons
in the course of one year. This statement is based only on observation
and not on picking records.
In all experiments so far conducted by this Office in the Riverside area,
the mulched-basin system on the heavier soil types has produced favora-
ble growth response in a few months. It usually takes longer to produce
appreciable response on the lighter soil types.
It would appear directly from the work here reported, and indirectly
from work elsewhere reported that the degradation products from
freshly decomposing organic substances are more effective in orange
production than the amount of "humus" formed. And the value of a
given mulch does not necessarily depend upon its being a legume or
nonlegume.
LITERATURE CITED
(i) Briggs, L. J., Jensen, C. A. and McLane, J. W.
1916. mottle-leaf of citrus trees in relation to soil conditions. In
Jour. Agr. Research, v. 6, no. 19, p. 721-739, 4 fig., 3 pi. (partly col.)
(2)
1917. THE MLTLCHED-BASIN SYSTEM OF IRRIGATED CITRUS CLT^TURE AND ITS
BEARING ON THE CONTROL OF MOTTLE-LEAF. U. S. Dept. Agf. Bul. 499,
31 p., I pi.
c 1 8 Journal of Agricultural Research voi. xii, No. 8
(3) Frear, Wiluam, and Hess, E. H.
1900. INFLUENCE OF SYSTEMS OF FERTILIZING VFON THE AMOUNT AND QUALITY
OF THE HUMUS OF THE SOIL. In Ftoc. 2ist Ann. Meeting Soc. Prom. Agr.
Sci., p. 60-69.
(4) GORTNER, R. A.
1916. THE ORGANIC MATTER OP THE SOIL: I. SOME DATA ON HUMUS, HUMUS
CARBON, AND HUMUS NITROGEN. In Soil Sci., V. 2, no. 5, p. 395-441, 2 pi.
Literature cited, p. 440-441.
(5)
I917. THE ORGANIC MATTER OF THE SOIL: III. ON THE PRODUCTION OF HUMUS
FROM MANURES. In Soil Sci., v. 3, no. i, p. 1-8. Literature cited, p. 8.
(6) Heinze, B.
1909. the formation and decomposition of humus in cultivated soils.
(Abstract.) hi Exp. Sta. Rec, v. 23, no. 7, p. 621. 1910. (Original article
in Landw. Mitt. Prov. Sachsen. u. Nachbarstaat. Halle, 1909, p. 145-146.
Not seen.)
(7) HacARD, E. W.
1914. soils: their formation, properties, composition, and RELATIONS
TO CLIMATE AND PLANT GROWTH. 593 p., 89 fig.
(8) Jensen, C. A.
191 7. EFFECT OF DECOMPOSING ORGANIC MATTER ON THE SOLUBILITY OF CER-
TAIN INORGANIC CONSTITUENTS OF THE SOIL. In Jour. Agr. Research, v. 9,
no. 8, p. 253-268.
(9) McBeth, I. G.
1916. STUDIES ON THE DECOMPOSITION OF CELLULOSE IN SOILS. In Soil Sci.,
V. I, no. 5, p. 437-487. Literature cited, p. 481-487.
(10) ScHREiNER, Oswald, and Shorey, E. C.
1909. THE ISOLATION OF HARMFUL ORGANIC SUBSTANCES FROM SOILS. U. S.
Dept. Agr. Bur. Soils Bui. 53, 53 p., 4 pi.
RELATION OF KINDS AND VARIETIES OF GRAIN TO
HESSIAN-FLY INJURY^
[PRELIMINARY REPORT]
By James W. McColloch, Assistant Entomologist, and S. C. Salmon, Professor of
Farm Crops, Kansas Agricultural Experiment Station
It has long been known that certain varieties of wheat (Triticum spp.)
are injured less than others by the Hessian fly {Mayeiiola destrticior).
Packard ^ mentions the Underhill, Mediterranean, Lancaster, and
Clawson varieties as being noted for resistance. He states that the
Underhill variety has been highly recommended for nearly a century.
Woodworth ^ made observations on 125 varieties of wheat grown at
the California Experiment Station in 1886, 1887, and 1889, and noted
the damage by Hessian fly in each. The Volo and Washington Glass
varieties were not injured. Forelle; bearded wheat from Missoyen,
Palestine; Polish; Bluegrass; Common March; Diamond; and Egyptian
Imported were practically free from injury.
Roberts, Slingerland, and Stone,'* in summarizing their observations on
Hessian-fly injury in New York, conclude that
the resisting power of varieties varies greatly
and that
those with large, coarse, strong straw are less liable to injury than weak-strawed and
slow-growing varieties.
Six varieties are mentioned which were not appreciably affected by
the fly in 1901, although numerous other varieties in the same neighbor-
hoods were much injured. These varieties were Dawson Golden Chaff,
Prosperity, No. 8, Democrat, Red Russian, and White Chaff Mediterranean.
Gossard and Houser^ made careful observations on 75 varieties of
wheat and other grains grown at the Ohio Experiment Station in 1904,
1905, and 1906. They determined the percentage of stalks infested and
of fallen straws. Their observations
give but little support to the idea that there are immune varieties,
and they suggest that cases of supposed immunity may be explained by
some other hypothesis. They state, however, that the most persistent
• Contribution from the Entomological Laboratory (Paper No. 30) and the Department of Agronomy
(Paper No. 13) cooperating. This paper embodies some of the results obtained in the prosecution of projects
No. 8 and 67 of the Kansas Asricultural Experiment Station.
* Packard, A. S. the hessian fly — its i^avages, habits, and the means of preventing its in-
crease. In 3rd Rpt. U. S. Ent. Com., p. 227-228. 1883.
» WOODWORTH, C. W. VARIATION IN HESSIAN FLY INJURY. In Cal. Agr. Exp. Sta., Rpt., iSgo, p. 312.
I89I.
< Roberts, I. P., Si,ingeri.and, M. V., and Stone. J. L. the hessian fly. its ravages in new
YORK IN 1901. N. Y. Cornell Agr. Exp. Sta. Bui. 194, p. 226-260, fig. 95-98. 1901.
' Gossard, H. A., and Houser, J. S. the herslan fly. Ohio Agr. Exp. Sta. Bui. 177, 39 p., 2 fig.,
I col. pi., map. 1906.
Journal of Agricultural Research, Vol. XII, No. 8
Washington, D. C. Feb. 25. 1918
mc Key No. Kans.— n
(SI9)
520 Journal of Agricultural Research voi. xii, no. s
search has never located a single Hessian-fly egg on oats, although eggs
were found on many grasses.
Since 1906 the relation between varieties and injury from Hessian
fly has received scant attention from investigators. Presumably the
work of Gossard and Houser has been accepted as disproof of the claims
of earlier observers that some varieties are resistant and others immune.
Recently claims of immunity put forth by growers of certain varieties,
general observations by farmers in eastern and central Kansas that
hard wheats are more susceptible to injury than soft varieties, and re-
sults of experiments at the Kansas Agricultural Experiment Station
indicate that the subject is at least worthy of further investigation.
Experiments have been outlined to determine (i) the relative infes-
tation and injury of different kinds, varieties, and strains of various
small grain, and (2) why certain kinds and varieties are resistant or
immune; or, if not, why they escape injury in some cases where others
are badly injured. This paper is concerned primarily with the first
problem.
EXPERIMENTAL DATA
The data reported in this paper were collected from 87 kinds and
varieties of wheat (Triticum spp.), oats {Avena saliva), barley {Hordeum
spp.), rye (Secale cereale), emmer {Triticxim dicoccum), einkorn {Triti-
cum, monococcum) , and spelt (Triticum spelta), planted in the Agronomy
Nursery of the Kansas Agricultural Experiment Station in the fall of
1 91 6. The different varieties were planted each in a row 50 feet long and
10 inches apart. Two plantings were made on each of two dates, Septem-
ber 15 and October i. The soil was in excellent condition, moisture
was plentiful, germination was prompt, and growth was normal in every
respect. Hessian flies were numerous, and, as far as known, there was
ample opportunity for all varieties to become equally infested. In this
paper all varieties are tabulated in the order in which they were planted.
Eight of the varieties tested were from Australia (rows i to 8, inclu-
sive, Table I), and had not been grown previously at Manhattan. The
spring varieties and about half of the soft winter varieties had been ob-
tained from various Experiment Stations in the United States in 1914
and had been grown at Manhattan for two years only before being in-
cluded in this experiment. All of the hard winter varieties and about
half of the soft winter varieties had been grown at Manhattan for several
years and were thoroughly acclimated.
The relative number of eggs deposited on each variety was determined
by taking five consecutive plants from the west end of each row and
counting the total number of eggs on the leaves. The first count was
made at the time of maximum deposition, September 25. Subsequent
determinations were made for the early sown plots on October 2 and 7,
and for the late sown plot on October 14. The total number of plants
of each variety examined was 20.
Feb. 2s. 1918 Relation of Grain to Hessian-Fly Injury 52 1
The relative number of flaxseeds in each variety was determined by
examining 50 consecutive plants of each row in each plot sown on Septem-
ber 15, and 25 consecutive plants of each row in each plot sown on
October I , making a total of 1 50 plants of each variety. All plants were
taken from the western end of the plots, or, in other words, adjacent
to those plants which were examined for eggs. The pertinent data for
each variety are given in Table I.
That the Hessian fly is apparently able to discriminate between kinds
and varieties of grain is shown by these data. For example, the total
number of eggs per 100 plants in the early sown plots ranges from 40 for
Culberson winter oats to 5,600 for Turkey winter wheat No. 2407. The
proportion of plants on which eggs were laid varied from 20 per cent for
Culberson winter oats and Michigan winter barley to 100 per cent for
Tennessee winter barley and most of the varieties of wheat.
The close agreement in determinations made at different times, and the
striking differences in the number of eggs laid on adjoining varieties indi-
cate that the difference can scarcely be attributed to experimental error.
Thus, Turkey winter wheat No. 2407, which showed the highest total num-
ber of eggs, also had the highest infestation on September 25, the second
highest on October 10, when the second determination was made, and was
among the highest on October 7, when the third determination was made.
On the other hand, Culberson Winter oats, Michigan Winter barley, and
einkom had the lowest total infestation and the lowest on each date.
Row 12 (Polish wheat) had a total of 2,040 eggs per 100 plants, as
compared with 280 for row 13 (einkorn). Row 15 (spring emmer) had a
total of 660 eggs per 100 plants, as compared with 1,740 for row 16
(Black Winter emmer).
On the whole, the Hessian fly appears to have shown a preference for
common wheat, as compared with barley, oats, einkorn, spring emmer,
spelt, and durum wheat. Black Winter emmer. Poulard wheat, and
Polish wheat were as heavily infested with eggs as many of the common
wheats. Rye showed a very heavy infestation, the total number of eggs
per 100 plants being 2,500, which is well above the average for all grains
included in the test.
Varieties of the hard winter wheat class were more generally infested
than soft winter wheat varieties. Thus, 27 varieties of hard winter
wheat averaged 2,737 ^ogs per 100 plants, as compared with an average
of 1,835 for 38 varieties of soft winter wheat. However, there are wide
variations in each class. For example. No, 2408 and Mealy, which are
soft, or semihard, varieties, were infested with 4,720 and 4,320 eggs, re-
spectively, per 100 plants, showing almost as high an infestation as the
most profusely infested varieties of hard wheat, and more than double
that of some varieties. On the other hand, certain varieties of hard
wheat, such as Defiance No. 2129, Red Winter No. 839, Improved Tur-
key No. 2382, and Pesterboden No. 205, had a lower infestation than the
average of the soft, or semihard, varieties.
522
Journal of Agricultural Research
Vol. XII. No. 8
0.0
•io
I-*
o
«i^
;2;.c'
ij- 'J- O M O N
0 •<t O M O •*
0000000000000000
OO00O-<i-OM00iN 'too 00 O 00
a
^
*>
10 r-. M N O t^oO -^00 i/^oo 'J- O 1^ O Tl-
O O >0 O to N
0000 «\0 N >riM O Tfoo 00 t>.t^r>.
O M O O O O
OONMrOM^ON«Ocot>.rO'* >0
■§X1 tfl *J
a tj M a >A.
0000000000000000
MMOOOf^OO^OOMO-t 'too 00 vO "t
row 'troiJ^OfCO O^'^O O O fl 10\0 J^
000000
't M 'O N 'to
M M 10 M ro
000000000000000
OCvOOOOvO-tvOvO OO't'tTtwO
Tj-rOO MM'OC< M HN
000000
't M VO ■^
M M \0
0000000000000000 'OOOOOO
O^OOMVOtNOOMN ■*00 C<M\000 iTfMPlOOO
0000M^0M0wvO'tt^C^0^^^'t't^o• NMrj-CNio
^
M M ro 't 100 «^0C 00 «~-
OOOOOOOOOOv
'tt't-^'t'tt'tM cs
1:13
«-1
t4 eq eq ffi u U ►S H W S
•§
be "5
.5 t«
B-o ■" S °
C O r^ VV
»-i in w
+^ "ti -^^
'^ ^ r-j <U
^f-l 5 <U C
- S "3 S -y >^ « •§
S <u o
o
^e
,:;:i ^ O ^ l2 P P ;-> Oh a.'>C^ <^
J3
c :
u O
•0^ MO-JJ
Feb. as. 1918 Relation of Grain to Hessian-Fly Injury
523
0000000
0000000000000000000
>0<N\OoiCivOONO'^NO "^OO 04 O vO o
N
'i-'^OcO i^.cO w-iO'O-
Tj- <N ro
ID 1000 ■<J- ►^ 0 t^O M 00 lOOO ro 0 LO 0
00 m f^ 0
0
f^
N fO M "1- ■* ro CO lOvO
10 Tt W
rO^Ot~OrO<^»^'!r''5'*N C^w N row 11
M 0 Ci 0
1-
w
■* ■* M 00 'too "t^O 0
M lOOO
•^M N -itlNCOOO^ OmDoOoOoO MCOO
0 0 t^cO 00
0
00
M \C 0 ^00 ro rooc 't
MUM
VOOO M lOvO rj OP) M lOTj-TtrorOroO
■* 0\vO CO CO
M 04 MM
00
I^OmioOmmO t^oO 10 vooO
10\0 t>. tOOO C4 tOOO M On N t-~ UOvO OfO'^'tt^-OO ■*
Tj- tJ- irj rf LOO toLO-^ •^•^■^ m r4C4iOM 10
00 M 04 lOO t~~MOO VOOnO-cOO
li^ t-^ t-~ ir.OO >J^ C7< ionO m CO O O
N M O))-(WMMC404C4
M M ro M "^ lOCC OOoO O M ro'Or-OroPi CT^cOm 10
p) vo t^ t^OO 0CP4<3l^>^O"*C^OMM mio t->.0 m
MMMMMC4C404M M M M
0
0
0
0
0
0
n
0
0
0
0
0
0
0
0
0
0
n
0
0
0
0
8
M
0
0
0
0
0
0
0
0
0
0
0
0
0
M
0
M
0
0
0
0
0
M
0
0
M
0
M
0 00
0
0
t-H
0
M
u
U
0
0
0
0
0
0
M
0
0 00
0
M
0
M
0
M
0 CO
M
0
M
0
M
0
M
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
0
0
0
0
0
0
0
0
Tl-OO
ct
0
-^
a)
0
PI
0
^0
0
0 00 vO CO
0 0 CO
1-0
'to
^0
0
0
ri
P4 00
0
to
n
vO
10
0
I^
l^
COCO
*^
C)
'^
LO
LO
00
N
Pt
»~~
^
CO
PI
CO
0
r^
0
CO
0^ t^
t^
W
r^
10
t
to
M
CO "^
C4
CI
•^
M
-*
CO
"
"
cs
M
CO
P4
M
C4
'-'
P)
P4
M
N
P4
-'
M
P4
"
M
"
CO
"
'-'
M
'-'
00000000
O 'toco Tj-'tP^ PI
00000000000000000000 o
piPtPfpfOOPi O00P40't0O't-t000Pi 00 4*
t-i P«MMMCOP4 MM *i
0000000000000
00PiOOP4 0O00000000
VOP4P4»OM't'tP4COP4C1PIP4
000000000000000000000
0'1-P4P40P4 000P40 'tCO O P4 O 00 ^tOO O O
H'l-MP4'ttOPinMCOP4't'rl"CO'tCOP4 T|-P1
^ ^
0000000000000 000000000000000000000
P» ■t 't P4 00 TfoO "to O P4 'too TJ-P4 PI T^P4000CO OOOOOOOOOOOO O O -too 00
00 P4 O O too CO COOO t CO P4 P4 lOOO M 100 PI On P4 PI lO -t t COCO CO O "t 0^ lO CO CO
Tti-rpi'TtpTi-rrfi-rcocoHri-rpr )-rpreri-ri-ri-ri-rprHrp4'"pr Hri-rhTi-ri-rpr i-Ti-t
r~~coMOO P4 10 o 00- coo 100 coOPiO'tMiocoOpiO'i-iococoOOcO'tco'o
O O I^ O CO 00 r^ coo mOvOm Ttco-tLOO^OOPici^ coo O t^oO O O O 0\ too
-t t O t CO CO 10,00 MMP4P4 -.a-OOOOMMMMMM MMtMCOCO
pT pT PI P4 M H PI PI p) P4 P4 04 cT cT cT cT pf pi'
Q Q
S a; « >^^ '^ 3 o ^ c
H C ^ W U) O H Pi S
3 ^ „
rt ^ _n "-M
«Pi<;P
3
J-.2T3
m
"3 ?=-a
be 5^
M be y
3 !=J3'Z3
4-> CS ^.j c!
cS flj rt lU
01 ^ « ^
'^ W ""^ Ut
Im HI Vh OJ
OJ Ui Cj Wi O
cBWcBM
<U M
•S'i
u o ^
rt <U c3
u V u
.5|.H
TS wo
II u 4)
t-^ I-'
* I
S :t8
to CO PO t t -"t Tl- •
524
Journal of Agricultural Research
Vol. XII, No. 8
i2 «J ff Si «
OnO OOOnnOOoO
O P) O O N o
TtOOvOcsOO'*
oooooooo
vO VO O 00 \0 rt-O 00
M M HI M M
oooooooo
8
a
.IS
i
,8
s
^
O >oO iot^t--.M t^MO O t^0"0
M O r^ O roO O 0>
g-oa.
t/i rt Qj
."■Sen
rt Qj N
0.W
i. ^^ ><^ o 4ij
> ^B'U^ Z 9
J? " Sj! Si «3~
^ 5^ ■" a Q.
So'
«3i2
a a
n _*j M Q, ^
5; ^ <"« "
00 O 00 vO ■* ■^OO O N >0 O M O 00
OnO ^00 r^ "^ r-- r^vO 0^0 ro fO O
N 00 M >0 00 00 1^0
00 00" rOOO' On ■^ N 0C5
■g O O O o o o
.000000
00000
00000
OOOQQOOOOOOOOO
o Tt-00 oooooo'OTl■ooT^o
M 00 i-i N O^vo 000 r^CM M r^ioro
O O O Q O
M VO \0 O ■+
On M LO C\ w
00000000000000
00 'J-oo 00 ■^mS rf P) pi \0 n m ■*\0
i-i^p<i-iMM p^H';^rowl-l
00000000000000
00 \0 00 vO ■* Tj-OO O O \0 O P< ^0 00
Q\\0 Looo 1^ '^^ r- t^ 10 00 ro ro O
OOOOOOOO
Ptoo 000000 p)vO
00 00 rooo Ch ^00 00
• n S K o
•<4- lOvO CN On fO OnnO O t^OO ■* O Os
000 w po ^n\0 t^ t^ t^oo On O
•* ^ Tj- M
MVOOO w PO'^PO't
mmOOOOpioO
l-IMIH(NPJP<C<0
•o>i Mo-a
w
oj HI 03
& C &
?: & a &
"O "O '^J TJ
Q»* C^ 'T' Q^ O O
- - i-i'O'a
■CO
:^c8
Feb. 25, 191 J
Relation of Grain to Hessian-Fly Injury
525
0000000
T^ N 00 o 00 c< N
8000000
00 -4- P) CO M 00
roO 00 fO r^ "^ "^
00 \0 t^ M Ov "^ O
00 10 iJOvO O r^oo
O M H M H M H
cT cT N M cf fT fT
•g ;3 ;3 W OJ
WfeCJOP
\C) t t 0 0 "^oo
Q 0 0 0 0 0 0
0 <N 'too 00 c< ■*
c5 N ro\d 'J- ro ro
000 'J-
f^O 00
M HI
N 00 N 00
CO cL LO lO
M M M
r^ rj- ro
H CI 0 (^
\0 M 1-1 ■*
M f» »O00 00 0 ro
0 ro^ 0 ■* •* <^
888
M H M
8888
H W H W
000
v6 0 '*
vo C^ 0
0000
W Tj-CO 00
t^ 0 CO t^
M M M
w N 1-1
000
P) 0 M
0000
too ■too
N H M
00^0000
tj tJ "^ "d "13 t3 13
526 Journal of Agricultural Research volxii, no.8
Perhaps the most significant result is the high mortality of the eggs
and larvae, and as a result the relatively low infestation with flaxseed of all
varieties and especially certain ones. The figures are open to criti-
cism, so far as the absolute mortality is concerned, since more eggs may
have been laid on the ends of the rows, where the egg counts were made,
than on the adjoining plants, which were used for the flaxseed determi-
nations. However, this criticism would not hold for the relative mortality
for the different varieties.
As will be seen, no flaxseeds were found in einkorn, spring emmer,
Culberson Winter oats, rye, and Illini Chief wheat. Very low infesta-
tions, 5 per cent or less, were recorded for Tennessee Winter barley and
for Beechwood Hybrid, Currell Selection, and Dawson Golden Chaff
wheats. The data for rye and for Illini Chief and Dawson Golden Chaff
wheats are especially significant, in view of the fact that 2,000 or more
eggs per 100 plants were laid on each.
Of the hard wheats. Red Winter No. 2132 is especially worthy of
mention, since it had only 30 flaxseeds per 100 plants, as compared
with over 200 for other varieties of the same class. Only 9 per cent of
the plants of this variety were infested with flaxseeds, as compared with
a range of from 28 to 62 per cent for other varieties of this class.
In general, the data show a low, or no infestation, with flaxseeds
for rye, barley, oats, durum wheat, Poulard wheat, Polish wheat, spelt,
emmer, and einkorn. The average number of flaxseeds per 100 plants
for 38 varieties of soft winter wheat was 76.1, as compared with 173.7
for 27 varieties of hard winter wheat, an increase for the latter of more
than 225 per cent.
These conclusions are based on the rows planted on September 15,
but essentially similar results were secured from those rows planted on
October i. The infestation on these plots was much lighter than in
the former case, and the data are less conclusive. It will be noticed,
however, that those varieties which were not infested or which show a
low infestation in the first plantings exhibit a similar characteristic in
the latter.
Field tests. — In 191 5 a bushel of Illini Chief wheat was secured by
the Entomology Department of the Kansas State Agricultural College
and planted in four localities in the State where the Hessian fly was
abundant.
The Illini Chief was practically free from injury in all cases. At
Manhattan it showed less than a i per cent infestation, while Turkey
wheat in an adjoining plot was infested practically 100 per cent.
An examination of the plants in the fall and spring indicated that
the fly showed no preference for either variety, the eggs being equally
numerous on both plots. In the Illini Chief the maggots were able to
work their way down to the crown of the plant, but at this point devel-
opment appeared to be arrested and the larvae died.
Feb. 25, 1918 Relation of Grain to Hessian-Fly Injury 527
At Winfield, Kans., the Illini Chief was sown in the center of a 40-
acre field of hard wheat. Determinations in the spring showed that 95
per cent of the plants of hard wheat were infested, as compared with
about 10 per cent for the Illini Chief.
In 1 91 6, a plot of Illini Chief wheat grown by the side of Turkey
wheat showed an infestation of from 3 to 5 per cent, as compared with
95 to 100 per cent for the Turkey variety.
While these tests appear to show that Illini Chief is somewhat resist-
ant to the Hessian fly, it should not be assumed that it is the best
variety to grow. In Kansas it is one of the least hardy of all winter-
wheat varieties and will survive none but the mildest winters.
CONCLUSIONS
The Hessian fly is able to discriminate between different kinds and
varieties of grain. Eggs were laid on all the kinds and varieties of
grain studied, but very sparingly on winter oats, winter barley, einkorn,
spring emmer, spelt, and durum spring wheat.
On the average, fewer eggs were laid on soft winter wheat than on
hard red winter wheat, but exceptions in both cases were found.
There appeared to be a large mortality of eggs or larvae on all kinds
and varieties studied. This appeared to be greatest for rye, einkorn,
spring emmer, winter oats, and Illini Chief wheat. Very few flaxseeds
were found on winter barley, and on Beechwood Hybrid, Currell Selec-
tion, and Dawson Golden Chaff wheats.
27811°— 18 5
ADDITIONAL COPIES
OF THIS PUBLICATIOK MAT BE PROCtniED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
■WASHINGTON, D. C.
AT
15 CENTS PER COPY
StmscRiPTioN Price, 33.90 Pee Yeab
A
Vol. ::vll ^^ARCH ^, 191S No. 9
JOURNAL OP
AL
RESEARCH
CONTKNTS
Page
Wilt Diseases of Okra and the Verticillium-V/ilt Problem 529
C. W. CARPENTER
( Contributioa from Burcsu oi Plant indtssJry )
Winter Cycle of Egg Production in the "Sthode Island Red
Breed of Domestic Fowl ~ - - - - - 547
II. D. GOODALS
( Coritdbutioa from Massachusetts Agrfculhirsi Kiperiirci:. f.i Uon")
Digestion of Starch hj the Young Calf - - ■ - 575
R. H. SHAW, T. £. WOODWARD, and E. P- ""■ :• i
(Contribution from Btixeaa of Aniraal ladusL;, ,,
Toxicity of Volatile Organic Compounr •; to I/isect Eggs - 579
WILLIAM MOORE and S;
( Contribulioa from Minnesota Agr; ; - . i )
Com-Stover Silage - - - ,. „ _ 559
J. M. SHERMAN, and ^, .. ..__..„-'',
(Ccntribulioa from Pennsylvania AgrJcuItarsl Excerimt
Weevils Which Affect the Irish Pc Potato, and
Yam ----- ., - 601
W. DV/IGHT ±'ii-r:i.-Ji;
( Coatrlbutioa frota Bureau of EnJom'>!'?gy )
PUBLISHED BY AUTHORITY OF THE SECRETilRY OF AGRIGJLTURB,
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPESBIENT STATIONS
WASHINOTON, r>* p.
wmhinqton: QovEnNMENT pRiKTiNA orFios ) ran
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCLATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPASTMENT
KARL F. KELLERMAN, Chairman
Physiologhl and Associate Chief, Bjtreau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Entomoloqisl and Assistant Chief , Bureau
of EniOTKology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Mains Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute cf Animal 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.
* Dr. Pearl has tmdertaken special work in connection with the war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
JOm£ OF AGMCPLTIAL ffiSEARCH
Vol. XII Washington, D. C, March 4, 191 8 No. 9
WILT DISEASES OF OKRA AND THE VERTICILLIUM-
WILT PROBLEM
By C. W. Carpenter,'
Scientific Assistant, Cotton, Truck, and Forage Crop Disease Investigations, Bureau of
Plant Industry, United States Department of Agriculture
INTRODUCTION
The investigation of the okra-wilt disease thus far reported has led
to some confusion as to the cause of this malady. Atkinson (jY men-
tions the fact that okra {Abelmoschtis esculentus) is sometimes attacked
by a disease similar to the wilt of cotton (Gossypium herbaceum). He
states that a fungus for which he proposes the name "Fusarium vasin-
fectum" is invariably found in the vascular system of cotton and okra
affected with this disease. Smith (25), in a more extensive study of the
wilt diseases of cotton, watermelon (Citrullus vulgaris), and cowpea
(Vigna sinensis), found an ascigerous fungus associated with the diseased
plants. He regarded this new genus as the perfect form of Fusarium
vasinfectum Atkinson, and renamed the latter fungus "'Neocosmospora
vasinfecta." Smith notes the occurrence of a wilt disease of okra, and
regarding A^. vasinfecta remarks (p. jj) :
It probably occvirs also on okra, although the identification is not complete, depend-
ing solely on the character of the symptoms, on the presence of similar macroconidea
and microconidia, and on the occurrence of the disease in the same localities, no cul-
tures or cross inoculations of the okra ftmgus having been made and no perithecial
fruits having been discovered.
Doubts as to the genetic relationship of Neocosmospora vasinfecta and
Fusarium vasinfectum are expressed by Higgins {12), Butler (5), and
Wollenweber {31, 32). It appears from the inoculation and pure-
culture work of the latter that A^. vasinfecta Smith is to be regarded as
a saprophyte entirely distinct from the vascular parasite F. vasinfectum
Atk.
Clinton {8, p. 268) notes the occurrence of a wilt disease of okra in
Connecticut which he considers as the same as that previously reported
by Atkinson. The wilt of okra as it occurs in North Carolina is described
' Since the completion of these studies, the writer has been transferred to the position of Pathologist,
Hawaiian Agricultural Experiment Station.
2 Reference is made by number (italic) to "Literature cited," p. 544-346.
Journal of Agricultural Research, Vol. XII, No. 9
Washington, D. C. Mar. 4, 1918
U , , KeyNo. G— IJ7
(529)
530 Journal of Agricultural Research voi. xii. No. 9
by Stevens and Wilson (25), with an additional statement of the disease
by Wilson (29). They attributed this disease to Fusarium vasinfectum.
No inoculations of okra plants with F. vasinfectum are recorded by
Atkinson, Clinton, Smith, or Stevens and Wilson. A species of Fusarium
was found constantly associated with the wilt of okra, and the disease
was regarded as the same as the cotton-wilt. Orton {16) states that in
his experience okra has contracted the disease when planted in fields
affected with cotton-wilt.
Wollenweber (jj, 32) reports a \vilt disease of okra similar to that
previously attributed to Fusaritim vasinfectum; but Verticillium albo-
atrum Reinke and Berthold, instead of the Fusariumu fungus, was con-
stantly present in the vascular ducts. Successful inoculations were secured
with V. albo-atrum, and the fungus was recovered from the wilting plants.
It would appear that there are two similar wilt diseases of this crop
induced by two quite different vascular parasites. It was for the purpose
of testing this latter hypothesis that the investigation herein recorded
was undertaken. While okra as a crop is restricted in general to the
home garden and is relatively of little importance, a study of the wilt
diseases of this plant is of general significance to the whole problem of wilt
disease.
THE GENERA VERTICILLIUM AND ACROSTALAGMUS IN RELATION TO
WILT DISEASES
Verticillium albo-atrum was first described by Reinke and Berthold (22)
in 1879 as the cause of a wilt disease of potato {Solanum tuberosum) in
Germany. This appears to be the earliest record of species of Verticillium
associated with wilt disease. The work of Reinke and Berthold was
generally overlooked until the investigation of the leafroll (Blattroll-
krankheit) and similar diseases of 5. tuberosum brought it again to the
attention of pathologists. Recently the presence of V. albo-atrum in
wilting potato plants has been noted by several investigators : Appel (2) ,
Spieckermann (24), Pethybridge (18), {19), Stormer (26), Wollenweber
and Schlumberger {34), Wollenweber (50, 31, 32), Orton {17). Wollen-
weber {31, 32) reports this organism as the cause of a wilt disease of a
S. tuberosum, S. melongena, and Abelmoschus (Hibiscus) esculentus. A
similar disease of snapdragon (Antirrhinum sp.) was described by
Brown (4) and the causal fungus determined by the present writer as
V. albo-atrum. V. dahliae, a species closely related to V. albo-atrum, is
described by Klebahn (14) as the cause of a wilting of dahlias. Aderhold
(i) also mentions a species of Verticillium in connection with thrombosis
of currants (Ribes spp. — "Johannisbeere") and gooseberry (Ribes spp. —
" Stachelbeere").
The causal organisms of a series of diseases of several host plants
variously described as wilt disease, thrombosis, bluestem, etc., have been
relegated by the respective authors to the genus Acrostalagmus. Owing
Mar. 4. 1918 Wilt Diseases of Okra 531
to the confusion existing in the genera Verticillium and Acrostalagmus, a
discussion of the interrelations of these form groups is necessary at this
time.
Among those writers who have obtained and described Acrostalag-
mus-like fungi may be cited the following: Van Hook (27) discusses a
wilt disease of ginseng (Panax quinque folium) which he found associated
with a vascular-inhabiting fungus. It is said to closely resemble Acro-
stalagmus alhus Preuss. Rankin (20) renamed the ginseng-wilt fungus
"Acrostalagmus panax" Rankin, without giving a technical description
of the fungus or stating reasons for the change. Whetzel and Rosen-
baum (28), in mentioning this disease of ginseng, attribute it to a species
of Acrostalagmus. Gueguen (10) describes a new species of Acro-
stalagmus causing a wilt disease of the China aster Reine-Marguerite
(Aster sp.), and to this fungus he gave the name A . vilmorinii. A variety
of this organism he later mentions (11) as associated with a disease of
fruits of the cacao (Theobroma cacao). Lawrence (15) attributes a new
disease of black raspberry (Rubus occidentalis) — that is, the bluestera
disease — to a species of Acrostalagmus, for which he proposes the name
"Acrostalagmus caulophagus." Rankin (21), in a preliminary report of
a thrombotic disease of maple trees (Acer spp.), mentions that a species
of Acrostalagmus is associated with the trouble.
Corda (9, p. jj) established the genus Acrostalagmus to accommodate
an organism, A. cinnabarinus , which was said to differ from Verticillium
by forming its spores in heads at the tips of the conidiophores. Hoffman
(is) has rightly explained the complicated structure of the sterigma as
represented by Corda, and likewise the collection in heads of the singly
abscissed conidia. The conidia are held together by virtue of an hygro-
scopic slime, which takes up water in the presence of a sufficiently moist
atmosphere, forming a sphere of water at the tip of each sterigma. In
these water droplets the conidia are observed to float freely. With ex-
cess moisture the drops rupture and the conidia slip down the conidio-
phore. However, in the absence of moisture the conidia cling tenaciously
to the tips of the sterigma in the form of irregular rounded masses of
varying sizes. If such material is placed under a dry cover slip without
being allowed to come in contact with the latter, and examined micro-
scopically while a small water drop is brought into the vicinity of the
dry spore head, the spore aggregate will be seen gradually to round up and
the conidia to float more or less actively within the spherule (PI. 17, B,D),
In the genera Acrostalagmus and Verticillium we now have the follow-
ing contradictory and imperfect characterization and limitation of forms.
In Verticillium the singly abscissed and characteristically singly borne
conidia may adhere in the presence of moisture, forming terminal heads —
that is, water droplets containing conidia. In Acrostalagmus the
conidia, characteristically united in heads, soon separate if the humidity
of the environment is in excess of the maximum for the moisture drops
532 Journal of Agricultural Research voi. xii, no. 9
to retain their form, leaving but one immature conidium on the sterigma
tip, as in Verticillium.
Furthermore, in the genus VerticilHum, characterized by singly borne
or readily separating conidia, we have the anomalous condition of a
section of the genus set aside for those species in which the conidia are
held together by slime — that is, section Gliocephalum ^ Saccardo (1886).
This section of the genus accommodates forms in which the conidia are
united in slimy heads — that is, Acrostalagmus.
The fact that the work of Reinke and Berthold {22) was so generally
overlooked is responsible for the confusion of the genera Verticillium and
Acrostalagmus. These investigators studied A. cinnaharinus, on which
Corda established his genus Acrostalagmus, and concluded that this
form genus must be united with the older genus Verticillium Nees.
They changed the name of Corda's fungus to Verticillium cinnaharinum.
In view of the above, there seems to be no doubt that the genus Acro-
stalagmus Corda should be united with the genus Verticillium Nees.
Klebahn {14) supports this view.
The species of Acrostalagmus, described as causing a vascular disease
of ginseng, China aster, and black raspberry, may prove upon further
work to be identical with each other and with V. alho-atrum, as proved
in this paper, for the strains of the latter fungus isolated from okra,
cotton, eggplant (Solanum melongena), and potato. Culturally, at least,
the strains of the species of Acrostalagmus from ginseng and raspberry,
in so far as they have been studied by the writer, are not to be dis-
tinguished from Verticillium alho-atrum. The organisms A. vilmorinii
Gueguen and V. dahliae Klebahn, as described by their authors, hardly
differ sufficiently from V. alho-atrum to be given specific rank. Possibly
this is the better treatment of such related forms until their identity
is established by careful cross-inoculation work with pure cultures.
The minor cultural differences found by the writer in cultures of V.
alho-atrum from different hosts are present in parallel cultures of the
same strain. Strains of V. alho-atrum which are morphogically indis-
tinguishable, isolated from different hosts, and capable of producing
the same s)Tiiptoms of disease by cross-inoculation, seemingly should
be considered identical.
The available data on the several strains of Verticillium and Acro-
stalagmus thus far described as plant parasites are brought together in
tabular form (Table I), in order that comparison may be readily made.
The hypothetical identity of all of these strains, suggested by the simi-
larity of this data, is strengthened if one consults and compares the
descriptions by the several authors. The existing differences seem to
be those of variety rather than of species.
1 Saccardo's section Gliocephalum of the genus Verticillium (Saccardo, P. A. svxlogb funoorum.
V. 4, p. 139. Patavii, 1886) is given as Gliocladium by Engler and Prantl (1900) apparently through error.
(Engler, A.,andPRANTi„K. naturlichen PFLANzeNFAMHiEN. Teil I, Abt. I**, p. 418,432. Leipzig,
1900.)
Mar. 4, 1918
Wilt Diseases of Okra
533
Table I. — Data of several species and strains of Verticillium and Acrostalagmus arranged
for comparison
Conidia.
Sclerotia.
Sclerotial cells.
Length of
Organism.
Limits of size. Average size.
Limits of
size.
Average
size.
Limits of
size.
Average
size.
conidio-
phore
branches.
V. clbo - alTum
strain 1717 Jrom
okra.
V. albo - atrum
5.1 to II by I 6.8by2. 4...
2. 1 to j.C. ;
4. 2 to 9. 3 by ! 6. 4 bv 2. 4. . .
/i.
21 to 91
31 to 70
31 to 87
28 to 70
34 by 44
45 by 54
SI by SI
39 by 57
7 to 10
s to 10
8. 4 by 8. 7
8. 2 by 8. 2
7- 3 by 9. 4
A".
13. 5 to 22.0
strain 2784 from
potato.
V. albo - atrum
I. 7 to 4. 2.
5. 1 to 8. 5 by
2. 1 to 4. 2.
6. s by 2. 6. . .
strain 1685 from
eggplant.
3 to 15
15. 0 to 38.0
strain 2985 from
snapdragon.
V. albo - atrum bv
5.0 to 12.0 by
2. 0 to 3. 5.
4. 0 to 6. 0 by
l.St0 2.5.
4. 0 to 7. 0 by
1.5 to 2.0.
2. 0 to 5. 0 by
1. 0 to 2. c.
3. 0 to 7. 0 by
2. 0 to 2- 0.
5. 0 to 7. 0 by
:. 5 to 3. 0.
6. 0 to 7. 0 by
3. 0 to 4. 0.
Reinke and
Berthold.
54
50
12. 0 to 45.0
16. 0 to 27. 0
la ot0 25. 0
given by Kle-
bahn.
V. dakliae by Kle-
Stc6
bahn.
Hook.
7 to 12
by Lawrence.
A. vilnn)rinii by
20 to 70
50 to 60
None.
Guegtien from
China aster.
cacao.
For the reasons above set forth, it is evident that the genus Acro-
stalagmus must be united with the older genus Verticillium. In view
of this condition, the following forms, if not identical with V. albo-
atrum, must at least be regarded as belonging to the genus Verticillium :
.4. albus from ginseng-wilt disease, described by Van Hook (27) and
renamed A. panax by Rankin (20); A. vilmorinii Guegen, from the China
aster and the fruits of the cacao; and A. caulophagus Lawrence, the
cause of the bluestem disease of black raspberry. Of the remaining
12 imperfectly characterized and apparently saprophytic species of the
genus Acrostalagmus some are no doubt identical with species of
Verticillium, and are just as incompletely delineated. The ubiquitous
V. lateritium Berk., commonly present on decaying potatoes, is possibly
the same as V. innabariniitn (Corda) Reinke and Berthold.
FUSARIUM VASINFECTUM AND VERTICILLIUM ALBO-ATRUM, CAUSES
OF THE WILT DISEASES OF OKRA
FUSARIUM VASINFECTUM
The species of Fusarium from okra- and cotton-wilt considered in this
work are regarded as identical with each other and with the species of
Fusarium isolated from "frenching" cotton by Atkinson (5) and named
by him "F. vasinjccium." The cross-inoculation work herein recorded
proves the casual relation of F. vasinfecium to the wilt disease of okra,
534 Journal of Agricultural Research voi. xn, Na^
and establishes the pathologic identity of the cotton- and okra-wilt
strains beyond reasonable doubt.
Certain strains of the species of Fusarium causing cotton-wilt were
observed by WoUenweber (ji, 32) to produce an aromatic odor (lilac)
when cultured on starchy media such as rice; other strains less com-
monly isolated lacked this property. These latter strains were desig-
nated " F. vasinfectum var. inodoratum" by WoUenweber. As a further
indication of the identity of the species of Fusarium on okra with the
species of Fusarium on cotton, it should be noted that both the odor-
forming and the non-odor-forming strains have been isolated several
times from okra. While the ability to generate this odor is of doubtful
specific value, since other species of the section Elegans (WoUenweber,
32) possess this property, and this ability has been observed to be lost
in culture {Carpenter 7, p. 206), the fact that the species of Fusarium
from okra and cotton agree in this character is significant.
Normal ^ cultures of this species of Fusarium develop in i to 3 weeks
at room temperature (PI. A). The best results are secured with plant
stems, potato cylinders, and other vegetable media. Morphologically
F. vasinfectum is scarcely distinguished from the other vascular parasites
of the section Elegans of this genus.
Potato-cylinder cultures (PI. A) develop an ocherous-salmon-colored
pionnotes with 3- to 5-septate conidia (PI. 17, L-M). Blue-gray sclerotia,
similar to those of F. oxysporum Schlecht. (7, PI. A, i). are generally
present on this medium. The slight violet color of the upper part of the
pionnotes as represented in Plate A, 3, has never been seen in other
closely related species and is possibly of differential value. Stems of
Meliloius alba and Gossypium sp. are useful in developing the sporodochia,
which are likewise of an ocherous-salmon color. The plate of F. oxy-
sporum in an earlier paper (7, PL A, 2, 5) illustrates stem cultures of F.
vasinfectum equally well. On steamed-rice medium a more or less
brilliant red color soon appears, later becoming tinged with various
shades of purple and blue, especially in subdued light. Normal conidia
are not usually present in rice cultures, the value of this medium being
the color reaction and the formation of an aromatic substance with an
odor suggesting lilac (7, p. 206). Chlamydospores (PI, 17, I) are formed
in large numbers on this medium.
The following measurements show the size and percentage of the
variously septate, normal, conidia (PI, 17, L, M) found in strains of F.
vasinfectum from okra and cotton :
F. vasinfectum, strain 2709, isolated from okra. Culture, 27-day-old
stem of Melilotus alba, without pionnotes. Normal triseptate conidia,
60 per cent. Limits of size: 25.5 to 40.8 by 4.2 to 5.1 ju. Largest nor-
mal triseptate conidium, 40.8 by 5.1 ix.; smallest, 25.5 by 4.2 /i. Average
' For a discussion of the idea "nonnal" and other special terms as used in relation to species of Fusarium,
see WoUenweber (jj, />. 255-257).
Mar. 4, 1918
Wilt Diseases of Okra
535
size of 10 normal triseptate conidia, 32.5 by 4.6 fx. One-septate conidia,
20 per cent. Abnormal conidia and microconidia, 20 per cent.
Normal conidia on rice culture 2H months old: Triseptate, 23.8 to
30.0 by 4.0 to 5.0 IX. Five-septate rare, 56.0 by 3.6 fx.
F. vasinfectum strain 3203 from okra. Pionnotes on a 25-day-old
stem culture of MeliloUis alba. Limits of size of 28 normal triseptate
conidia: 23.8 to 34.0 by 3.8 to 4.2 tx. Average size, 28.4 by 3.9 /u.
Largest triseptate conidium, 34.0 by 4.2 fx. Smallest triseptate coni-
dium, 23.8 by 3.8 (x.
F. vasinfectum strain 3242 from okra. Normal triseptate conidia,
34.0 to 41.0 by 3.4 to 5.0 fx. Four-septate conidium 40.8 by 4.2 [x.
F. vasinfectum var. inodoratum strain 3257 from okra. Thirty-four-
day-old Meliloius stem culture with pionnotes. Normal triseptate coni-
dia 60 per cent. Limits of 17 spores, 32.3 to 45.9 by 3.4 to 4.8 /x. Aver-
age, 37.4 by 4.2JU. Four-septate, 35 per cent. Limits of 13 spores, 34.0
to 44.2 by 4.2 to 4.8 fx. Average, 39.0 by 4.2 m- Five-septate, 5 per
cent. Limits of four spores, 37.4 to 45.9 by 4.2 to 4.9 /u., average four
spores 42.0 by 4.4 m-
F. vasinfectum var. inodoratum, strain 3258 from okra. Thirty-four-
day-old pionnotes culture. Normal triseptate conidia 76 per cent.
Limits of 9 spores, 27.2 to 41.6 by 3.4 to 4.6 /u. Average, 34.o»by
4.2 ju. Four-septate conidia, 19 per cent. Limits of 13 spores, 35.7 to
42.5 by 3.4 to 5.0 IX. Average, 37.8 by 4.1 ix. Five-septate, 5 per cent.
Limits of 6 spores, 34.0 to 45.9 by 3.8 to 4.7 ix. Average, 40.0 by 4.2 /u.
Parallel cultures of F. vasinfectum, strain 1855, from cotton and strain
3592 from okra were prepared upon cotton stems as a medium. At the
age of 45 days, 50 normal triseptate conidia from a pionnotes of each
culture were measured to obtain the relative sizes. The maximum
minimum, and average length and width of conidia in each culture are
shown in Table IL There is a close correspondence of the measurements
of conidia from the two strains, considering that these measurements
were made from but one culture of each strain.
Table II. — Comparative size of normal triseptate conidia of Fusarium, vasinfectum from
okra and cotton
Maximum length.
Minimum length .
Maximum width.
Minimum width .
Average length . . .
Average width . . .
F. vasinfectum
1855 from
cotton.
5-1
3-4
37- o
4.2
F. vasinfectum
3592 from
okra.
44.
27.
5-
3-
35-
4-
536 Journal of Agricultural Research voi. xii. no. 9
Fusarium vasiofectum Atkinson.
Sporodochia and perfect pionnotes present, in mass ocherous-salmon colored, the
conidia being of the Elegans type, 3- to 5 septate, sickle-shaped, constricted at the
apex and pedicillate at the base (PI. 17, L,M). Conidiophores verticillately branched.
Normal triseptate conidia present up to 100 per cent, 23.8 to 46.0 by 3.4 to 5.1 m-
Four-septate conidia up to 35 per cent, 34.0 to 44.0 by 3.4 to 5.0 m- Five-septate
conidia up to 5 per cent, 34.0 to 56.0 by 3.6 to 5.0 n. Microconidia may be present
in subnormal cultures up to 100 per cent, 4.0 to 14.0 by 2. o to 3.5 m- in size. Chlaray-
dospores (PI. 17, 1) ellipsoidal to round, terminal, intercalary and conidial; when
unicellular measvtring from 8 to 1 5 m- Blue-gray sclerotia on potato cylinders. Strong
lilac odor on rice and other starchy media. Vascular parasite, cause of a wilt disease
of Gossypium Jierbaceum, G. barbadense, and Abelmoschus esculentus.
VERTICILLIUM ALBO-ATRUM
Veriicillium albo-atrum Reinke and Berthold is classified by Engler
and Prantl in the section Eu-Verticillium of the genus Verticillium of
the Mucedinaceae-Hyalosporae-Verticillieae. The conidophores (PI. 17,
A-D) are verticillately branched, conidia which readily fall being formed
at the tips of all the branches. The distinction between the three sec-
tions of the genus — that is, Eu-Verticillium Sacc, Oncocladium Wallr.,
and Gliocephalum Sacc. is not sharply drawn. In the latter the conidia
are held together by slime, while in Eu-Verticillium and Oncocladium
this is not the case. In the opinion of the writer, V. albo-atrum should
more appropriately be placed in the section Gliocephalum, for this is
where it naturally belongs, if its characters are determined on the sub-
stratum (PI. 17, B, D). If examined in water mounts, rarely more than
one conidium would be found on each sterigma tip (PI. 17, C), and the
fungus would erroneously be placed with the section Eu-Verticillium.
DESCRIPTION OF THE FUNGUS
The conidia are ellipsoidal (PI. 17, A), unicellular, 4.0 to ii.o by 1.7
to 4.2 n, abscissed singly from the tips of verticillate-branched conidio-
phores. They may or may not cling to the tips of the sterigma in rounded
masses. In the absence of sufficient moisture in the air, relatively dry,
rounded aggregates of spores accumulate ; but with more moisture present
spherical drops appear on the sterigma tips by virtue of the hygroscopic
slime in which the conidia are embedded (PI. 17, B). With additional
moisture the drops rupture, leaving one immature conidium clinging to
the sterigma. These masses measure from 3 /* to a size where the water
drop breaks. The verticillate branches of the conidiophores are i to 7 to
a whorl or virtel, more commonly 3 to 5, and these in turn may bear
secondary branches in virtels. The branches are from 13 to 38 ^ long,
disposed in virtels 30 to 38 fx apart along the conidiophore. Conidiophores
consisting of a terminal branch and two primary virtels, are about 100
to 120 M in length, while those bearing four primary virtels measure 250
to 300 fi. Conidiophores with 7 to 8 primary virtels are occasionally
Mar.4, i9i8 WUt Diseases of Okra 537
seen in petri-dish cultures. The terminal branch is usually i to 3 times
longer than the virtel branches and measures from 1 5 to 60 ju.
The mycelium is hyalin at first, becoming brownish with age. It is
septate, 2 to 4 jit in diameter, but is often swollen, all gradations from
slightly swollen threads to large, thick-walled, and knotted sclerotia,
and chlamydospore-like cells being present (PI. 17, G). The sclerotium
is at first but swollen, closely septate mycelium, which enlarges and
knots itself into a variety of forms. The measurements of sclerotia of
strains 1685, 1717, 2784, and 2985, recorded in Table I, were made only
from the more or less rounded aggregations. Such measurements of
irregular formations varying so greatly in size are only of general sig-
nificance. Certain strains of V. albo-atrum cultivated by the writer
produce sclerotia abundantly in a few days in petri-dish cultures. Macro-
scopically, their presence is manifested by a beautiful black zonation of
the colonies, as illustrated in Plate 19. In other strains, and sometimes
in other cultures of the same strain, these sclerotial rings either do not
develop at all or only after a long time ; yet these forms produce abun-
dant sclerotia if cultivated on potato cylinders and other vegetable media.
The entire growth then frequently consists of a black confluent layer of
sclerotia and hyphae. Parallel cultures of the same strain differ suffi-
ciently with respect to the characters of the conidiophores and sclerotia for-
mation so that specific determinations based on slight differences of these
characters and unsupported by inoculation tests are of doubtful value.
Verticillium albo-atnun Reinke and Berthold.
Conidia ellipsoidal, iinicellular, 4.0 to 11. o by 1.7 to 4.2 ft, abscissed singly from
the sterigma tips of verticillate conidiophores. Primary whorls or virtels of branches,
I to 8 in number, 30 to 90 n apart, sometimes bearing secondar}- virtels. Branches
I to 7, usually 3 to 5 in number, 13 to 38 /x long, tapering, straight to slightly bowed.
Conidiophores 100 to 300 ^ or more in length. The terminal branch of the conidio-
phore is from 1 5 to 60 m long. Conidia may or may not collect in heads on the sterigma
tips. Mycelium septate, hyalin to brown with age; may become swollen into
chlamydospore-like chains of closely septate, knotted masses. These aggregates
constitute the sclerotia of this fungus. Vascular parasite, cause of a wilt disease of
okra, potato, eggplant, cotton, snapdragon, and probably of species of Abutilon and
Xanthium, ginseng, black raspberry, China aster, and dahlia. V. albo-airum may
prove to be the cause of the Verticillium-wilt disease reported on currants and goose-
berries by Aderhold (/).
OCCURRENCE OF FUSARIUM VASINFECTUM .\ND VERTICILLIUM ALBO-
ATRUM IN WILT DISEASES. OF OKRA
VerticUlium albo-atrum was found constantly inhabiting the vascular
system of wilt-diseased okra plants (PI. 27) in New Jersey, where this
crop is of considerable importance. This organism was also isolated from
similar material from Monetta, S. C, Birmingham, Ala., Middle River,
Cal., and Medford, Oreg., specimens from Middle River and Medford
having been collected by Dr. Wollenweber.
On the other hand, a Fusariura indistinguishable from the cotton-
wilt Fusarium was constantly obtained from wilting okra collected at
538 Journal of Agricultural Research voI.xii.no.?
Florence, Sumter, and Charleston, S. C, and Wrightsboro, N. C. Accord-
ing to Clinton (8), the Fusarium-wilt of okra occurs in Connecticut.
Fusarium vasinfectum was not obtained by the writer from okra-wilt in
New Jersey; and VerticUlium alho-atrum was isolated from this host in
the South only in the few mentioned localities.
V. alho-atrum was isolated from a wilt disease of the weeds AhutUon
sp. and Xanthium sp. in New Jersey ; and from spontaneous wilt of cotton
plants in rows adjoining the experimental plot at Arlington, Va.
The wilt diseases of the several plants brought about by F. vasinfectum
and V. albo-atrum manifest the same symptoms, so that the real cause of
the trouble is safely to be determined only by cultural means. There is
a lack of turgor in the leaves first in evidence in those parts farthest
removed from the veins (PI. 20-23). The lower leaves are first affected,
wilt, and drop ofif one or two at a time. Frequently the plant does
not die for a long time, but continues a dwarfed existence. This is
especially true of the Verticillium-wilt. If the plants are cut longitudi-
nally and crosswise, it will be seen that the vascular tissue is brown or
black (PI. 18), the discloration being traceable from the small roots to the
top of the stem and into the petioles. Microscopic examination of thin
sections of this material shows that the vessels are plugged with the
mycelium of the parasite, which interferes with the conduction of moisture
to the aerial portions of the plant (PI. 17, E). The host tissues do not
appear to be invaded or broken down, the vascular inhabitant merely
living as a saprophyte on the fluids of the vessels, and injuring the host
plant only by mechanical obstruction of the latter. Whether there are
injurious products secreted in the metabolism of the fungus detrimental
to the plant in other ways is yet to be demonstrated. The parasitism of
these fungi is, then, a mechanical interference with the nutrition of the
host and not our usual conception of this term.
METHOD OF TESTING PARASITISM
In order to demonstrate the ability of V. albo-atrum and F. vasinfectum
to produce wilt diseases of okra and to gain a knowledge of the relation
of these organisms to other host plants, approximately i ,000 inoculations *
and cross-inoculations with pure cultures were made. Strains of V. albo-
atrum, isolated from okra, eggplant, potato, and snapdragon, were used
to inoculate okra; strains from okra, snapdragon, and eggplant were
used to inoculate eggplant; and the strain from okra was used to inocu-
late cotton. Similarly okra plants were inoculated with F. vasinfectum
isolated from okra and cotton ; and cotton was inoculated with this fungus
isolated from cotton and okra.
The general method used in the inoculations may be summarized as
follows: Selected okra seed, or seed of other plants to be used, were disin-
fected in a solution of formalin, rinsed in sterile water, and planted in
> The writer is indebted to Mr. J. M. R. Adams for faithful assistance in the inoculation tests.
Mar. 4. 1918 Wilt Diseases of Okra 539
sterilized soil in the greenhouse, or in soil new to these crops in field
plots. Soil inoculations were made by pouring a few cubic centimeters
of a sterile-water spore suspension of the organism to be used on the
steam-sterilized soil, either before or after planting the seed, or in the
vicinity of the seedlings. The majority of the inoculations were made
through wounds at the hypocotyl below the soil level. The wounds
were made with a sterile scalpel, and after the inoculum was intoduced,
the soil was replaced to prevent drying. A few inoculations were made
through wounds made by breaking off leaves and pods of the okra plants.
Approximately as many control plants were prepared as plants for
inoculation. Half of these were wounded or otherwise subjected to the
treatment employed on the plants inoculated, and the other half were
left as imwounded controls. For the wounded controls, sterile water
was used for inoculating, in lieu of the spore suspension. Wilting plants
were examined microscopically for mycelium in the xykm of roots,
stem, or petioles; and pure cultures of the organism used were reisolated
and identified as a control on the work. Frequently such reisolated
strains were used for subsequent inoculations.
In but one experiment was there any wilt in the controls. In this
case after six weeks 6 per cent wilted, and the organism used in the
experiment was recovered from the interior of the plants. However,
this was a field test and there was in the duration of the experiment
an opportunity for infection from inoculated plants in the adjoining
rows on either side.
INOCULATION OF VARIOUS ECONOMIC PLANTS WITH VERTICILLIUM
ALBO-ATRUM AND FUSARIUM VASINFECTUM
Inasmuch as the inoculation experiments were carried on in a uniform
manner and controlled by a large number of wounded and unwounded
plants, as well as by reisolation and identification of the organisms
causing the disease, the results are comparable. For convenience these
results are brought together in tabular form (Tables III, IV, V) and the
discussion of the results of the inoculations are arranged according to the
host plant and the parasite used.
HISTORY OF THE STRAINS OF VERTICILUUM ALBO-ATRUM AND FUSARIUM
VASINFECTUM USED FOR INOCULATION
The history of the various strains of V. albo-atrum and F. vasinfectum
used for inoculating the several host plants is as follows :
F. vasinfectum strain 1855. Reisolated from the vascular system of a
cotton plant, which was wilting as a result of inoculation with strain 1733,
a reisolation of strain 1635, which was in turn a reisolation of the original
strain 1485, isolated from the discolored vascular system of a wilting
cotton plant at Florence, S. C, in 191 2.
F. vasinfectum strains 2708, 2709, 3203. Isolated from the discolored
vascular system of wilting okra plants collected at Florence, S. C. at
540
Journal of Agricultural Research
Vol. XII, No. 9
different times. Strain 3592 is a reisolation from an okra plant inoculated
with strain 3203.
F. vasinfecium strain 3210. Isolated from the vascular system of an
okra plant collected at Sumter, S. C.
V. albo-airum strain 1717. Isolated from the interior of the stem of a
wilting okra plant at Monetta, S. C. Strains 2943, 3075, 3076, and 3156
are reisolations from okra plants inoculated with strain 171 7.
V. albo-atrum strain 2821. Isolated from wilting okra plant collected
at Middle River, Cal.
V. albo-atrum strain 1685. Isolated from wiltmg plant of Solanum
melongena.
V. albo-atrum strain 2784. Isolated from wilting potato plant.
V. albo-atrum strain 2985. Isolated from wilting snapdragon plant
by Brown (4).
Table III. — Results of inoculating okra plants with Verticillium albo-airumfrom various
hosts
Num-
ber of
plants
Age.
Locality.
Method of
inoculation.
Species and strain of organism.
Incu-
bation
pe-
riod.
Per-
cent-
age
ulti-
mate-
ly
wilt-
ing.
Time.
40
60
Days.
17
n
37
46
90
30
46
46
46
90
46
46
30
60
60
60
60
60
Greenhouse .
do
Soil
V. albo-atrum, okra
(1717)-
do
Days.
12
40
40
II
13
18
12
15
13
13
18
88
59
55
100
81
100
92
66
17
40
ICO
0
0
0
75
71
36
73
45
Days.
21
do
60
60
do
do
V. albo-airum, Irish
potato (2784).
V. albo-atrum, okra
(1717).
do
60
40
TlS
do
do
Wounds: hy-
pocotj'l.
.... do
30
^O
do
do
do
^0
13
do
do
do
10
do
do
V. albo-airum, okra
(3075)-
V. albo-atrum, okta
(3076).
V. albo-airum, okra
(2943)-
V. albo-atrum, okra
(2821).
V. albo-atrum, snap-
dragon (2985).
V. albo-atrum, okra
(1717).
do
30
17
40
9
40
40
26
do
do
do
do
do
do
do
do
30
30
24
30
do
do
Wounds: ped-
icel.
....do
30
45
43
23
58
14
19
26
Field
.. .do.. .
Wounds: hy-
pocotyl.
do
do
10
14
10
10
10
V. albo-atrum, okra
(3156).
V. albo-airum, eggplant
(1685).
V. albo-atrum, Insh
potato (2784).
V. albo-atrum, snap-
dragon (2985.).
do
do
43
do
do
43
22
. do... .
do
43
" Soil inoculated before planting the seed.
Mar. 4, igi8
''ill Diseases of Okra
541
Table IV. — Results of inoculating okra plants with Fusarium vasinfectum from cotton
and okra
Num-
ber of
plants
Age.
Days.
40
17
40
17
40
37
99
21
40
37
40
37
30
24
50
49
100
10
20
60
17
21
Locality.
Greenhouse
do
.... do
.do.
.do.
.do.
.do.
.do.
.do.
.do.
.do.
_ Method of
inoculation.
Soil.
.do.
Wounds: hy-
pocotyl.
do
do
.do.
.do.
.do.
.do.
.do.
.do.
Species and strain of organism.
F. vasinfectum, okra
(2709).
F. vasinfectum., cotton
(1855)-
do
.do.
F. vasinfectum; okra
(2709).
F. vasinfectum, okra
(2708).
F. vasinfectum, okra
(3203).
do
do
F. vasinfectum, okra
(3210).
F. vasinfectum, okra
(3592)-
Incu-
bation
pe-
riod.
Days.
17
17
Per-
cent-
ape
ulti-
mate-
ly
wilt-
ing.
50
55
Time.
Days.
56
56
5f
36
56
56
15
30
18
23
36
Table V. — Results of inoculating cotton, eggplant, and Brabham coivpeas with tite okra-
wilt organisms through wounds at the hypocotyl
Host.
Num-
ber
of
Age.
plants.
Days.
14
60
20
60
18
21
43
21
7
45
5
45
7
45
16
24
Ixxality.
Species and strain of
organism.
Per-
Incu-
ba-
tion
peri-
od.
cent-
age
ulti-
mate-
ly
wilt-
ing.
Days.
0
14
80
17
89
17
87
10
85
10
80
10
100
0
Time.
Cotton
Do
Do
Do
Eggplant
Do
Do
Brabham cow-
peas.
Field
....do
Greenhouse .
....do
....do
do
....do
....do
F. vasinfectum , okra
(3210).
V. albo-atrum, okra
(3156).
F. vasinfectum, cotton
(1855).
F. vasinfectum, okra
(3592).
V. albo-atrum, okra
(1717)-
V. albo-atrum, snap-
dragon (2985).
V. albo-atrum, egg-
plant (1685).
F. vasinfectum, olcra
(3203).
Days.
23
23
36
36
15
15
15
60
^42 Journal of Agricultural Research voi. xii, no. 9
INOCULATION OF OKRA
No difficulty was experienced in securing typical wilting okra plants
(PI. 21, 22) by the inoculation of the soil or of the plants through wounds
at the hypocotyl with pure cultures of V. albo-airum isolated from okra,
eggplant, potato, or snapdragon. By the inoculation of the soil in the
vicinity of 17-day-old plants with strain 171 7 isolated from okra, 88
per cent of wilting plants were obtained in 2 1 days. Inoculations of the
soil with this strain before planting the seed resulted in 59 per cent of
wilt. Inoculation of S7- to 90-day-old okra plants with this strain through
the hypocotyl gave from 75 to 100 per cent of wilting plants. The
reisolated strains of V. alho-atrum 171 7 gave the following results: Strain
2943, 40 per cent; strain 3075, 66 per cent; strain 3076, 16 per cent;
strain 3156, 71 per cent. V. alho-atrum, strain 2821, isolated from a
wilting okra plant from Middle River, Cal., produced 100 per cent of
wilting plants through hypocotyl inoculations.
The introduction of V. alho-atrum into wounds of the pedicel and the
stem of okra plants was without apparent effect, thus indicating that
there is no danger of carrying the wilt disease from plant to plant by the
cutting knife.
The inoculation of okra plants with V alho-atrum from hosts other
than okra gave the following results: Strain 2985, isolated from snap-
dragon, produced from none to 45 per cent of wilt through hypocotyl
inoculation; strain 2784 from potato gave 73 per cent through hypocotyl
wounds and 55 per cent through soil inoculation previous to planting the
seed; strain 1685, isolated from eggplant, gave 43 per cent of wilt through
the inoculation of 60-day-old plants at the hypocotyl.
The results of inoculating okra plants with F. vasinfectum were for
some time disappointing. However, after an insight was gained into
the conditions necessary to induce the Fusarium-wilt of okra, better
results were obtained. With the' species of Fusarium from okra in two
trials, 50 and 55 per cent of wilting plants were secured by inoculation
through the hypocotyl (PL 23). But with the species of Fusarium from
cotton only i per cent was obtained. However, the cotton inoculated
with the species of Fusarium (3592) from okra and the strain (1855) from
cotton gave a high percentage of wilting plants in a parallel test of the
two strains (PI. 24-26).
INOCULATION OF COTTON
Fusarium vasinfectum strain 1855 from cotton, when introduced into
wounds at the hypocotyl of 21-day-old plants, gave 50 per cent of wilt
in 17 days and a total of 89 per cent in 5 weeks (PI'. 25). Similarly, in a
parallel test, F. vasinfectum strain 3592 from okra gave 42 per cent of
wilt in 17 days and a total of 87 per cent in 5 weeks (PI. 26). In contrast
to these results are those secured with these strains on okra in the parallel
Mar. 4. 1918 Wilt Diseases of Okra 543
test where but i and 6 per cent of wilt was secured. Verticillium albo-
airum stram 3156 inoculated through wounds of the hypocotyl gave 80
per cent of wilt in 14 days.
The Verticillium- wilt of cotton, as observed in these tests can not be
distinguished from the very destructive Fusarium-wilt of this crop imless
a critical examination be made. It is possible that this wilt is present
in the cotton fields of the South and has been overlooked, owing to the
similarity of the two diseases.
INOCULATION OF EGGPLANT
Eggplant was found to be susceptable to the strains of V. alho-atrum
isolated from okra, eggplant (PI. 20), and snapdragon. Strain 171 7
from okra gave 85 per cent of wilt in 15 days; strain 1685 froni eggplant
gave 100 per cent in the same time, and strain 2985 from snapdragon
80 per cent correspondingly. The inoculations were made through
wounds of the hypocotyl on 45-day-old plants.
INOCULATION OF BRABHAM COWPEAS
Twenty-four-day-old seedlings of Brabham cowpeas were inoculated
through the hypocotyl with F. vasinjectum strain 3203 from okra.
Although held under observation for 60 days, no cases of wilt were
detected.
CONCLUSIONS
There are two similar wilt diseases of okra, caused, respectively, by
Fusarium vasinfectum and Verticillium alho-atrum. These diseases
can be differentiated only by isolating the causal fungi. The Fusarium-
wilt in general is more serious in the southern range of okra growing,
while the Verticillium-wilt is more serious in the northern range of this
crop. However, the former has been reported in Connecticut, and the
latter occurs in South Carolina and Alabama.
The wilt diseases of the Irish potato offer a parallel case of two organ-
isms producing the same disease symptoms. Here, the Fusarium-wilt
is induced by F. oxysporum and the Verticillium-wilt by V. alho-atrum,
as in okra. Similarly, it is demonstrated in this paper for the first time
that cotton may have both wilt diseases. No doubt there are several
other economic plants, which act as host to vascular parasites of the
genera Fusarium and Verticillium — for example, China aster, ginseng,
eggplant, brambles.
V. alho-atrum was isolated from the discolored vascular system of
wilting plants of okra, eggplant, potato, and species of Abutilon and
Xanthium, and was demonstrated in these studies to be the cause of a
wilt disease at least of okra, eggplant, and cotton. It was identified
from a wilt disease of snapdragon after its causal relation to this disease
had been established by Miss Nellie A. Brown, of the Bureau of Plant
544 Journal of Agricultural Research voi. xii, N0.9
Industry. Okra is susceptible to inoculation with V. albo-arttim from
okra, snapdragon, eggplant, and potato; and to F. vasinjectum from
okra. Eggplant is susceptible to V. alho-atrum from eggplant, okra,
and snapdragon. Cotton is susceptible to the strains of this fungus
from okra and to F. vasinjectum from okra and cotton.
Thus, it is apparent that the species of Fusarium causing the wilt
disease of okra is identical with F. vasinjectum. Likewise it is evident
that V. alho-atrum is a serious vascular parasite of a number of different
economic plants. In view of the fact that the genus Acrostalagmus
should be combined with the earlier genus Verticillium, it seems probable
that V . alho-atrum is the fungus described by Van Hook under the name
''Acrostalagmus alhus," the cause of ginseng-wilt; by Lawrence as A.
caulophagus, the cause of the bluestem disease of black raspberry; and
by Gueguen as A. vilmorinii, the cause of a wilt disease of China aster
and associated with a disease of cacao fruits.
V . alho-atrum and F. vasinjectum are readily cultivated artificially,
indicating that both are capable of persisting saprophytically in the soil
for an indefinite period in the absence of the preferred host. There
was no indication that either fimgus is carried from field to field or from
year to year by the seed or from plant to plant by the cutting knives. .
As a control measure it may be suggested that seed be selected only
from healthy plants. If extreme precaution is to be taken, the seed
may be disinfected in a formalin solution (i to 240) for two hours.
Since okra, eggplant, potato, cotton, snapdragon, and the weeds
Xanthium spp. and Abuiilon spp. are all susceptible to the Verticillium-
wilt, as well as ginseng, China aster, and black raspberry, as seems
probable, these plants should be taken into consideration in planning
a rotation to eliminate wilt diseases. Similarly, okra and cotton are
hosts of F. vasinjectum and should not follow each other in rotation if
best results are expected.
LITERATURE CITED
(i) Aderhold, Rudolf.
1907. UBER EINE THROMBOSE DES JOHANNIS- UND STACHELBEERE. In Mitt.
K. Biol. Anst. Land- u. Forstw., Heft 4, p. 26-27.
(2) Appei<, Otto.
1909. EiNiGES UBER DIE BI.ATTR0LLKRANKHE1T DER KARTOFPEL. /n Jahresbcr.
Ver. Angew. Bot., Jahrg. 6, 1908, p. 259-265.
(3) Atkinson, G. F.
1892. SOME DISEASES OP COTTON. Ala. Agf. Exp. Sta. Bui. 41, 65 p., 25 fig.
(4) Brown, Nelue A.
1914. A SNAPDRAGON WILT DUE TO VERTiciLUUM. In Phytopathology, V. 4,
no. 3, p. 217.
(5) Butler, E. J.
1910. THE WILT DISEASE OP PIGEON-PEA AND THE PARASITISM OP NEOCOSMOS-
PORA VASINFECTA SMITH. Mem. Dept. Agr. India, Bot. Ser., v. 2,
no. 9, 64 p., 6 pi. (part col.). Bibliography, p. 63-64.
Mar. 4. i9ig IVilt Diseases of Okra 545
(6) Carpenter, C. W.
1914. THE vERTiciLLiuM w^LT PROBLEM. (Abstract.) In Phytopathology, v. 4,
no. 6, p. 393.
(7)
1915. SOME POTATO TUBER-ROTS CAUSED BY SPECIES OP FUSARIUM. In JoUT.
Agr. Research, v. 5, no. 5, p. 183-209. Literature cited, p. 208-209.
(8) CUNTON, G. P.
1906. NOTES OX FUNGUS DISEASES, ETC., FOR 1905. In Cotui. AgT. Exp. Sta.,
29th Ann. Rpt., [1904/05], p. 263-303, fig. 8-9. Literatiire, p. 301-303.
(9) CORDA, A. C. I.
1838. icoNES FUNGORUM HucusQUE coGNiTORUM. t. 2. Prague.
(10) Gu6guen, Femand.
1906. acrostalagmus vilmorinii n. sp., muceidinee produisant une mala-
DiE a scljSrotes du collet des reines-marguerites. In Bui. Soc.
Mycol. France, t. 22, fasc. 4, p. 254-26^, 5 fig., pi. 16.
(11)
1910. SUR UNE MALADIE DU FRUIT DE CACAOYER PRODUITE PAR UNE MUC^-
din6e ET SUR LE mecanisme DE l'infection In Compt. Rend. Soc.
Biol. [Paris], t. 68, no. 5, p. 221-222.
(12) Higgins, B. B.
1911. IS neocosmospora vasinfecta (atk.) smith, the perethecial stage
OF THE FUSARIUM WHICH CAUSES COWPEA WILT? In N. C. AgT. Exp. Sta.,
32d Ann. Rpt. [i9oS]/o9, p. 100-116, 16 fig.
(13) Hoffmann, Hermann.
1854. SPERMATIEN BEI £INEM FADENPILZE. In Bot. Ztg., Jahrg. 12, No. l6,
p. 249-254, no. 16, p. 265-269.
(14) Klebahn, H.
1913. beitrage zur kenntnis der fungi imperpscti. In Mycol. Centbl.,
Bd. 3, Heft 2, p. 49-66, 15 fig.
(15) Lawrence, W. H.
1912. BLUESTEM of THE BLACK R.^SPBERRY. /w Wash. Agr. Exp. Sta. Bui.
108, 30 p., fig. 21-31, I pi.
(16) Orton, W. a.
1900. THE wilt DISEASE OF COTTON AND ITS CONTROL. U. S. Dept. Agr. Div.
Veg. Physiol, and Pathol. Bui. 27, 16 p., 4 pi.
(17)
1914. POTATO WILT, LEAF-ROLL, AND RELATED DISEASES. U. S. Dept. AgT.
t Bul. 64, 48 p., 16 pi. Bibliography, p. 44-48.
(18) Pethybridge, G. H.
1910. potato diseases in IRELAND. In Dept. Agr. and Tech. Instr. Ireland,
Jour., V. 10, no. 2, p. 241-256, 8 fig.
(19)
1911. investigations on potato DISEASES, (second report). In Dept.
Agr. and Tech. Instr. Ireland, Jour., v. 11, no. 3, p. 417-449, 14 fig.
(20) Rankin, W. H.
1910. root rots of ginseng. In Special Crops, v. 9, no. 94, p. 349-360, 14
fig. Bibliography, p. 359-360.
(21)
1914. THROMBOTIC DISEASE OP M.\PLE. (Abstract.) In Phytopathology, v. 4,
no. 6, p. 395-396.
(22) Reinke, J., and Berthold, G.
1879. DIE zersetzung der kartoffel durch pilze. Untersuch. Bot. Lab.
Univ. Gottingen, Heft 1, 100 p., 9 pi.
38324°— 18 2
546 Journal of Agricultural Research voi. xii, no. 9
(23) Smith, E. F.
1899. WILT DISEASE OF COTTON, WATERMELON, AND COWPEA (nEOCOSMOS-
PORA NOV. GEN.)- U- S. Dept. Agr. Div. Veg. Physiol, and Pathol.
Bui. 17, 72 p., 10 pi. (part col.). Previous literature, p. 50.
(24) SpiEckermann, a.
191 1. BEITRAGE ZUR KENNTNIS DER BAKTEIUENRING UND BLATTROLL-KRANK-
HEiTEN DER kartoffelpflanze. /« Jahrcsbet. Angew. Bot., Jahrg
8, 1910, p. 1-19, 173-177-
(25) Stevens, F. L., and Wilson, G. W.
1912. okra WILT (fusariose), fusarium vasinfectum, and clover rhizoc-
TONiosE. In N. C. Agr. Exp. Sta., 34th Ann. Rpt., [igioj/n, p.
70-73, fig. 15-18.
(26) Stormer, K.
1910. OBSTBAUMSTERBEN UND kartoffelblattrollkrankheit. In Jahres-
ber. Ver. Angew, Bot., Jahrg. 7, 1909, p. 1 19-170, 15 fig., pi. 5.
(27) Van Hook, J. M.
1904. SOME DISEASES OF GINSENG. N. Y. Cornell Agr. Exp. Sta. Bui. 219,
p. 165-186, fig. 18-42.
(28) WhETZel, H. W., and Rosenbaum, J.
1912. THE DISEASES OF GINSENG AND THEIR CONTROL. U. S. Dept. Agr. Bur.
Plant Indus. Bui. 250, 44 p., 5 fig., 12 pi.
(29) Wilson, G. W.
1913. FUSARIUM OR vERTiciLLiuM ON OKRA IN NORTH CAROLINA? In Phyto-
pathology, V. 3, no. 3, p. 183-185.
(30) WOLLENWEBER, H. W.
1911. UNTERSUCHUNGEN XJBER DIE NATXJRLICHE VERBRElTUNG DER FUSARIEN
AN DER KARTOFFEL. In Mitt. K. Biol. Anst. Land. u. Forstw., Heft
II, p. 20-23.
(31)
(32)
(33)
1913. PILZPARASITARE WELKEKRANKHEITEN DER KtrLTURPFLANZEN. In Ber.
Deut. Bot. Gesell., Bd. 31, Heft i, p. 17-34-
1913. STUDIES ON THE FUSARIUM PROBLEM. In Ph3rtopathology, V. 3, no. I,
p. 24-50, X fig., pi. 5. Literatture, p. 46-48.
I914. IDENTIFICATION OF SPECIES OF FUSARIUM OCCURRING ON THE SWEET
POTATO, IPOMOEA BATATAS. In Jour. Agr. Research, v. 2, no. 4, p.
251-286, pi. 12-16 (part col.). Literature cited, p. 284-285.
(34) and SCHLUMBERGER, Otto.
191 1. INFEKTIONSVERSUCHE MIT KARTOFrELBEWOHNENDEN PILZEN. In Mitt.
K. Biol. Anst. Land. u. Forstw., Keft 11, p. 15-17.
PLATE A
Fusarium vasinfecHvm on vegetable media:
1-3. — Growth on steamed potato. Both potato cultures show pionnotes.
2 , 4. — Growth on rice .
Cultures i and 2 were grown in a strong north light; 3 and 4 in a subdued light.
Wilt Diseases of Okra
y
Journal of Agricultural Research
Plate A
Vol. XII. No. 9
PLATE 17
A-H. — Verticillium albo-atrum:
A, Simple conidiophores and conidia. X 1,000.
B, Same showing, respectively, the collection of the conidia on the sterigma in
irregular aggregations in dry air, and in water drops in humid air. X 1,000.
C, Verticillate conidiophores bearing one and three whorls, or virtels, of branches,
respectively. X500.
D, Verticillate conidiophore having conidial heads, from humid environment —
that is, moistiu-e drops in which the conidia float as in figure B. This is the so-called
Acrostalagmus type of conidial head. X500.
E, Mycelium of V. albo-atrum in the vascular ducts of an okra plant inoculated
with this fungus. X250.
F, H, Germinating conidia: F, Xsoo; H, Xi,ooo.
G, Swollen, sclerotia-like mycelium. X500.
I-M. — Fusarium vasinfectwm:
I, Terminal, intercalary and conidial chlamydospores. i, Germinating terminal
chlamydospore. 2, Free mature chlamydospore. X 1,000.
K, Germinating macroconidium. X 1,000.
1-,, F. vasinfectumirova.okxa-WAt. Twelve macroconidia. X 1,000.
M, F. vasinfectum from cotton-wilt. Four macroconidia. X 1,000.
Wilt Diseases of OPcra
Plate 17
Journal of Agricultural Research
Vol. XII. No. 9
Wilt Diseases of Okra
Plate 18
Journal of Agricultural Research
Vol. XII, No. 9
PLATE i8
Longitudinal section of an okra plant naturally infected with Verticillium albo-
atrum, showing the typical appearance. The vascular elements are discolored from
the roots to the pedicels and petioles. About natural size.
PLATE 19
Verticillium albo-atrum:
Two-weeks-old colony on potato agar, showing the concentric rings of black scle-
rotial bodies. X4.
Wilt Diseases of Okra
Plate 19
Journal of Agricultural Rcacarcii
Vol. XI I, No. 9
Wilt Diseases of Okra
Plate 20
Journal of Agricultural Research
Vol. XII, No. 9
PLATE 20
Solarium melongena, showing effect of wilt:
A. — Control plant of the same age as the wilted plant (B). Photographed at the
same time as figure B. X2/5.
B. — Wilted plant photographed two months after inoculation at the hypocotyl
with Verticillium albo-airum isolated from wilted eggplant. The organism was
recovered from the stem of the small plant 10 cm. above the point of inoculation.
X2/5.
PLATE 21
Abelmoschus escuknttis, showing effect of wilt:
A. — Control plant. Photographed at the same time as figure B. X (about) 1/2.
B. — ^Wilted plant photographed two weeks after inoculation at the hypocotyl with
a pure culture of Verticillium albo-atrum. X (about) 1/2.
Wilt Diseases of OPcra
Plate 21
Journal of Agricultural Research
Vol. XII. No. 9
Wilt Diseases of Okra
Plate 22
Journal of Agricultural Research
Vol. XII, No.9
PLATE 22
Abelmoschus esculentus, showing effect of wilt:
A. — Wilted plant inoculated with Verticilium albo-atrum. X (about) 1/3.
B. — Control plant of the same age as wilted plant. X (about) 1/3.
Both plants were photographed two months after the wilted plant had been inocu-
lated.
PLATE 23
Abelmosckus esculentus, showing the effect of wilt as a result of inoculation with
Fusarium vasinfecium isolated from okra-wilt. The inoculation was made at the
hypocotyl and the organism was recovered from the vascular tissues of the petioles
two weeks afterwards — that is, immediately after the photograph was taken. X2/3.
Wilt Diseases of Okra
Plate 23
Journal of Agricultural Research
Vol. XII, No. 9
Wilt Diseases of Ol<ra
Plate 24
Journal of Agricultural Researcli
Vol. XII, No. 9
PLATE 24
Gossypium lierbaceum (Columbia variety):
Control plants 35 days old. Photographed 15 days after having been wounded at
the hypocotyl. Natural size.
PLATE 25
Gossypium herhaceum (Columbia variety), showing effect of wilt:
Wilting plants photographed 15 days after inoculatiou at the hypocotyl with Fusa-
rium vasinfecium, strain 1855, isolated from wilting cotton plants. The plants are
the same age as the control plants in Plate 24. Natural size.
Wilt Diseases of Okra
Plate 25
Journal of Aj/ricultural Rusuarch
Vol. XII, No. 9
Wilt Diseases of Ol<ra
Plate 26
Journal of Agricultural Research
Vol. XII, No. 9
PLATE 26
Gossypium kerbaceum (Columbia variety), showing effect of wilt:
Wilting plants photographed 15 days after inoculation at the hypocotyl with Fusa-
Tium vasinfectum, strain 3592, isolated from wilting okra. The plants are the same
age as those in Plates 24 and 25. Natiural size.
PLATE 27
Abelmoschus esculentus, showing the characteristic symptoms of the wilt produced
by Verticillium albo-atrum. Photographed in a field in New Jersey.
Wilt Diseases of Okra
Plate 27
Journal of Agricultural Research
Vol. XII, No. 9
WINTER CYCLE OF EGG PRODUCTION IN THE RHODE
ISLAND RED BREED OF THE DOMESTIC FOWL
By H. D. GooDALE,
Massachusetts Agricultural Experiment Station
INTRODUCTION
The winter cycle of egg production is one of the internal factors
concerned in determining total production. It was first recognized by
Pearl and Surface in Barred Plymouth Rocks. They state (7, p. gg-
100) : ^
It will undeniably be advantageous, in studying certain phases of the problem of
egg production, to endeavor to use a time unit which conforms to the natural
periodicity displayed by hens [italics are mine — H. D. G.]. . . .
The plan followed at the present time in the investigations in progress at the
Maine Station breaks the laying year up into four parts. The first of these includes
the months of November, December, January, and February. Broadly speaking,
this is the period of winter laying and is so designated. . . .
The justification for the conclusion that this division of the year is in general a
natiu-al one and corresponds to a real cyclical periodicity in egg production is in con-
siderable measure to be found in the facts regarding mean monthly egg production
and variation in this character set forth in this and the preceding section of the paper.
The winter-laying period is a period characterized by rapid increase in mean pro-
duction associated with a relatively equally rapid decrease in variability, both
absolute and relative. In this period a large part of the flock falls in the A compo-
nent of the monthly distributions (see p. 142). This laying period is, strictly speak-
ing, not a part of the natural or normal reproductive cycle of the hen. Egg laying
in this part of the year is something which diu^ing domestication has been added on,
as it were, to the natural reproductive activity of the wild Callus. It is a result of
"forcing" or special stimulation. From the evolution standpoint, egg production
in these months is a comparatively recent acquisition. Such being the case, the
greater variability observed in winter laying is only what would be expected.
The limits of this artificial winter cycle of egg production are fairly well defined.
It begins with the beginning of the laying year. Its other limit is marked by the
slacking up in egg production, which occurs in February (see Fig. i). This slacken-
ing up in February, which appears to be a characteristic of egg production, generally,
is to be explained, we believe, chiefly if not entirely as the result of the ending of
the winter cycle by the majority of birds which have laid during the early winter.
Such birds rest for a period at about this time before beginning the spring laying
cycle. Of cotu-se it must be understood that these statements are made only with
reference to what might be called the general or average course of events. Particular
birds may form exceptions in their laying. Many birds, of course, have no proper
winter cycle of laying at all. They begin to lay for the first time in January or Feb-
ruary, and keep on laying without any large break straight through the spring cycle.
' Reference is made by number (italic) to " Literature cited," p. 574.
Journal of Agricultural Research, Vol. XII, No. 9
Washington, D. C. Mar. 4, 1918
mf Key No. Mass. — 3
38324°— 18 3 (547)
548 Journal of Agricultural Research voi. xii. no. 9
In discussing the monthly egg distribution Pearl and Surface (7, p.
89-90) state:
Considering the form of the polygon somewhat more in detail, we note that the
line starts from a low point in November and rises rather rapidly and in almost a
straight line to January. The slope of the line from January to February is down-
ward. In other words, there is an indentation in the ascending limb of the egg-pro-
duction polygon in the month of February. This is a very characteristic featiire of
the distribution of egg production, not only observed with the birds here under
discussion but also in published records from other sources.
A study of Table i shows that this is generally true for every year covered by the
investigation. While the February mean production is not necessarily lower than the
January, though this is true in many cases, there is a perceptible slowing of the rate
of increase in egg production which has obtained up to that time. The most prob-
able interpretation of this appears to us to be that the February indentation in the
egg-production curve represents a rest or reaction after the winter laying and in
anticipation of the heavy March and April production. It marks the completion of a
laying cycle on the part of those birds which have been laying dturing the winter
months.
The mean egg production for February, however, seems to rest on a
basis of 28 days. If reduced to a basis of 31 days, the mean production
is 12.03 eggs, a value, however, that produces a pronounced indentation
in the upward slope of the polygon. Pearl believes that this change in
the slope of the polygon, disregarding entirely any actual drop in pro-
duction, is indicative of a winter cycle. However, a change in the
slope of such a polygon would occur if a flock of birds began to lay in
some given month, gradually increased in production for a definite
period of some length until they reached a maixmum and maintained
this maximum for a period of several months. Pearl, moreover, is
inclined to believe that the change in mean temperature that occurs in
March bears no causal relation to the increase in egg production that
is observed at this season. Now, Avhile it is clear that there are other
factors than the change in temperature that increases production at
this season,, it is not at all clear that temperature can be entirely elimin-
ated as a factor. In the imaginary case just mentioned it seems pos-
sible that the maximum production under one set of environmental
conditions might be different from that observed under another set.
Hence, it seems entirely probable that the maximum production
possible in January and February would not be as high as in March,
and that the indentation noted in the curve of production merely means
that the maximum production possible for midwinter conditions has
been reached.
In several other papers, notably those of Pearl (5, 6), the winter cycle
is again discussed. In the former paper he states (5, p. 173-174) :
(2) The upper limit of the winter period at March i is arbitrary, and only approxi-
mately coincides with the biological reality. Actually with most birds the spring
or reproductive cycle of production (cf. 37) begins in the latter part of February.
In handling the material it has been found necessary (for reasons which will be
Mar. 4. 1918 Winter Egg Production of Rhode Island Reds 549
obvious upon consideration of the matter) to take a fixed date for the beginning of
the spring cycle of laying and the ending of the winter cycle. The records of the
station prior to 1908 are tabulated only for months (the daily records unfortunately
having been destroyed before I took charge of the work), and on this account it is
necessary to take the working limit of the winter cycle at the end of a calendar month.
Since March i comes the nearest to the biological limit of any date which is also the
beginning of a calendar month it has been chosen. The error introduced by taking
this arbitrary date for a point which really shifts within rather narrow limits is, on
the average, small. However, it must b.e recognized as a disturbing element in the
individual case. Thus, some birds which really lack any genetic factor for winter
production will begin to lay in the last days of February, and consequently on the
arbitrary "March i" basis will actually be credited with a small winter production.
This will tend to make the number of zero birds observed smaller than that expected
on theory.
in the latter paper, entitled "Measurement of the Winter Cycle in the
Egg Production of Domestic Fowl," a comparison is made betvs^een the
egg production of a pullet during the first 300 days of its life v^^ith its egg
production up to March i of the pullet year. He finds that there is very
little difference in the value of these two measures of production
The evidence for a winter cycle published by Pearl is all mass evi-
dence. The possibility that a flock might be heterogeneous in respect
to the winter cycle is not considered by him. It is true that he speaks of
particular birds forming exceptions to the rule in their laying, but the
sentence immediatel)- following seems to indicate that Pearl has in mind
birds that begin to lay late in the winter (7, p. 100) rather than birds
that lay throughout the winter. However, a few records that form real
exceptions to the rule can be found in the records of individual Barred
Plymouth Rocks published by Gowell (j, 4).
Pearl's mass evidence of the existence of a winter cycle is supported
by a study of the records described belov/. While a cursory examina-
tion of the data on egg production of Rhode Island Reds at this Station
indicates that a considerable percentage of our records are without vis-
ible indications of a winter cycle (since many birds that begin to lay in
November continue to lay without noticeable slackening of production
straight through the winter and spring), at the same time there are
many instances in which the existence of a winter cycle is indisputable.
Records of both sorts may also be obsers^ed in the reports from various
egg-laying contests. Further, an examination of Gowell's records of the
monthly egg production of individuals (j, 4) shows in most instances a
marked decrease in the production for February over January, indicat-
ing that the birds either stop laying entirely for a time during one or both
months or that they slow down in their daily rate. The former alterna-
tive, in view- of Pearl's statements, appears to represent the facts. A
few records published by Miss Curtiss in another connection also show
the same thing.
550 Journal of Agricultural Research voi. xii.no. 9
Since inspection of our records shows the possibility of the existence of
two types oif birds with respect to a winter cycle — namely (i) those
exhibiting such a cycle, and (2) those that lay but give no evidence of a
cycle — it becomes necessary to examine the matter in detail, and espe-
cially to endeavor to discover some criterion by which any individual may
be suitably classified. This has been attempted by a study of individual
records associated with a study of the length and seasonal distribution
of pauses in production as well as "a study of the monthly rate of
production,
MATERIAL AND METHODS
The material studied consisted of the daily egg records of three flocks
of Rhode Island Red pullets, hatched, respectively, in 1913, 1915,
and 1 91 6. While Pearl has never published individual records bearing
on this point, Gowell, who initiated the v/ork at the Maine Station, has
published the individual monthly egg records of the flocks making
their pullet records in 1899-1900, 1900-1901, 1901-2 (j, 4). These have
been used for comparison with our records. Further comparisons are
outside the scope of this paper.
It is obvious that a cycle in production should appear on the record in
one of two ways. Either thiere is a period of continuous egg production
followed by a period in which no eggs whatsoever are laid, or else the
period of production is followed by a period in which eggs are produced
at a less rapid rate than previously. It is well known to poultrymen
that hens often lay well for a while and then enter on a resting period
of variable length. The egg-producing period may be designated as a
"litter." This is, of course, one form of cycle, but is to be distinguished
in most cases from the winter cycle through the fact that the period of
egg production is in the latter relatively long and may be composed of
more than one litter. As a rule the only difficulty encountered is the
case where a single litter extends over the entire winter. In this in-
stance one can not tell whether one has a winter cycle, or a very long
litter, or whether the two coincide.
Since any cycle consists of a productive phase and a nonproductive
phase, we have put the original records into tabular form (Table
I), to which the reader's attention is directed in lieu of a detailed
account of individual histories and variations. In studjdng this table
we have come to pay particular attention to the pauses in production,
since these serve as our visible indexes of cycles,
EXPLANATION OP TABLE I
In Table I the details of the records of several families are shown. Practically
all the different sorts of pauses are illustrated. Pauses of less than three days in
length have not been included. The arrangement of the tables by families brings
out the strong resemblances betw^een the members of the various families. The
record of the mother is printed in italics at the head of the list of her daughters.
Mar. 4. iQis Winter Egg Production of Rhode Island Reds 551
Daughters that became broody before March i are given in bold-faced type. Among
other things listed in the column headed "Remarks," pauses occiuring after March
I and extending through April 30 are listed and are to be read as follows: The date
of the beginning of a pause is given, followed by its length, and in case more pauses
occur it is indicated by the number of eggs (inclosed in parentlieses) followed by
the length of the succeeding pause. Thus, No. 9277 in the family of male 3617 by
female 6003, a pause of 7 days length began April 7, when the bird laid one egg
followed by a 3-day pause.
Families sired by male 3617. — The progeny of one father and two mothers. None
of the members of tliese families shows a well-defined winter c)''cle. The February
egg production, in most cases where production began January i or before, is greater
than for January. The families are also noteworthy because of the number of birds
without any pause exceeding two days in length.
Family sired by male 5240. — In this family several individuals have a well-
defined winter cycle; others lack such a cycle.
Family sired by male 4786. — This family is characterized by the existence of a
. single short pause in each record.
Family sired by male 6781. — This family is characterized liy numerous pauses of
varying lengths.
F'amily sired by male 4723. — This is a heterogeneous family.
552
Journal of Agricultural Research
\-ol. xn. No. 9
iis
<i
5r,
S
s«.
><
C
n
8
•2
Q
B
fi
^
CO
PX)g
o^^-g
O 7-
00 OO M\00<^^00000 ^, ^o^
ss^2?::rS3
Q 00 TTf^^^wMCTOO
O to O M "-MO
00 O OOwcir^-Ot"'
O & T -^30 ^«0
< O "100 O O O O •-■ t^ <
o o o o "-"J"
2 *.2S
^ >-^it '-■5
•Si? «
I +++ I ++++ +
++++++7
-^ go -a
< ii a
a t; " o Ji >" 2
d ?; 41'— J3 I- S
53 3^
a u VI <u ;h
H Si
d aj -r
d-o-9
> > (J d o e H^
3 i 3
oooooooo
o o o o O M-<
.5 53 Si"
05
r^ t>. O 00 O '
O O 11 Si O 01 5 1) 5 5
u o
0:2:
o 9 o 9 fc oj «
t-t ■ TO
S. "^ !i >• ^ >>>,>, o >.o >. o 5,v Suo
O >. O O >^ O >>
O* <*"0 w ro f^ O* O •-•
CO O^O'^QsO^O^O'C*
o c
CO OQ
O -o n- -T -T -y-^
t'T o o o -r rrr-
00 Ot i7> 9t 0< QtQQ
Mar. 4. 1918 Winter Egg Production of Rhode Island Reds
553
o O r*
O. CO 00 HI O i-t '■n
I + 1 + r +-f
5 U 0) tj O «J u o «
)-«t.fO»o<*50»0 cnO
(J5o o
J J > >
U o O C
f-00 00
W 00 » 90
« Qi— »
Ot-.f»5CtOP< ^01
1 o 'T M * e>it-
°^ TTOOMOfO 00k
^ vO t^ CI 00 -O O
to M w Ti<eij
I I I II + + + + +
Tf «o r- w MO
"•"
■<
7
^
.1
1
a a
sj id
I CO ^ tata
t CO ^ to«o
c^ inno OlO
»^
»^
■o
»^ J
" I-' i^ C 0 >. >,
M P«
?-a.
SaaaT) a 3
0,al
X
■<<< ;^:::S
<»
^
2.S S^"2 r
S??{
r-00 00 00 a> »
>^
30 e»
554
Journal of Agricultural Research
Vol. XII, No. 9
^ W
>»
<»
S
tJ
la
>
^
a
a
K
tj
Ki
<j
r/l
a
-w
>
o
,J
-<A,
g
Oi
<
C^
U<
<»
)>.
S
?i
.2^
'T^
<
1
Cl" "^
o
o
^
(
"'Mi, to 4>
M
>^ "■ . ^-^
Jo
■ts
U
^o<a Om
d
o
1)
0
s
O
oH 2^— o
or
fA
1
fr- rt
o
a
O
O O
O^
O
is'^-s
r~ r, 1-
PO Oi
00
»H r* w
^ill
o
3
u t>, >- - «
T*; ?) t^
1/1 OG
r>.
00
t«
O O t-
lO Tf
00
vO
n>-
M cj
w
B .
■>»
cm lo
c< 00
t^
o
4) J3
0
Febru-
ary pro-
duction
greater
or less
than
Janu-
ary.
+
o ■«■
+ +
1
+
Aver-
age
numbe
of eggs
be-
tween
pauses
>^
^ "
^0
V,
"j-
Aver-
age
length
of
pauses.
v; ^
Q
Num-
ber
of eggs
before
the
first
pause.
M H
t^
p«
•* w
"
,.^_— — —
.
1 •« <U rl m
N >-■ M
M
: " " Si
r-MH^
JHHM
N M >ovo a
K • vO H 00
VO to
Nu
ber
eggs
twe
pau
tjj:
a a
«s
N
■^
"S*
"^
^2a
■ 2
o oc
fo
Xi
n 3
c>S
: u
BX
« >.
:§
ttj - Jo * ro OJ " M
o
^•^Q jtiH.:?^ ?^„
^«f-^fe
o 6 .
^•osi
O O 1 A
. ??l « r 1 1 i^i< VT
.^ifM/^S'P^M - ^1,
/-S-^/-
.vir
u-tt 3
>-> H <\) IM
M M ^
Ua
8 8-p-c
ISpHtJjOOQQ^tntH
•g^lSSti^a'Sifl
s ° ^ ^' s
si g d
0
fa^l^SQOO^CQ "
.2,:z;Qp£
QQ^^
^t^O^O •<t'tt-i O -'tT
1 to to M
S •" -
r^ w tH DO w M
«
a
i-t"o a
Q
.
,
_.
.
Total
length
of time
spent
in
pauses.
M ^ M
00 H
o
t-
1 "
00 N t
t^ vt
S o^
^j-
N -If lO
■* -O
■*
to
3S3
^•^a
i-«5 "
M M
^
3fe S)S
•n
to -a-
to
to
:5^!fs
fn
O PO
t-) 0
V)
o
O M
•-'
<^
05
Q
^ O
o
o
(U
0
N
H
m « Ov
^ :
o
rt-S
1-^
?« rt S
C o
o
Q-si
a -a
•0
0.
j:
■^
^3 S
< :
<
"0 .
^
0 ^ 0
>o o
o
s
P O
n^
u%
00
»
'
«
00
1
Mar.4. i9i8 Winter Egg Production of Rhode Island Reds 555
^ <
'O O ■* O « Ov O
-O OO^OOvO 10 t^ O^
00 f^ 11 o 00
00 O <^ 'O M 0\ o
+ I
1 +
+11 + ++
'"fc!^
i y a> (U I
j > > > ^ rt^ o
' o o o 9 9ti o'
H f^f^, \0 f^lt^f'lfi"
jJ i > o >
o rt o a< o
_0 C >;
•O Q ft
OC 00 00 00
T3 to
556 Journal of Agricultural Research voi. xii, no. 9
As inspection of these records shows that there are pronounced family
resemblances in the number and length of pauses among the members of
each of several families, and since environmental influences have been
excluded, as far as possible, it is evident that our material is not homoge-
neous from the biometrical standpoint. Consequently most of the usual
biometrical constants have not been calculated, since in this case such
constants tend to conceal the differences we are endeavoring to find.
MEANS BY WHICH A WINTER CYCLE MAY BE RECOGNIZED
A. — PAUSES
LENGTH AND SEASONAL DISTRIBUTION
Two questions arise in regard to the length of the pauses: First, is
there any significant difference in the length of pauses? Second, is there
any difference in the distribution of pauses of varying length through the
winter? Table II shows the percentage distribution of pauses beginning
at two days in length, grouped according to the month of origin. The
combined values for the three years are shown in the last column. It
is clear that the pauses fall into two classes, short and long, the dividing
line between the two groups falling at about 10 days. Further, it is
evident from the table that the long pauses originate mainly in December
and January, and to a less extent in February, but that the short pauses
are distributed in a fairly uniform manner throughout the winter.^ An
examination of the pauses occurring in March and April not due to
broodiness shows that while all pauses are less numerous than in the
winter, and while a few long pauses may be found, most are short ones
occurring in practically the same proportions observed in the other
months.
' The increase in long pauses originating in December and January automatically decreases the percent-
age of the short pauses so that relatively they are somewhat less numerous in spite of the percentages
being about equal.
Mar. 4.I9I8 Winter Egg Production of Rhode Island Reds
557
Table II. — Length of pauses distributed according to month in which they originate, for
IQIJ-14, IQIS-16, igi6-i7 «
1913-14
1915-16
Length of pause.
Novem- Decem-
ber, ber.
1
Janu-
ary.
Febru-
ary.
March.
Novem-
ber.
Decem-
ber.
Janu-
ary.
Febru-
ary.
March.
Days.
2
Per d. Per d.
Per d.
SI- 54
19-23
6. 15
3- 8s
Per d.
53-33
20.00
9.17
4.17
P.r d.
48.81
13.09
13.09
4. 76
Per d.
71- IS
9.62
3-8s
Per d.
57-01
14.91
3- SI
2-63
Per ct.
53-85
8.24
4-95
3-85
Per d.
53-81
15-71
8-57
1.90
Per d.
59-69
12.25
7-14
8. i6
7.84
17-65
1.96
2-5
80.77
4.63
3.08
I- 54
3-85
86.66
5-83
2.50
1.67
2.50
.83
79-76
10. 71
1. 19
3-57
1. 19
2. iH
84.62
1.92
78.07
S-26
70.88
6.59
3- 30
i-iO
2. 20
2. 20
1. 10
1.65
2. 20
2. 20
•55
2. 20
•55
•55
80.00
S-71
2.86
2.38
1.90
3-81
1-43
•95
87-25
6. 13
3- 06
3-92
11-15
16-20
1
1.92
1.92
3-85
.88
t-7S
.88
2.63
.88
.88
1-75
I- 75
.88
1-75
.88
1
31-35
6. IS
. . . .
12.50
1.92
41-45
• SI
46-50
1
' ■ ■ 'i
51-SS
• 1. .
■^\
s6-<5o
1.92
1.92
61-65
66-70
71-7^
.48
76-80
1
•51
81-85
..::;;::r. ;:::■: .:::•■■■
1-75
1
86-90
1 1
91-95
1 :.l :: ....:;:: ..:.:.'y.'.vv'
•55
96-100
. . . . 1 .
1
101-105
"1
1
■
.
i
1
1
126-150
I. 19
Length of pause.
Days.
1916-17
Novem-
ber.
Per d.
57-78
H. II
6.67
7^40
Decern- Janu-
ber. ary.
Per
48.
13^
Per d.
51. 16
15. 61
S^65
3^65
Febru- March,
ary.
Per d. I Per d.
49. 06 j 64. 16
14-98 I 12.44
7. 12 i 7- 83
3-37 I 4- ij_
Three years combined.
Novem- Decem-
ber. I ber.
Per d.
62.03
10. 26
5- 64
Per d.
52.90
13- 12
8.39
2. 80
Janu-
ary.
Per ,
52- <
Febru-
arj'.
Per d.
51-59
16. 25
8.04
3-02
March.
Per d.
59-76
12.47
8.45
5-84
2-S
6-10
11-15...
16-20. . .
21-25. •.
26-30...
31-35.--
36-40...
41-45 ■ . -
46-50...
51-55- -.
56-60.. .
61-65...
66-70. . .
71-75- -■
76-80...
81-85...
86-90...
91-95- ••
96-100. .
101-105.
121-125.
126-150.
82.96
6.67
2. 96
1.48
76.08
6.64
2.99
I. 00
2-33
2.66
3-32
1.66
.66
1-33
74-53
8-99
4- 12
2. 25
3-75
3-75
1-50
•75
•37
•33
1.48
88.48
6. 91
2.30
2.30
83.08
5^13
2. 56
1-54
1-54
2.05
1^03
77- 20
4-95
1- 51
•65
2- IS
1.08
1.72
.65
.86
1.08
1.72
I. 29
I. 72
1.08
78.89
7. 20
3^35
2.18
2.85
3-18
I- 17
■67
- 17
7.24
2.41
I. 81
.60
.80
a Broody pauses are not included.
Table III gives certain values of interest in connection with Table II.
In particular, the ratios of the number of pauses to the number of birds
558
Journal of Agricultural Research
Vol. XII, No. 9
laying, in spite of marked differences in the value of the ratios from year
to year, are in agreement (with one partial exception) in having lower
ratios at each end of the season than in the middle.
Table III. — Number of birds laying and number of pauses
Number of birds laying
Total number of
pauses
Ratio of pauses to
number of birds lay-
ing (expressed in
percentage)
Number without any
pause
Percentage without
any pause
Number with one or
more pauses
Percentage with one or
more pauses
34-7
65. 22
34-78
104
51
49.0
57
54.81
47
45- 19
1915-16
1916-17
161
130
81.0
83
SI- 55
78
48-45
176 aSj 60
120 S3
95
53-98
100 86. 6
35 34
42. 17 56. 67
81 48 26
46.02.57.8345.33
128
114
89.0
61
47.66
67
52-34
loi. 7; 100. c
88
41. 9042. II
104 121
58. 10^57. 89
145
13s
89. I 93- I
61
;o. 91 42. o
ic8
266
300
33-46
177
49-0957-9366.54
303
301
99-3
133
43-89
170
56. II
352
267
75-9
172
376
217
57- 7
213
56-65
163
o A part of this flock was not trap-nested after March i.
SIGNIFICANCE OK THE PAUSES
It appears from a consideration of the data presented in the preceding
paragraphs that pauses differ in their significance. It seems clear that
three classes at least can be distinguished : First, the long pause that is
clearly indicative of the presence of a winter cycle. The actual length
of this pause in exceptional instances need not exceed three or four days,
provided it comes at the proper season of the year, and is preceded by a
considerable period of egg production. Ordinarily, however, it exceeds
10 days in length. Second, short pauses which occur at frequent inter-
vals in the records of particular individuals because of their very
number must be regarded as having considerable significance (see Table
I, sire 6781). This type may be called "multipause" provisionally
and is to be distinguished from the type in which production is essen-
tially continuous, even though no sharp dividing line can be drawn
between these two groups. It is possible that certain records of the
multipause type represent a particular genotype, since there is a pro-
nounced tendency for many multipause records to occur in the same
families. This tendency has been particularly marked in a small flock
of Brown Leghorns. Another miultipause type results from intermittent
egg production during the winter pause.
A third class includes those individuals which exhibit only one or two
short pauses. Some individual records probably represent extreme
variants of the multipause type, others pauses at the end of the winter
cycle, but the majority are clearly without particular significance.
Mar. 4. i9i8 V/inttr Egg Production of Rhode Island Reds
559
NUMBER OP PAUSES PER INDIVIDUAL
In Table IV is shown the number of pauses above two days in length
per individual, for each of the three years. The data for each of the
three years are divided into three groups, determined by the time at
which egg production began. The chief points of interest are: (i) the
large percentage of records without a pause even when production begins
before January, (2) the progressive diminution in number of individuals
as the number of pauses rises, and (3) the agreement with expectation —
viz, that those birds that begin to lay early should have more pauses
than those beginning to lay later in the season.
Table IV. — Number of pauses per individual {disregarding I- and 2-day pauses), arranged
in three periods, according to the time egg production began (>■
Number of individual pauses in^
1913-14
1915-16
1916-17
Number cf pauses.
Where pro-
duction
began —
"a
1
0
Where pro-
duction
began —
"a
1
G
Where pro-
duction
began —
"a
1
IS
a g
4
la
bet:
3 9
be aj
.SH
2i2
II
a nj
■n 3
^1
.ss
a
0
u
V
59
34
19
4
I
3
29
9
s
0
I
12
\
0
I
0
90
S4
29
9
2
4
47-87
28. 72
15-43
4- 79
I. 06
2-13
33
61
40
20
5
3
■!2
16
II
3
21
16
I
0
0
0
86
93
52
23
6
4
0
I
32-45
35-09
19. 62
8.68
2. 26
I. 51
■■'.'38'
66
III
57
38
18
8
I
5
I
3
26
19
13
s
s
I
0
0
0
0
20
6
I
0
0
0
0
0
0
0
112
136
71
43
23
9
I
5
I
3
292
404
27.72
33-66
17- 57
10. 64
S-69
2. 23
•25
1.24
•2S
•74
72.28
6
0
0
8
9
98
188
52- 13
179
26s
67- 55
Total number of in-
o A few records in which production began very early in the fall liave been omitted.
In Table V the number of instances of different lengths of pauses is
shown.
Table V. — Number of instances
of pauses j
0 5 days
trt length
(5 ^0 70,
and
II u
P
1913-14
1915-16
I9I6-
17
^
^
u
Number of pauses.
>.
>.
a
3
i
c
3
>.
a
3
nl
"0
fl
0
a
•0
r1
0
«
T3
a
u
b
•a
-t-
0
•a
-h
0
•O
+
■^
0
1^
«
1
Jn
m
"
03
.^
pq
^
n
a
^^
-1-
—
+
^-'
+
^-^
-t-
-f-
^
-t-
<
a
<
0
<
<
n
<
0
■<
<,'
m
<
0
<
H
1
36
13
49
28
77
63
24
87
80
167
89
so
139
96
23 s
479
2
3
4
13
4
I
0
0
J4
4
5
3
0
0
17
4
5
19
6
2
5
I
0
24
7
2
18
2
0
42
9
2
39
IS
10
II
I
0
so
16
10
36
9
0
86
2S
10
I4S
38
17
5
6
0
I
0
I
4
0
4
0
4
S
0
7
8
I
0
I
0
I
I
Total
58
14
72
31
103
91
30
121
100
221
153
62
220
141
n6o
Journal of Agricultural Research
Vol. XII. No. 9
TOTAL LENGTH OF TIME SPENT IN PAUSES
In Table VI the percentage of the flock spending the specified number
of days in pauses is given. It should be noted that pauses of one or two
days in length have been disregarded in compiling this table. Thus, the
row marked zero does not mean that the birds laid continuously or that
they had no 2-day pauses, but that the total time spent in pauses, dis-
regarding those of one or two days, was nothing. The same is true
for all the other values for the column headed "Total time spent in
pauses." An examination of the individual records shows that the 2-
day pauses, which might be taken as the lower limit instead of 3 -day
pauses, are distributed proportionately.
Table VI. — Percentage distribution of the total time spent in pauses
[Records beginning after January i are excluded]
Total time spent in pauses.
o. ..
3-7-
13-17...
18-22...
23-27...
28-32...
33-37- ••
38-42...
43-47- • •
48-52...
53-57- ••
58-62 . . .
63-67...
68-72...
73-77- •-
78-82...
83-87..-
88-92...
93-97- ••
98-102. .
103-107.
108-112.
113-117.
118-122.
123-127.
128-132.
133-137-
138-142.
143-147-
Days.
Percentage of flock pausing.
March
hatched.
60.87
13. 04
8. 70
4-35
2. 17
6.52
2. 17
2. 17
April and
May-
hatched.
53-331
16. 67
10. 00
I. 67
6.67
I. 67
3-33
1. 67
I. 67
3-32
1915-16
March
hatched.
16. 22
10. 81
4-05
4-05
5-41
1-35
». II
^■35
8. II
6.76
2. 70
5-41
5-41
1-35
2. 70
1-35
5-41
1-35
I- 35
1-35
1-35
1-35
1-35
1-35
April and
May
hatched.
30.88
II. 76
5-88
8.82
5.88
5-88
5.88
7-35
2-94
2.94
1.47
4.41
2.94
1-47
1.47
1916— 17
March
hatched.
'5-38
7.69
2. 56
2. 56
2. 56
2. 56
2. 56
2.56
7.69
2.56
2. 56
7.69
5-13
2. 56
10. 26
2.56
2.56
7.69
2.56
2. 56
2. 56
■2.56
April and
May
hatched.
34.08
17.49
93
93
69
93
59
93
14
14
24
79
14
14
69
45
90
90
45
45
Mar. 4, i9i8 WiuieY Eqq Production of Rhode Island Reds 561
Several features of this table require comment. First, the large per-
centage of the 1 91 3- 1 4 flock that falls in the zero class, and the low
value of the pause of maximum length. It should be noted that this
flock was not put into the laying houses until late October, and none of
them began to lay until after November i. Second, the high percentage
of the flocks of the April and May hatches of 191 5-1 6 and 191 6-1 7 that
spent practically no time at all in pauses. Thus, the sums of the first two
rows are 42.64 and 51 .56 percent, respectively. This high value is the more
remarkable for the season of 1916-17, since the birds began to lay early
in the fall. Contrasted with this is the third point — viz, the relatively
low percentage of the March-hatched pullets of these two years that fall
into these rows. This may perhaps be explained by the tendency of
March-hatched pullets to begin to lay early in the fall and to undergo
a winter molt.^ That the 191 3-1 4 flocks shows a radically different dis-
tribution is probably due to its delay in beginning production. This
raises the question as to whether or not the time a bird begins to lay
may not have a considerable effect on the appearance or nonappearance
of the winter pause. The individual records, however, prove that no
necessary relation of the sort postulated exists.
B. RATE OF PRODUCTION
VALUE OF MONTHLY EGG PRODUCTION AS AN INDEX OP A WINTER CYCLE
A cycle in egg production may be indicated by a lessened daily rate
of production as well as by a pause. Specifically one would expect of
the winter cycle, if it were delineated only by a change in rate, either
that production should begin at a relatively high rate to be followed
during the latter part of the winter by a period of lessened production
or that the production should start at a slow rate, rise to a maximum, and
then decline. Following Pearl, we should anticipate that February
would have less eggs than January.
In handling the data given in Tables VII to XI we have proceeded as
follows, unless otherwise specified. Each lot of Rhode Island Reds is a
group of individuals selected for the following reasons: (i) A laying
period of considerable length in order that ample time should be allowed
for the completion of the entire cycle — that is, both egg production and
pause. Thus, birds beginning to lay after December i are excluded from
the tables. (2) Records containing broody pauses are excluded. (3)
Each month should be equal in length. We utilize, therefore, three
periods of 31 days each — viz, the months of December, January, and
February, including in the latter the first three days of March. (4)
The data, furthermore, are divided into two groups — viz, March-hatched
' Possibly the large percentage of JIarch-hatched pullets with pauses may be interpreted as being due
to a greater opportunity for the appearance of the winter cycle, since they begin to lay early in the season.
Pearl, however, does not mention an association of a molt with the winter pause.
562
Journal of Agricultural Research
Vol. XII, No. 9
pullets and those hatched in April and May, the hatching date of the
latter being the same as for the Maine birds.
The data for Barred Plymouth Rocks and Wyandottes have been
extracted from the Maine Station Bulletins 79 and 93 (j, 4) and have
been handled in essentially the same way. As these records have been
given for months only, only those birds that laid at least one egg in
November have been included in the tables. Dr. Pearl has stated to
the writer that his Barred Plymouth Rocks do not go broody during
the winter, so that we have assumed that broodiness does not enter as a
disturbing factor. The February records have been corrected to a
basis of 31 days.
Table VII. — Mean monthly egg production of three breeds of pullets beginning to lay
before December I
[Broody records are excluded. No record is included where production began later than December i.
No Barred Plymouth Rocks or White Wyandottes that failed to lay in November are included. Febru-
ary records are reduced to a 31-day basis]
BARRED PLYMOUTH ROCKS
Flock of—
Platched.
Num-
ber of
indi-
vidu-
als.
Decem-
ber.
January.
Feb-
ruary.
38
25
17-3
16. 2
17.2
19.4
15-9
17.2
13-4
17. 6
II. 2
Total
96
16. 9
X7.6
14-3
WHITE WYANDOTTES
1899-I9OO .
I90O-I9OI .
I9OI-I902.
Total.
34
S8
103
12. 7
15-8
15.0
17.4
16. 9
13. 2
16. 7
14.7
13-4
12.7
I3.8
RHODE ISLAND REDS
1913-14-
I915-16,
I916-17 ,
Total.
fMarch
1 April and May.
fMarch
\ April and May.
|March
\ April and May.
10
21. I
21.5
13
21. 2
21. 9
5&
17.0
14. 0
24
19.9
14.9
32
II. 6
7-9
122
19-3
15.0
257
18. I
14-5
20. 6
21. 6
12. 9
12.8
12. 9
16.3
15-2
From Table VII it is clear that the egg production of February is usually
lower than that for January. Some exceptions are to be noted, especially
in the Rhode Island Reds, and also for 1 900-1 901 in the Barred Plymouth
Mar. 4. i9i8 Winter Egg Production of Rhode Island Reds
563
Rocks. The Rhode Island Reds, compared with the Barred Plymouth
Rocks and the White Wyandottes, show a less difference between the
two months. In this connection it may be noted that the mean value
of the February egg production for eight years, as given for the Barred
Plymouth Rocks by Pearl (7) if reduced to a basis of 31 days is 12.03,
approximately, or slightly higher than for January, which is 11. 71, The
mean egg production for February (28 days) for the entire flock of our
Rhode Island Reds has always been higher than for January.
Table VIII is similar to Table VII, but we have utilized the larger
numbers made available by including in the January and February
means the records made by birds beginning to lay as late as January i.
The essential result is the same as shown in Table VII.
Table VIII.
-Mean monihly egg prodticiion of pullets from the same flocks as Table VII ,
but including those beginning to lay as late as January i
1899-1900.
1900-1901.
1901-1902 .
Total .
BARRED PLYMOUTH ROCKS
Hatched.
Number of in-
dividuals
laying in —
Decem-
ber.
January.
Febru-
ary.
Flock of—
Decem-
ber.
Janu-
ary and
Febru-
ary.
38
25
55
56
45
17-3
16. 2
17.2
18.7
15-5
15-4
II. 8
16. 5
12.4
Total
96
156
16. 9
16.6
13.7
WHITE WYANDOTTES
34
50
18.8
18.3
58
77
12.7
16. 9
II
31
15-8
16.5
103
158
15.0
17.2
14. o
14. 6
12.7
14. o
RHODE ISLAND REDS
I913-14
I915-16
I916-17
Total,
/March 10
lApril and May j 13
fMarch ' 56
\ April and May 24
/March 1 32
\April and May 122
256
46
62
90
70
44
230
542
21. I
21. 2
19.9
17. I
II. 6
19-3
17.7
20. 5
20.6
14-3
16.8
8.3
16. I
14.8
19.7
19.9
13-5
15-3
II. I
17. o
16. o
For the Barred Plymouth Rocks and White Wyandottes the December
production varies in the individual years, being sometimes greater,
38324°— 18 4
j64
Journal of Agricultural Research
Vol. XII. No. 9
sometimes less than for January. For the Rhode Island Reds the
December production is considerably greater than for January and also
for February.^ -
Table IX. — Mean monthly egg production for all Rhode Island Reds that began to lay
by December i, separated into two groups, according to length of pauses
[Any record having a single pause of more than jo days in length is included in the group of pauses of
II days or more. A record having several short pauses of less than 1 1 days each, even though aggregating
more than lo days, is included in the groups of pauses of less than ii days]
Flock
cf-
Grouping of records.
Hatched.
Num-
ber of
indi-
vidu-
als.
De-
cem-
ber.
Janu-
ary.
Febru-
ary (31
days).
I9I3-I4
All birds
23
9
II
21. I
21. 0
22. 6
21. 7
22. 0
23.1
20. 9
22. 2
22.3
Records with pauses
not exceeding 10
days in length.
Records with pauses
of II days or more.
All birds
fMarch
\ April— May
Hatches combined
("March
[April-May
20
21. 9
22. 6
22.3
I
2
22. 0
13.0
17.0
15-5
I. I
16.5
Hatches combined
3
16. 0
16.0
II. 4
I9I5-I6
80
20
II
17.9
19.4
19-5
14-3
19.9
19. 2
12. 9
20. 0
Records with pauses
not exceeding 11
days in length.
Records with pauses
of 1 1 days or more .
All birds
/March
[^ April— May
18.4
Hatches combined
JMarch
31
19-5
19. 6
19.4
36
13
15-7
20.3
10.8
II. 2
9.0
8.0
\April— May
Hatches combined
49
16.9
10. 9
8.7
1916-17
154
7
69
17.8
21.7
22. 4
13.6
19.7
21. I
15- 5
20. 7
20. 9
Records with pauses
not exceeding 11
days in length.
Records with pauses
of 1 1 days or more.
fMarch
\ April— May
Hatches combined
/March
\April-May
76
21.7
19.7
20. 7
25
53
8.8
15-6
4.6
7.2
8.8
10.7
Hatches combined
78
13-4
6.1
10. I
If the data for Rhode Island Reds are analyzed further, as is done in
Table IX, certain interesting facts come to light. In this table the
records have been divided into two groups, one consisting of those records
' A comparison was attempted between November and December egg production, but no satisfactory
results could be obtained, because of the small number of birds which had complete records for November.
Mar. 4, i9i8 IVinier Egg Production of Rhode Island Reds 565
that have no pauses exceeding 10 days in length, and the second of those
that have pauses of more than 10 days. In the former group the egg
production remains at essentially the same point in all three months.
In the second group, however, the January and February production is
markedly lower than that of December. The production for January
may or may not exceed that of February.^
DIFFERENCES IN MONTHLY EGG PRODUCTION DURING DECEMBER, JANUARY, AND FEB-
RUARY FOR RHODE ISLAND REDS, BARRED PLYMOUTH ROCKS, AND WHITE WYAN-
DOTTES
In Table X we have examined the differences in the monthly egg
production for the same set of birds shown in Tables VII and VIII. For
the Barred Plymouth Rocks and White Wyandottes it is noticeable that
for each year but one the number of birds laying more eggs in January
than February is greater than the number laying more eggs in February
than January (in the exception the number is equal) and that in five out
of six instances the average number of eggs laid by the birds having an
excess January production is greater than for those having an excess
production in February. The percentage of birds having an excess
January production is 75.29 for Barred Plymouth Rocks and 68.23 ^or
the White Wyandottes.
The data for the Rhode Island Reds are quite different in character.
Only 51 .39 per cent of the March-hatched pullets lay more eggs in January
than February, while for April and May hatched pullets the value is 47.16
per cent, although in two years out of three the number of birds with a Janu-
ary excess is greater than for February. In only one instance, and that
with a small number of birds involved, is the number much more than
twice the number having an excess February production. This is to be
compared with the Maine records, where in every instance the number of
birds with a January excess is greater than those with a February excess
and, except in two instances, are two to three times as numerous (in
one case 1 6 times). For the Rhode Island Reds hatched in April and May
the average number of excess eggs in two years out of three is greater for
those pullets with an excess February production. For the March-
hatched pullets conditions are reversed.
' It is possible that the two groups, high producers and mediocre producers, should be separated in an
analysis of this sort. So few of the latter were available, however, that it seemed unwise to attempt the
separation.
566
Jouryial of Agricultural Research
Vol. XII, No. 9
^
a S
•2^
*§
>^
1 -
« ■ A
to
9.
w
0
K
10
M
sR
^
otc a ..•
a
vO •«
0> vn
v 2
5<i 3 .J,
0 to
«
M 10
f^
t^ 10
c2
■0 •
a
ca
>->
'^
^
rt Q.J, t ,,;
to H
•*
0 0
0
w xS
to r^ t^
t^ 0 ^
3
„ ^
00 1 M
§
A
OJ ^ 3I
to «
(«
t>.\0
06 -"i-
vn
t^
M Tt
>o 00
S^-
,° S^
00 M
« 0
rti *^
&I
^
t^cn
in
.n r.
3
d
t
■<t
^^ii
to >o
00
10 0
>o
t^ 0
„
0
>—•
3 , •
c3« 3 ^
0 0
0 -c
-c
S>8
s
0 ^
0
5
w 2
1-2
05
to ■*
• to
VOvO
lO
00 tf
"'
u>
2« s >.
O — .Si 0
o.
Cl
0 <^
00 1^
"
"A
go.i - ^
■" "^
CO
00 ■*
-
0 t^
■*
t!.T
3
"" X «
10
0 lO
1 "
^o -o
00
0
«
to 0^
■«• to
.0
0 0
0
0 in
-1-
fe .
1 1 7i
0
't to
t^
1- J3
1'^
j2-e
to w
10
M •*
"-
tH ^O
00
M 0
-
M M
o\
0
V
oJ,iS
to «
VO
C H
to
0 -0
>o
00
«
^^•ss
3 , <
f.
to to
0 0
0 to
"
vO t<-
0) ^
g-0-3"
to to
to
tovO
t
lo f
to to
8 a
2(3 3 ^
0 »
""
-0 H
" "
lii^i
.0 u-
0 to t^
05
00
^ i- S
OJ
2§
S " ^
" '^
"
*o t~
t^
v6 ■■6
vO VO
,0 0
0 0
»o^o
„
°"2
r^
0 0
^»->
{-, X
M
0"
■ 1
0 M
„
%o td
0 II 0 .. 1
1 ::?-^
1 -
O^iS
n
C* H
to
1 °
II 0
^^^s
1
■a
J3
i
3
J2
f^
"3
«
•3 a ^
g
c
fii
H
c
(-1 f-i|l<
s
1'^
.
a c
-0
r.
.^ ,
■^
"
8i
M
»
»
so
'
c
c
1
t^
u- 0 ^,
to o> tr
"• :
t^ t^ w
.0 :
"
" :
« tt m
w 0
^
"
0\ 0 t^
« moo
■0
(-1 0 CT.
r..
N m t-
■«• 0
to M M
r-.
M (M w
•r
t-i
■^
vo r~ f
-0
0 ti 00
0
^0 0 .-
0 to I'
00 T
M HI
Cl »..
•r I
to 1
m t-cc
■o- 0 0
00 >n Q
'"
to « !/■
o\
"
0
M •^- Cl
t>.
to r, c.
to 0 0
tr
00 t-
0
t^ W M
0.
r..
^
" -!t«
m 0
troo
m
00 0 00
o\
m t-^ «/
. -n
0 0 I
tH m t.
1-1 l-l w
•^ ^
^
ti d n 0
>
^>
>
a ca 0
3 1
sss
5^
•c-o-c
a S S
HOh
^H .-^ HH
"C C w-
a ac
<;<<
0 H W
0 0 c
'oh I
" H
1
Mar. 4. 1918 Winter Egg Production of Rhode Island Reds
567
rJ-OO »^
1^ '^'O
CO ;
N 00 r^
r^ .
r^O -"T
1^0
000
0\ rt ^
»^0 OC
•6 '■
-
&§"'
-C' •
0
^rt 0 t-
"
0
0 0 0>
^
t^O f^
-0 0
1. 00 0
Ci H 0
06 0 M
0
C>0 vo
00 fO >n
0
CO
f1
ro <S l/^
0 ^
(M Tf 0
0
•3- 0 0
\0 i^ 0
1-
H"^
T
"-"
so 0
rr 0
vO 10 0
0 00 0
r- i>^ r^
0
M CO T
r^
>0 t<5
vO 0
10 T M
0
>
«
r
c
<
n
(a
-d
a
a
J, G
1
3
\
^
568
Journal of Agricultural Research
Vol. XII, No. 9
A consideration of the number of individuals laying an excess number
of eggs for December over January (and also February) shows that
more of the Rhode Island Reds have a December excess than a January
(or February) excess. For the Barred Plymouth Rocks and White
Wyandottes more individuals have a January production in excess of
that of December, but a February deficiency. The difference in the
case of the Barred Plymouth Rocks, however, is not great. Neverthe-
less it is clear that the egg production of these two varieties does not
undergo a period of depression at the same season as that of the Rhode
Island Reds.
TablB XI. — Distribution of individuals having a greater egg production in January
over February, or vice versa, arranged according to the maximiim length of any single
pause
January production greater than
February.
February production greater than
January.
1913-14
1915-16 1916-17
1913-14
1915-16 I916-I7
Length of pauses, .days. .
0
3-10
II-
up.
0
3-10
II-
up.
0
IS
6
7
3
2
I
3-10
6
6
7
3
I
2
II-
up.
5
3
I
2
2
3
3
0
7
S
S
S
....
3-10
3
I
2
3
3
II-
up.
2
I
2
I
I
0
5
2
6
3
I
3-10
: 2
3
3
2
II-
up.
0
II
II
8
I
I
3-10
10
7
9
3
3
2
II-
up.
EXCESS NUMBER OF EGGS .
II
8
4
I
4
I
3
2
4
3
3
2
I
2
I
I
3
3
I
3
6
2
2
3
3
3
3
4
3
2
I
3
4
3
5
I
8
I
I
9
6
I
I
8
I
I
3
I
'°
I
3
5
3
I
2
5
3
3
I
I
40
30
Total
24
II
9
9
17
44
34
26
22
12
10
17
12
32
36
6l
Sum of first two columns
35
S
26
0
60
6
34
29
68
Number with the same
production in each
IS
0
6
4
18
9
SUMMARY
Production equal.
February produc-
tion greater than
January.
January production
greater than Feb-
ruary.
Length of pause days. .
0
3-10
ii-up.
0
3-10
ii-up.
0
3-10
II-Up.
39
II
so
13
71
60
131
103
67
S4
121
93
Sum of first two columns in each sec-
Sum of all three colimms in each sec-
63
334
214
^
Mar. 4. 1918 Winter Egg Production of Rhode Island Reds 569
DrFFERENCE IN JANUARY AND FEBRUARY EGG PRODUCTION, TAKING THE LENGTH OP
PAUSE INTO CONSIDERATION
The results of an examination of this point are shown in Table XI,
which gives no evidence that a difference in the egg production of an
individual for the two months is more likely to occur in one month than
the other, since the number of individuals falling in the different cate-
gories is approximately equal, although fluctuating somewhat from year
to year.
These data, taken in connection with those previously discussed
regarding differences in monthly production and with the observations
on rate of production, show that a lower rate of production for Feb-
ruary over January in the individual case is without significance, unless
associated with a definite pause. This information is of particular value
for those instances in which the production for February is only a few
eggs less than for January and which, from Pearl's data, might be con-
sidered to exhibit a winter cycle.
Table XII. — Mean production of two groups of pullets hatched in April, igi6,for the
periods after the first egg designated in the headings
BIRDS BEGINNING TO LAY OCT. S-DEC. 13, INCLUSIVE (80 TO 72 INDIVIDUALS^)
Days of production . . .
33-63
63-93
Mean
March
produc-
tion.
Mean number of eggs
per period
Mean date of begin-
ning and end of
each period
16. 96
Nov. 9
to
Dec. 10.
Dec. II
to
Jan. 10.
14.56
Jan. II
to
Feb. 10.
Nov. 9
to
Nov. 19.
17-63
Nov. 3o
to
Dec. 3o.
15-87
Dec. 31
to
Jan. 30.
Jan. 31
to
Feb. 30.
BIRDS BEGINNING TO LAY DEC. 14-JAN. 25, INCLUSIVE (69 TO 55 INDIVIDUALS «)
Mean number of eggs
per period
Mean date of begin-
ning and end of
each period
Jan. 3
to
Feb. I.
Feb. 3
to
Mar. 3.
Mar. 4
to
Apr. 4-
Jan. 2
to
Jan. 12.
Jan. 13
to
Feb. 12.
19-38
Feb. 13
to
Mar. 16.
Mar. 17
to
Apr. 16.
o The variation number of individuals is due to broodiness.
EGG PRODUCTION IN PERIODS OF DEFINITE LENGTH, BEGINNING WITH FIRST EGG
In the preceding paragraphs we have used records that begin before
a definite date — viz, December i. On being examined from the stand-
point of the first egg it will be observed that the production for any one
month, say December, is made up of the production of birds that have
been laying for varying intervals of time, including those well along in
production and those just beginning. In Table XII the rate of pro-
duction has been examined from another standpoint. Three periods of
57© Journal of Agricultural Research voi. xii.no.9
31 days each, beginning with the date of the first egg, and three
periods of 31 days each beginning with the eleventh day of production
have been employed in studying the production of two groups of April-
hatched pullets. One group has a mean date of first egg on November 9;
the other, on January 2. The reason for employing two sets of three
periods of 31 days, each differing by 10 days, for each lot, lies in the fact
that egg production sometimes is extremely slow and erratic at the start
and that this may reduce the egg production disproportionately for the
first 3 1 -day period.
The following points are shown by Table XII : First, the egg produc-
tion of the first group is somewhat inferior to that of the second. Note
particularly the March production of the first group compared with that
period of the second that extends from March 4 to April 4. Second,
while there is a fall from the first 31 -day period through each of the two
successive periods in the first group until March, the second group shows
a constant rise from period to period, which may mean that the time of
year in v/hich the various periods fall is concerned with the drop in pro-
duction, for it will be noted that the first period of the, second group
nearly coincides with the third period of the first group.
CONCLUSIONS REGARDING THE CRITERIA FOR THE WINTER CYCLE
IN THE INDIVIDUAL
A consideration of the data presented in the preceding pages leads
to the following conclusions regarding criteria by which the winter cycle
can be recognized in the individual record.
First. The rate of production, as shown by the monthly egg records,
fails to furnish a satisfactory index of the existence of a winter cycle in
the individual Rhode Island Red pullet.
Second. The best criterion of the existence of a winter cycle in the
individual is the existence of a pause in production beginning in Decem-
ber, January, February, or, rarely, March, following a period of con-
tinuous egg production, and usually exceeding 10 days in length. A
single pause in some instances may be replaced by series of short pauses
separated by only one or two eggs.^
Third. In some instances a short pause — that is, 10 days or less in
length — occurring in February or March and following a period of several
weeks of continuous egg production may delimit the winter cycle.
It seems clear that the period of low flock production for the Rhode
Island Reds, for birds beginning to lay sufficiently early in the season
may come earlier in the winter than at the Maine Station. In some
' The second part of the two recent bulletins from the Utah Station— viz, Ball, Alder, and Egbert (i),
and Ball and Alder (2)— was received after the manuscript of this paper had been completed. Only a very
brief comment can be made on their discussion of the "'-winter' egg-laying period" in White Leghorns.
They conclude "that there is no apparent biological ground for either the beginning or end of this period
..." This conclusion, which rests on mass statistics, needs reexamination before it can be considered of
universal applicability to all White Leghorns.
Mar. 4. 1918 Winter Egg Production of Rhode Island Reds 571
individuals the pause comes comparatively early in the winter so that the
following [spring ( ?) cf . Pearl and Surface (7)] cycle of production may
begin as early as the middle of January. This may mean, perhaps, that
the winter pause in Rhode Island Reds is not homologous with that of the
Maine Station birds. Whether or not this is so, it would seem desirable
to look at it from a somewhat dififerent standpoint. It may be that this
pause follows an initial cycle of production. It may be, too, that we are
not dealing wholly with an inherent pause but with a pause that depends
in part on the environment for its manifestation, which is due to a
difference in resistance on the part of individuals to the weather con-
ditions at this season of the year.
NUMBER OF EGGS LAID BEFORE THE WINTER PAUSE AND LENGTH
OF WINTER PAUSE
In those instances where the winter pauses could be determined with
some degree of accuracy, we have determined the range and mean num-
ber of eggs laid before the pause and the same constants for the pause
itself, for the pullets of the April and May hatches shown in Table XIII.
It should be borne in mind, however, that in most instances the limit
specified earlier has been used for the lowest number of days indicative
of the winter pause.
For the number of eggs before the pause, the range has a value of
2 to 96,^ with a mean of 35.98 eggs. The length of the pause has a
range of 8 to 72 days, with a mean of 34.23 days. For the birds listed
under section B of Table XIII, the values are for eggs; range i to 96,*
mean, 36.98, and for length of pause range, 3 to 104, mean 24.87.
The possibility of a correlation between the number of eggs laid before
the pause and the length of the pause has been examined and found
to be practically nonexistent.
MODE OF INHERITANCE OF THE WINTER CYCLE
From the data that have been presented in the preceding pages it is
clear that some individual Rhode Island Reds exhibit a definite winter
cycle, while others as definitely show no winter cycle. For purposes of
description we may describe the former as "winter cycle," the latter
as "no winter cycle." It is clear, moreover, from an examination of
the family records (Tables I and XIII) that the character is inherited
and that segregation takes place. When the ratios are examined, how-
ever, no evidence of an entirely satisfactory character is afforded us as
to the mode of inheritance of the winter cycle, although there is some
evidence that the winter cycle is inherited according to the simple
1 The rather absurd values for the lower end of the range result from the inclusion of a few records that
are obviously out of the ordinary but which can not be excluded. The next lower value is 9.
572
Journal of Agricultural Research
Vol. XII, No. 9
Mendelian scheme, "no winter cycle" being the dominant allelomorph,
"winter cycle" the recessive.
Before considering the evidence for this scheme there are certain
pecuUarities of the character that should be listed.
First. The character does not come to visible expression in the male;
hence, his gametic composition can be made out only through a con-
sideration of his maternal ancestors, his daughters, and his sisters.
Second. The character is a physiological one, subject to possible in-
fluences by the environment, and perhaps to other internal factors con-
cerned with &gg production.
Third. Pauses distinct from those indicative of the winter cycle but
likely to be mistaken for them may occur.
Fourth. Difficulties are encountered in classifying certain individual
records, such as records with several very short pauses, a single very
short pause, a simple slackening in rate of production without any defi-
nite pause, and pauses occurring at the limits of the season. In such
instances we have proceeded somewhat arbitrarily. There have been ex-
cluded, first, all birds that have laid less than 20 eggs after the first of
January; second, all birds showing broody pauses; third, March-hatched
birds that begin to lay early in the season, and which molt.
TablB XI 1 1 . — Progeny of individual pairs of birds of flock of igi6-iy, showing the number
with a winter cycle (P), those without (N), and those on which a definite determination
could not be m,ade {X)
Presence
or absence
Method A.
Method B.
(X4
u
0
Presence
or absence
of winter
cycle in
mother's
record as
deter-
mined by
two
methods.
Method A.
Method B.
u
lU
0
of winter
cycle in
mother's
record as
deter-
mined by
two
methods.
P
N
X
P
N
X
"3
0
P
N
X
"a
P
N
X
"3
1
4723
SS8i
507 7
6265
4354
43"
4473
5476
5463
4178
4845
4138
4553
4533
5S95
2564
5771
4529
2565
5703
6840
6767
5719
5159
5982
4518
5412
4S28
5090
4030
4254
49
NX
NP
NP
NN
PP
PP
XP
XP
XP
XP
XP
XX
XX
XP
NN
NP
NP
XP
NN
PP
XP
XP
XP
NN
XP
NN
6
I
2
3
3
4
0
0
2
3
3
0
I
1
0
2
I
I
0
I
I
2
0
0
0
I
5
9
3
3
8
6
I
5
3
3
4
0
0
2
0
3
4
4
4
3
7
6
8
0
0
0
12
0
2
4
I
S
6
0
2
5
2
6
2
I
4
0
7
I
3
4
I
2
2
»
I
7
23
10
7
10
12
IS
7
S
8
6
3
4
4
5
12
6
7
8
9
10
10
2
I
8
12
2
3
7
5
5
0
0
2
6
4
3
I
I
0
3
4
I
3
3
3
5
3
0
0
2
4
8
3
1
6
6
I
5
3
2
3
0
0
2
I
2
4
4
2
3
6
4
6
0
0
0
7
0
I
2
I
t
0
2
3
I
i
I
3
0
4
I
2
2
°
I
I
2
I
6
23
10
7
10
12
IS
7
S
7
II
8
6
\
12
6
7
8
9
10
10
2
I
8
6781
6373
4378
4786
3617
S477
5240
5776
4882
4509
3719
3482
3600
174
284
4043
4754
4012
4592
6684
5881
4409
4716
5230
6751
6003
4844
6366
6404
6232
5098
2332
PP
NX
NN
XX
XP
XX
NN
PP
NN
PP
XX
XX
PP
NP
NP
NN
NN
XX
NN
XX
PP
NN
XP
Total...
I
I
3
2
2
I
2
0
4
I
0
0
0
0
0
I
0
2
I
3
0
0
0
61
8
2
7
2
0
I
1
0
2
0
0
6
4
8
2
5
9
2
3
4
5
0
13
175
2
3
4
7
2
3
4
8
5
3
2
2
I
2
I
I
I
7
0
0
3
2
13
158
II
6
14
II
4
5
7
8
II
4
2
8
5
10
3
7
10
II
4
7
8
.1
394
s
2
6
5
2
I
3
0
4
I
0
I
3
I
I
I
3
2
I
4
2
2
4
132
4
I
5
2
0
I
0
0
2
0
0
5
I
7
2
5
6
2
3
3
3
0
12
140
2
3
3
4
2
3
4
8
5
3
2
2
I
2
0
I
I
i
3
0
10
122
II
6
14
II
4
5
7
8
II
4
2
8
S
10
3
7
10
II
4
7
8
2
26
394
Mar. 4. 191S Winter Egg Production of Rhode Island Reds
573
In Table XIII is shown the classification of the flocks of 191 6-17,
arranged by families. Each individual is classified twice. According to
one classification (method A) our endeavor has been to ascertain as
closely as our best judgment would permit, the true status of each in-
dividual. In method B, however, we have given the benefit of any di .ubt,
to the winter cycle and have listed every bird in the positive column
that could possibly be considered as having a winter cycle.
Section A gives the ratio of 61 individuals with a winter cycle to 174
without. This is close to the i to 3 ratio — viz, expected 58%" to 176X —
for a simple Mendelian case of inheritance. According to method B,
however, the ratio approaches closely to equality.
Table XIV gives the ratios for the flocks as a whole for 1913-14 and
191 5-1 6, in addition to 1 916-17, and the grand total. The years vary
somewhat, but the total, 147 to 388, is perhaps merely a deviation from
the expected 1 33^ to 401 X> the deviation being in the direction expected,
on the assumption that birds without the genes for a winter cycle may
exhibit a false cycle.
Table XIV. — Number of individuals classified according to the presence or absence of the
winter cycle for the years igij-14, igi^-i6, and igid-iy
. Method A.
Method B.
Flock of—
Birds
with a
winter
cycle.
Birds
with-
out a
winter
cycle.
Unde-
termin-
able.
Total.
Birds
with a
winter
cycle.
Birds
with-
out a
winter
cycle.
Unde-
termin-
able.
Total.
IQI'?— 14.
25
61
61
114
100
85
139
158
224
300
394
53
104
132
99
79
140
72
117
122
224
300
394
ioic;-i6
1016—17
Total
147 ^
388
401X
289
318
(?)
Expected.
For method B the observed ratio for the three seasons is 289 to 318.
Mass figures of the sort just given are merely suggestive, since the
ratio I to 3 holds only under certain conditions. However, an examina-
tion of the proportions in which the two types occur among the progeny
of a single female, and in some instances of the progeny of one male by
several females, favors the suggestion given above. At the same time,
the gametic constitution of the parents can not be made out with a
satisfactory degree of accuracy. To be sure, one can assign a gametic
constitution to many individuals, but it is impossible to check these by
reference to preceding years, mainly because the number of progeny
from a single pair in the earlier years was too small to afford critical
evidence. Since prospective matings are likely to furnish critical
evidence on the point in question, it seems advisable to defer any attempt
at a solution of this phase of the problem for the present.
574 Journal of Agricultural Research voi. xii, N0.9
SUMMARY
(i) An examination of the data published by Gowell confirms the
statements of Pearl and Surface (7) and Pearl (3,6) regarding the presence
of a winter cycle in Barred Plymouth Rocks.
(2) The winter cycle is much more characteristic of the Maine flocks
as a whole than it is of our Rhode Island Reds, where it can be demon-
strated in only a portion of the flock.
(3) The period of decreased flock egg production for Barred Plymouth
Rocks and White Wyandottes comes in February. For Rhode Island
Reds it may come in January as well as in February.
(4) A pause, or series of pauses, usually exceeding 10 days in length
and following a considerable period of regular egg production, is the best
index of the existence of a winter cycle in the individual Rhode Island
Red.
(5) The rate of production does not furnish a satisfactory index of the
presence or absence of a winter cycle.
(6) Evidence is presented which indicates that the winter cycle may
be inherited in some definite but undetermined manner.
LITERATURE CITED
(i) Ball, E. D., Alder, Byron, and Egbert, A. D.
1916. BREEDING FOR EGG PRODUCTION. PART I. A STUDY OP ANNUAL AND
TOTAL PRODUCTION. Utah Agr. Exp. Sta. Bui. 148, 60 p., 10 fig.
Bibliography, p. 59-60.
(2)
1917. BREEDING FOR EGG PRODUCTION. PART II. SEASONAL DISTRIBUTION OF
EGG PRODUCTION WITH SPECIAL REFERENCE TO " WINTER" EGG PRO-
DUCTION. Utah Agr. Exp. Sta. Bui. 149, 71 p., 28 fig. Bibliography,
P- 55-56-
(3) Gowell. G. M.
1902. BREEDING FOR EGG PRODUCTION. In Maine Agr. Exp. Sta. Bui. 79, p.
26-40.
(4)
1903. BREEDING FOR EGG PRODUCTION. In Alainc Agr. Exp. Sta. Bui. 93, p.
69-76.
(5) Pearl, Raymond.
I912. THE mode of INHERITANCE OF FECUNDITY IN THE DOMESTIC FOWL. In
Jour. Exp. Zool., V. 13, no. 2, p. 153-268, 2 fig. Literature cited, p.
266-268. Also published as Maine Agr. Exp. Sta. Bui. 205, p. 283-394.
1912.
(6)
I915. MEASUREMENT OF THE WINTER CYCLE IN THE EGG PRODUCTION OF
DOMESTIC FOWL. In Jour. Agr. Research, v. 5, no. 10, p. 429-437.
Literature cited, p. 436-437.
(7) and Surface, F. M.
I911. A BIOMETRICAL STUDY OF EGG PRODUCTION IN THE DOMESTIC FOWL.
II. SEASON.'^L DISTRIBUTION OF EGG PRODUCTION. U. S. Dept. Agt.
Bur. Anim. Indus. Bui. no, p. 81-120, 30 fig.
DIGESTION OF STARCH BY THE YOUNG CALF
By R. H. Shaw, T. E. Woodward, and R. P. Norton, of the Dairy Division, Bureau
of Animal Industry, United States Department of Agriculture
PREVIOUS INVESTIGATIONS
There is considerable evidence that young animals thrive on a ration
containing starch, but a rather extensive search in the literature failed
to discover any data concerning the question as to how scon after birth
the calf can begin to digest starch. The investigation here described was
undertaken for the purpose of ascertaining how early in its life the calf
can utilize starch or starch -containing feeds. The practical application,
of course, is in supplementing or supplanting the milk ration of the
young calf with other feed.
The literature contains many accounts of feeding experiments with
young animals where starch alone or as the principal component of
some feed has been used. The purposes of these experiments, however,
have been largely to determine the effect of starch upon the health, the
rate of gain in weight, the cost of raising, or the digestibility of some
other component of the ration rather than the actual digestibility of
the starch itself. The record of but one experiment was found in which
the feces of young starch-fed calves were tested for the presence of
starch.
Ewing and Wells ^ report the use of starch in combination with corn
silage and cottonseed meal in the ration of 12 -month-old steers on di-
gestion trial. In their summary they state that when as much as
47.3 per cent of the net energy of the ration was supplied in the form
of starch the iodin test did not indicate the presence of starch in the
feces.
There are recorded in the medical literature on the diet and hygiene
of children several investigations in which the actual digestibility of
starch by children was studied. Kerley, Campbell, and Mason ^ report
the examination for starch of 324 stools, collected under controlled con-
ditions at the New York Infant Asylum from 60 children, all under i
year of age, who had been fed either wholly or in part on barley water.
The barley water was prepared by boiling raw barley flour for 1^4 hours.
The stools were examined for starch by the Von Jaksch iodin test, with
1 Ewing, Perry van, and Wells, C. A. the associative digestibility of corn silage, cotton-
seed meal, and STARCH IN steer RATIONS. Ga. Agt. Exp. Sta. Bui. 115, p. 269-296, 7diagr., 1915.
2 Kerley, C. G., Campbell, W^ C, and Mason, H. N. a further contribution to the study of
STOOLS of starch-fed INFANTS. In Jour. Amer. Med. Assoc, v. 47, no. 10, p. 763-765- 1906.
Journal of Agricultural Research. Vol. XII. No. 9
Washington. D. C. Mar. 4, 1918
mg
(575)
Key No. A — 35
t^y6 Journal of Agricultural Research voi.xii, no. 9
Lugol's solution. Of the 60 children, 33 always gave negative iodin
tests, indicating complete utilization of the starch. Among the re-
mainder, 8 usually gave a negative test, 12 usually gave a positive test,
and 7 always a positive test for starch. Of the 41 children showing a
good capacity for starch utilization one 19-day -old child received 9.2
gm. of barley flour daily for 2 days; one 21 -day-old child received 14.6
gm. every 24 hours; one child i month and 22 days old received 25.9 gm.
the first day and 25.3 gm. a day for the following four days; and another
child I month and 19 days old received 12 gm. daily for three days.
Heubner,* in a paper presented before the Berlin Medical Society,
describes an investigation conducted at Leipzig for the purpose of
determining the digestibility of starch in the food of artificially fed
children. The children received during i-day and 2-day periods a care-
fully prepared starchy gruel which was fed in place of milk, at the same
intervals and in the same quantity as the milk feeding. Carbon was
used to identify the experimental stools, and the feces from each child
were assembled, dried, and analyzed for starch. A 7-weeks-old child
received 24.6 gm. of rice flour during a 25-hour period, and no starch
was found in the feces. Another child 14 weeks old received 53 gm.
of rice flour during a 39-hour period, and 0.1689 g"^- o^ starch was found
in the feces. A third child i year old received, in addition to 72 gm.
of butter, 133 gm. of rice flour during a 48-hour period, and 0.2804 g^'
of starch was found in the feces. A fourth child, 14 weeks old, received
57 gm. of a specially prepared oatmeal during a 34-hour period, and
0.261 1 gm. of starch was found in the feces.
In an elaboration of the work done by Heubner at Leipzig, Carstens '
gives the results of digestion experiments on eight children from 5 to 14
weeks old. Some of these children received starch from rice flour, some
from a prepared oatmeal flour, and some from two different proprietary
infant foods. The same methods were followed as in the Heubner
investigation. The quantity of undigested starch varied from a trace
in the feces of two children, one 9 weeks and one 15 weeks old, respec-
tively, to 5.08 gm., or 6.23 per cent of the amount ingested by a child
6)4 weeks old.
Kriiger * who worked with fetal and newborn calves found that the
ptyalin is secreted in the salivary glands as early as the seventh month
of fetal life, but that while the quantity increases up to birth, even at
that time it is too small to be of any importance in the digestion of food.
' Heubner, O. ueberdieadsnutzung des mehls im darm jungersaugunge. /ra Berlin. Klin.
Wchnschr., Bd. 32, No. 10, p. 201-204, 1S95. Literatur, p. 204.
'Carstens, J. H. weitere erfahrungen iber die ausnutzung des mehls im darme junger
Saugunge. In Verhandl. Gesell. Kinderheilk., Bd. 12, p. 169-176. 1895.
' Kruger, Friedrich. die verdauungsfermente bhim embryo und neugeborenen. 80 p.
Wiesbaden, 1891. Literatur, p. 79-80.
Mar. 4, 1918 Digestion of Starch by the Young Calf 577
EXPERIMENTAL WORK
Two male calves, each 4 days old, were selected. Each was fed
5.44 kgm. of whole milk a day in two feedings. Beginning at 4 days of
age, each calf received 40 gm. of ordinary cornstarch per feeding, in
addition to the milk, for a period of three days. The starch ration was
prepared as follows: The weighed quantity of cornstarch was placed
in a pail and mixed with a little milk, then the bulk of the milk was
added and the mixture well stirred. The calf consumed the mixture
with eagerness and without any apparent digestive disturbance. To
make sure that all the starch was consumfd, the pail was rinsed once
or twice with milk and the calf permitted to drink the rinsings. The
starch-feeding period was followed by a rest period of about five days,
during which only whole milk was fed; then the calves again received
starch in addition to their whole-milk ration exactly as in their first
3-day period. The length of the periods and the duration of the
Fig. I.— Bag for receiving feces and harness for supporting it.
experiment are shown in Table I. The cornstarch fed to calf i con-
tained, according to analysis, 77.02 per cent of pure starch, and that
fed calf 2, 76.32 per cent. The feces were received in a closely fitting
rubber bag supported by a harness as shown in figure i. Collections
were made during the three starch-feeding days and the three days
following. The feces were removed from the bag each day and imme-
diately dried on the premises in an electric oven at about the temperature
of boiling water. The dried feces, representing a starch-feeding period
and the three subsequent days, were united and ground in a mill. The
analyses were made according to the methods adopted by the Association
of Official Agricultural Chemists, the malt-diastase method being selected
for the starch determinations.
The feeding was conducted at the Bureau of Animal Industry's Ex-
periment Farm, Beltsville, Md., and the analytical work was done at
the Dairy Division laboratories in Washington. The results are given
in Tables I and II.
578
Journal of Agricultural Research
Vol. XII, No. 9
Table I. — Composition of feces of calves
Animal and period.
Calf i
I. .. .
2 . . .
3- ••
4- ■ ■
Calf 2
I . . .
2« . .
3- ••
4. . .
Age of
calf
when
feces
were
col-
lected.
Days.
4
12
20
30
39
4
14
Weight
of dry
feces.
Gin.
288. 98
270. 28
244. 87
179. 40
202. 00
283. 49
190. 21
152-95
Nitrogen in feces.
Gm.
14.07
13-74
14. 20
11.66
17-25
14.27
13-54
10. 28
Per cent.
4.87
5.08
5.80
6. 50
8.54
5-03
7. 12
6. 72
Ether extract infeces.
Starch in feces
Gin.
6.36
Per cent.
2. 20
Gm.
144.2
"•95
13-75
7-36
14.71
4.42
5.62
4. 10
7.28
101. 5
67.8
IS- 2
2. 2
12.08
4. 26
146. 0
24. 25
17-58
12.75
11.49
6-7
1.6
Per ci.
49.91
37-56
27. 70
8.49
I. 07
51.48
3-53
1.03
a Sample was lost at the farm.
Table II. — Proportion of starch digested by calves
Animal and period.
Starch fed.
Starch in
feces.
Starch digested.
Calf i:
I
Gm.
184.9
184.9
184.9
184.9
184. 9
183.2
183.2
183.2
183.2
Gm.
144.2
loi. 5
67.8
15-2
2. 2
146. 0
Gm.
40.7
83-4
117. I
169.7
182.7
37-2
Per cent.
22. 02
2
45- II
2
63. 34
A . .
9*1. 79
98.81
20.30
e
Calf 2:
I
2 a
•1
6-7
1.6
176.5
181. 6
96. 32
A... . . .
99.10
<» Sample was lost at the farm.
CONCLUSIONS
The figures in Table II for digested starch show that the calves when
from 4 to 7 days old were able to digest about one-fifth of the quantity
consumed; in one case 22.02 per cent and in the other 20.30 per cent.
When calf i was 12 to 15 days old, the percentage of starch digested
had more than doubled and when 3 weeks old it had nearly tripled,
while at 4 weeks in the case of calf i and at 3 weeks in calf 2, the per-
centage of starch digested was well over 90.
While it is quite probable that a calf but a few hours old can not
digest an appreciable amount of starch, it can readily be seen that the
quantity of starch-splitting enzyms must increase very rapidly in the
first few days of life, for the calves under experiment, when only 3 to 4
weeks old, were able to digest a ration nearly 10 per cent of the dry
matter of which was starch.
These results indicate that the milk ration of a calf but a few days
old may be supplemented with a starchy food and that the starchy
material may be rapidh' increased as the calf grows older.
TOXICITY OF VOLATILE ORGANIC COMPOUNDS tO
INSECT EGGS^
By William Moore, Head of Section of Research in Economic Zoology, and Samuel A.
Graham, Assistant in Entomology, Minnesota Agricultural Experiment Station
INTRODUCTION
A general sur\"ey of the literature has failed to reveal any extensive
study of the toxicity of different materials to insect eggs. Certain spray
solutions have been studied, but they have been considered individually
and not in comparison with other related compounds.
Cooley (j)^, working with the oyster-shell scale {Lepidosaphes tdmi L.)
has shown that linseed oil, cottonseed oil, and lime-sulphur were effective.
In the cases of the oils some of the eggs were killed, while other eggs
hatched, but the young insects died during or shortly after emergence.
Lime-sulphur failed to kill the eggs, but the young were killed very soon
after hatching. Pure kerosene apparently had no effect on the eggs.
Gillette (2), working with aphid eggs of different species, has shown
that eggs treated with kerosene emulsion containing less than 25 per
cent of keresene were unaffected by the spray. Scalecide, Thompson's
Soluble Oil, lime-sulphur, and different soaps had little effect unless used
in very large doses. Tobacco extracts containing nicotine or nicotine
sulphate were found to be very effective.
Safro (9), in the study of lime-sulphur as an ovicide for the codling
moth (Carpocaspa pomenella L.), shows that this material is only effective
to eggs in which the embryo is almost fully developed.
Woodworth (lo) has studied the toxicity of hydrocyanic-acid gas
to the eggs of scale insects, but does not consider the factor of age.
Recently particular attention has been given to the study of the
toxicity of volatile organic compounds to the eggs of lice (Pedictdus
capitis and P. corporis, and Phthiris pubis). Kerosene has been used
against head lice and their eggs for many years, but in the recent work
many other materials have been recommended.
Postnikov (7) recommends amyl alcohol, ethyl alcohol, benzene,
chloroform, carbon tetrachlorid, methane, and birch tar for the destruc-
tion of head lice and their eggs. Gasoline has been used for the destruc-
tion of the eggs of the clothes louse, and Klinloch (4) claims that im-
mersion in this material for one minute will kill, while exposure to its
vapor is fatal in one-half hour. He considers that benzene, toluene,
1 Published, with the approval of the Director, as Paper 88 of the Journal Series of the Minnesota Agri-
cultural Experiment Station.
* Reference is made by number (italic) to "Literature cited," p. 586-587.
Journal of Agrictiltural Research, Vol. XII, No. 9
Washington, D. C. Mar. 4, 1918
™^ (579) ^^^ ^°' ^^""i- — 24
38321°— 18 5
580 Journal of A gricultural Research voi. xii, no. 9
and acetone are as toxic as gasoline. Von Prowazek (8) recommends
xylol and ether for the destruction of lice and their eggs.
In view of the work of the senior author (5, 6) showing that the toxicity
of organic compounds to insects is related to their volatility, of which
the boiling point is a general index, it was thought advisable to make
a similar study of the toxicity of a series of volatile organic compounds
to insect eggs.
METHOD OF EXPERIMENTATION
For these experiments it was considered desirable to use eggs which
were not protected from the action of the chemical by any covering.
Eggs of the bedbug and the clothes louse were considered; but, owing
to the fact that many of these eggs were found to be infertile under
the artificial conditions of breeding, and, further, that they were hard
to obtain in large enough quantities for the purpose of the experiments,
they were discarded. Potato-beetle eggs (Leptinotarsa decemlineata
Say) were finally decided upon as fulfilling all requirements. As many
as 50,000 eggs were used in these experiments, and it was found that
in every case untreated eggs hatched 100 per cent. They were also
convenient to use, as a cluster of 20 to 30 or more eggs could be easily
treated as a unit. The effects of the chemicals were studied in three
different ways:
1. By dipping the clusters in the chemical to be tested.
2. By spraying the clusters with the chemical by means of an atomizer.
3. By exposing the eggs to the action of the vapor of the chemical.
In the exposure to the vapor the eggs were fumigated in a similar
manner to that employed in the study of the toxicity of the vapor to
houseflies recorded in a previous paper (6). It was found necessary,
however, to use a longer time limit than 400 minutes, 15 hours being
finally selected for this purpose.
Eggs were fumigated for 15 hours with varying quantities of the
chemical, after which they were removed from the flask and placed in
open pasteboard pill boxes until they hatched or were undoubtedly
dead. The smallest dose necessary to kill the eggs in this length of
time was thus determined and reduced to millionths of a gram-molecule,
making possible an accurate comparison of the different chemicals used.
In experiments where the eggs were dipped or sprayed they were placed
in open pill boxes after treatment and handled in a manner similar to
those fumigated.
RESULTS OF THE EXPERIMENTS
The results of dipping and spraying the eggs are given in Table I.
The compounds used are arranged in the order of their boiling points,
from the lowest to the highest. It will be noted that in general the
eggs treated with compounds having the lowest boiling point — that is,
the most volatile compounds, permitted most, if not all, of the eggs to
hatch. Exceptions may be noted of compounds extremely active
Mar. 4. 1918 Toxicity of Organic Compounds to Insect Eggs
581
chemically, such as allyl alcohol, which contained ammonia as an im-
purity, and chlorpicrin. In general those sprayed showed a higher
percentage of hatching than those dipped. Some compounds of this
series, the vapor of which had been previously shown (6) to be non-
toxic to houseflies, owing to the fact that they formed gummy masses
on exposure to the air, were found to be toxic to the insect eggs. Pinene,
terpineol, and geranyl acetate are examples of such chemicals. Kurther,
it is noted that compounds which are so slightly volatile, owing to
their high boiling point, that they were ineffective against flies, were
found to be toxic to the insect eggs. Such chemicals were eugenol,
alpha naphthol, ethyl ether, and trimethylene cyanid. These com-
pounds are only effective, however, when brought into actual contact
with the eggs, as in spraying or dipping, and are no more effective as a
fumigant against the eggs than they are against adult insects.
Table I. — Relation of the boiling point to the toxicity of organic compounds used in
dips and sprays for potato-beetle eggs
Organic compound.
Ethyl ether
Ethyl mercaptan . . .
Carbon bisulphid . . .
Petroleum ether . . . .
Acetone
Chloroform
Methyl alcohol
Carbon tetrachlorid .
Ethyl alcohol
Gasoline
Benzene
Thiophene
Allyl alcohol
Amyl nitrite
Nitromethane
Propyl acetate
Toluene
Chlorpicrin
Pyridin
Acetic acid
Chlorbenzene
Amyl alcohol
Xylene
Amyl acetate
Bromoform
Pinene
Ethyl malonate
Allyl isosulphocyan
ate
Furfural
Butyric acid
"C.
35
36.2
46
40-70
56-3
61
66. s
78.1
78.4
70-90
80.3
84
97
98
lOI
IC2
III
112
116. 7
119
132
137
140
148
151. 2
160
160
161
162
163
O v.
2-oS
g bio
100
97
95
85
100
44
100
100
.S-d
•S >.
CO u
100
ICO
100
100
100
ICO
100
100
100
100
100
0
100
0
0
16
100
0
0
0
20
74
100
ICO
20
Organic compound.
Trimethylene bromid
Terpineol
Benzonitrile
Thiophenol
Benzaldeliyde
Anilin
Ortho-bromtoluene . . .
Valeric acid
Ortho-creosol
lodobenzene
Salicylic aldehyde. . .
Para-cresol
Meta-cresol
Nitrobenzene
Benzyl alcohol
Kerosene
Ortho-nitrotuluene. ..
Bromxylene
Citral
Quinolin
Eugenol
Nitroxylene
Nicotine
a-Napthol ethyl
ether
Trimethylene cyanid
Geranyl acetate
Brbmmethylphenyl-
ketone
Ethyl aceto-acetate. .
°C.
165
168
170
172.5
179. I
182
182
184.5
190
193
196
201.8
202. 8
205
206. 5
150-300
223
225
225
239
247-5
250
274
Si
0 »- .
i >.
bt'2'3
.a c3 ce
C4 O H
01 o. n
M-- —
Ssfa
17
582
Journal of Agricultural Research
Vol. XII, No. 9
Kerosene, although having a high boihng point of 150'' to 300° C,
allowed 83 percent of the dipped eggs and 100 percent of the sprayed
eggs to hatch.
Table II. — Toxicity of various organic compounds to eggs of different ages when dipped
for -various periods
Compound.
Ether.
Carbon bisulphid.
Methyl alcohol.
Chloroform .
Carbon tetrachlorid .
Benzene .
Toluene .
Acetic Acid.
Xylene.
Ortho-bromtoluene .
Age.
Fully developed
embryos
Partially developed
Freshly laid — . ...
'Fully developed. . . .
Partially developed
.Freshly laid
Fully developed . . . .
Partially developed
Freshly laid
'Fully developed ...
Partially developed
.Freshly laid
I Fully developed ...
Partially developed
Freshly laid
Fully developed. . .
Partially developed
Freshly laid
Fully developed. .
Partially developed .
Freshly laid
Fully developed. . .
Partially developed
Freshly laid
Fully developed
Partially developed
Freshly laid
Fully developed. . .
Partially developed
Freshly laid
Percentage hatching after dippmg for-
I sec-
ond.
100
100
IOC
100
100
100
100
100
100
100
100
100
100
100
100
100
100
S sec-
ond.
10 sec-
onds.
100
IOC
ICO
100
100
100
3
100
100
100
100
o
100
100
100
100
o
100
100
100
100
100
100
15 sec-
onds.
100
100
100
100
30 sec-
onds.
o
100
100
100
100
o
100
100
100
60 sec-
onds
120
sec-
onds.
100
100
5
100
100
100
100
100
100
O
In the experiments it v^as noted that compounds with low boiling
points evaporated from the surface of the eggs very quickly, this giving
the material very little time to penetrate. On the other hand, com-
pounds with high boiling points remained upon the eggs for hours, or
even days. In view of this fact, a series of experiments was conducted
with the lower boiling-point compounds, in which the eggs were dipped
Mar. 4, 1918 Toxicity of Organic Compounds to Insect Eggs
583
for periods from i to 120 seconds, eggs of different ages being used.
The results are given in Table II. This table shows that in general with
compounds of very low boiling points the freshly laid eggs or those con-
taining embryos only partially developed were more easily killed than
those in which the embryo was fully developed. Gortner and Banta (5),
in working with the toxicity of certain phenolic compounds to amphibian
eggs, found that the youngest eggs were more susceptible than the older
eggs. This may be due to disturbances in the permeability of the egg,
and compounds with low boiling points such as ether would have more
influence on the permeability than compounds with higher boiling points.
With an increase in the boiling point it was found that the eggs in which
the embryo was fully developed were most easily killed. Compounds
with very high boiling points, such as nicotine and kerosene, often
remained on the eggs and killed the larvae in the act of hatching. It
seems that compounds with high boiling points are not able to penetrate
the egg as readily as compounds having low boiling points. In general a
slightly longer exposure to the chemical will result in the death of a larger
number of eggs.
In these experiments the writers they have been unable to remove
compounds of high boiling points from the surface of the egg without
injury to the egg, and, hence, have no data as to the length of time
necessary to kill with these compounds.
Table III. — Toxicity of kerosene to eggs of various ages
Age of eggs.
1 day
2 days
3 days
4 days
Freshly laid
Slightly developed
Well developed
Spiracles visible through shell .
Within I day of hatching
Within 2 days of hatching
Within 3 days of hatching
Within 4 days of hatching
100
100
100
33
o
100
100
100
100
Percentage hatching after dipping for-
I min. sniin. loxnin. is jnin.i aojnin. Brand
13
100
100
100
100
100
100
100
o
100
o
100
100
100
100
o
100
100
Unknown.
Do.
Do.
Do.
Do.
Do.
Do.
Do.
Pearl oil.
Do.
Do.
Do.
A special study was made of the action of kerosene on the eggs, inas-
much as it was a high-boiling-point compound which did not always
kill. The results of dipping eggs of different ages in kerosene are given
in Table III. Eggs i day old in general did not hatch after being dipped.
Eggs classed as freshly laid, which may have been i or 2 days old, hatched
in every case, while eggs with well developed or fully developed embr3'os
failed to hatch. This is interesting in that by killing the young eggs it
584
Journal of A gricultural Research
Vol. XII, No. 9
acts in a similar manner to compounds with a low boiling point, while by
killing the eggs with fully developed embryos it acts as a compound hav-
ing a high boiling point. The Pearl oil used in the experiments is a prod-
uct from the California oil fields and therefore differs chemically from the
other kerosenes used.
In order to obtain further data on the toxicity of kerosene to eggs, two
brands of kerosene were broken up into fractions of different boiling
points. The two brands used were Pennsylvania Best sold by the Pure
Oil Co. and Perfection oil put out by the Standard Oil Co. of Indiana.
The data obtained by this experiment are shown in Table IV. The
fraction of Pennsylvania Best boiling between 140° and 187° C. killed
only freshly laid eggs. The fraction between 187° and 234° C. killed
both the fully developed and freshly laid eggs, while the fraction between
234° and 280° C. killed all the eggs. Apparently the lower fraction did
not remain long enough upon the fully developed eggs to produce any
influence. It was noted in the kerosene experiments that the higher
fractions killed at the time of hatching. This may explain the fact that
the partially developed eggs treated with the second fraction allowed
100 per cent to hatch. These eggs being too far developed were not
effected by changes in permeability such as would kill a freshly laid egg,
while the compound had sufficient time to evaporate from the surface of
the egg before the emergence of the larva.
Table IV. — Toxicity of different fractions of kerosene on eggs of different ages
Brand.
Pennsylvania Best. .
Age.
Fully developed ...
Partially developed
Freshly laid
General
do
Percentage hatching after dippinur for 5 seconds ia
fractions with boiUng points of —
140 -187
C.
100
100
o
100
100
187 -234
c.
100
o
100
100
234 -2 So
C.
O
o
o
100
Residue
above
280° C.
Undis-
tiUed.
O
O
100
50
Brand.
Age.
Percentage hatching after dipping for s
seconds in fractions with boiling
points of —
i35°-i83° C.
i83°-23i'' C.
23i°-28o'' C.
Perfection
Fully developed . . . .
Partially developed
Freshly laid
Mar. 4. i9i8 Toxicity of Organic- Compouuds to hisect Eggs
585
The Perfection oil killed both the fully developed and freshly laid
eggs, leaving the partially developed uninjured except for the fraction
between 231° and 280° C. Other experiments with these two oils, which
will be published later, have shown that the Perfection oil is more toxic
than the Pennsylvania Best. Kerosene therefore has oils of low boiling
points which kill the freshly developed eggs and oils of high boiling points
which destroy the fully developed eggs, while the partially developed
eggs are most likely to hatch.
Nineteen compounds were tested to determine the toxicity of their
vapor to insect eggs. The results of these experiments are given in Table
V, in which the compounds are arranged according to their boiling
points, from the lowest to the highest. The increase in toxicity with an
increase in boiling point is indicated by the smaller portion of a gram-
molecule necessary to kill in 15 hours. That the boiling point is not so
good an index of the toxicity of a chemical as its volatility is shown in
Table VI, where the compounds are arranged according to their volatility,
from the most volatile to the least volatile. A comparison of the two
tables shows a better correlation between volatility and toxicity than
between boiling point and toxicity. As was found in the work with the
housefly (5), chemicals having low boiling points are more valuable
for fumigation purposes due to the fact that a larger amount of the
vapor may be contained in the air.
Table V. — Toxicity of the vapor of certain organic compounds (in millionths of a gram-
molecule required to kill in 75 hours' exposure). Compoundf arranged according to
their boiling points
Compound.
Ethyl ether
Carbon bisulphid . .
Petroleum ether. . .
Acetone
Chloroform
Methyl alcohol ....
Carbon tetrachlorid
Ethyl alcohol
Gasoline
Thiophene
Boiling
point.
°C.
35
46
40-70
56.3
61
66.5
78.1
78.4
70-90
84
Toxicity
(mil-
lionths of
a gram-
molecule).
3, 468. 2
555-5
243. 2
558-0
I, 232. o
404. o
284. o
41. 1
344- o
Compound.
Toluene
Chlorpicrin . .
Chlorbenzene
Xylene
Bromoform . .
Furfural
Nitrobenzene
Nitroxylene . .
Nicotine
Boiling
point.
°C.
Ill
112
132
140
151-
162
205
250
250
Toxicity
(mil-
lionths of
a gram-
molecule).
104. O
2.8
160.8
29-5
3-9
1.8
2.7
3-3
1.8
586
Journal of Agricultural Research
Vol. XII, No. 9
Table VI. — Toxicity of the vapor of certain organic compounds {in millionihs df a gram-
molecule required to kill in 15 hours' exposure). Compounds arranged from the m-ost
volatile to the least
Compound.
Toxicity
(millionth s
of a gram-
molecule).
ComiX)und.
Toxicity
(millionths
of a gram-
molecule).
Kthyl ether
3, 468. 2
243. 2
1,232. 0
I; 363-0
555-0
558.0
404. 0
284. 0
2.8
344- 0
Chlorbenzeiie
160 8
Petroleum etlier
Toluene
104. 0
29-5
41. I
3-9
I 8
Methyl alcohol
Acetone
Xylene
Gasoline
Carbon bisulphid
Chloroform
Bromoform
Furfural
Carbon tetrachlorid
Nitrobenzene . . .
2.7
1.8
Ethyl alcohol
Nicotine
Nitroxylene
Chlorpicrin
Thiophene
3-3
Fumigation in a saturated atmosphere with ether, ethyl mercaptan,
carbon bisulphid, benzene, carbon tetrachlorid, and chloroform will kill
all the eggs in one hour.
SUMMARY
(i) In general, compounds with high boiling point and slight volatility
are more eiffective in dipping and spraying insect eggs than compounds
with low boiling point and high volatility.
(2) Compounds with low boiling points kill freshly laid eggs more
readily than eggs in which the embryo is partially or fully developed.
(3) Compounds of higher boiling points are more toxic to eggs with
fully developed embryos than they are to eggs in which the embryo is
only slightly formed.
(4) Kerosene containing both high and low boiling points is destruc-
tive to both young and old, but is only slightly toxic to partially devel-
oped eggs.
(5) The toxicity of the vapor of organic compounds to insect eggs is
related to the boiling point and the volatility. As the boiling point
increases and the volatility decreases, the toxicity increases.
LITERATURE CITED
(1) COOLEY, R. A.
1910. NOTES ON SPRAYING EXPERIMENTS FOR THE OYSTER SHELL SCALE IN
MONTANA. In Jour. Econ. Ent., v. 3, no. i, p. 57-64.
(2) Gillette, C. P.
I910. SOME insecticide tests for the destruction op APHIDIDAE AND
THEIR EGGS. In Jour. Econ. Ent., v. 3, no. 2, p. 207-210.
(3) GoRTNER, R. A., and Banta, A. M.
1914. notes ON the toxicity of dilute solutions of CERTAIN PHENOLIC
COMPOUNDS AS INDICATED BY THEIR EFFECT ON AMPHIBIAN EGGS AND
EMBRYOS, TOGETHER WITH REFERENCES ON MODIFICATIONS OP PIG-
MENT DEVELOPMENT. In Biochem. Bui., v. 3, no. 11/12, p. 357-368.
Literature cited, p. 368.
Mar. 4, i9i8 Toxicity of Organic CompouTids to Insect Eggs 587
(4) KiNLOCH, J. p.
191 5. AN INVESTIGATION O? THE BEST METHODS OF DESTROYING LICE AND
OTHER BODY VERMIN. In Brit. Med. Jour., 1915, no. 2842, p. 1038-
1041.
(5) Moore, William.
I917. toxicity OV various BENZINE DERIVATIVES TO INSECTS. In JotlT. Agr.
Research, v. 9, no. 11, p. 371-381, 4 fig. Literature cited, p. 380-381.
(6)
1917. VOLATILITY OP ORGANIC COMPOUNDS AS AN INDEX OP THE TOXICITY OP
THEIR VAPORS TO INSECTS. In JoxiT. Agr. Research, v. 10, no. 7,«p.
365-371. 7 fig-
(7) POSTNIKOV, A. I.
1915. ON THE QUESTION OF THE dbNTROL OP LICE IN THE ACTIVE ARMY.
(Abstract.) In Rev. Appl. Ent., s. B, v. 3, pt. 8, p. 122-123. 1913.
(Original article in Proc. Conf . Bacteriologists and Representatives of
Medical-Sanitary Authorities on the Campaign against Infectious Dis-
eases in Connection with the War. p. 70-71. Moscow. Not seen.)
(8) Prowazek, S. von.
19x5. BEMERKUNGEN UBER DIE BIOLOGIE UND BEKAMPFUNG DER KLEIDER-
LAUS. In Mtinchen. Med. Wchnschr., Jahrg. 62, No. 2, p. 67-68.
(9) Safro, V. I.
1912. LIME-SULPHUR WASH AN INEFFIOENT OVICIDE FOR CODLING MOTH. In
Jour. Econ. Ent., v. 5. no. 5, p. 385-395.
(10) WOODWORTH, C. W.
1915. TOXICITY OF INSECTICIDES. In Science, v. 41, no. 1053, p. 367-369.
CORN-STOVER SILAGE
By J. M. Sherman, Bacteriologist, and S. I. Bechdel, Assistant Dairy Husbandman,
Pennsylvania Agricultural Experiment Station
INTRODUCTION
The ensiling of com stover and cured com fodder is not a new idea.
In a few localities farmers have followed this practice to a limited extent
for some years, and there have appeared several Experiment Station
publications which deal briefly with the process. Experimental data
on the subject are, however, very meager; little has been established with
reference to the practicability of stover silage, while nothing is known
concerning the nature of the fermentation which takes place and the
factors operative other than what may be deduced from knowledge
of ordinary silage. The present necessity for more economic produc-
tion, with the conservation of concentrates and the utilization of more
roughage in the live-stock and dairy industries, makes a reconsidera-
tion of corn-stover silage of especial pertinence.
The object of the present study was partly to test the practicability
of ensiling stover, and partly to determine the nature of the fermenta-
tion which takes place therein.
PRACTICABILITY OF ENSILING CORN STOVER
The stover used in this experiment was ensiled early in April, 191 6;
the material had been kept in a shed since fall and was quite dry. The
condition of the material was not good; it was moldy in spots and on
the whole represented an inferior grade of stover. This stover was
run through a silage cutter and packed in the silo by means of tramp-
ing. Water was added in a continuous stream through a hose which
was cairied and the water distributed by the man who did the packing.
A water meter was attached to the hose so as to enable the regulation of
the amount used. A wooden-stave silo 16 feet in diameter was filled
with 32,000 pounds of stover to which were added 66,000 pounds of
water.
vSamples of the silage, which were taken at frequent inter\'als, were
examined for general appearance, texture, and aroma. The stover
was soon observed to undergo a fermentation with the formation of a
product quite similar to normal silage made with green com. The
material softened, regained a slightly greenisli color, and developed an
aroma simulating that of normal silage, though inferior in all these
respects to silage made in the usual way from green com.
Feeding tests made at the end of the experiment showed that cattle
ate this silage with little waste and apparently with a relish. While
Journal of Agricultural Research, Vol. XII, Ko. 9
Washington, D. C. Mar. 4, 1918
mi Key No. Pa. — 5
(589)
590 Journal of A gricultural Research voi. xn, No. 9
it is not believed that the stover silage is as palatable as is that made
from fresh com, it did prove to be a very acceptable feed which was
preferred by cattle to any of the dry roughages furnished. Some ex-
periments in which the actual feeding value of stover silage is deter-
mined in comparison with ordinary silage and other roughages are
desired, and it is hoped that such tests may be carried out at this Station
in the near future.
The keeping quality of stover silage appears to be excellent, pro-
vided sufficient water is added. Although, as is the case with other
types of silage, the surface material undergoes a moldy spoilage
accompanied with heat formation, this condition does not extend to
more than the ordinary depth. The silage made in this experiment
was not all used during the following winter, and the remaining material,
at the time of this writing nearly 1^2 years old, is still in excellent con-
dition. In view of the very satisfactory results obtained wdth such
an inferior raw product, we do not hesitate to predict success in the
ensiling of any stover which is in reasonably good condition.
Probably the most important consideration for the successful pro-
duction of stover silage is the amount of Vv^ater to add. This obviously
will vary according to the quantity of water contained in the stover, and
this factor should be taken into consideration. While it would be
more scientific and exact to determine the most desirable amount of
water to add by means of moisture tests on the stover, such a recom-
mendation would find no place in farm practice. In our experimental
silo the proportion was about 2 parts of water to i of stover, but the
stover was probably somewhat drier than would usually be the case.
As may be seen from the moisture determinations, which are reported in
another section (Table I), the quantity of water added was none too
much; in general appearance and to the touch some of the samples
seemed to be considerably below the most desirable point.
A laboratory test was carried out upon this point by making stover
silage in small jars with varying quantities of water and examining
after about one month for general appearance and condition of moisture.
The stover used in these tests was very dry. Samples which were
made with equal parts by weight of water and stover, as well as those
made with i }4 parts of water to i of stover were too dry to undergo a
typical fermentation and form good silage. Those which had water in
the proportion of 2 parts to i of stover made good silage, but did not
appear to have as much moisture as would be best. Samples put up
with 2)4 and with 2^ parts by weight of water to i part of stover were in
good condition when opened and apparently did not contain an excess
of water.
It seems that in general, when reasonably fresh stover is used, about
2 parts of water by weight to i of stover would be advisable, while for
older and drier stover a slightly larger proportion of water may be more
Mar. 4, i9i8 CoYn-StoveY SUagc 591
desirable. It should be kept in mind that these tests were all made
with rather dry stover. In the case of ensiling soon after the corn is
husked, 2 parts of water might be too much. However, it appears
from our observ^ations that there is less danger of adding too much
water than of getting too small an amount, and that considerable water
may be added above the required amount without injury to the product.
The water should be added uniformly as the silo is being filled so that
all the dry cut stover becomes thoroughly wet down. If this precau-
tion is not taken, the water may follow channels down through the silage
and waste away at the bottom of the silo. In such an instance spoiled
silage might result in some parts of the silo because of an insufficient
amount of water.
FERMENTATION OF STOVER SILAGE
To obtain information on the nature of the fermentation which takes
place in silage made from stover, determinations were made of the vola-
tile and nonvolatile acids, temperatures, and numbers and types of
bacteria at various stages of the ripening process.
Samples for examination were obtained by means of a 2-inch auger
provided with an extension shaft of 8 feet, thus making it possible to
penetrate to the center of the silo. By repeatedly boring in a short dis-
tance and withdrawing the auger until the center of the silo was reached,
no difficulty was experienced in securing sufficient material for the
different tests. The sample obtained in this way represented the silage
mass from the wall to the center of the silo. Not more than one boring
was made in one place, the different samples being removed at points
all the way around the silo and from 3 to 8 feet from the ground. The
material so obtained was subjected to pressure in an ordinary lard
press and sufficient juice collected for the various examinations made.
Acid formation. — There can be no doubt that the amount and char-
acter of acids in silage influence its quality profoundly. Since it has
been the general experience that extremely green com produces a very
sour silage, while com more nearly mature produces a silage with less
acid and of a much better quality, the acid fermentation in silage made
from dry material is of interest.
The volatile acidity was determined by subjecting a loo-gm. sample
of juice to steam distillation under reduced pressure until 4 liters of
distillate were secured. These were titrated directly after collection with
Njio barium hydrate, with phenolphthalein as the indicator. The non-
volatile acidity was obtained by the difference between the volatile
acidity and a total acid determination made by the titration of 20 gm.
of juice, diluted to 500 c. c. with carbon-dioxid-free water, against
Njio barium hydroxid. In Table I the nonvolatile and volatile acids
(calculated as lactic and acetic, respectively) are reported in terms of
percentage of air-dry material.
592
Journal of Agricultural Research
Vol. XII, No. 9
Table I. — Acid formation in stover silage
Age.
Week
3/7
1
2
3
4
5
6
8
10
12
Total
solids."
Per cent.
28.9
35- 5
26. 7
26.8
27. I
29.7
31.2
27-5
32. I
25.8
Acid in total solids.
Nonvola-
tile.^
Per cent.
Trace
o. 16
•95
1.36
1-54
1-51
2. 00
2-53
2. 64
3-15
Volatile, c
Per cent.
0.51
87
36
41
49
55
69
92
82
24
Ratio of
nonvolatile
to volatile.
5- 40
1-43
I. 04
•97
I. 02
•85
•7S
•75
•71
u Air-dried.
b Calculated as lactic acid.
c Calculated as acetic acid.
It is seen from the figures in Table I that for the first week the vola-
tile acidity was greatly in excess if the nonvolatile portion ; that from the
second to the fifth week the two were apparently present in about equal
amounts; and that from the sixth week on, the nonvolatile acids were in
excess, the proportion of nonvolatile acidity increasing to the end of the
experiment. The total acidity obtained was somewhat lower than is
usually found in ordinary silage. This is probably to be expected, in
view of the chemical differences in the raw materials. In this con-
nection, however, it should be noted that in this experiment samples
were not taken after the twelfth week. It is not unlikely that the acidity
increased some after the last sample was secured. In regard to the pro-
portion of nonvolatile to volatile acids, if we accept the ratio i to 0.75
reported by Dox and Neidig (5)^ as representing a general average for
ordinary silage, it will be seen that our results on com stover silage in-
dicate a remarkable agreement in this respect between these two types of
ensilage.
Fermentation temperatures. — Temperature records were obtained
by means of four resistance bulbs with about 60 feet of insulated cable
attached to each which w^ere buried in the silage as the silo was filled.
The bulbs were located as follows :
No. I. About 2 feet from the bottom and in the center of the silo.
No. 2. About 6 feet from the bottom and about 3 feet from the center of the silo.
No. 3. At the same height as No. 2 but about 3 feet from the center in the opposite
direction.
No. 4. About 12 feet from the bottom and in the center of the silo.
The ends of the cables were located at a convenient place on the out-
side of the silo so as to allow easy attachment for temperature readings.
Table II gives the temperatures obtained from April 4, the day the silo
was filled, until June 16.
• Reference is made by number (italic) to "Literature cited," p. 6co.
Mar. 4, 1918
Corn-Stover Silage
593
Table II. — FeTtnentaiion temperatures of stover silage
Date.
Apr, 4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
May I,
2,
4
6
8
16
22
31
June 16.
Temperature CF.).
Bulb I. Bulb 2
42. o
42.5
46.5
46. 5
47.0
47-5
49.0
49.0
49-5
50.0
50-5
50.0
50.0
50.0
50.0
50.0
50.0
50. o
49- 5
49-5
50.0
50. o
49-5
49-5
50. o
49-5
49.0
50.0
50.0
50.0
50-5
50-5
51.0
52.0
42
49.
SC-
52-
54-
54-
54-
54
56-
56.
56.
56.
56-
56.
57-
57-
57-
57-
57-
57-
57-
57-
57-
57-
57-
57-
58-
58-
58.
58.
58.
58.
58.
Bulb 3.
. 0
42.
•5
47-
-5
48.
. 0
48.
-5
49.
. 0
51-
.0
51-
•5
51-
5
52-
. 0
52-
. 0
53-
. 0
53-
. 0
53-
. 0
53-
•5
54-
.0
54-
.0
54-
. 0
54-
. 0
54-
•5
55-
-5
55-
•5
55-
•5
55-
-5
55-
•5
55-
. 0
55-
-5
55-
•5
50.
•5
5^
. 0
5f
. 0
sf
. 0
50
. 0
57
•5
57
Bulb 4.
46.
47.
48.
49.
SO'
SI'
52'
52'
55'
55'
55'
55'
56.
56.0
56.0
55- o
56. o
57- o
57- o
57-0
57-5
57-5
57
58. 5
59-
59-
59-
59-
59-
59-
61.
63-
Atmos-
phere.
33'
42.
43'
46,
33
33
37
42
48.0
58.0
51.0
49.0
50. o
52.0
49-5
46.0
55-0
49
49
41,
47
47
48,
48. 5
47.0
66.5
63-5
53-5
60.5
65-5
65.0
55- o
64. o
63-5
Recent investigations have furnished abundant proof that high tem-
peratures are not essential in silage preservation, and, in fact, do not
occur except at the surface, which undergoes an aerobic decomposition.
Bechdel (3) has recorded an instance in which the maximum temperature
attained in the center of a concrete silo during the curing period was only
60° F. As has been shown by Eckles, Oshel, and Magruder (6), the
atmospheric temperature at time of filling influences greatly the tem-
perature attained during the fermentation of the silage.
Table II shows that the temperature at the start was 42° F. and gradu-
ally increased until the readings were discontinued. The maximum
temperature attained was 63° F. in the case of bulb 4; but bulb i, which
was buried to a depth slightly below the surface of the surrounding
soil, showed a maximum temperature of only 52° F. An examination
of the column giving the atmospheric temperature during this period
suggests that the continued increase in the silage temperature during
the latter part of the time may be accounted for by a similar increase in
594
Journal of Agricultural Research
Vol. XII, No. 9
the outside air. However, the more rapid increase during the early
part of the period was entirely independent of this factor. A compari-
son of the temperature records secured in this experiment with the data
which have been obtained on ordinary silage at this and other Experi-
ment Stations indicates that there is no wide difference, if any, between
the rate and amount of increase in temperature in silos in which ordi-
nary silage and corn-stover silage have been stored.
BactERIOIvOGICAIv observations. — A quantitative bacterial exami-
nation was made on each of the samples taken. The juice obtained was
plated in proper dilutions and the counts obtained were reported as
numbers per cubic centimeter of juice. Lactose agar was used, and the
plates were counted after six days' incubation at 33° C. As may be
seen from Table III, it appears that the bacterial comit increases during
the first week and is followed bv a continued decrease thereafter.
Table III. — N timber of bacteria in stover silage at different stages of curing
Age.
Number of bacte-
ria per cubic
centimeter.
Age.
Number of bacte-
ria per cubic cen-
timeter.
Weeks,
oh
528, 000, 000
3, 630, 000, 000
I, 850, 000, 000
975,000,000
400, 000, 000
Weeks.
c
510, 000, 000
235, 000, 000
186, 000, 000
98, 000, 000
71,000,000
I
6
2
8
•J
10
4
12
Direct microscopic examinations of the silage juice were made in
order to follow in a general way any marked changes which take place
in the bacterial flora during the curing process. This at best could only
give suggestive data, but such examinations are sometimes important
in connection with cultural studies. At first a great variety of cells
were observed. During the first two weeks rods and cocci were appar-
ently present in about equal numbers, after which the rods became
increasingly predominant. Toward the end of the experiment prac-
tically nothing but rods were found in the microscopic preparations.
Because of the high acidity it is not likely that the cocci were active
nearly as long as they appeared under the microscope. But it is prob-
able that the acid medium would tend to peserve the cells so that they
would appear for some time after they were inactive or even dead.
A qualitative bacterial study was also carried out. All of the colonies
from a representative lactose-agar plate from each sample were isolated
and subjected to a cultural study. For the present purpose they may be
divided roughly, according to their action on litmus milk, into acid-
forming, casein-digesting, alkali-forming, and inert groups. The acid
formers may be further divided according to whether they produced
sufficient acid to cause coagulation of the milk. The distribution of
these groups in silage at various stages of its fermentation is shown in
Table IV.
Mar. 4, 1918
Corn-Stover Silage
595
Table IV. — Groups of bacteria present at different stages of curing
Age.
Percentage of total number.
Acid-non-
coagulat-
ing group.
Weeks.
3/7
I
2
3
5
6
8
10
12
10
53
14
57
20
71
II
63
21
67
60
30
68
32
60
40
--,t>
39
Casein-
Alkali-
digesting
fonniug
group.
group.
10
4
9
6
0
0
5
0
0
0
0
0
0
0
0
0
5
0
Inert
group.
23
14
9
10
O
o
o
Table IV shows that the rather complex bacterial flora which is present
at the beginning of the process gives way to one which is almost entirely
acid-producing as the fermentation progresses. The proportion of
acid-forming and coagulating organisms to the noncoagulating ones
also increases as the curing period advances. A comparison of these
figures with those given in Table III indicates that the change in flora
is not to be accounted for by an actual increase in the high acid-forming
organisms during the latter part of the fermentation period, but rather
to the fact that they do not decrease as rapidly because of their greater
resistance to the unfavorable hydrogen-ion concentration.
The division of the acid-forming organisms into coagulating and non-
coagulating types, though convenient and significant for the present
purpose, probably does not separate them into natural groups. From
early in the fermentation the predominating organisms were acid formers,
most of which probably belonged to the same general group'. We have
found cultures which were apparently identical, as indicated by the
cultural and fermentative reactions studied, but which varied in the
amount of lactic acid produced in milk from only 0.3 to more than 2.0
per cent. All of these probably belonged to the same general group as
the aciduric bacteria which have previously been noted as occurring
abundantly in silage. In the first two samples examined organisms of
the colon-aerogenes group were found and ^Iso a few cultures which
were probably Streptococcus lacticus, but tests were not applied which
would definitely identify the latter.
NATURE OF SILAGE FERMENTATION
Until recently the cell-respiration theory of silage fermentation estab-
lished by Babcock and Russell (/, 2) has not been seriously challenged.
During the past year, however, several publications ha\e appeared in
support of the bacterial explanation of this phenomenon. In view of
the recent contributions to the subject, it is not out of place to examine
38324"— 18 6
596 Journal of Agricultural Research voi. xii, no. 9
critically the present status of the question. As the older literature
has been reviewed so many times that further elaboration is not neces-
sary, we shall pass in review only those papers of very recent date.
Hunter and Bushnell (8) have demonstrated the presence in silage
of large numbers of high acid-producing bacteria, and have furnished
strong evidence that these organisms are mainly responsible for the
acid fermentation. Although their work is a most valuable one, it should
be borne in mind that the evidence is circumstantial and perhaps not
conclusive. It is rather doubtful if the data submitted justify the positive
conclusion that —
The present investigation warrants the statement that acid production, common
to ail normal silage, is largely the result of fermentation by the Bulgarian group of
bacteria.
The fact that these bacteria formed considerable acetic acid when
grown in alfalfa extract to which was added i per cent of glucose hardly
warrants the assumption that —
Although these organisms evidently do not produce all of the acetic acid found
in normal silage, they must be responsible for a large per cent of it.
Sherman (75) also noted the presence of large num.bers of the aciduric
bacilli in silage and reported some observations which indicated that
they are of significance in the fermentation process. The evidence,
however, was not direct and by no means conclusive.
Hunter (7) in his work on heat production in silage has added further
weight to the bacterial theory of silage fermentation. Unfortunately
it is not possible to evaluate properly some of the interesting points
contained in his paper, as they are obscured by insufficient description.
For example, some data are given which show the difference in heat
formation between green kafir heated and green kafir inoculated with
Bacillus bulgaricus. The exact treatment in this case is not clear: If
the inoculation v\^as made into heated kafir, the results are of utmost
significance; if, on the other hand, as the caption of the graph would
indicate (see 7, fig. 9), the inoculation was made into unheated kafir
and that compared with heated kafir, the test contributes nothing to
the solution of the moot question. In his assumption that cell respira-
tion can play no part in the fermentation of silage made from dry for-
age. Hunter has arrived, we think, at conclusions which are, in part at
least, erroneous.
In a very interesting paper Lamb (9) concludes that both factors
are of importance, btit that microorganisms play the larger part, espec-
ially in the production of acid. Following the suggestion of Rahn {10),
he has attempted to determine the cause of the process by the rates
of change in the fermentations studied. The course of the curve obtained
when such data were plotted was interpreted as indicating whether
the action was of bacterial or enzymic origin.
Mar. 4, igis Corn-Stover Silage 597
Although it is true that purified enzym preparations acting under
favorable conditions give time curves which follow, with certain modi-
fications, the law of mass action, while cur\res representing bacterial
action take an entirely different course because of an increase in the
active mass with the multiplication of the organisms, we are inclined
to believe that Lamb has placed undue confidence in this method,
especially when we consider the complexity of the material studied
and the factors concerned. A few considerations will suffice to illustrate
some of the possibilities of error in such a method. In the first place
conditions in silage are not constant, but are undergoing continual
change. For example, the temperature and acidity, which are of utmost
significance in enzym action, increase as the fermentation progresses.
In view of the great increase in activity of some enzyms as the temper-
ature is increased, and the stimulating effect on some enzyms of an
increased hydrogen-ion concentration (within certain limits), it is not at
all impossible that these factors might so modify the course of action
of an enzym as to produce a curve resembling that typical of bacterial
action.
Again, the phenomenon of adsorption and the action of the so-called
antienzyms in many cases may so suppress the activities of an enzym
during the early stages of the reaction as to cause it to follow a course
not at all characteristic of enzym action. This has been beautifully
illustrated by Rosenthal (11) in his work on the antitryptic action of
egg albumen. It was shown that the trypsin was at first suppressed,
but gradually regained its power and increased in activity so as to give
the appearance of an increase in the "active mass" as indicated by a
curve convex to the axis of abscissae (the typical bacterial curve). On
the other hand, the same trypsin preparation when acting on egg albu-
men which had been previously heated to destroy the antitrypsin gave
a time curve characteristic of enzym action.
These illustrations will suffice to show the fallacy of such a method,
but the number of possibilities of error in its application to such a com-
plex mixture as silage might be increased almost indefinitely. That the
limitations of this method of studying biochemical phenomena were
appreciated by Rahn (10) is shown in the following paragraph from
his valuable paper:
It is hardly necessary to mention that the curve of a process will be an absolute
means of discussion only in case of pure cultures. In natiu-al fermentations, there
is always the possibility that different processes taking place at the same time de-
stroy the regular form of the curve. A simple example would be the growth of an
acid-producing and an alkali-producing organism in the same liquid. It is also
possible that an enz;jTnic curve imder certain conditions shows the form of a fermen-
tation curve : We can imagine that an enzym is acting slowly at first, because of an
imsatisfactory acidity of the medium. By a chemical or microbial process, inde-
pendent of the enzymic action, the acidity may be made more suitable for the en-
zyme, and this will cause an increased rate of action of the enzyme and give the type
598
Journal of Agricultural Research
Vol. XII, No. 9
of a fermentation curve without the presence of organisms. The value of the curve
is, therefore, not an absolute one and no conclusions ought to be drawn without con-
sideration of the possibilities of error.
It would indeed seem that the application of such methods to the
study of silage fermentation is entirely without foundation.
To the evidence in support of the bacterial theory of silage fermenta-
tion should be added the very suggestive observation of Clark (4) that
the hydrogen-ion concentration of mature silage is coir;cident with the
limiting hydrogen-ion concentration obtained in cultures of BacUlus bui-
garicus, which organisms are considered by some workers as of para-
mount importance in the ripening process. In support of the respiration
theory, on the other hand, it is pertinent to call attention to the recent
work of Round (12), which indicates that cell respiration is of greater
importance in the fermentation of sauerkraut than has been generally
recognized.
It was thought at the beginning of this experiment that a study of
the fermentation in silage made from dry stover would throw much
light on the nature of the process in ordinary silage. The belief was
held that the activity of the plant cells (which have been demon-
strated to play an important part in the fermentation of silage made
from green com) would be eliminated, to a large extent at least, in the
stover silage. But, as may be seen from an examination of the fore-
goiag results, the fermentation of ensiled stover appears to resemble,
in its main characteristics, that which takes place in green-corn silage.
In an effort to determine the factors responsible for the fermentation,
laboratory tests were made by ensiling stoVer under different condi-
tions. Glass jars containing 175 gm. of cut stover and 400 gm. of water
were used. Some were untreated, some put up with antiseptics, while
some were sterilized in the aiitoclave and reinoculated with i per cent
of raw-silage juice. The results of this study are given in Table V.
Table V. — Fermentation of stover ensiled under different conditions
Sample;
No.
Treatment.
Untreated
....do
Sterilized and inoculated
...do
2 per cent of toluene
....do
2 per cent of ether
do
Aj;e.
Acidity.
Weeks.
Per cent.
4
15-5
4
10.4
7
2.4
5
6
6
3-3
4-3
4.6
5
10.8
8
12.4
Bacterial count.
210, 000, 000
240, 000, 000
320, 000, 000
36, 000, 000
180, 000
460, 000
7, 500, 000
Not made.
The main points brought out by this test are that stover silage is
capable of undergoing a fermentation when preser\^ed with ether, while
bacteria alone are apparently unable to produce the typical fermenta-
tion, even though conditions are favorable for their active development.
The predominating organisms in the sterilized and inoculated samples
Mar. 4, 1918 Corn-Stover Silage 599
were of the same type as those characteristic of normal silage. The
organisms found in the samples preserved with antiseptics, on the other
hand, were a more miscellaneous group; and, although many probably
belonged to the same group as the aciduric bacilli of normal silage, the
cultures isolated were mostly very weak acid producers. The fact that
fermentation took place under ether indicates that the activity of the
plant cells, whether it be called "respiration" or "autolysis," is present
in silage made from dry stover. As silage preserv^ed with ether fer-
mented, whereas in that kept with toluene the process was checked
suggests that conclusions drawn from experiments conducted with only
one antiseptic are of doubtful value. When opened, the ether-preserved
samples, after the evaporation of the ether, appeared to resemble the
untreated material, while the sterilized and inoculated silage were
"flat" and lacking in the characteristic aroma. The results with ether
were checked by another test in which triplicate samples were pre-
served, and again an active fermentation took place as was indicated
by the development of acidity in each case.
Although our results would tend to support the respiration theory
of silage-curing rather than the bacterial, we do not feel that the data
thus far collected warrant definite conclusions on this point. It is
difficult to believe that such active acid-forming organisms should
occur in silage in large numbers without taking some part in the acid
fermentation; perhaps they supplement in some important way the
action of the plant cells. It is not inconceivable that a preliminary
cleavage due to cell respiration is an essential prerequisite for the
vigorous action of the aciduric bacteria. In fact, the continued increase
in the ratio of nonvolatile to volatile acidit}' as the fermentation pro-
gressed (see Table I) might lead one to suspect that such was the case.
On the other hand, the great increase in the nonvoladle acidity from
the fifth to the twelfth week, during which time the bacterial count
was rapidly decreasing, might be interpreted as strong evidence against
that view. It is clear that microorganisms are not solely responsible
for the fermentation of silage, and the weight of evidence at the present
time, in our opinion, indicates that their role is not as important as that
of the plant cells.
Although not committing ourselves definitely qn the nature of silage
fermentation in general, in regard to the present problem we do maintain
that the fermentation which takes place in stover silage is similar in its
essential points to that of ordinary silage and is caused by similar
factors. SUMMARY
Com stover when ensiled with a suitable quantity of water undergoes
fermentation with the production of a palatable silage of good -keeping
quality, which resembles ordinary com silage in aroma and appearance.
The fermentation which takes place in corn-stover silage appears to
be essentially the same as that of silage made from green com, as is
5oo Journal of Agricultural Research voi. xii. N0.9
indicated by the total acidity developed, the ratio of nonvolatile to
volatile acids, temperature observations, and bacterial studies.
From a review of the present status of the question as to whether
bacteria or plant cells are mainly responsible for silage fermentation,
it is concluded that the data thus far published are inconclusive.
Although the results of the present study tend to support the cell-
respiration theory, conclusions on this point are withheld.
LITERATURE CITED
(i) Babcock, S. M., and Russell, H. L.
1900. CAUSES OPERATIVE IN THE PRODUCTION OF SILAGE. In Wis. Agr. Exp.
Sta. i7tli Ann. Rpt., [1899V1900, p. 123-141, fig. 17.
(2)
1901. CAUSES OPERATIVE IN THE FORMATION OF SILAGE- (SECOND PAPER.)
In Wis. Agr. Exp. Sta. i8th Ann. Rpt., [i90o]/i90i. p. 177-1S4, fig. 44.
(3) BechdEL, S. I.
1917. STUDIES IN THE PRESERVATION OF CORN SILAGE. 3° P- Reprint (separate
No. y)from Pa. Agr. Exp. Sta. Ann. Rpt. 1915/16.
(4) Clark, W. M.
191 7. THE ACID PRODUCTION OF BACILLUS BULGARicus. (Abstract.) In Abs.
Bact., V. I, no. i, p. 59-60.
(5) Dox, A. W., and Neidig, R. E.
1913. lactic ACID in CORN SILAGE. lowa Agr. Exp. Sta. Research Bui. 10,
P- 363-37S. 4 fig-
(6) EcKLES, C. H., OsHEL, O. I., and Magruder, D. M.
1916. SILAGE investigations: normal temperatures and some factors
INFLUENCING THE QUALITY OF SILAGE. Mo. Agr. Exp. Sta. Research
Bui. 22, 32, p., 7 fig.
(7) Hunter, O. W.
i917. microorganisms and heat production in silage fermentation.
In Jour. Agr. Research, v. 10, no. 2, p. 75-83, 10 fig. Literature
cited, p. 82-83.
(8) and BushnELL, L. D.
1916. SOME IMPORTANT FERMENTATIONS IN SILAGE. Kan. Agr. Exp. Sta.
Tech. Bui. 2, 32 p. Bibliography, p. 32.
(9) Lamb, A. R.
1917. THE RELATIVE INFLUENCE OF MICROORGANISMS AND PLANT ENZYMS ON.
THE FERMENTATION OF CORN SILAGE. In JouT. Agr. Research, v. 8,
no. 10, p. 361-380, 13 fig. Literature cited, p. 37S-380.
(10) Rahn, Otto.
I910. THE USEFULNESS OF CURVES IN THE INTERPRETATION OF MICROBIAL
AND BIOCHEMICAL PROCESSES. Mich. Agr. Exp. Sta. Tech. Bui.
5, 29 p., 18 fig.
(11) Rosenthal, Eugen.
I9IO. UNTERSUCHUNGEN UBER die ANTrPROTEOLYTlSCHE WIRKUNG DES
BLUTSERUMS. In Folia Serol., Bd. 6, p. 285-300.
(12) Round, L. A.
I917. THE BACTERIOLOGY OF SAUERKRAUT, A FURTHER STUDY. (Abstract.)
In Abs. Bact., v. i, no. i, p. 50.
(13) Sherman, J. M.
I916. A CONTRIBUTION TO THE BACTERIOLOGY OF SILAGE. In JoUT. Bact., V. I,
no. 4, p. 445-452. Bibliography, p. 452.
WEEVILS WHICH AFFECT IRISH POTATO, SWEET
POTATO, AND YAM
By W. DwiGHT Pierce,
Entomologist, Southern Field Crop Insect Investigations, Bureau of Entomology, United
States Department of Agriculture
INTRODUCTION
In a previous article ^ the writer discussed three important Andean
weevil pests of the potato tuber {Solanum tuberosum). In the present
paper a fourth potato tuber weevil is described and notes are presented
on three weevils which attack the tubers of sweet potato (Jpomoea
batatas) and one which attacks the tubers of the yam {Dioscorea batatas).
WEEVILS WHICH AFFECT IRISH POTATO TUBERS
The native home of the Irish potato is the west coast of South America,
and here we find that the crop has a series of characteristic pests which
may be easily disseminated in shipments of potatoes. As stated above,
the writer has described three of these species in a previous paper. The
description of the larva of one of them is now added and a new species
described. In order that these weevils may be easily distinguished one
from another by the man in the field the following table has been con-
structed :
TABLE OF IRISH POTATO TUBER WEEVILS
1. Prostemum grooved for reception of beak; mandibles without deciduous piece;
pronotum with, a deep median furrow widened angularly at middle and also
behind Rhigopsidius tucumanus Heller.
Prostemum not grooved for reception of beak; mandibles with deciduous piece . . 2
2. Mandibles with tooth beneath; scrobes abruptly and broadly terminated, not
extending beneath ?
Mandibles without tooth beneath; scrobes narrowing and extending beneath; pro-
thorax not as wide as elytra, angulate and broadest in front of base
Premnotrypes solani Pierce.
3. Prothorax broader than elytra, subquadrate, with sides parallel to apical third,
thence strongly narrowed ; sides of elytra smooth . . Trypopremnon laiithorax Pierce.
Prothorax acutely angulate at sides, widest before base; sides of elyixa tuberculate,
Trypopremnon sanfordi, n. sp.
' Pierce, W. D. new potato weevus from andean south America. In Jour. Agr. Research, v. i,
no. 4, p. 347-352, 3 fig., pi. 39-41- 1914.
Journal of Agricultural Research, Vol. XII, No. 9
Washington, D. C. (601) Mar. 4, 191S
ms Key No. K— 61
6o2 Journal of Agricultural Research Voi. xii.No. 9
FAMILY PSALIDURIDAE PIERCE (1914)
Rhigopsidius tucumanus Heller (1906)
This weevil has been recorded by the writer ^ from Tucuman, Argen-
tina; Cuzco, Temuco, and Arequipa, Peru; Oruro, Bolivia; and Ancud or
San Carlos and Castro Islands, Chile. The weevil belongs to the sub-
family Rhytirhininae and the tribe Rhytirhinini.
FAMILY PSALLIDIIDAE PIERCE (1916)
Premnotrypes solani Pierce (1914) ^
This weevil was described from the mountain districts of Peru. It
belongs to the subfamily Entiminae and the tribe Ophryastini.
Trjrpopremnon latithorax Pierce (1914) ^ (PI. 29, 30)
This weevil was described from Cuzco, Peru. It belongs to the same
tribe as Premnotrypes. The ventral tooth of the mandible does not
belong to the deciduous piece as stated in the original description.
On June ii, 1914, Mr. H. L. Sanford found several larvae of this species
in potatoes from La Paz, Bolivia, collected by Mr. H. T. Knowles, under
Federal Horticultural Board No. 2475. On June 20-26 pupae were
noticed, and an adult emerged on June 26.
This enables the writer to describe these stages.
Larva (PI. 29). — Length 12.5 mrn. when crawling, 10 mm. when slightly curved.
It is typically rhynchophorid in form, white, with light reddish brown head and dark
mandibles. The essential diagnostic characters are illustrated by the author in
Plate 29.
From the base of the head a median pale line passes forward. This is the epicranial
suture. It divides behind the frons and forms the two frontal sutures. The frons is
subtriangular, rounded behind and margined in front by the epistoma. The epicranial
areas are the two large areas at each side of the epicranial suture, further bounded by
the frontal sutiu-e, the pleurostoma, and the hypostoma. In front of the frons is the
clypeus, and in front of this is the labrum. The clypeus and labrum partly overlap
the mandibles which arise at the side of the clypeus based on the pleurostoma. Below
the hypostoma at the sides of the mandibles arise the maxillae, of which the cardo is
a Very large basal area. Located on the median line below the mouth opening, which
is covered by labrum and mandibles, is a shield-shaped area known as stipes labii.
Below and around this is the large basal area consisting of mentum and submentum.
There is a small abortive branch of the frontal suture extending back on the epicra-
nium, on each side of and not far from the epicranial suture. It is terminated by a
setigerous puncture. On the epicranium there are setae arranged as follows on each
lobe: One at terminus of branch of frontal suture, one on the frontal suture, two oppo-
site middle of frons, one basal, two discal , one opposite base of mandible, two on hypo-
stoma. On the frons there are three pairs of setas, the two posterior pairs being about
equidistant, the anterior close to the antennae. At base of clypeus there are four
tiny hairs. On labrum are four subbasal, six subapical, and six marginal hairs. The
mandibles have one hair each. The maxillae are provided with two-jointed palpi
(PI. 29, F), and a very broad setose lacinia, two setae near base of palpi and one
near base. Some of the hairs are clavate as shown in the illustration, but this is not
always true, some specimens having normal hairs. The stipes labii has one pair of
hairs. Each lobe of the mentum has one pair.
> Pierce, V»'. d. Op. cit.
Mar.4. i9i8 WeevUs Affecting Potatocs 603
The pronotum is simple, undivided; the mesonotum and metanotum are composed
of praescutum and scutoscutellum. The first six abdominal sclerites are composed
of a spindle-shaped praescutum, a transverse scutum terminated by the spiracles, a
spindle-shaped scutellum, and a transverse postscutellum very greatly narrowed on
the dorsum. The praescutum has a few hairs. The scutellum has a row of hairs.
Just above each spiracle is a tiny hair. On each epipleural lobe beneath the spiracles
there is one hair. There are eight abdominal spiracles and one on the mesothorax.
The seventh and eighth segments are more crowded than the preceding. The ninth
and tenth are small and reduced.
Pupa (PI. 30). — Length 10 mm., white. The most interesting features of this pupa
are the rudimentary wing pads seen only when the elytra are spread. The elytral pads
are not as large as often found in Aveevil pupae. The antennae are not geniculate.
The beak is short. There are five pairs of hairs located on the head and beak as
illustrated. On the thorax, which is subquadrate with truncate angles, there are
setigerous tubercles as follows: Four on anterior margin, two antemedian and two
postmedian on the disc, two pairs of antemedian and two pairs of postmedian on lateral
margin. Mesonotum and mettootum with one pair of setae each. The- first abdominal
segment has two pairs of setse, and the remaining segments have a long line of setigerous
tubercles. Each femur has two apical hairs, and a few ventral hairs are found as
illustrated. It is interesting to note that the processes of the ninth segment are acute
but reduced almost to the size of the tubercles. The tenth segment is ventral to the
ninth.
Trypopremnon sanfordi, n. sp. (PI. 28)
Described from a single specimen collected in quarantine by Mr. H. 1,.
Sanford September 24, 191 5, from a potato tuber sent by Mr. O. F. Cook
from Cuzco, Peru.^ The excellent illustrations of the type (PI. 28) were
made under the writer's supervision by Mr. H. B. Bradford.
Length 8 mm., greatest breadth 4.5 mm. Beak longer than head and narrower
than eyes; the dorsal squamose portion being gradually narrowed from eyes to nasal
plate. Alae strongly flared, making the scrobes open above. Head tumid above
the eyes. Median line slightly depressed on head, strongly in frontal fovea, and
very faintly on beak except just behind nasal plate. Lateral depressions on beak
strong. Apex of beak brownish black, -with nasal plate polished, convexly raised
around margin, emarginate at apex. Mandibles shining brownish black; deciduous
piece reddish brown, lightest at tip, moderately long, arcuate, with sharp edges;
the ventral tooth is not as acute as in T. latithorax; there is a slight denticle on the
right deciduous piece, and the left mandible is denticulate as shown in the figure.
The antennal scrobes are strongly flexed downward, very much broadened and
evanescent behind; scape clavate; fimicle with first two joints elongate, the others
progressively shorter, the last moniliform; club as long as the foiu- preceding joints.
Head, beak, and scape densely clothed with fine silky-bronze scales, and with scat-
tered white setae; funicle sparsely setose; club minutely pubescent, sparsely setose.
Prothorax basally truncate, slightly broadly emarginate at middle; apically sinuate;
with very strong supraocular lobes, which have vibrissae on the inner surface; surface
coarsely irregularly punctured, finely densely squamose with golden metallic scales,"
sparsely setose with white curved setae; surface very uneven, with median depression
bordered by antemedian ridges and two postmedian tubercles; sides prominently
produced by two angulate tubercle's; widest at posterior tubercles.
Elytra at base narrower than thorax; humeri tuberculate; sides subparallel but
very roughly tuberculate, abruptly narrowed at posterior declivity which is nearly
perpendicular. Scutellum triangular. Surface densely minutely squamose, sparsely
' Recorded under Federal Horticultural Board No. 4348.
6o4 Journal of Agricultural Research voi. xii, N0.9
setose; striae irregular, with strong isolated punctures; entire surface covered
with tubercles which are largest on the third, fifth, and seventh intervals.
Front coxae contiguous. Prostemum strongly arcuately emarginate. Meso-
stemal coxae narrowly separated. Intercoxal piece of first abdominal segment
broad and deeply punctate. Second segment as long as third and fourth together.
Type. — Cat. No. 21613, United States National Museum.
WEEVILS WHICH AFFECT SWEET-POTATO TUBERS
At least four species of weevils attack the tubers of the sweet potato —
namely, Euscepes batatae Waterhouse, Cylas formicarius Fabricius, C.
turcipcnnis Boheman, and C. jemoralis Faust. The adults of Euscepes
and Cylas can not be confused. Those of Cylas are differentiated
as shown in the "Table of sweet-potato weevils of the genus Cylas," p.
605. The pupffi of Cylas can be recognized from the fact that the direction
of the appendages is anteriad, while in Euscepes it is posteriad. Refer-
ence to the illustrations will be of great assistance in separating them.
The larvse can not be so easily distinguished, as both are of the same
general shape. It will be noticed, however, that they have quite a
different system of abdominal folds. The larva of Euscepes is more com-
pact. That of Cylas, when alive, is often attenuate and tightly drawn
so that no folds can be distinguished. When killed in a liquid which
shrinks it slightly, however, it will be noticed that the praescutal areas
are proportionately larger and often subdivided transversely. The
praescutum of Euscepes is not subdivided. This sclerite is the anterior
sclerite of a segment and almost always has a few tiny hairs. The only
other dorsal sclerite with hairs is the scutellum. In Cylas the scutellum
adjoins the praescutum, and the scutum is only lateral. In Euscepes
the scutellum is separated from the praescutum by the scutum.
FAMILY APIONIDAE LE CONTE (1876)
Subfamily Cyladinae Pierce (1916)
genus cylas latreille (1802)
Cylas Latreille, 1802, Hist. Nat. Gen. et Part. Crust, et Insects, t. 3, p. 196.
Type. — Cylas brunneus Fabricius, monotypic.
This genus contains twenty named species, of which two are widely
known under the names formicarius and turcipcnnis. There is consid-
erable confusion about these two species, due in part to the claims of
lyC Conte and Faust that they are synonymous. Fabricius described a
piceous-brown Indian species with reddish thorax as formicarius; Olivier
illustrated the species as almost pink but described it as brownish;
Schonherr cited it as piceous; Gyllenhal described a species from Java
with greenish-blue elytra, red thorax, and black head as turcipcnnis;
Labram and Imhoff illustrated a blue species under this name. Finally
Wagner presented an illustration of a species with green elytron as
turcipcnnis, and he probably is right. The bluish species was named
elegantulus by Summers. It is a common sweet-potato weevil. If it
Mar 4, iyi8 WeevUs Affectiug Potatoes 605
should prove to be a variety of formicarius as here treated, after exam-
ination of the type, it must still be considered very distinct specifically
from iurcipennis. The National Museum collection contains six species.
Sketches have been made of the side view of the head and thorax ^of
the three species which presumably attack sweet potato, formicarius
variety elegantulus (Pi. 31, A), femoralis (Pi. 31, F), and iurcipennis
(PI. 31, B), and also of brunneus (PI. 31, C-E), which was erroneously
recorded by the writer from sweet potato in the Manual of Dangerous
Insects.^
TABLE OP SWEET POTATO WEEVILS OP THE GENUS CYLAS
1. Male club twice as long as funicle or longer; antennae as long as head and thorax;
head not more than one-fifth shorter than beak; elytra greenish, thorax red,
head black, legs red with dark band (PI. 31, B) iurcipennis Boheman.
Male club not twice as long as funicle 2
2. Male club half to three-fourths longer than funicle, female club almost one-third
shorter than funicle; male antennas almost as long as head and thorax; head
one-fourth to one-third shorter than beak; el^-tra bluish, thorax red, head black,
legs red (PI. 31, A) .formicaritis Fabricius, var. elegantulus Summers.
Male club half longer than funicle ; head as long as beak ; antennae as long as thorax
plus head behind eye; el;>i;ra black with blue or green luster, suture piceous;
thorax black, margins piceous; head black; legs dark red with black ring on
femora (PI. 31, F) .femoralis Faust.
Cylas formicarius Fabricius (1798)
Brentus formicarius Fabricius, 1798. Sup. Ent. Syst., p. 174, no. 5.
Fabricius - gave the following description :
Habitat Tranquebariae.
Parvus in hoc genere. Rostrum cylindricum, atrum, antennis rufis, monili-
formibus: articulo ultimo longiori, cylindrico, clavato. Thorax rufus, antice
globosus. Elytra laevia, atra, nitida. Pedes rafi, femoribus clavatis, at inermibus:
annulo nigro.
In altero sexu antennarum clava brevior, ovata.
Olivier and Schonherr described the species as piceous with ferru-
ginous thorax, antennae, and legs. It hardly seems possible that this
can be the same species as the common sweet potato wee\'il with shiny
blue-black elytra, red thorax and appendages, and black head and beak.
For this reason it is considered best to apply to the sweet-potato
weevil a name which certainly applies to it — elegantulus Summers.
In order that economic entomologists may not be inconvenienced greatly,
and in deference to the many writers who have assigned Fabricius's
name to the sweet-potato weevil, elegantulus may be considered as a
variety of formicarius until there can be an examination of the type.
Cylas formicarius elegantulus Summers (1875), the Sweet Potato Weevil (PI. 31,
A; PI. 32, A, B; PI. 2>Z< E-H; PI. 34, A-D)
Otidocephalus elegantulus Summers, 1875, in New Orleans Home Jour., Jan. and Dec.
,Cylas formicarius Le Conte, 1876, in Proc. Amer. Phil. Soc., v. 15, p. 327. O . elegantulus is QVLOted in
synonymy.
> Pierce, W. D. a manual of dangerous insects ... p. 209. 1917- Published by the United States
Department of Agriculture, Office of Secretary.
2 Fabricius, J. C. systema eleutheratorum ... v. 2 p. 549. Kiliae, 1801.
6o6 Journal of Agricultural Research voi. xii, N0.9
This is the common sweet-potato weevil (PI. 31, A; 32, A, B) with
bluish elytra, red thorax and appendages, and black head. The illus-
tration of the adult is drawn from a New Orleans specimen. The side
view of head and thorax is from a Hawaiian specimen. There is quite
a r^nge of difference in measurements of the species but the analysis of
these differences is reserved for a more technical paper now in prepara-
tion. The immature stages are described from specimens collected at
Victoria, Texas.
Larva. (PI. 34, A-D). — The larva of this species measures from 5 to 8 mm. in
length and is white, with light brownish head and darker brown mandibles. The
head shield is slightly angulately emarginate behind. From the center of the emar-
gination on the median line the epicranial suture passes forward, separating the epi-
cranium into two parts; this suture divides behind the frons and forms the two frontal
sutures. The frons is subtriangular, emarginate at anterior angles for antennae,
and emarginate along epistoma for attachment of clypeus. The median line is im-
pressed and darkened. The frons has three pairs of large setae, the posterior pair
being closest and the median pair but little more separated. The anterior pair are
located very close to the antennal fossae. A tiny pair of setae are located in such a
way as almost to form an equilateral triangle with the posterior and median setae.
The epicranial areas are located on each side of the epicranial suture. Each lobe
bears the following setae: One very close to the apex of frons, one slightly posterior
to this and farther from the median line, one opposite the middle of the frons, one
a little farther from the median line on the same line as the preceding, one toward
the base of the frons, one opposite the mid(Jle of the mandible, one opposite the hy{X)S-
tomal angle of mandible, one on hypostoma near base of mandible, one opposite
but distant from mandible, and three forming a triangle on disc of epicranium.
The antenna is a fleshy, two-jointed appendage located at the lateral angle of the
frons. The mandibles are very bluntly bidentate. Each mandible has a tiny hair
about the middle. The clypeus is attached in front of the frons and is broadly trans-
verse. It bears on the epistomal margin four tiny hairs. The labrum is not as broad,
is rounded in front, and has a row of four setae behind the middle, a seta on each side
in front and closer than the outermost setae, and four marginal setae. The maxillae
are elongate, terminated by a two-jointed palpus and a setose lacinia. The maxillae
are provided with four setae, two near palpus, one at middle, and one at base. The
stipes labii is cordate, bearing two-jointed palpi and a single pair of setae. Each
lobe of the mentum is provided with two pairs of setae.
The entire surface of the body is covered with tiny pubescence.
The prothorax is dorsally not divided but has the praescutal and scutal areas in-
dicated by rows of setae. The mesothoracic spiracle is located on a lobe very close
to the prothorax. The praescutum of the prothorax and that of the mesothorax are
provided with a few small hairs. The scutellum is marked with a row of hairs.
The first eight abdominal segments are normal and each bears a spiracle. The
praescutal area is more or less transversely divided, and its posterior lobe is marked
with a few tiny setae. The scutellum is large and prominent and provided with a
row of setae. The scutum is only lateral and has just above the spiracle a tiny seta.
Each epipleural lobe is provided with a single seta, and each hypopleural lobe with
two setse. The coxal lobes of the thorax bear several setae, and those of the abdomen
a single seta each. The last two abdominal segments are simple and provided with
a number of setae.
Pupa (PI. 35, E-H). Elongate, about 6 ram. long, white. This pupa is especially
characterized by the nongeniculate antennae which lie parallel to the legs. The
antennae and two anterior pairs of legs are directed cephalad instead of posteriad.
Mar.4. i9i8 WeevUs Affecttfig Potatoes 607
as in the Curculionidae. Another characteristic is that the femora and tibiae of the
posterior pair, being directed in the same manner, are completely covered by the
wings. The head and beak are elongate and provided with setigerous tubercles
as follows: One pair between the eyes at base, one pair immediately behind eyes,
two tiny pairs between eyes, and two pairs on beak; the posterior pair being close
to the eyes, and the anterior behind the middle. The antennae are roughly tuber-
culate.
The prothorax is margined anteriorly by four pairs of setigerous tubercles and has
one pair of discal setae. The mesothorax has two pairs of lar-ge setigerous tubercles
and a lateral pair of tiny setae. The femora bear two or three setae. The knees of
the posterior femora are visible dorsal ly only.
The mesothorax bears three pairs of small setae between the bases of the elytra.
The metathorax is provided with two rows of setae on tiny tubercles, the anterior row
having two pairs and the posterior row six pairs. The abdominal segments have
dorsally five pairs of setigerous tubercles near the posterior margin, a pair of tiny
setae near the middle of the segment, and lateral setigerous tubercles.
The ninth segment is provided with two large curved processes. The tenth seg-
ment is ventral to the ninth.
Cylas turcipennis Boheman (1833) (PI. 31, B)
Cylcs turcipenfiis 'Bohew.a.B, 1833, f« Schonherr, Gen. etSpec. Cure. v. i. p. 369-370.
The brief preliminary diagnosis of the species presented by Boheman
is as follows :
Elongatus, viridi-coerulescens, nitidus, antennis thorace pedibusque rufis, capite
cruciatim impresso, rostro punctulato, elytris modice convexis, subtiliter striato-
punctatis. Habitat in Java, in India orientali.
The following dimensions are included in the detailed description: Length 3 lines
(6 mm.); antennae as long as thorax and head; club of male antenna longer than pre-
ceding joints; beak not longer than head; elytra twice as wide as thorax at base, and
twice longer than wide. The color description is as follows: Head obscurely viridi-
cceruleous; beak almost black; antennae rufo-ferruginous; thorax shining rufous;
elytra coerulescent-virescent; thorax beneath rufous, remainder of body beneath
coeruleo-virescent; legs rufous; tarsi beneath fulvous, spongy; female with femora
in middle annulate virescent.
Two specimens from Palembang, Sumatra, collected by Mr. M. Knappert,
are here considered as this species. They differ only in having the beak
slightly longer than the head, and a statement to th\s effect might have
been made if the description had been based on examination with a low-
power lens. Two other specimens are at hand from Bay Laguna Prov-
ince, Philippine Islands, collected by Mr. P. L. Stangl. A part of a
body of a weevil from Guatemala, collected by Mr. D. G. Eisen, is also
undoubtedly this species. Pascoe records the species from Sarawak,
Java, and India.
Cylas femoralis Faust (1898). (PI. 31, F)
Cylasfc-moralis Faust, 1899, in Deut. Ent. Ztschr., p. 24.
This species was collected by Mr. Rolla P. Currie at Mount Coffee, Liberia,
in February to April, 1897, and he has informed the writer that it was a
serious sweet-potato pest in that country. It is described from Kamerun.
In the Manual of Dangerous Insects ^ this species was referred to as
C. brunneus by mistake.
'Pierce, W.D. a m.\nuai, of dangerous insects ... p. 209. 1917. Published by the United States
Department of Agriculture, OfKce of Secretary.
6o8 Journal of Agricultural Research voi. xii.Ng.g
FAMILY OROBITIDAE PIERCE (1916)
Subfamily Orobitinae Pierce (19 16)
genus euscepes schonherk (1844)
Euscepes Schonherr, 1844, Gen. et Spec. Cure, v. 8, pt. i, p. 4:9- Type— porcellus Boh. by original desig-
nation.
Euscepes Lacordaire, 1866, Gen. Coleop., v. 7, p. loo-ioi. Type. — porcellus Boh.
Hyperonwrpha Blackburn, 1885, in Sci. Trans. Roy. Dublin Soc, s. 2, v. 3, p. 182-183. "Typt— {squamosa.
Blackbum)=fcato<ae Waterhouse.
Euscepes Champion, 1905, in Biol. Centr.-Amer., Coleopt., v. 4, pt. 4, p. 496-49S. Type.— porcellus Boh.
Lacordaire caused a confusion of genera by wrongly interpreting the
number of funicular joints, of which there are seven. This error was
corrected by Champion. The two genera Euscepes and Hyperomorpha
are strictly congeneric ; in fact, the two type species differ principally in
size. A large series of porcellus from various parts of Central America is
at hand. These have been carefully compared w^th Blackburn's descrip-
tion of Hyperomorpha, but no generic difference can be found.
The rostral canal extends along the prosteraum and ends in a meso-
stemal pocket. The beak when at rest fits tightly into this canal. The
prothorax is lobed to cover the eyes when at rest. The body is elongate.
Euscepes batatae C. O. Waterhouse (1849), ^Iie Scarabee of the Sweet Potato (PI. 32,
C, D; PI. Z2,> A-D; PI. 34, E-H)
CryptorhynchusbataiaeWnterhouse, 1849, in Trans. Ent. Soc, London, v. 5, p. LXix.
Hypercjiwrpha sguamosa Blackburn, 1885, in Sci. Trans. Roy. Dublin Soc, s. 2, v. 3, p. 182-183.
Euscepes bataiae Champion, 1905, in Biol. Centr.-Amer., Coleopt., v. 4, pt. 4, p. 497.
This weevil (Pi. 32, C, D) is one of the most serious cosmopolitan pests
of the sweet potato, although hitherto it has been recorded only from
Barbados, St. Vincent, and Antigua, St. Kitts, Nevis, and Hawaii. In
all of these places, however, it is reported as damaging sweet potatoes.
The receipt of tvv^o specimens from Dr. Da Costa Lima, of Brazil, with
the statement that they were injuring sweet potatoes at Rio de Janeiro,
caused the writer to make a search through the undetermined collections
of the National Museum with the result that the known distribution of
the species is hereby greatly extended. Specimens are at hand from
Barbados, injuring sweet potatoes May 22, 1900, and more recent
material; Hope, Kingston, Jamaica, on sweet potatoes, Mr. S. F. Ashby;
Campinas, Brazil, injuring sweet potatoes, August, 191 3, Mr. A. Hempel
(No. 100); Rio de Janeiro, Brazil, injuring sweet potatoes, July, 1917,
Carlos Moreira; Honolulu, Oahu, Hawaii, bred from sweet potato ; Kaimuki,
Oahu, Hawaii, bred from sweet potato; Guam, on sweet potato, Mr.
®. T. Fullaway; Norfolk Island, New Zealand, March, 1883, Mr. P. H.
Metcalfe; Mayaguez, Porto Rico, injuring sweet potatoes, 1912, 1914,
1 91 7, Mr. C. W. Hooker, Mr. R. H. Van Zwaluwenburg.
This extensive distribution indicates that there are probably many
other countries where sweet potatoes are grown that may have the
weevil. If, fortunately, it should prove to be absent in other countries,
rigid quarantines should be put into effect, such as that recently estab-
Mar. 4. I9I& Weevils Affecting Potatoes 609
lished by the United States. In fact it has been only because of tke
excellent system of quarantine inspection in California that the species
has not already come into the United States with Hawaiian potatoes.
Many shipments of infested sweet potatoes have already been inter-
cepted at the California ports.^
The following is a description of the species redrawn to include all
the material at hand. The variations of color will be mentioned in
subsequent paragraphs.
Length about 4 mm. Brown, mottled with lighter areas, especially by a transverse,
irregular, postmedian band on the elytra. Squamose, bristling with upright setse.
Beak curved, carinate and laterally bifurcate. Front foveate. Front and beak
bristling with erect scales. Vertex provided with more decumbent scales. Prothorax
constricted in front, laterally impressed on disk behind, mottled with erect scales,
except on posterior margin which is provided with smaller, more decumbent scales.
Elytral strise composed of rather distant punctures, each bearing a small scale ; stu^ace
closely set with overlapping scales and each interspace with a single series of elongate
squamiform seta. Undersides more sparsely clad with semierect scales. Legs
provided with scales and seta. Rostral canal deep, terminating in a prominent
pocket of the mesostemum. Intercoxal process broad, angulate on anterior margin.
First segment behind coxae subequal to the second, which is but slightly longer than
the subequal third and fourth segments. The femora are minutely toothed.
•
Mr. Bradford's excellent illustrations will be very helpful in identifying
this weevil. His illustration of the adult is from Hawaiian material.
The species varies from very light brown to almost black and on the darkest speci-
mens the mottling and the postmedian vitta have practically disappeared.
On light specimens the scales of the thorax are mostly dark brown, with flecks of
pale scales and with the basal scales orange colored. The scales of the eljixa are
mottled in many shades of brown. The postmedian fascia extends to the fifth inter-
space and is bordered by very dark scales and divided by a wa\^ dark line. The
erect setae are mixed dark and white. The ventral scales are pale, but on the legs
they are mottled dark brown and pale. This description fits best some of the
Jamaican, Barbados, and Brazilian specimens. Almost black material comes from
Brazil, Jamaica, and Guam. The Hawaiian specimens are a duller brown, and the
New Zealand material is the lightest of all. There is, however, no doubt of the specific
identity of the entire series.
In order that the immature stages may be readily distinguished from
those of Cylas jormicarius a series of very careful drawings of the essen-
tial characters of the larva and pupa have been made by Mr. Harry
Bradford under the writer's direction. The drawings of the vertex and
face are by the writer. Barbados material was used for these drawings.
Larva (PI. 34, E-H). The larva of this species measures about 5 mm. in length
and is white, with a yellowish head and reddish brown mandibles tipped with black.
The maxillae and labium are slightly tinged with brown.
The head shield is broadly, angulately emarginate behind; from the center of the
emargination on the median line the epicranial suture passes forward, separating the
epicranium into two parts. This suture divides behind the frons and forms the two
' Whitney, L. A. the small sweet potato weevil (crvptorhynchus batatae waterh.). In Mo
Bui. State Com. Hort. [Cal.], v. 4, no. 3, p. 162-164, fig. 24-28. 1915.
6jo Journal of Agricultural Research voi. xii, no. 9
fj-ontal sutures. The frons is sub triangular, rounded at anterior angles, and slightly
emarginate for antennae; its front margin is the epistoma. There are three pairs of
large setae on the frons, the posterior pair being located rather close together and near
the apex of the triangle. The second pair are farther apart and halfway to the front.
The anterior pair are located near the lateral angles just behind the antennae. A tiny
pair of setae are located just in front of the posterior pair of large setae. A tiny pair
are located slightly behind and outside of the median pair of large setae. The median
line of the frons is impressed from the posterior angle almost to the middle.
The epicranial areas are the two large areas on each side of the epicranial suture
further bounded by the frontal suture, the pleurostoma, and the hypostoma. The
following setae occur on each lobe of the epicranium: One opposite the apex of the
frons, one on tlie disk of the epicranium in the line with the preceding; one opposite
the middle of the frons and near the frontal suture; one near the hypostoma toward
the base of the mandible; one as close to the hypostoma and opposite the base of the
maxilla; one near the pleurostoma; the last three forming a triangle. Forming a
semicircular line with the setse opposite the frons are a tiny seta, a longer one opposite
the antenna, and a long one opposite the pleurostomal seta.
The antenna is a small, fleshy, two-jointed appendage at the angle of the frontal
suture and pleurostoma. The mandibles are bluntly bidentate and have two small
setae. The clypeus is attached in front of the frons and is broadly transverse. It
bears on the epistomal margin four tiny hairs. The labrum is not as broad; it has a
pair of median setae and three pairs of marginal setae. The maxillae attached at the
side of the mandibles are terminated by a two-jointed palpus and a setose lacinia.
The maxillae are provided with three setae, two near the palpi and one toward the base.
The stipes labii is appendiculate, bilobed, bearing two-jointed palpi and a single pair
of posterior setae. Each lobe of the mentum is provided with one seta. The thoracic
segments are simple, being composed of only praescutum and scutoscutellum. In
the prothorax there is no separation of these parts, but they are indicated by the
arrangement of the setae. The abdominal segments are dorsally composed of four
single sclerites, namely, praescutum, scutum, scutellum, and postscutellum.
The thoracic spiracle is located on a lobe of the meso thorax, very close to the head.
The abdominal spiracles are located on a small lobe at the side of the scutum on the
first eight segments. The ninth and tenth abdominal segments are smaller and
considerably modified.
Setae are arranged as follows : Four pairs of tiny hairs on anterior margin of pro-
thoracic praescutal area; a row of longer hairs on the scutal area of the prothorax;
each segment from the mesothorax back with two pairs of long lateral setae, near
which are located smaller and inconspicuous setae; each praescutum with a single
pair of setae; on each segment of the abdomen the lateral lobe of the scutellum is
provided with a tiny seta, behind and close to the spiracle; each epipleural lobe
below the spiracle with two setse; each coxal lobe with several setas which are more
conspicuous on the prothorax.
Pupa (PI. 33, A-D). — Length 4 mm., white. This is a normal, characteristic
ciu-culionid pupa with geniculate antennae and legs turned posteriad. Head oval,
beak short. The head bears four pairs of basal setigerous tubercles, two pairs of
interocular tubercles, and one pair of tiny setae at base of beak. Prothorax with
two pairs of antero-marginal setigerous tubercles; one pair of antero-lateral and one
pair of postero-lateral setigerous tubercles; four pairs of dorsal tubercles and one
pair of ventral. The femora are apically armed with two setae. Mesonotum and
metanotum each provided with two pairs of setae. Each abdominal segment bears
four dorsal and one or more lateral setae. The ninth segment is armed with two very
large processes. The tenth segment is very small and located on the venter of the
ninth.
Mar. 4, i9is Weevils Affecting Potatoes 6ii
A WEEVIL WHICH ATTACKS THE TUBERS OF YAMS
Palaeopus dioscoreae, n. sp. (PI. 32, E, F)
Described from two specimens reared from tubers of Dioscorea batatas,
Hope, Kingston, Jamaica, by Mr. S. F. Ashby in April, 191 4. Belongs
in the same subfamily as Euscepes.
Length 4.5 mm., breadth 1.75 mm. Piceous black, with reddish brown append-
ages. It is sparsely clad with dark brown or whitish oblong decumbent scales
and erect, longer truncate scales of variable color. The punctuation is very coarse.
Head smooth, beak separated from head by strong transverse constriction. Beak
longitudinally five-striate and bristling in basal half with erect brown scales; apical
half smoother, with confluent pimctures and flat scales. Scrobes beginning beyond
middle, diagonal, reaching eyes beneath. Eyes lateral, separated by width of beak,
covered when at rest by pronotal lobes. Antennal funicle seven-jointed, first joint
a little longer than second. Pro thorax broad, depressed, convexly rounded on
sides, bisinuate at base, lobed over eyes, trtmcate at apex which is about half as
wide as base; median line broadly elevated; punctuation very coarse. Elytra 10-
striate, the tenth striae abbreviated; strial ptmctures. large, rounded, well separated,
and setigerous; interspaces not wider than striae, clad with a single row of erect squa-
mose setae of variable color. Scutellum indistinct. Elytra broader at base than
thorax; with distinct humeri.
Sternal canal deep, sharply margined, limited by a cuplike depression of the
mesostemum. Posterior coxae very broadly separated. Metastemum at middle as
long as first abdominal segment behind coxae. Intercoxal piece of first abdominal
segment angulate at middle. Second segment not as long as third and fourth
together. *
Femora dentate and canaliculate beneath. Tibiae curved at base, strongly hooked
at the apex. Tarsal claws simple.
Type. — Cat. No. 21612, United States National Museum.
38324°— 18— 7
PLATE 28
Trypopremnon sanfordi: Adult from Cuzco, Peru
A. — Dorsal view. Actual length 8.025 mm.
B. — Face of same. Actual length of head and beak 3.5 mm.
C. — Side view of thorax and head.
D. — Ventral view of adult.
Drawn by Mr. H. B. Bradford.
(612)
Weevils Affecting Potato and Yam
PLATE 28
Journal of Agricultural Research
Vol. XII, No. 9
Weevils Affecting Potato and Yam
Plate 29
Journal of Agricultural Research
Vol. XII, No. 9
PLATE 29
Trypopremnon latithorax: Larva from La Paz, Bplivia
A. — Prothoracic spiracle.
B. — Larva, lateral view.
C. — Lateral view of head.
D. — Right side view of apex of labium.
E. — Corresponding hair on left side.
F. — Maxillary palpiger and palpus, lateral view.
G. — Face.
Drawn by the author.
PLATE 30
Trypbpremnon laiiihorax: Pupa from La Paz, Bolivia
A. — Dorsal view.
B. — Ventral view.
C. — Enlarged sketch of eighth, ninth, and tenth abdominal segments.
Drawn by Mr. H. B. Bradford.
Weevils Affecting Potato and Yam
Plate 30
Journal of Agricultural Research
Vol. XII, No. 9
Weevils Affecting Potato and Yam
Plate 3 1
Journal of Agricultural Research
Vol. XII, No. 9
PLATE 31
Species of the genus Cylas:
' A. — Cylas formicarius elegantulus from Honolulu, Hawaii, side view of head and
thorax.
B. — Cylas iurcipennis from Sumatra, side view of head and thorax.
C. — Cylas brunnetis from East Africa, dorsal view of thorax.
D. — Cylas brunneus, side view of head and thorax.
E. — Cylas brunneus, ventral view of thorax.
F. — Cylas femoralis , side view of head and thorax.
The abbreviations used on this plate are as follows: pr, Presegmental ring; sc,
scutum; si, scutellum; psl, postsegmental ring; pi, pleurum; in, trochantin; c, coxa;
bs, basistemite; stl, stemellum; cstl, centrostemellum.
Drawn by the author.
PLATE 32
Sweet-potato and yam weevils:
A. — Cylas formicarius eleganfulus, female, from sweet potatoes, New Orleans, La.
B. — Same, head of male.
C. — Euscepes batatae, from sweet potatoes, Hawaii.
D. — Same, side view of head.
E. — Palaeopus dioscoreae, from yams {Dioscorea batatas), Jamaica.
F. — Same, side view of head.
Drawn by Mr. H. B. Bradford.
Weevils Affecting Potato and Yam
Plate 32
Journal of Agricultural Research
Vo'. XII, No. 9
Weevils Affecting Potato and Yam
Plate 33
Journal of Agricultural Researcn
Vol. XII, No.y
PLATE 33
Pupae of sweet-potato weevils :
A. — Euscepes batatae, Barbados, venter (length 4 mm.).
B. — Same, latere- ventral view of fifth to tenth segments.
C. — Same, dorsal view.
D. — Same, venter of seventh to tenth segments (length of this portion i mm.).
E. — Cylas formicarius elegantulus , Victoria, Texas, ventral view of sixth to tenth
segments (length of this portion i mm.).
F. — Same, ventral view (length 6 mm.).
G. — Same, latero- ventral view.
H. — Same, dorsal view.
Drawn by Mr. H. B. Bradford.
PLATE 34
Larvae of sweet-potato weevils :
A. — Cylas formicaritis elegantulus, Victoria, Texas, lateral view.
B. — Same, dorsum of head.
C. — Same, face.
D. — Same, side of head.
E. — Euscepes batatae, Barbados, dorsum of head.
F. — Same, face.
G. — Same, side of head.
H. — Same, lateral view of larva.
Drawn by Mr. H. B. Bradford.
Weevils Affecting Potato and Yam
PLATE 34
Journal of Agricultural Research
Vol. XII, No. 9
^
vtmBKammnammum
Vol. XII IVIARCM 11, 191S No. lO
JOURNAL OF
AGRICULTURAL
RESEARCH
CONXENXS
Page
Sterility in the Strawberry - - - - - -613
W. D. VALLEAU
( Contrlbutigo bom Minnesota Agrictilttiral Exj>etimeat Station )
Effect of Nitrifying Bacteria on the Solubility of Trical-
cium Phosphate - - - - - - -671
W. P. KELLEY
(Contiibation from California Agricultura) Bzpezbneol: Station)
PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE,
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICLXTURAL COLLEGES AND EXPERIMENT STATIONS
WASHINOXON, D. C.
WAtHIKOTON t OOVERMMENT PRINTINO Omce :1916
EDITORUL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
KARL F. KEI/LERMAN, Chairman
Physiologist and Associate Chief , Bureau
of Plant Industry
EDWIN W.AXtEN
Chief, Ojfflce of Experiment Stations
CHARLES L. MARLATT
Entomologist and Assistant Chief, Bureau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agrictdtural ExperinunI
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
the University of Minnesota
All correspondence regarding articles from the Department of Agriculture should be
addressed to Karl F. Kellerman, JoiU"nal of Agricultural Research, Washington, D. C.
*Dr. Pearl has undertaken special work in connection with the war emei^ency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
JOINAL OF AGRiaiLTIIAL RESEARCH
Vol. XII Washington, D. C, March ii, 1918 No. 10
STERILITY IN THE STRAWBERPvY ^
By W. D. Valleau
Research Assistant in Fruit-Breeding Investigations, Agricultural Experiment Station
of the University of Minnesota
INTRODUCTION
This paper is a report of studies on the sex condition in strawberries
(Fragaria spp.) which have been carried on during the past four years.
The study of pistil steriHty and anther abortion in the cultivated varieties
and wild species, which are the result of the strong tendency of this
genus toward dieciousness, has received considerable attention; but the
primary object of the investigation has been to find, if possible, some
satisfactory explanation for the phenomenon of pollen abortion, which
is so prevalent among heterozygous plants or plants of hybrid origin.^
MORPHOLOGY OF THE FLOWER PARTS
INFLORESCENCE
The inflorescence of our cultivated strawberry and of Fragaria vir-
giniana, which it closjely resembles, is a dichasial cyme or sometimes,
especially in certain cultivated varieties, a pleiochasium. The two
lateral branches of the relatively main axis are not always equal either
in size — that is, number of flowers borne — or in time of flowering. The
pedicel of the primary flower is generally inserted a short distance from
the joint of the two secondary branches and on the smaller of the two.
The primary flower of the largest lateral branch usually opens directly
after the primary flower and before that of the smaller lateral branch.
The arrangement of the flowers and order of blooming are shown in
figure I.
Variations from these types are not uncommon. In some cases the
primary flower is lacking ; in others, the primary stalk seems to be made
1 Published, with the approval of the Director, as Paper 94 of the Journal Series of the Minnesota Agri-
cultural Experiment Station.
2 The work was begun at the suggestion of Dr. M. J. Dorsey, of the Minnesota Agricultural Experiment
Station , and I wish to express my thanks for the help and encouragement which he gave diuing the progress
of the work. I also wish to express my appreciation of the assistance given by Dr. C. E. Allen, of the
Department of Botany, University of Wisconsin, in a portion of the cytological studies and for space
kindly furnished me in his laboratory during a month in 1915.
Journal of Agricultural Research. ' Vol. XII, No. 10
Washington, D. C. Mar. 11, 1918
mj Key No. Minn, aj
(613)
6i4
Journal of Agricultural Research
Vol. XII, No. lo
up of two which separate at varying distances from the ground, thus
producing two typical cymes from one main stalk. The peduncle and
pedicel lengths vary greatly within a variety, but there is apparently a
rather constant negative correlation between the two — that is, as the
peduncle or primary stalk decreases in length the pedicels or secondary
branches increase, resulting in a rather constant ratio between fruit
stalk length and leaf petiole length or height of plant.
Opposite the point of insertion of the small secondary branch is
usually a large bract. This may be and usually is in the wild forms
a monophyllous leaf, while in manv
m 01 the cultivated vaneties it may
be a well-developed di- or tri-phyl-
lous leaf. The bracts or bracteoles
subtending the branches of lesser
degree are usually only rudimentary
structures, being made up of the
stipules often much reduced, but
with an occasional slight broaden-
ing of the midvein to form a small
leaf blade.
The flowers are hypogynous, and
typically pentamerous with regard
to all parts except the carpels.
The perianth consists of three
whorls of members, the outer five
epicalyx lobes alternating with the
five sepals and opposite to the five
white petals.
STAMENS
Stamen arrangement. — The
stamens in typical flowers are ar-
ranged in multiples of five. The
number is not constant in pure
species or within a variety. The
stamens are arranged in three
whorls. The outer consists of lo
parapetalous stamens located at
either side of the base of the petals (fig. 2, a, PP). These have long
filaments. Their number is the most constant of any of the whorls. The
second whorl consists of five antipetalous stamens located opposite the
petals and inside of the parapetalous whorl (fig. 2,a,AP). The filaments
are shorter than those of the outer and inner whorls. The third whorl
consists of five antisepalous stamens inserted opposite the point of inser-
tion of the sepals, and inside of the two other whorls (fig. 2, a, AS).
0 PR/MARY AXIS
P/^IMAfiy FL.
'SECONDARY
«7-
'TE.RTIARY
43
Fjg. 1. — Diagram showing the arrangement of
flowers of the strawberry and the order of blos-
soming. The approximate order of opening is
indicated by the figures.
Mar. II, i«i8 Sterility in the Strawberry 615
Variations in stamen number from the above arrangement, if slight,
are usually due to the addition or loss of one or more stamens from the
antipetalous series. If a definite increase of five takes place, it may be
the result of an increase in one of two places: either the 5 single anti-
petalous stamens may be replaced by 5 pairs to form a whorl of 10 (fig. 2,
b, AP), or the 5 antisepalous stamens may have been replaced by 5 pairs
of parasepalous stamens located at the same points as the whorl of 5
(fig. 2, c, PS). A further increase in the number of antipetalous stamens
may consist in the development of a pair located on either side and
slightly inside of the single antisepalous stamens (fig. 2, d, AP). These
are characterized by the same short filament that is found in the anti-
Fig. 2. — Flower diagrams of Fragaria spp., showing stamen arrangement: A.S represents antisepalous;
AP, antipetalous; PP, parapetalous; and PS, parasepalous stamens, a represents the 20-stamen
arrangement found commonly in F. vhginiana and F. americana and many cultivated varieties; 6 and c,
a 25-stamen arrangement found in F. virginiana and some cultivated varieties; d, a 35-stamen arrange-
ment sometimes found in cultivated varieties; while e and/represent a 10 and 15 stamen arrangement
found in some clones of F. ainericana,
petalous whorl of 5. This increase, plus the 20-stamen arrangement,
gives a 30-stamen arrangement, or with either of the two 25-stamen
arrangements, gives 35.
Rydberg {34, p. loY has pointed out that the antipetalous stamens which
appear to be the middle whorl in Fragaria spp. are truly an itmer whorl
which has been pushed out to form apparently a middle one (fig. 2,d), and
that the outer parapetalous stamens are in reality younger with regard to
development than the antisepalous or inner series. A study of the
position of the accessory stamens of the antipetalous series (which can
readily be distinguished by their short filament) and of the order of
stamen development indicates that this view is correct.
1 Reference is made by number (italic) to "Literature cited," p. 666-669.
6 16 Journal of Agricultural Research voi. xii. No. lo
In F. americana, lo, 15, and 20-stamen arrangements are common.
The 20-stamen arrangement is the same as that described for F. vir-
giniana and the cultivated varieties. When, however, a decrease
below 20 to 15 takes place, it is due to the loss of the short filamented
middle whorl of antipetalous stamens (fig. 2, /) — further proof that
this is truly an inner and not a middle series. The next series to be
lost is that consisting of the inner long filamented antisepalous stamens,
thus leaving the parapetalous stamen arrangement (fig. 2,e). This seems
peculiar, in view of the fact that they are older than the parapetalous
stamens and therefore should remain longest. This might be consid-
ered as proof that —
the parapetalous stamens must be regarded as abnormal supernumary parts,
as Rydberg {34, p. 11) considers them. The genus Fragaria differs
from other species of the Potentilleae in this respect, as the more usual
order of loss is first, the parapetalous, followed by the antipetalous
stamens {34, p. 11), the long filamented antisepalous stamens being
the most permanent. When stamens are dropped in Fragaria spp.,
they are lost completely and do not form the staminoids or partially
developed stamens which are found in the pistillate flowers, so that a
decrease in stamen number can not be considered as a step toward
dieciousness.
Staminodia. — Typically F. virginiana and, as will be shown later,
some other species of strawberry are diecious, although the separation
into strictly staminate and pistillate forms is not complete. The flowers
of pistillate plants bear staminodia showing varying degrees of develop-
ment (fig. 3, 11-16), which never, as far as I have observed, produce
pollen.^ The staminate plants develop pistils which, as far as outward
appearances are concerned, are normal, but which do not set fruit. As
a result of this inconiplete separation of the sex-bearing organs, there
are variations in the stamen and pistil condition on individual clones
and also on the flowers of an infloresence within a clone.
The most common condition in the staminate plants is that in which
all of the flowers produce normal stamens bearing good pollen. Occa-
sionally clones are found in which the first flower bears only staminodia
in place of the normal stamens. In flowers of this type the pistils are
practically always fertile and produce normal fruits. On the other
hand, the primary flower may produce normal stamens and no fertile
pistils while one or both of the secondary flowers may be of the pistillate
1 A possible case of pollen production in a pistillate variety is that of the Crescent. Plants of it which
I have observed 'are strictly pistillate, although producing very large prominent staminoids (fig. 3, 16)-
which are entirely devoid of reproductive tissue (PI. 36, A). Castle (7 ,p. 150) states that in England "it
produces perfect flowers and sets its fruit most readily, cropping heavily in favorable seasons." As most
other English varieties are hermaphrodites, it is very possible that large crops might set as a result of cross-
pollination, and that the presence of the extremely large staminoids has been misleading with regard to
the exact sexual condition of these plants.
Fletcher (75, p. 132) also states that Crescent may vary in stamen condition becoming "a true stami,
nate on rich soils," but gives no ftuther evidence on the point.
Mar. II, 1918
Sterility in the Strawberry
617
type, in which case they set fruit. A few cases have been found in which
one side of a flower produced normal stamens and sterile pistils while
the other half produced staminodia and set fruit. A similar range of
conditions with regard to stamens has been noticed in seedlings of cer-
tain cultivated varieties. Figures G and H, Plate 35, are from photo-
graphs of primary and secondary flowers of the seedling Minnesota 1017 X
Progressive 32-1, both of which bear only staminoids, while I and J repre-
sent secondary and tertiary flowers of the same variety, I producing
both staminodia and normal anthers and J producing only normal
anthers. The production of pistillate flowers on the primary and on
Fig. 3. — Outline camera-lucida drawings of perfect and intermediate anthers and staminodia of straw-
berry: I, 2, and J, Normal anthers from tertiary flowers of a seedling of Minnesota 1017 X Progressive,
Progressive, and another seedlingof Minnesota 1017 X Progressive (40-1), respectively. 4, Staminodiuin
from a primary flower of Minnesota loi 7 X Progressive (40-1 ). 5 and 6, Normal anthers from Miimesota,
3 and a late primary flower of Minnesota loi 7, respectively. 7, Staminodia from a primary flower of Minne-
sota 1017 produced early in the spring. 8 and p, Intermediate anthers from primary flowers of Minnesota
3. JO, Intermediate anther from a primary flower of Progressive. 11. Staminodia from a pistillate flower
cf F.virginiana. J 2. 13, and 1$, Staminodia from pistillate flowers of seedlings of Minnesota 1017 X Pro-
gressive, 2-25, 13-40, and i:-59, respectively. 14, A staminodimn from a flower of Haverland.a pistil-
late variety. 16, Staminodia from a flower of Crescent, a pistillate variety which produces extremely large
and prominent abortive anthers.
some of the secondary flowers throughout the season seems to be the nor-
mal condition in a number of seedlings of the cross 1017 Minnesota X Pro-
gressive (fig. 3, 3, 4), while among the commercial varieties there are a
number which produce pistillate primary flowers early in the flowering sea-
son, while those produced later are all perfect. A few varieties which
show this peculiar condition early in the spring, are Brandywine, Minne-
sota 3, Bederwood, Tennessee Prolific, and Staples. In figure 3, 6
represents an outline drawing of a normal anther from a primary flower
of Minnesota 1017, produced late in the season, while 7 represents
staminoids of the same variety taken from flowers produced early in the
flowering season.
5i8 Journal of Agricultural Research voi. xii.no. lo
Although typically there are two rather distinct types with regard to
stamen development in both cultivated and wild clones of the straw-
berry— namely, the perfect stamens bearing normal pollen and the
staminodia of the pistillate varieties — there are apparently a series con-
necting these two conditions.
Figures ii, 12, 13, 14, 15, and 16 of text figure 3 show the range of de-
velopment in size of the staminodia on some pistillate plants, and i, 2, j,
5, and 6 show the relative size of normal anthers. In actual size the two
types closely approach one another. With regard to development of re-
productive tissue, there is considerable difference. The normal stamens
naturally carry pollen development through to completion. The stami-
nodia, on the other hand, never produce normal pollen, but show some
variations in the extent to which development is carried. Figures A,
B, C, and D, Plate 36, are photographs of cross sections of staminodia,
figures A and B being cross sections of two loculi of staminodia of Crescent
and Columbia, respectively, in both of which varieties the staminodia are
rather prominent. Plate 36, C, is from the seedling Minnesota 1017 X
Progressive, 1 1-59, which produces very large and prominent staminodia.
Here there are distinct evidences of early degeneration of the repro-
ductive tissue, probably pollen mother cells, although the early stages of
this variety have not been studied. Figure D (PI. 36) is from a stami-
nodium of a pistillate variety which produces extremely small stami-
nodia. There is no evidence of any reproductive tissue whatever having
been produced.
Janczewski {24) has studied the stamen condition in some of the
diecious species of Ribes and finds that in the pistillate flowers small sta-
mens develop. Their internal development soon ceases and abortion of the
reproductive tissue takes place. He considers that the small dark stain-
ing mass which he found in the staminodia was made up of the decom-
posed pollen mother cells. Often the cavity left by the breaking down
of the pollen mother cells was filled with parenchyma which had grown
in from the wails. Gates (17) found the same condition in some anthers
of Oenothera lata. Apparently the parenchymatous tissue filling the
staminodia of Crescent and Columbia is not of this origin, as early stages
show no signs of the formation of pollen mother cells.
Intermediate anthers. — In studying the anther types of F. virginiana
certain clones were found which on first examination appeared to be
producing normal stamens, but on closer examination were found to
contain either a dark staining disintegrated mass or completely aborted
microspores, the walls of which, in some cases, were disintegrating.
These are apparently intermediate types between the staminodia and
normal anthers. Similar types of anthers are" not infrequently found
in the primary flowers of many wild staminate clones.
A study of the intermediate stamens of F. -virginiana indicates that
pollen development is generally carried to the homceotypic division
Mar. II, 1918 Sterility in the Strawberry 619
or to the formation of the tetrads when degeneration occurs. This be-
comes apparent first through degeneration of the mother-cell wall and
the cytoplasm, if the homoeotypic division is taking place, leaving the
spindles and chromosomes standing out sharply in this degenerate mass;
or if the tetrads have already been formed, the material in which the
microspores are embedded disintegrates and is follov/ed directly by the
disintegration of the microspores. Plate 36, E, shows degenerating
tetrads; F shows a later stage of the same thing in which the micro-
spores have completely degenerated; and G shows the condition found
in mature anthers of this type.
Occasionally development may proceed to the formation and libera-
tion of the microspore when, following a slight thickening of the walls,
degeneration of the contents and disintegration of the microspore walls
takes place. The same type of degeneration is found here as where
earlier abortion takes place. The walls become thickened and, as degen-
eration proceeds, show a beaded appearance and finally break up into
drops of a yellow oily appearing substance which makes up the mass
shown in figure G (PI. 36) . In other clones of F. virginiana development
proceeds to the liberation of the microspores from the tetrad when, fol-
lowing a slight development of the microspore wall, degeneration of the
cell contents takes place, leaving aborted pollen of the type so charac-
teristic of hybrids.
In the cultivated hermaphroditic varieties which produce staminodia
on the early primary flowers (Pi. 36, D, G), and on some other varieties,
such as Lovett, Glen Mary, and Minnesota 1017 X Progressive 9-24 (PI. 36,
B, E), these same types of anthers characterized by being small, shrunken,
and bleached yellow or deep ocher are common. They show both types
of degeneration — i. e., complete disintegration of the anther contents
and abortion of the microspore contents following their liberation from
the tetrad. Figure H, Plate 36, shows a section of a whitish yellow
anther of the type shown in figure 3, 8 and 9, from a primary flower of
Minnesota 3, a variety which for the most part produces normal stamens.
Jeffrey and others have recently given emphasis to the relationship
between aborted pollen and hybridity and have attempted to correlate
any considerable amount of pollen abortion with a hybrid condition of
the plant. Apparently in the strawberry the above type of pollen
sterility and the tendency toward dieciousness are very closely related.
As all degrees of stamen development may be found on a single culti-
vated variety of the strawberry, and on some wild plants also, from the
small staminodia to well-developed stamens bearing normal pollen, it
seems safe to conclude that these intermediate stamen types bearing 100
per cent aborted pollen and found in apparently pure F. virginiana are
not the result of hybridization but are' really the expression of various
degrees of dieciousness.
Whether the clones of F. virginiana bearing these intermediate anther
types are able to develop fruit, thus indicating whether they have been
620
Journal of Agricultural Research
Vol. XII, No. lo
derived from the pistillate forms continuing pollen development in the
staminoids beyond the usual time or whether they are staminate forms
in which pollen development is inhibited has not been determined ex-
perimentally. However, they appear to be of the latter type as they
have been seen in flower a number of times and have shown no signs of
setting fruit.
RELATION OF FLOWER
PART NUMBER TO
SIZE OF FRUIT
It is generally rec-
ognized by growers
that toward the end
of and in fact during
the whole progression
of the picking season
of strawberries, there
is a progressive de-
crease in the size of
berries produced, but
the relationship be-
tween this decrease
and the position of
the flowers on the in-
florescence producing
these smaller berries
has not been so gen-
erally recognized. As
has already been men-
tioned, strawberry
flowers are typically
pentamerous, but un-
der cultivation there
has been an increase
in the number of
parts in a portion of
the flowers. This in-
Fig. 4.— Graphs showing the relation between sepal nmnber and flower cj-g^sg jg mOSt Strfk-
position in the seedling varieties No. 373. 968, and 1006. The sepal ... _.
number is indicated on the abscissas and the frequencies on the ing lU the primary and
ordinates. sccoudary flowcrs and
is only apparent to a very slight degree in the later ones. An increase in
calyx-lobe number is practically always accompanied by an increase in both
petal and epicalyx lobe number and as the stamens are arranged with regard
to petal position there is necessarily an increase in stamen number also.
Figure 4 shows the relationship between flower position and sepal num-
ber in three seedling strawberry varieties. These show a condition typical
Mar. II, 1918
Sterility in the Strawberry
621
of our cultivated varieties. It will be seen that there is a very direct
relationship between flower position and number of flower parts. In
these same varieties the relationship between calyx-lobe number and
size of berries was studied. The results are shown in Table I. These
results indicate that there is a high correlation between flower-part number
and fruit size, and as a definite relationship has been pointed out between
flower position and flower-part number, it follows that the larger fruits
will be developed on the early blooming primary flowers and that, as the
season progresses, there will be a decrease in fruit size due to their being
borne on later-blooming flowers of a higher order.
Table I. — Relation of fruit size to calyx-lobe number in strawberries
Popu-
lation.
Calyx-Iobe number.
Diameter of berry.
Variety.
Range.
Mean.
Stand-
ard
devia-
tion.
Kange.
Mean.
stand-
ard
devia-
tion.
Coefficient of
correlation.
Seedling 1006. . . .
Seedling 373----
Seedling 968
274
288
275
4-8
5-9
5-9
6-350
6.827
6-313
0.925
I. 040
•956
Cm.
I. 4-3- 8
I- 4-3- 5
I. 7-3. 8
2.326
2.403
2.427
0-343
•532
.4361
0. 482 ±0. 031
• 667 ± . 022
•5i9± -03
Even a casual observation of normal large and small berries of any
variety will indicate that there must be a relationship between berry size
and carpel number. Table II shows the direct relationship which exists
between fruit position, size of fruit and achene or carpel number in all of
the fruits produced on the inflorescences studied. The fact that there is
such a definite relationship between size of berry and flower position
should be constantly kept in mind in the selection of breeding stock.
This is apparently the point that Mr. Hubach, a southern strawberry
breeder, has in mind in selecting for stalks which bear only one fruit
per inflorescence {see Darrow, 9).
Table II. — Relationship between fruit position, number of achenes, and size of fruit
in the strawberry
Pr
imary.
Secondary.
Tertiary.
Quaternary.
A-n-
<n
u
fc
u
in
u
<Kv
J3
J3
J3
X
Variety.
^ 3
0
Eg
ci
0
is
ng
V
0
ci
0
0
I/)
J3
w
^
Vi
t/i
<UU
vlj
h
DO
u 0
ti ™
M
s
•^ C
""S
a
«-,
a
•O
«_
a
XI
ts_
ts
C 0
s
S°
S
a
fco
to
H
ijo
i
S
fco
Ui
>
>
3
>
>
3
>
>
>
>
iz t:
z
<
<;
•z
<
<
2
<
<
Z
<
<
A/w.
Mm.
Mm.
Mm.
Minnesota 3
6
7
3^2. 28
28.0
14
224. 27
17-7
21
150. 9
9.8
7
92.42
7
Wiidwood
7
4
2 29- 70
23-8
17
142.7
19.0
20
88.15
II. 7
3
72-7
7
F . virginiana 9
I
I
112. 00
15.0
2
116. 5
13- 5
4
77.0
8-3
I
70. 0
6
622 Journal of Agricultural Research voi. xii, no. io
PISTILS
MoRPHOivOGY. — The carpels bearing a single orthotropous ovule are
arranged in a spiral on the fleshy cone-shaped receptacle. At maturity
they form dry achenes either set on the surface of the receptacle, as in
F. aynericana, or in shallow or deep pits, as in F. virginiana. The style
is inserted laterally on the inner side of the carpel and extends well above
the upper portion of the ovary. The pistil number is not constant on
the flowers of an inflorescence but is directly dependent upon the posi-
tion of the flower.
Pistil sterility and dieciousness. — As has been previously men-
tioned, F. virginiana is, in the wild, typically diecious, the stamens having
been reduced to staminodia in the pistillate plants and the pistils, although
present and apparently normal, as far as can be seen superficially, in the
staminate flowers are nearly always functionless. This condition of
dieciousness has apparently remained unrecognized by systematists of
Fragaria. Apparently dieciousness is not confined to F. virginiana alone,
but is typical of most of the American species, except F. americana,
which is hermaphroditic, as is also the European species F. vesca.
As early as 1760 dieciousness was recognized in F. elatior by Duchesne
{see Fletcher, j(5), who showed that the apparent sterility of the Hautboy
was in reality due to the weeding out of the unproductive male plants.
He also recognized partial separation of F. chiloensis into male and
female plants. A study of herbarium material of some of the American
species of Fragaria indicates that F. chiloensis from Alaska to Bolivia,
F. cuneijolia on Vancouver Island and in Washington and F. platypetala
from the north moraine of Sir Sandford Glacier are all diecious, at least
some plants produced pistillate flowers bearing staminodia, while others
bore flowers with well-developed stamens and apparently normal pistils
which showed no signs of setting fruit. On two plants of F. chiloensis
from Lake Merced, California, the primary flowers of staminate inflores-
cences were found to have set fruit, while the remaining flowers, although
well beyond the fruiting stage, showed no signs of setting. Supposedly
hermaphroditic plants of F. chiloensis from Alaska, grown at University
Farm for a number of years, produced few, if any, fruits, although they
blossomed profusely.
Georgeson (19, p. ij), in speaking of F. chiloensis, which he used in his
hybridization experiments, says :
There is a decided variation among the plants; some are much more productive
than others, and some appear to bear only staminate flowers, though, as a rule, the
flowers are perfect.
and again (20, p. 11) :
The flowers are large and white and many of them staminate and sterile.
The first plants of F. chiloensis brought to Europe by Frezier were all
pistillate and had probably been selected by him because of their fruiting
propensities.
Mar. II, 1918 Sterility in the Strawberry 623
Richardson (jj, p. 1-6) mentioned receiving male plants of F. virgin-
iana from America. I have grown plants of F. virginiana illinoensis
from near Dresden, Ohio, which proved to be strictly diecious. The
literature on strawberry growing in this country during the early part of
the last century makes many references to the necessity of growing
staminate varieties for the purpose of fertilizing the pistillate forms and
to the fact that these plants were unproductive of fruit, but very pro-
ductive of plants and would soon dominate the garden if attention was
not paid to them. Apparently they were using staminate F. virginiana
plants as pollen producers.
A study of F. virginiana in various sections of Minnesota shows that for
the most part this species is diecious, although some few hermaphroditic
plants may be found. Of a total of 1,615 pistillate flowers of this species
borne on 304 plants located in four distinct regions of the State, 1,180
set fruit, while 393 were still in the bud or blossom stage, thus indicating
that practically all of the flov/ers of the pistillate forms are fertile if
pollinated. One pistillate clone, composed of 11 plants was found,
however, which produced a total of 57 flowers, 18 of which were still in
the bud or flower stage; of the remaining 39 only one set seed. Another
clone of this same type was found in a different region. Material of both
these has been saved to test further their fertility. Plants of both of
these clones, when grown under conditions more favorable to pollination,
proved fertile.
In contrast to the striking fertility of most of the pistillate forms is
the condition in the apparently hermaphroditic plants. Of 1,640
flowers of this type borne on 381 fruiting stalks of separate plants, 403
were in the flower or bud stage, while only 152 of the remaining 1,237
set fruit, leaving a total of 1,085 flowers which were definitely sterile;
286 plants of the 381 studied bore no fruit.
The position on the inflorescence of the flowers which set is interesting
in connection with the problem of nubbins and pistil sterility in our
cultivated varieties. The fact has previously been mentioned that a
few flowers borne on male plants may bear only staminodia in place of
stamens and that these are generally fertile. Of the 152 fruits which
set, 31 were developed from this type of flower. The other flowers on
these stalks w^ere of the usual staminate form and were generally sterile,
although a few instances were noted in which one of the flowers bearing
normal stamens set a few achenes. Of these 31 fruits, 17 were borne on
primary, 10 on secondary, 3 on tertiary, and one on a quinary flower.
Five of the 152 which set were borne on flowers bearing the intermediate
type of anther and of these 2 were primary and 3 secondary. The other
flowers of these inflorescences bore normal anthers and were sterile.
Fifteen resulted from flowers which produced some staminodia and
anthers either distinctly segregated in definite portions of the flower or
mixed indiscriminately; of these, 12 were from primary and 3 from
624 Journal of Agricultural Research voi. xii, no. i©
secondary flowers. The remaining loi fruits were borne on flowers
bearing a full quota of normal anthers. Of these flowers 54 were primary,
40 secondary, and 7 tertiary. While most of the fruits which set on the
pistillate plants were of a regular shape, indicating a perfect or nearly
perfect set of achenes, those borne on the staminates were for the most
part very irregular in shape, as the achenes which set were often few in
number and irregularly scattered. Often not more than one or two
achenes per flower developed. Where only a few achenes developed,
the typical nubbins which are so common in the latter part of the
picking season in commercial plantings were produced. These results
prove that F. virginiana is a species well on its way toward diecious-
ness, and, reasoning from analogy with F. virginiana and F. elatior,
it may be concluded that those other American species which produce
two types of plants — that is, pistillate and somatic hermaphrodites — are
also diecious.
Recent investigations by Bunyard (6) and Fletcher (74) into the origin
of our cultivated strawberries tend to show that they have originated
from hybrids of F. virginiana and F. chiloensis, both of which are ap-
parently diecious. If this is the case, it raises the question of the origin
of our cultivated hermaphroditic forms. A study of the pistil ste-
rility in these forms seems to indicate that they may have been derived
from males which have varied in regard to pistil fertility.
Table III shows the relationship between flower position and pistil
sterility in 10 hermaphroditic and 4 pistillate varieties. This table was
prepared regardless of the degree of setting, whether perfect or whether
the resulting fruit was a nubbin, all flowers which set any achenes being
put under the heading "Set." The lower horizontal row under each of
the two groups indicates the percentage of the flowers of each position
which set fruit. It shows the very great increase in sterility from the
first flowers to the last in both the pistillate and hermaphroditic forms,
being greater in the latter than the former. This is the condition which
would be expected if the cultivated hermaphrodites have been derived
from males of the wild type, as the males which do set fruit in the wild
exhibit a high percentage of their low fertility in the primary flowers.
The conclusion that the hermaphrodites have been derived from stam-
inate forms rather than from pistillate is in keeping with the results
found in other species, as Lychnis spp. (55) and the grape {43). A further
study of pistil sterility was made in 1 5 other varieties of hermaphrodites
and 3 pistillates to determine the relationship between nubbins or irregu-
larly set fruit and flower position.
Nubbins and lack of setting of flowers have been attributed by horti-
culturists for the most part to lack of proper pollination or to frost injury.
The first of these factors may be eliminated, however, as pollen is usually
very plentiful and in a mixed planting, such as the data given in Tables
in and IV were taken from, was always abundant and especially so when
Mar. II, 1918
Sterility in the Strawberry
625
the later more sterile flowers were in blossom. Further, the fact that
pistillate varieties, grown in proximity to hermaphrodites, set fruit even
in the early part of the season, when pollen is admittedly scarce, would
argue for pollination having little to do with nubbin formation. To
those who have noticed the effect of frost on strawberry flowers it will
be clear that this factor may also be eliminated as a cause of irregularly
set fruit, as frost, if it injures the flower at all, will blacken the entire
receptacle. The possibilities of the primary flowers being "frostbitten"
are much greater than the later ones, but it is the latter which generally
form nubbins or are entirely sterile.
Table III. — Relationship between flower position and number of fruits set in hermaphro-
ditic and pistillate varieties of strawberries
Variety.
Ses.
Num-
ber of
stalks
Primary.
Set.
Not
set.
Secondary.
Set.
Not
set.
Tertiary.
Set.
Not
set.
Quater-
nary.
Set.
Not
set.
Quinary.
Set.
Not
set.
Reasoners 324.
Seedling 947...
Orem
Lovett
Seedling 893.. .
Seedling 1023. .
Abington
Everbearing. .
Glen Mary . . . .
Seedling 924. . .
Total . . .
Per cent .
Paul Jones .
Marie
July
Wildwood .
Total . . .
Per cent .
66
III
79
72
83
77
90
6j
74
87
55
68
77
S8
90
82
107
6
1-3
68 s
65.8
356
34-2
IS5
65-7
5
83.3
48
286
90.8
177
66.5
80
47 I
6
100
Table IV shows the relation between flower position, imperfectly
developed fruit or nubbins, and complete pistil sterility. It corroborates
what has already been pointed out, namely, that the first flowers of an
inflorescence are much more fertile than the later ones. With regard
to nubbins the same relationship is shown — that is, there is a gradual
increase in the percentage of nubbins formed from the primary to the
last flowers which open. This condition can hardly be construed as
indicating anything but a morphological sterility of a portion of the
pistils in those flowers which result in nubbins, if viewed with the facts
in mind of the condition shown in the hermaphrodites of the wild parent
species, the unquestionable sterility of many of the later flowers, and the
fact that the greater percentage of these partially sterile flowers are in
bloom when pollen is most abundant. If it were a question of pollina-
tion, we would expect the pistillate forms to exhibit much more sterility
than the hermaphrodites, whereas they exhibit decidedly less, both with
regard to the actual number of sterile flowers as well as nubbins.
626
Journal of Agricultural Research
Vol. XII, No. lo
Table IV. — Relationship between flower position and the degree of setting in hermaphro-
ditic and pistillate varieties of strawberries
i
to
0
Primary.
Secondary.
Tertiary.
Quaternary.
Quinary.
Variety.
1
>-•
2;
01
0
a
IS
3
2;
1
0
a
3
t
0
1
■§
-4-1
•4-t
pi
a
3
Dorman
20
20
20
20
24
20
20
20
20
20
20
20
20
20
20
17
17
21
15
36
19
21
18
22
37
30
26
27
25
24
2
2
0
5
18
I
I
5
1
I
0
0
5
53
42
84
36
27
60
45
47
64
73
68
39
73
64
67
13
II
7
16
40
12
16
23
2
4
4
9
2
3
"46
3
4
43
74
62
74
3
74
35
33
70
63
95
59
85
72
88
23
23
24
28
0
51
19
47
25
2
6
4
I
17
5
45
I
35
30
12
48
33
6
7
9
7
2
21
8
32
13
28
II
II
IS
14
55
6
39
5
Haverland o
Parson's Beauty
Bederwood
Steven's Late Champion
Helen Davies
31
9
16
26
I
24
II
7
6
29
10
II
16
16
27
44
36
10
Senator Dunlap
Pride of Delaware
Minnesota 3
Seedling 924
Seedling 937
8
I
I
4
I
Seedling 947
I
14
8
7
Seedling 876
Seedling (number lost)
s
2
1
Seedling 778
Total
355
89-4
19
15
20
5-1
35
8.8
0
4
1.8
842
79.1
62
46
56
164
82. 8
164
IS- 4
16
8
10
34
17-2
59
5-5
930
('3-3
95
52
90
237
69-5
284
19-3
44
27
21
92
27
256
17.4
4
2
6
12
3-5
241
39-4
46
24
57
127
62
119
19-5
19
10
10
39
19
251
41. I
23
10
6
39
19
Per cent
Enormous
9
9
9
20
20
20
==;;
Warfield
Crescent
2
2
40
1
I
20
Total
4
6-9
Per cent
" Not certainly true to name.
The most characteristic type of nubbin is that in which all of the
achenes set except those situated at the tip, thus, producing a berry
with a dead, dry tip. This raises the question as to why the tip pistil
should be more sterile than those at the base.
We have already seen that when a reduction in stamen number takes
place it is the youngest which are lost first, and apparently this is true
also of the pistils. The pistils of the oldest flowers are decidedly more
fertile than those of the later flowers, and it seems logical that the older
pistils Avithin a flower should be the more fertile. Observations on a
seedling everbearing variety which in the spring produced only stamens
and a small, white dome destitute of pistils in place of the ordinary
receptacle, but which later in the season gradually produced normal
flowers, illustrates this point. It was noticed that the first flowers which
produced pistils developed only a few normal ones around the base of
the receptacle, while those above gradually decreased in size until at
the tip there were none. The later fruiting stalks increased the number
of normal pistils until in the last fruiting stalks the entire receptacles of
the first flowers were covered with normal pistils. This series of flowers,
although an extreme with regard to sterility, still indicates the portion
of a flower in which there is the greatest likelihood of its appear^ance.
The question of pistil sterility should be kept in mind in the selection
of breeding stock, as it is most certainly inherited in the pure species
and apparently is in the cultivated varieties, as selfed seed of Glenville,
a variety which rarely sets more than an occasional primary flower.
Mar. II, 1918 Sterility in the Strawberry 627
produced a number of seedlings which were as sterile as the parent.
(All the progeny have not yet flowered.)
Sufficient pistillate varieties have not been studied to indicate defi-
nitely whether there is always a distinct difference between the ability
of these and of the hermaphrodites to set fruit on the later flowers;
however, a comparison of the percentage of fruit set on the tertiary
and quarternary flowers of the pistillate and staminate varieties reported
in Tables III and IV would seem to indicate that the former are de-
cidedly more fertile than the latter, as is also the case in the wild forms.
Darrow (9) reports that Mr. Hubach will use only pistillate varieties
as the fem.ale parents because of the decrease in fruit production when
hermaphrodites are used as both parents.
A cytological comparison of the sterile pistils of wild males and the
sterile pistils of hermaphrodites may give further evidence as to the
origin of sterility in the varieties which produce many nubbins and
entirely sterile flowers. Strasburger (jp) has shown that in a male
resulting from a cross between F. mrginiana 9 and F. elatior 3- , the
pistils which are apparently ready for fertilization already, in longi-
section, show a mass of degenerating material which contains the embryo
sac mother cell. This may or may not be the condition in the pure
forms of these species and in the sterile pistils of cultivated varieties.
To the commercial grower of strawberries as well as to the breeder
the question of variation of fertility under varying conditions of environ-
ment or culture is of importance. Evidence which indicates that fer-
tility is affected somewhat by environmental or seasonal conditions is
given by the seedling which in the spring produced strictly male flowers
but which in the summer and early autumn produced fertile pistils as
well as stamens. Further, it is a matter of common observation that a
bed of berries, if allowed to fruit more than one year, will produce an
increasing number of small berries and nubbins. Of actual observational
evidence the following indicates that growth conditions have something
to do with sterility: A variety, named "Glenville" for convenience, was
sent to the Station with inquiries as to why it did not set fruit. Plants
of it were grown in the greenhouse during the late winter, and, although
they produced numerous fruiting stalks ancl an average of 13 flowers
per stalk, only an occasional primary or secondary flower set fruit.
Some of these plants, after having been grown in benches, were put
into pots and given little attention. On June 5 they were fruiting,
and a count was made of the flowers which had set. On 6 fruiting
stalks there was a total of 43 flowers, an average of slightly over 7 per
stalk. Of these, 22 set fruit. Of 6 primaries there was i which set, of
12 secondaries 10, of 19 tertiaries 11, and of 6 quaternaries none set.
The previous 3'car some plants of this variety had been planted in the
field ; and 20 days after taking the above notes, observations were made
on the field plants. Of 105 fruiting stalks examined, bearing a total of
1,292 flowers, an average of 12.3 flowers per stalk, there was a total of
628 Journal of Agricultural Research voi. xii. no. io
20 fruits, of which 15 were borne on primary flowers, 4 on secondaries,
and I on a tertiary. Although this variety shows an extreme case of
sterility, the condition found as regards variability of sterility may be
an indication of what will be found when a thorough study is made of
this point in our cultivated varieties.
Thus far the study of sterility has dealt mainly with those types of
sterility induced by a decided tendency toward dieciousness in species
of Fragaria. Another type of sterility very prevalent in cultivated
varieties and undoubtedly a factor in pollination is expressed in the
appearance in ripe pollen of varying amounts of defective grains. It is
with this type of sterility that the remainder of this paper deals.
POLLEN DEVELOPMENT AND STERILITY
A careful cytological examination of the pollen condition in the straw-
berries, both wild and cultivated, was made with the objects of deter-
mining (i) the amount of viable pollen in cultivated varieties and its
relation to the setting of fruit and (2) the cause of pollen abortion in
plants of hybrid origin.
The material used as a basis in determining the general pollen condi-
tion in Fragaria spp., consisted of (a) F. virginiana from various parts of
Minnesota, (b) F. americana, (c) a considerable number of cultivated varie-
ties, and (d) seedlings under test in the course of the fruit-breeding work.
The cytological study was carried on principally on the self -fertile variety
Minnesota No. 3, a cross of Senator Dunlap X Pocomoke, recently intro-
duced by the Minnesota Agricultural Experiment Station. It produces,
on an average, about 50 per cent of aborted grains and so furnishes
desirable material for the study of normal and abnormal pollen develop-
ment . The stages in normal development we re also studied in F. virginiana.
POLLEN CONDITION IN WILD FORMS
The recent work of Jeffry and his students on the pollen condition in
wild forms puts under suspicion the genetic purity of the Rosaceae in
general. The forms which have been studied most intensively,
Onagraceae (25), Crataegus spp. {37), Rubus spp. (23), and Rosa spp. (8),
show, in some species, a relatively large proportion of aborted pollen
and the appearance of many subspecies, some of which appear to be
hybrids. Because of this fact and with a view to comparing the pollen
condition of the wild with the cultivated forms, pollen of F. virginiana
and F. americana was examined.
The methods used in determining the amount of abortive pollen were
as follows: Fresh flowers were collected and either were allowed to dry
or were kept with their pedicels immersed in water until the anthers
had dehisced. The pollen was then transferred to slides by holding the
flower over a slide and giving it a few sharp taps. In this way the
anthers dehisced completely onto the slide. A drop of lactic acid was
then added, and a small cover slide placed over the drop, forming a
Mar. II, 1918 Sterility in the Strawberry 629
fairly permanent mount if handled carefully. The lactic acid has the
advantage over water or alcoliol for this purpose, as it is not volatile
and seldom, if ever, breaks the pollen grains through osmotic pressure.
It readily enters and expands the normal grains, while it leaves the
aborted ones collapsed. It has another very distinct advantage over
the more mobile liquids, in that its viscosity holds both the normal and
aborted grains in place until the cover has settled firmly, while, if water
is used, the empty grains have a decided tendency to wander toward
the edge of the cover slide, thus invalidating the count. Where no
aborted pollen was present, of course no actual counts were made; but
where present, counts were made varying from 200 to 2,000, the object
in each case being to include enough grains to indicate within a close
range the percentage of abortion. A record was kept of the position
on the inflorescence of the flower from which the pollen was collected
in order to determine whether a similar relationship existed between
flower position and abortion of pollen as was found to exist between
flower position and stamen type in certain clones.
The results of the pollen counts on 223 flowers of F. virginiana are
shown in Table V. They indicate that F. virginiana produces for the
most part morphologically perfect pollen, although a few plants were
found in which the percentage of aborted grains was high. If pollen
condition may be taken as a criterion of species' purity it may be said
that F. virginiana in this region is nearly a pure form. The appearance
of considerable amounts of aborted pollen in a few plants might be
considered as the results of the "conditions" under which these plants
have been grown, but the fact that several flower types have been found
in the wild may indicate that abortion is due to a condition arising
from a slight degree of hybridity consequent on the intercrossing of
these forms. The one primary flower which produces 100 per cent
aborted pollen bore intermediate anthers and abortion was more likely
due to the partial suppression of stamens, in some w^ay connected with
the tendency toward dieciousness, than to other causes which may result
in sterility. This was not included in the average percentage of aborted
pollen in the primary flowers for this reason.
A comparison of the pollen condition in the flowers borne on various
positions shows that pollen i.,bortion is in no way related to flower position
and thus to dieciousness as are the various anther types and sterile pistils
before mentioned.
In the hermaphroditic species F. americana nearly the same condition
exists with regard to the degree of perfection of pollen as in F. virginiana.
The result of pollen counts on 49 flowers taken from an equal number of
plants are shown in Table VI and they indicate that the pollen condition
of F. americana is normal. As it is very difficult to cross F. virginiana
and F. americana, it is probable that where these two species are growing
in close proximity they remain pure.
38325°— 18 2
630
Journal of Agricultural Research
. Vol. XII, No. 10
Table V. — Pollen condition in wild plants of Fragaria virginiana with reference to the
position of ike flower on the inflorescence
Aborted percentage.
Number of
primary
flowers.
Number of
secondary
flowers.
Number of
tertiary
flowers.
Nmnber of
quaternary
flowers.
Number of
flowers
position
not
recorded.
0
9
15
16
6
4
5
I
0
I
3
4
2
I
I
II
18
5
5
3
2
3
4
2
2
I
I
I
5
6
4
I
I
14
II
2
2
6
4
I
I
5
3
7
2
1;
6
7
8
2
I
0
10
3
6
11-15
16—20
3
I
4
2
21-25
26-30
31-35
36-40
100 . .
I
2
i
I
I
a T
i
Total
72
4.9
58
3-2
23
4-3
2
?> 28.0
68
Average percentage of
abortion
6.9
a Not included in average because borne in an intermediate type of anther.
b Of no significance because of few flowers.
Table VI. — Pollen condition in wild plants of F. americana with reference to the position
of the flower on the inflorescence
Aborted percentage.
Number of
primary
flowers.
Number of
secondary
flowers.
Number of
flowers
position
not
recorded.
0
I
I
7
2
4
4
6
2
3
2
•5
4.
3
c
6
I
I
I
I
I
7
8
Q
I
ic
II
I
2
12
12
I
2
14
JC
I
Total
14
2.8
13
4-4
22
Average percentage of abortion
5' 2
Mar. II, 1918
Sterility in the Strawberry
631
Observations on the pollen condition in herbarium material of other
species than those above mentioned are not conclusive with regard to
the species examined because of the scarcity of material. Plants of
F. chiloensis from Sequin, Washington, from two localities in San
Francisco County, California, and from Bolivia, South America, pro-
duced perfect pollen. A single plant from Vark Hill, Cal., produced
small amounts of defective grains while two plants of a clone from Lake
Merced, Cal., which set fruit on the primary flowers of an otherwise
staminate cluster, produced in the neighborhood of 50 per cent of aborted
grains. A plant of F. chiloensis var. Scouleri, from Klantaak Island,
Yakutat Bay, Alaska, produced perfect pollen. F. platypetala from Mt.
Carleton, Wash., also produced perfect pollen. Although flowers of
comparatively few plants of these two species have been examined, the
facts seem to indicate that the pollen condition is much the same as that
found in F. virginiana and F. americana.
POLLEN CONDITION IN CULTIVATED FORMS
In contrast to the nearly normal pollen condition of the wild species
is the variable condition found in the cultivated forms in which practi-
cally all have a larger or smaller percentage of aborted grains. The
pollen conditions in varieties, controlled seedlings of varieties, and some
species-variety hybrids are given in Table VII. The percentages given
are based on an average of over 600 grains per count, and indicate fairly
accurately the pollen condition of the flowers studied.
Table VII. — Percentage of aborted pollen in flowers, of various positions, from 120
cultivated varieties, l8 controlled seedlings of cultivated varieties, J3 selfed seedlings of
one of these, 3 F, plants of F. chiloensis X Wilson, 7 Fj plants of F. cuneifolia X
Magoon, and 10 Fg plants of F. vesca X F. cuneifolia. The percentages are based
on an average of over 600 grains per count
Variety.
Position of flower.
Primary.
Second-
ary.
Tertiary.
Position
not
recorded.
Cultivated:
Abington . . .
Abundance .
Amanda .
Arizona . . . .
Aroma
Barrymore .
Beacon
Bederwood.
Bradley
Brandywine . . .
Brown Beauty.
Charles
Chesapeake . . . .
26. 5
63.0
1-5
13-6
90.8
37- o
20. O
10.5
33- o
4.8
I. o
97.0
10. s
39- o
47.0
7.0
19. 8
2. 8
10. 0
1 16.3
1 l\
22.4
9-7
99-3
88.7
71.4
14. I
9-7
36.7
{
31-7
94-3
100. 0
{
12- S
S-9
632
Jouryial of Agricultural Research
Vol. XII, No. 10
Table VII. — Percentage of aborted pollen in flowers, of various positions, from 120
cultivated varieties, 18 controlled seedlings of cultivated varieties, jj selfed seedlings of
o-ne of these, 5 Fj plants of F. chiloensis X Wilson, 7 Fj plants of F. cuneifolia X
Magoon, and 10 F2 plants of F. vesca X F. cuneifolia — Continued
Variety.
Position of flower.
Primary.
Second-
ary.
Tertiary.
Position
not
recorded.
Cultivated — Continued:
Clara
Climax
Clyde
Commonwealth
Cooper
Corsican
Darlinsiton
Deacon
Dewdrop
Dorman.
Duncan
Early Giant
Early Jersey Giant.
Early Ozark
Ekey
Enhance
Evcrbearin?
Ewell's Early
Excelsior
Fendell
First Quality
Frances E. Willard.
Fitting Eclipse
Candy
Gill
Glen Mary.
Glenville
Gold Dollar
Gold
Good Luck
Goree
Grand Marie
Hanbeck Beauty.
Haverland o
Ideal
Indiana
Jas. Todd
Jerome
Jessie
Jewell Improved.
Jewell
Jocunda
Joe
Kevitt Wonder . .
King Edward
Klondike
Late Jersey Giant.
Lea
Longfellow
Lovett.
45-5
13- O
42-5
2.0
27. o
56.0
1.8
38.5
23. 2
38-0
50.0
i-S
o. o
7-9
79.6
61. S
S4-0
4.9
f 87.4
\ S9-I
63-7
63.1
56.0
S3-0
I. o
69.9
27. I
63.0
39-7
20.5
83.3
8.6
6.S
95
39- S
60.0
80.8
33-8
73-7
S4-0
6.8
2-7
86.6
44.9
97-7
46.0
99.0
84-3
76.9
84-3
5-6
70. 2
46.5
83-9
24.8
61.4
32-3
9.4
16.6
35- 6
6.0
42-3
10.6
10.8
1-3
68.4
34-2
5-2
61.0
48.0
46.7
SO.O
44.0
42.9
42.9
36. S
56.0
43-.';
IS. 8
69.6
38. S
72.4
33-1
12.6
50.1
13- o
6.8
54- 7
43-5
38.2
10.3
39-6
38.5
Si-S
44.9
100,0
97-3
roo. o
98.0
17.6
3-0
67.8
Si-o
14-3
53-7
II. 8
iS-o
100. o
100. o
95-0
92- S
85.8
84.0
3 Not certainly true to name.
Mar. II, 1918
Sterility in the Strawberry
633
Table VII. — Percentage of aborted pollen in flowers, of various positions, from 120
cultivated varieties, 18 controlled seedlings of cultivated varieties, JJ selfed seedlings of
one of these, j Fj plants of F. chiloensis X Wilson, 7 F^ plants of F. cuneifolia X
Magoon, and 10 Fg plants of F. vesca X F- cuneifolia — Continued
Variety.
Position of flower.
Primary.
Second-
ary.
Tertiary.
Position
not
recorded.
Cultivated — Continued :
Magoon
Maiinda
Manhattan .
Marstiall .
Mascot
Michel! Early.
Miller
Minnesota 3 .
Missionary . .
Model
New Home. .
New York . . .
Nick Ohmer .
II-3
32-6
44-5
44.1
82.0
65-6
97-5
28.7
41.9
47-5
45-3
i-S
47-4
3-5
Ohio Boy.
Orem
Oswego
Palmer
Panama
Pan American . . .
Park Beauty . . . .
Parson's Beauty.
Pearl
94.2
10. 2
96-3
Pennell .
Pineapple.
Pitcher Eclipse.
Progressive
Prolific
Pride of Delaware
Providence
Purcell Early
Purcell
Reasoner 324
Rewatisco
Ridgeway.
Sample.
Saratoga
Saunders
Seedling 373.
Seedling 585.
Seedling 702.
Seedling 753.
Seedling 776.
75-3
26. 7
24.0
29.6
62.0
7.0
85.1
54.0
12.5
60. 4
14.2
38.0
26.4
8.6
SO. 8
43-5
21. 7
18.2
47.1
41.8
41.9
39- S
57-5
81.4
4.0
67. 2
50-S
i.o
15- o
45- S
25.0
19-8
72-5
11.5
10.3
29.6
22.0
53-8
43- S
39-9
42.9
7.8
H
SS-7
S3- 5
91- S
23-1
19. 2
61. 7
46.7
3-0
3-9
6.0
24.8
15-5
33-3
38-4
50.0
56.1
9.0
18.0
16.9
15.4
49.8
63-8
28.1
14- S
62.0
100. o
100. o
35-6
37-4
40.0
45-4
82.6
44-3
60.5
29.6
84.4
85-9
634
Journal of Agricultural Research
Vol. XII, No. lo
Table VII. — Percentage of aborted pollen in flowers, of various positions, from 120
cultivated varieties, i8 controlled seedlings of cultivated varieties, 33 selfed seedlings of
one of these, 3 F^ plants of F. chiloenis X Wilson, / F^ plants of F. cuneifolia X
Magoon, arid 10 F^, plants of F. vesca X F. cuneifolia — Continued
•Variety.
Position of flower.
Primary.
Second-
ary.
Tertiary.
Position
not
recorded.
Cultivated — Continued:
Seedling 778
24.8
Seedling 813.
Seedling 823.
Seedling 893.
Seedling 908.
79.8
Seedling 923
Seedling 924
Seedling 927
Seedling 947
Seedling loio
Seedling 1017
Seedling 1043
Seedling 104s —
Senator Dunlap.
Son's Prolific
South Dakota
Splendid
Splendid X Bunlap.
9.0
13- o
18.5
Staples.
Steven's Late ChaxajMon.
St. Louis.
Success . .
Sweetheart ,
Tennessee Prolific.
Texas
Three Ws
Twilley
Uncle Jim
Warren
Warfield a
Wm. Belt
Wilson
Winner
Wolverton
Wonder
Average.
Selfed seedlings of Seedling 778:
24-4.
25-1.
25-2.
25^-
25-6.
26-1.
26-4.
27-5-
28-3.
S3-4-
54-2.
77.0
17.0
49.2
52-3
24.5
23. S
o Not certainly true to name.
25-4
15-9
55-4
25.0
77- S
S4-I
61.8
29.6
IS- 2
31.0
10.0
S9-3
14-3
13-8
4S-9
20. o
34-9
46-3
28.6
S7-9
1. o
52-1
69. o
75-8
40. o
45-2
4.6
2. 2
10. 2
80.6
9S
1.6
I. 2
5S-4
7.8
38.3
47.1
52-4
27.9
Mar. II, 1918
Sterility in the Strawberry
635
Table VII. — Percentage of aborted pollen in flowers, of various positions, from 120
cultivated varieties, 18 controlled seedlings of cultivated varieties, jj selfed seedlings of
one of tliese, j F^ plants of F. chiloensis X Wilson, 7 Fj plants of F. cuneifolia X
Magoon, and 10 F^ plants of F. vesca X F. cuneifolia— -Continued
Position of flower.
Variety.
Primary.
Second-
ary.
Tertiary.
Position
not
recorded.
Selfed seedlings of Seedling 778— Continued:
2S-3
9-9
f 22.5
I 19- 4
/ 52-4
I 42.5
66-1
32-0
66-2
94.2
99.0
43-9
38.1
66-3
67-1
25.8
41-3
99.0
9.1
f 10.6
\ 22. 3
21-3
71-5
46.0
100. 0
33-1
25.9
0.4
49.1
SO. 8
52- 7
34-9
72-6
1.8
62.8
. 34-7
61. 0
Average
47.1
36.1
32.7
3 Fi plants of F. chiloensis X Wilson:
77- X
58.0
76.7
7 Fi plants of F. cuneifolia X Magoon:
S-2
IS-O
2.0
SI. 4
18.2
16.4
29.1
..1
II. 2
7-7
8-3
8.0
19-3
7-3
7.0
8
II. 9
S-2
3- I
8.9
The most striking fact exhibited by the counts in Table VII, other
than the general presence of some aborted pollen, is the variability of
the pollen condition within a variety. Abington, for instance, shows a
range of from 7 to 37 per cent of abortive pollen. Abundance from 2
to 33 per cent, and Bederwood, which usually produces a high per-
636
Journal of Agricultural Research
Vol. XII, No. 10
centage, in one instance produced as low as 31.7 per cent of abortive
grains. Numerous other equally striking variations in pollen abortion
will be evident by referring to the table.
In view of the variability shown above, the pollen condition was deter-
mined in all of the flowers from two inflorescences of Minnesota 3
to determine what variations occur in flowers grown under so nearly
similar conditions. The results of these counts are presented in Table
VIII and exhibit as great variability within the flowers of a cluster as
is found between flowers from separate plants. In one stalk the range
is from 31.2 to 91. i per cent, in the other it is from 21.4 to 40.3 per cent,
while the range in other counts of the same variety (Table VII) from
different plants is from 14.2 to 60.4 per cent.
Table VIII. — Percentage of aborted pollen in all of the flowers from two stalks of
Minnesota j
Stalk.
Primary.
Secondary.
Tertiary.
Quaternary.
Stalk I
40.3
(a)
/ 29.9
I 30.6
f 91. I
63.5
1 70-7
I 7^-9
21. 4
Stalk II
42.3
68.2
31. 2
49- 7
42. 6
42.7
o Anthers intermediate.
In view of the variability shown between different plants of the same
variety and between different flowers of a single inflorescence, a study
was made of the variation in pollen conditions of individual anthers
within a single flower. The pollen of nine anthers, from a flower of
Seedling 778 was studied, four of which were from the outer para-
petalous series, three from the middle antipetalous series, and two from
the inner long filamented antisepalous series. The results, given in
Table IX, show a greater range of variability between the anthers of a
single flower than was exhibited by the five separate flowers of the same
variety reported in Table VII. Because of the extremely great varia-
bility in pollen condition shown by some varieties, as Abington, Abun-
dance, Aroma, Darlington and others, which often produce nearly perfect
pollen (as far as can be determined by this method of study), while at
other times, apparently under the same conditions, high percentages of
abortive pollen are produced, too much stress should not be laid on a
few scattered observations in determining whether a species is pure or
of hybrid origin until more is known of the factors which produce such
great variability. It may, of course, be argued, and logically, that a
large number of the grains in the anthers, which produce nearly perfect
pollen, are truly abortive, but have developed beyond the stage where
Mar. II, 191S
Sterility in the Strawberry
637
degeneration can be discovered by a superficial examination. The
possibility of this being the case will be shown later. Nevertheless the
factors causing a relatively high degree of pollen abortion in apparently
pure species have been so little studied that to assign hybridity as the
only cause is, to say the least, presumptive.
Table IX. — Variation in percentage of pollen abortion in anthers from one flower of
Seedling yjS
Stamen position.
Parapetalous.
Antipetal-
ous.
29.18
50.00
41.70
34.61
33-65
38.65
17.09
Antisepal-
ous.
27. 27
41.87
POLLEN GERMINATION TESTS
In view of the possibility of many of the apparently normal grains
being in reality abortive, attempts were made to determine the exact
amount of fertile pollen without regard to its apparent morphological
condition. The usual method of pollen germination in Van Tieghem
cells was employed. The results were disappointing, so far as a deter-
mination of actual condition of individual grains was concerned; but
nevertheless some suggestive conclusions may be drawn from them. In
all, 450 tests were made, comprising 28 of F. americana, 45 of F. virginiana,
and 377 of cultivated varieties. The results obtained under carefully
controlled conditions were very erratic. Solutions of cane sugar in
distilled water were used in concentrations var}dng from 7 to 60 per cent.
Pollen germinated to a very slight degree at both of these extremes,
but the optimum concentration ranged between 35 and 45 per cent.
Temperature is an important factor in pollen germination, since at
ordinary room temperature there was practically no germination, while
if the cultures were placed in a warmer portion of the room, at a tempera-
ture of about 90° F. sometimes a high germination resulted. Other
tests, carried on under as nearly identical conditions as possible, often
gave entirely negative results. Tests made in an electric oven, at 95° F.,
showed at times a high percentage of germination, while at others the
germination was very low. Although the proper conditions for germi-
nation could not always be produced, yet enough evidence was obtained
to conclude that wherever any normal pollen is present, a portion of
it is likely to have the power of germination, for in several cases where
over 95 per cent of the pollen was abortive some of the morphologically
perfect grains produced normal tubes.
638
Journal of Agricultural Research
Vol. XII. No. 10
BAGGING TESTS
The final test for the fertility of pollen is its behavior when used in
pollination. The simplest method of testing pollen fertility is to bag
the flowers before they open. This method has been used very exten-
sively by several investigators in self-sterility studies of the pear, apple,
cherry, peach, and grape and has been found to be efficient under favor-
able conditions. Observations by growers upon large blocks of any of
the hermaphroditic varieties of strawberries agree that strawberries are
all self-fertile, physiologically, wherever normal pollen is produced, so
that the question of self -sterility does not enter into the problem.
Bagging tests were made on 106 varieties and 40 unnamed seedlings
produced at the Minnesota Fruit Breeding Station. The detailed results
and summary of these tests are given in Table X, and show (i) that in
the hermaphroditic varieties studied no physiological self -sterility exists;
(2) that wherever morphologically normal pollen is present fertilization
takes place; and (3) that the extent of fertilization is dependent upon
the percentage of normal pollen produced.
Table X. — Degree of setting of the fruits of various positions on the inflorescence on g8
hermaphroditic varieties and jg seedlings of strawberries when bagged
Variety.
Primary.
Secondary.
Tertiary.
Quaternary.
Total.
Abington
Abundance
Amanda
Arizona
Arojna
Barryittore
Bederwood
Bradley
Brandywine
Brown Beauty. . . .
Charles I
Chesapeake
Clara
Climax
Clyde
Cooper
Commonwealth
Corsican
Darlington
Dorman
Duncan
Early Jersey Giant
Early Ozark
Ekey
Enhance
Ewell Early
Excelsior
First Quality
Candy
Gill
Glen Mary
Gold Dollar
Good Luck
Goree
Grand Marie
Hanbeck Beauty . .
Haverland "■
a Not certainly true to name.
Mar. II, 1918
Sterility in the Strawberry
639
Table X. — Degree of setting of the fruits of various positions on the inflorescence on
hermaphroditic varieties and jg seedlings of strawberries when bagged— Continued
Variety.
Primary.
Secondary.
Tertiary.
Quaternary.
Total.
Ideal
Indiana
Jas. Todd
Jerome
Jessie
Jewell Imp
Joe
Kevitt Wonder
King Edward
Klondike
Late Jersey King
Lea
Longfellow
Lovett
Manhattan
Marshall
Mascott
Michell Early
Miller
Missionary
Model
Molinda
New Home
New York
Nich Ohmer
Ohio Boy
Orem
Oswego
Palmer
Pan American
Parson Beauty
Pearl
PenneU
Pineapple
Pitcher Eclipse
Pride of Delaware
Pride of Minnesota
Prolific
Providence
Reasoner 324
Rewatisco
Ridgway
Sample
Saratoga
Saunders
Senator Dunlap
Son's Prolific
South Dakota
Staples
Steven's Late Champion
St. Louis
Success
Tennessee Prolific
Three Ws
Twilley
Uncle Jim
Warfield
Warren
Wm. Belt
Winner
Wolverton
Minnesota 3
Seedling 14
Seedling 114
Seedling 130
Seedling 97
Seedling 15
Seedling 123
Seedling 123
Seedling 89
38
2S
17
3
19
8
25
IS
37
27
12
9
8
I
16
17
17
1 One bag split or tip open, allowing the possibility of cross-pollination.
6 Of a total of 11 flowers, none set. The pollen of this variety has not been ejcamined.
640
Joitrnal of Agricultural Research
Vol. XII, No. 10
Table X. — Degree of setting of the fruits of various positions on the inflorescence on
hermaphroditic varieties and jp seedlings of strawberries when bagged — Continued
Variety.
Seedling 40 - - - - ' 2
Seedling 168 | 2
SeedlingsSs ' 2
Seedling 753 [ " 2
Seedling 876
Seedling 703
Seedling 778
Seedling 937
Seedling 1065
Seedling 1009
Seedling 845
Seedling 901
Seedling 1018
Seedling 1017
Seedling 585
Seedling 702
Seedling 753
Seedling 778
Seedling 776
Seedling 908
Seedling 923
Seedling 924
Seedling 937
Seedling 947 -
Seedling loio
Seedling 101 7
Seedling 1023
Seedling 1026
Seedling 1045
Splendid XDunlap
Total....
Per cent .
Primary.
113
26.8
48
II. 4
Secondary.
28s
24.9
624
54.6
234
20.5
Tertiary.
337
21-5
587 646
37. 441. 1
Quaternary.
70
II. 8
114
19. 2
Total.
a Two bags split or tip open, allowing the possibility of cross-pollination.
Pollination in unbagged flowers is for the most part dependent upon
bees and small insects and upon the anthers becoming erect and partially-
folding about the receptacle while dehiscing, thus dropping the pollen
on the stigmatic surfaces. In the bagged flowers the insects are elimi-
nated, and thus the most efficient natural means of pollination is lost.
As a consequence many of the bagged varieties produced many nubbins,
but no case of complete self-sterility was found where morphologically
perfect pollen was present.
If the percentages of total perfect fruits, nubbins, and sterile flowers
of each flower position in Table IV are compared with those in Table X, it
will be seen that (i ) the setting under bags is very much poorer than in the
open, both with regard to the number of perfect fruits set and also with
regard to the actual number of flowers which set any achenes, and (2) the
primary flowers are decidedly more fertile than the later ones, as, even
under the adverse conditions of pollination within the bags only ii.o
per cent of the primary flowers were sterile, while 20 per cent of the
secondary. 40.7 per cent of the tertiary, and 68.5 per cent of the quater-
nary flowers were sterile, in spite of the very great increase in amount of
Mar. II, 1918 Sterility in the Strawberry 641
pollen present in the bags while the later flowers were in condition for
pollination.
In order to determine to what extent parthenogenesis or parthenocarpy
might possibly enter into the above results, bags were put over 67 clusters
of 22 pistillate varieties. Of a total of 661 flowers covered, 55 set some
achenes. Of these, 52 were found in 6 bags which had been accidentally
split, thus accounting for the probable pollination by insects. Of the
remaining three fruits, which developed in apparently tight bags, two bore
I achene each and the other 7. Significance can hardly be attributed
to the setting of these few achenes, since the chances for accidental
pollination, to this extent, are relatively great. It may therefore be
concluded that parthenogenesis does not exist in the cultivated strawberry.
A condition which might possibly be attributed to parthenocarpy
occurred in the Buster variety, in which 9 flowers of the 22 bagged
showed a very decided development of the achenes with no accompany-
ing development of the receptacle. These achenes contained no embryos.
In the strict sense of the word parthenocarpy in the strawberry could
only be applied to a development of this kind. A more comprehensive
use of the term might include the development of the fleshy receptacle.
Ordinarily, flowers which set only one or a few achenes develop the
fleshy receptacle only at corresponding points, due probably to the
stimulus of fertilization. In perfect varieties receptacles are often
found in which development has taken place not only at the base of
the pistils but also about the base of the stamens. In 3 out of 10
bagged flowers of the pistillate Red Bird variety, the fleshy receptacle
developed about the base of the staminodia forming a red fleshy circle
about the dried pistils. Two flowers of Crescent, also an imperfect
variety, developed normal fleshy berries, one bearing one achene and
the other none. Apparently these receptacles developed without the
stimulus of fertilization in the same way as that at the base of the stami-
nodia in Red Bird.
POLLEN DEVELOPMENT
As Mendelian and, in fact, most genetic results are dependent upon
the segregation of determiners during the formation of gametes and to
their recombination again at the time of fertilization, any processes
which interfere with the normal procedure should be carefully studied
and, if possible, their nature determined.
There are a number of ways in which the normal order may be dis-
turbed, at least there are various outward expressions of them. The
condition in the Phylloxera spp., as pointed out by Morgan, in which
half of the spermatids degenerate regularly, while the other half con-
tinue and form normal spermatozoa, and the relationship between
degeneration and the absence of the accessory chromosome is so well
known that it needs no comment. Gates (iS) has shown that in
642 Journal of Agricultural Research voi. xii, no. 10
Oenothera lata the early abortion of the male generative organs and
partial abortion of the female is in some way associated with the presence
of an extra chromosome, while in Oenothera semilata an extra chromo-
some is present, but only a portion of the pollen grains abort. Morgan
has shown that slight mutations are continually occurring in Drosophila
spp., which inhibit the development of the 2X individual, while Bridges
(5) has shown that certain chromosome combinations can not bring
about normal development in the zygote. Is it too much to expect
that like conditions may affect the i X generation also ?
Dorsey {12) pointed out a different type of pollen degeneration from
that which expresses itself in the production of empty grains. He
showed that in functionally pistillate grapevines pollen developnemt
proceeds normally through the microspore division and the formation
of the normal content of cytoplasm found in mature fertile grains.
During the period of development following the microspore division,
one or both of the nuclei of a portion of the pollen grains aborted, leaving
the grain normal as far as cytoplasm was concerned. Associated \^^th
the complete sterility of the pollen of the functionally pistillate varieties
were the reflexed type of stamens, an entire lack of sutures and germ
pores in the mature pollen, and dieciousness. It is probable that the
lack of germ pores is the direct cause of stei-ility in the numerous grains
which otherwise appear entirely normal. Tischler, Rosenberg, and others
have shown that pollen abortion in hybrids may follow either normal or
irregular reduction divisions, when the parents have both an equal and
unequal number of chromosomes. Shull (35, 36) has recorded some
consistent irregularities in sex ratios in Lychnis dioica which will later
be shov/n possibly to have been due to slight mutations causing pollen
abortion. Goodspeed {21, 22) gives further evidence on the sterility of
hybrids oi Nicotiana spp,, when A^ sylvestris is used as one parent andshovv^s
that not only the pollen is sterile but that the F^ plants are incapable of
forming any very appreciable amounts of seed. Rimpau {31) and
Jesenko (26) have shown that in hybrids between wheat and r5'e there is
complete male sterility, while some of the egg cells are able to produce
viable seed if pollinated with either wheat or rye pollen.
The cytological investigation of pollen development in the strawberry,
reported in this paper, has two main objects in view: (i) the determina-
tion of the mechanism and nature of pollen abortion and \vith these
facts at hand, (2) the determination to which of the many categories of
sterility the very prevalent pollen abortion in the strawberry varieties
belongs.
Material and methods. — The variety used as the basis of this study
is Minnesota 3. As was previously stated, it furnishes desirable ma-
terial for this type of study, as about 50 per cent of its pollen aborts,
while the remainder develops normally. The egg cells in this variety
and in the cultivated varieties in general do not show a corresponding
Mar. II, 1918 Sterility in the Strawberry 643
degree of abortion, as it is very common for practically all of the achenes
to develop on perfectly formed strawberries.
The material was prepared for microscopic examination according
to the ordinary cytological methods. Carnoy's, Flemming's strong,
medium, and weak, and chromacetic-acid fixing solutions were used.
All g'ave very good results, except Carnoy's fluid. Sections were cut
from 4 to 20 ^t thick, the best results being obtained from those 4 to 6 /i
thick. The triple stain and Haidenhain's iron-alum-hymatoxylin stains
were used, both giving good satisfaction.
The drawings in Plates B to E were outlined w^th the help of an Abbe
camera lucida. All, with a few exceptions noted, are drawn to the same
scale, in order that comparisons of cell size and cytoplasm content may be
readily made.
Anther tissues. — The walls of the young anthers are made up of
four oblong layers of cells of about equal size ; the outer epidermal layer
and three inner layers. Inside of these layers there is usually one
layer of tapetal cells and about five layers of pollen mother cells, both
of which at this time are easily distinguished from the wall cells by
their large size and different staining reaction.
The growth and development of the wall layers should be followed
because of its relation to the increase in size of the pollen mother-cell
cavity during the formation of the tetrads. As the pollen mother cells
prepare for reduction, and during the division, rapid cell division is tak-
ing place in the parietal cells, so that by the time the heterotypic division
is complete there is an appreciable increase in the size of the anther
cavity. Gradually the two inner layers of cells flatten out, owing prob-
ably to the growth of the outer layer, the cells of which rapidly increase
in size, although showing no further cell divisions.
At the tetrad stage the inner layer is very much flattened, while the
middle layer is still plainly visible and the cells are still full of cytoplasm.
The cells of the outer layer are now very appreciably larger than the
epidermal cells, which have also grown slightly. After the liberation
of the microspores and while they are increasing in size, the epidermal
cells sometimes collapse, as their contents have become scant. By the
tirhe the microspores have nearly completed their growth and have
begun to divide, the cells of the outer layer have become deeper than
long and are nearly as large as the tapetal cells. At this time they show
distinctly the spiral thickenings which have to do with dehiscence. Both
of the inner layers have now collapsed or show very scant cytoplasm.
Before dehiscence the walls separating the members of the two pairs of
loculi break down, leaving two large loculi in each anther. The relation
between the increase in size of the anther cavities both during prepara-
tion for the first meiotic division and subsequent to it, and a diff'erence
which Fragaria spp. shows from some other forms in the history of the
mother-cell wall will be pointed out later.
644 Journal of Agricultural Research voi. xii, no. 10
The; tapetum. — The tapetum, usually one cell layer thick, is com-
posed of large angular cells similar in size, and staining reactions to the
pollen mother cells. The tapetum in Fragaria spp. differs in its greater
persistence from most other forms reported. Division in the tapetal
cells begins at about the time of synapsis of the pollen mother cells and
has been observed as late as the metaphase of reduction division. The
divisions are all mitotic, no evidences of amitotic divisions having been
observed. Following reduction division the tapetal cells are binucleate
and remain so until the liberation of the microspores, when they degen-
erate and completely disappear. The disappearance of the tapetal
layers is gradual. They first separate from the wall layers and then
proceed to dissolve, the wall which was in contact with the anther wall
first disappearing, followed by gradual dissolution of the entire cell
layer. During this process the walls become thick and laminate and in
places the middle lamella dissolves, partially freeing the individual cells.
When the microspores are three-quarters grown the tapetal cells have
entirely disappeared. This persistence of the tapetal cells will be shown
to be correlated with a like persistence of the pollen mother-cell walls.
The history of the tapetum in F. virginiana is identical with that just
described which is of Minnesota 3.
Pollen mother cells. — The study of pollen development was begun
with the so-called resting stage of the pollen mother cells between the
last archesporial division and the first meiotic division. They do not,
however, show the characteristic chromatin and linin condition found
in true resting cells of Fragaria spp. (compare fig. i, PI. B and fig. 2,
PI. D). The cells are angular and contain a large nucleus (PI. B, i).
The contents of the nucleus are irregular dark-staining, very small masses
of chromatin held in a network of linin fibers. The number of chro-
matin bodies is very much larger than the number of chromosomes.
Ordinarily one large nucleolus is present, although it is not uncommon
to find two.
Synapsis. — ^The first indication of the onset of the prophase is to be
seen in the gradually increasing size of these chromatin bodies which
still, however, appear very flaky and irregular. The linin and chromatin
during this period are so indistinct that it is impossible to determine
whether there is any definite pairing of the individual particles and
threads as has been shown to exist in Lilium spp. by Allen (2) and in
several forms by Overton (jo). These larger masses gradually move to
one side of the nuclear cavity and congregate about or near the nucleolus
in a loose indefinite mass (Pi. B, 2). At this time a few rather defi-
nite threads appear in the mass, some extending out from it as loops.
Where the loops are long enough, they are seen to be distinctly double
(PI. B, 2, 3). This is the only evidence of any pairing during the
presynaptic stages. Gradations between the conditions shown in
figures I and 2 occur within a single loculus of an anther and are proof
Mar. II. 1918 Sterility in the Strawberry 645
that these stages bearing bivalent loops are presynaptic. Gradually the
synaptic mass tightens until it is close and compact, occupying a very
small portion of the nuclear cavity (PL B, 4). During this contrac-
tion there are refractive particles present both in the mass, some of which
are in contact with the nucleolus, and outside of the nuclear membrane,
which give the same staining reactions as the nucleolus. Similar masses
to these Digby (jj) has considered to be synaptic extrusions.
The synaptic stage is of long duration. Gradually the chromatin in
synapsis takes on the appearance of being made up of a closely tangled
mass of threads. Soon loops are pushed out from it, which are bivalent
often for their entire length (PI. B, 5). The fact that these loops are
often double from the point at which they leave the mass to the point of
entrance, and can sometimes be traced through a portion of the synaptic
mass gives the impression that they are made up of two continuous
threads which closely approximate each other over their entire length.
This view is supported by the later stages, especially those at and follow-
ing segmentation, which in Fragaria spp. are very clear.
Loops continue to push out from all sides of the synaptic mass, often
shifting it to the center of the nucleus. Gradually the bivalent thread
becomes more or less regularly distributed about the nuclear cavity,
usually having, however, a somewhat tangled center near the nucleolus.
The spireme thread at this stage often appears to be a single strand
due to the dose approximation of its univalent portions (PI. B, 6).
However, no anthers at this stage of development have been found which
do not contain many portions of the spireme which are double for eon-
siderable distances. It is probable that, during the post synaptic stages
up to segmentation of the spireme into chromosomes, the univalent
portions of the thread are never fused throughout their whole length to
form a single spireme. It is even possible that no fusion takes place,
but that the imivalent threads only approach each other so closely that
in such delicate threads the line of demarcation can not be distinguished.
There is no distinct second contraction, but there is a semblance of
one following the loosely-distributed spireme stage. The thread con-
tracts gradually but unevenly throughout its whole length, its univalent
portions as a consequence becoming separated from one another and
appearing thicker (PI. B, 7). The portions of loops which are in
contact with the nuclear membrane remain so and often extend long
distances on the periphery and then turn at relatively sharp angles and
again extend in fairly straight lines toward the central mass, still situ-
ated usually near the nucleolus. Many of the loops in this way form
equilateral triangles. There is no doubt at this time of the double
nature of these loops. The paired threads are evidently identical with
those which appeared while passing out of synapsis and may be identical
with the bivalent loops seen extending out from the loose presynaptic
38325°— 18 3
646 Journal of Agricultural Research voi. xii, no. 10
mass. Continued contraction of the bivalent spireme results in seg-
mentation.
Segmentation. — In order to determine definitely whether a telosyn-
aptic or a parasynaptic arrangement of the univalent chromosomes
prevails during the synaptic and postsynaptic stages, it would seem nec-
essary to determine the exact procedure from the bivalent condition just
previous to segmentation, through segmentation to the paired condition
in diakenesis. If it can be shown that the bivalent threads appearing
during the pre and post segmentation stages are identical in Fragaria
spp., it will be a strong argument in favor of the parasynaptic arrange-
ment of the chromosomes. As Fragaria spp. is not complicated by a
second contraction and, as the segmentation stages are rapid, all stages
from that shown in Plate B, figure 7, to diakenesis being found within
a single loculus; and as the stages during this period are unusually
distinct, such a determination is not difficult.
Digby {11) has recently presented the results of a very detailed study
of the cytology of Crepis virens in which the conclusion is reached that
there is an end-to-end arrangement of the chromosomes during the
synaptic stages. The details from the loosening of the synaptic mass to
segmentation are very similar to those in Fragaria spp., although the
chromatin in C. virens is apparently much more viscous and gives less
clear-cut images than does Fragaria spp. Her figures 76, 78, and 79
may be considered as in the same stages as my figure 7 of Plate B and to
present the same condition — that is to say, a split or a double spireme the
pairs of which are somewhat twisted about one another. She considers
that these figures do not show two univalent threads lying side by side,
but that the bivalent loops are due to the folding back upon each other
of univalent segments during second contraction. At this stage the
chromatin mass became so viscous that —
it is generally impossible to individualize the future three bivalent chromosomes.
The chromosomes are in fact evolved out of what appears to be a tangle of viscous
nuclear contents.
It appears hardly logical to conclude from this evidence that there is
an end to end rather than a side to side pairing of the univalent
chromosomes.
The nimiber of loops present it segmentation in species of Fragaria
is always less than the number of chromosome pairs which appear at
diakenesis. It has been mentioned that these loops extend for long
distances on the periphery of the nucleus, forming more or less regular
equilateral triangles. When segmentation takes place, it is usually at
the outer angles of these loops and at or near the nucleolus, which gen-
erally forms more or less of a center from which the loops radiate. Thus
the bivalent loops are often divided into three pairs of bivalent chromo-
somes (PI. B, 9). The pairs continue to contract, those attached to
Mar. II. 1918 Sterility in the Strawberry 647
the periphery remaining in that position and those which have one end
attached to the nucleolus assuming a position alongside of it (PI. B,
II, 14). Occasionally one pair may be attached both to the nucleolus
and to the periphery ; when there is evidence of considerable force exerted
by the pair in contraction (PI. B, 11), The pairs continue to contract
(PI. B, 12, 13), forming various figures which have often been described
in other forms, but very few circles have ever been seen at this time.
The contraction continues until the typical diakenesis stage is reached
when it is often difficult to distinguish between the two univalent
chromosomes of a pair (PI. B, 14, 15). Apparently they often fuse,
as in the multipolar spindle stage they sometimes appear as single entities.
At diakenesis 26 chromosome pairs can readily be counted. At this
stage in Minnesota No. 3, of 22 counts made, 19 showed definitely 26
chromosome pairs and three others showed, respectively, 24, 25, and 27.
A very similar condition to the diakenesis of the pollen mother cells is
shown in the prophase of the tapetal cell divisions. Here, however, in
place of the 26 pairs of chromosomes 52 pairs appear arranged about the
periphery of the nucleus. Five counts made at this time showed in
three cases 52, and in two 50 and 54 pairs, respectively.
Heterotypic division. — Diakenesis in the pollen mother cells is of
somewhat long duration, but the period between it and the metaphase
of the heterotypic division is extremely short, usually not more than two
or three multipola'r spindles appearing in a loculus simultaneously. The
small oval chromosomes now arrange themselves on the equatorial plate.
Whether there is any definite order or arrangement could not be deter-
mined, as the chromosomes appear identical. They are arranged close
together and, while their number can not be readily determined, 26 have
been counted on one plate and 24 on another.
The chromosomes are then gradually pulled apart and drawn to the
poles. No irregularities in cell division or extrusion of chromatin matter
have been seen during this process. The daughter chromosomes show
only slight evidence of fission for the following division. The disk-
shaped daughter nuclei are soon formed (PI. C, i), and directly after
prepare for the second meiotic division.
HoMEOTYPic DIVISION. — The two spindles of the homeotypic division
may be parallel to one another or their axes may be at right angles.
The metaphase of the division is also characterized by great uniformity,
the daughter chromosomes separating and advancing toward the poles
simultaneously. After separation they could be readily counted and
showed in 7 counts 26 chromosomes (PI. C, 2).
The daughter nuclei are soon formed, and walls are laid down between
them, dividing the cytoplasm evenly. The cells gradually split apart,
separating the four microspores and allowing the entrance between
them of the viscous material which has up to this time surrounded either
648 Journal of Agricultural Research voi. xn, no. »
partially or entirely the original mother cytoplasm (PL B, 9; PL C,
I, 3, 4)-
Pollen mother cell wall. — The history of the pollen mother cell
wall is of interest as it differs somewhat from that generally reported for
the higher plants.
In the lily (Allen, i), grape (Dorsey, 12), and in many other plants in
which pollen development has been studied, it is usual, during prepara-
tion for the first meiotic division, for the pollen mother cells to separate
from one another, due apparently to dissolution of the middle lamella
and to growth of the anther walls, forming a greater space into which
the cells can round up and float free from one another. Allen (/, p. 200)
considers that the separation is due to a dissolution of the cell walls
from between the mother cells, and that each is "surrounded only by a
plasma membrane." Following separation, a very decided thickening
of the material surrounding the cytoplasm takes place (Pi. B, 9;
PI. C, i). This, Allen (j), Tischler {41), Stevens {38), and others speak
of as a thickening of the mother cell wall. Following the formation of
the tetrads, this material increases and, as the cells of the tetrad separate
from one another, flows between them. This material is usually of a
rather firm nature and in buckwheat {38) often persists for some time
after the liberation of the microspores from it.
There is evidence in Fragaria spp. which indicates that this material
is entirely distinct from the mother cell wall and is in no way dependent
on it for its increase in volume, thus appearing to be more of the nature
of the gelatinous sheath which surrounds groups of cells in many of the
algae.
In Fragaria spp., in place of the pollen mother cells rounding up just
before or during reduction division, while there is taking place a rapid
growth of the anther walls and a consequent increase in size of the
anther cavities, the cytoplasm separates at the angles from the walls
and rounds up independently (Pi. B, 9, 14, 15; Pi. C, t). The walls
remain in contact with one another and adjust themselves to the
increasing space by stretching. As soon as evidence of rounding up of
the cytoplasm appears, a gelatin-like material is secreted unevenly
about the cytoplasmic mass (PI. B, 9; Pi. C, i). This material is
apparently identical with that laid down about the plasmic mass in
the grape, lily, and forms like them in which the mother cell wall rounds
up, supposedly following the dissolution of the mother cell wall. In
these cases it is generally spoken of as the thickening of a new mother
cell wall. In the strawberry this material increases in amount until at
the completion of tetrad formation and before liberation of the micro-
spores, the spores are completely embedded in it (PL C, 3, 4). The
mother cell walls are still present, but simply divide the anther cavity
into large spaces, which are only partially filled by the tetrad (PL C,
3). An examination of analogous stages to these in the lily and
Mar. II, 191S Sterility in the Strawberry 649
grape shows the mother cell wall lying closely about the gelatinous
material in which the spores are embedded, it having not disappeared
at the period of rounding up of the pollen mother cells, as is generally
assumed. At the time of rounding up of the pollen mother cells in these
two plants, there is no evidence of old walls being left behind, but there
seems rather to be a separation of the contiguous walls, due to the dis-
solution of the middle lamella and rounding up of them with the plasma
masses. The walls give a slightly different staining reaction from the
thick secreted mass and so can be readily distinguished from it. During
the liberation of the spores in the grape, it is not uncommon to see this
gelatinous sheath completely disappear, leaving portions of the original
mother cell wall about the spores. This gradually dissolves, liberating
the spores. In the strawberry, the procedure is much the same. The
gelatinous sheath, which shows no wall closely about it but only a limit-
ing membrane, dissolves, liberating the spores into the large cell bounded
by the mother cell wall. This soon disappears.
At the time of liberation the spores have a distinct wall about them-
selves, which is independent of the surrounding sheath. Apparently
the condition in the lily, grape, and strawberry is identical, as far as
structures are concerned, but differs primarily in the separation of the
walls of contiguous cells in the former plants, while in the latter only
the cytoplasm of the pollen mother cells rounds up, leaving the walls
and middle lamella intact.
Following the liberation of the tetrads and the disappearance of the
mother cell walls, degeneration of the tapetum takes place. Tapetal
degeneration seems to be in some way correlated with the disappearance
of the middle lamella from between the mother cell walls, for in those
forms which show an early rounding up of the mother cell walls there
is a correspondingly early degeneration of the tapetal cells.
Growth of the; microspores. — Following the liberation of the
microspores, there is a period of very rapid growth in their walls. At
first this causes the microspores to become very irregular in outline
(PI. C, 7), but as growth continues and the wall becomes thicker the
cells become more spherical. The growth of the microspore wall is
so rapid that there is not a corresponding growth of the cytoplasm,
a condition which results in a large vacuole occupying the greater portion
of the cell cavity. When gr«wth of the microspores is nearly complete,
the original cytoplasmic content of the spore is spread out over the
periphery of the cell and about the nucleus (PI. C, 8). The compara-
tive size and consequently the very great decrease in relative cytoplasmic
content, of the newly liberated microspores and those ready for the
microspore division may be realized by comparing figures 5 and 8 of
Plate C, which are drawn to the same scale.
Early in the development of the liberated microspore the wall is
differentiated into the extine and intine (PI. C, 7). The extine gradually
650 Journal of Agricultural Research voi. xii, no. 10
thickens and a series of scales are formed over its surface. The ex-
ternal appearance of the extine is shown in Plate C, figures 15 and 16;
Plate D, figures 6 and 15. Soon after the microspore division the extine
development is practically complete.
Division of the microspore nucleus. — ^With the growth of the
microspore there is an apparent decrease in chromatin content; for,
in the nuclei which are just about to divide, the chromatin is distributed
in small particles about the periphery of the nucleus and appears very
scant. Transition stages between that seen in Plate D, figure 8, and
the completion of the spireme have not been observed. The spireme
(PI. D, 9, 10) is a continuous, heavy, dark-staining thread. It follows,
more or less, the periphery of the nucleus and surrounds the large
nucleolus which has at this time begun to break down. The nucleolus
takes on a very irregular outline, which is in some way related to the nu-
merous threadlike processes which extend from it to the spireme thread
(PI. D, 10). The nucleolus is now very light-staining and seems for the
most part homogeneous, but it may contain one or more small vacuoles.
Soon after the segementation of the spireme into its 26 constituent parts
the nucleolus completely disappears.
Following segmentation of the spireme the nuclear membrane dis-
appears, the chromosomes are drawn to one plane, and the spindle
is formed (PI. D, 11). The most usual position of the spindle in the
microspores of many plants is near the wall and perpendicular to it;
the pole which is to form the generative nucleus being nearly in con-
tact with it. As a result of this position the generative cell usually lies
against the wall and the vegetative nucleus in the cell cavity (PI. D, 15).
This arrangement of the spindle is also found commonly in Fragaria
spp., but is not universal. Often spindles are found which lie parallel
to the wall (PI. D, 12) and result in the arrangement of the nuclei shown
in Plate D, figure 14, both of which lie against the wall.
The spindle is always broad at the poles and short.' The chromo-
somes on the equatorial plate are small and oval in shape and may
readily be counted if a section can be obtained in which the heavy wall
of the microspore is removed both above and below the chromosomes.
In these sections 26 chromosomes are plainly visible. The daughter
chromosomes are drawn to the poles simultaneously (Pi. D, 12), no
instances having been seen in which chromosomes lagged behind on
the spindle or in which there was an extrusion of chromosomes.
Directly after the rounding up of the daughter nuclei and the dis-
appearance of the spindle fibers there is visible no sign of a cell wall or
limiting membrane (PI. D, 13) between the two nuclei, which is even-
tually to set off the generative nucleus in a separate cell (Pi. D, 14, 15).
Soon, however, the wall appears and the generative cell begins to round
up (PI. D, 14), eventually to lie free in the cytoplasm of the pollen grain
(PI. C, I).
Mar. II, 1918 Sterility in the Strawberry 651
Pollen maturity. — After the microspore division there is again a
period of growth of the pollen grain and a very marked increase in the
amount of cytoplasm. Eventually the pollen grain is completely filled
mth cytoplasm of a distinctly alveolar structure. During this time the
cytoplasm nearly disappears from about the microspore nucleus, leaving
the microspore cell wall loosely surrounding the nucleus, the chromatin
of which is graduallv taking on the condition typical of resting nuclei
(PI. D, 2).
When pollen formation is complete, changes take place in the anthers
preparatory to dehiscence. These consist primarily in breaking down
the wall between each pair of loculi, thus throwing all of the grains of
one-half of the anther together. There is also a general drying-out
process which results in the disappearance of the liquid material which
previously surrounded the developing grains, and of a considerable
portion of moisture from the pollen grains, thus causing them to collapse
along three meridial sutures which fold in, thus giving in cross section
somewhat of a clover-leaf shape, while the general shape of the grain is
long-oval. The three germ pores are located at the middle points of the
sutures. On placing the dry pollen in water it soon swells to form a
sphere.
The position of the generative cell and vegetative nucleus within the
collapsed grains could not be determined, for as soon as the killing fluid
penetrated the anthers the grains immediately swelled. In the swelled
grains the nuclei lie in various positions with regard to one another, but
usually in close proximity. The vegetative nucleus is generally spherical
while the generative cell contents are' fusiform, owing to the folding of
the dried pollen grains and are not closely surrounded by the cell wall.
The chromatin of both nuclei now shows the typical resting condition
(PI. D, 2).
Up to and including the liberation of the microspores from the tetrad,
the cells have shown marked regularity in division, the stages of short
duration proceeding regularly from one end of a loculus to the other,
while during the stages of longer duration the cells of a loculus all show
the same degree of development. Up to the time of liberation of the
microspores, the development which has taken place has depended
entirely on materials furnished by the sporophyte, the one group of
chromosomes merely being the tools by which the materials of the
pollen mother cell were divided into four parts, the microspores. So
far as can be seen up to this point, no growth process resulting directly
in an increase of cell material can be directly attributed to the one aggre-
gation of chromosomes. At the time of liberation from the tetrad
the spores are strikingly alike in size and all other visible characters.
There has been, up to this time, no differentiation in rate of development
of individual microspores; and the spores, as liberated, are all normal.
652
Journal of Agricultural R,esearch
Vol. XII. No. 10
DEGENERATION OF THE MICROSPORE
Liberation of the microspore from the tetrad places these individuals
upon their own resources for future development. It is true that the
spores are surrounded by a nourishing medium furnished by the sporo-
ph3rte, but the abihty to use this material depends upon the individual
cell metabolism of the microspores.
A study of the actual change in cytoplasmic content from the pollen
mother cell stage to the stage of complete formation of the pollen grains
indicates the degree to which the young spores are dependent upon
their own metabolism from the time of liberation from the tetrad to
maturity of the male gametophyte or mature pollen grain. Measure-
ments were made of the diameter of the spherical cytoplasmic mass at
the following stages during pollen formation: Rounded pollen mother
cell at diakenesis, young microspores which have just been liberated
from the tetrad, microspores which have completed growth and in
which the nuclei are either dividing or are about to divide, and the
mature pollen grains. A summary of the measurements is given in
Table XI. It will be seen by comparing the volume of the pollen mother
cells and of the microspores at liberation that the latter show slightly
less than one-fourth the volume of the former, indicating that during
reduction division and the period subsequent to the liberation of the
microspores no increase in cytoplasm has occurred. Following libera-
tion from the tetrad, however, when the microspores are floating in
the nourishing medium of the loculus as independent units, very rapid
wall growth takes place. From <the time of liberation to division of
the microspores nucleus the volume of the cells increases about 6.4
times, but shows no corresponding increase in cytoplasm. At the
mature pollen stage the cells have increased to nearly 7.5 times that
of the liberated microspore. The changes in volume and cytoplasmic
content from the pollen mother-cell stage to mature pollen are also
well illustrated by Plate B, figure 14; Plate C, figures '4 and 8; and
Plate D, figure 2, all of which are drawn to one scale.
Table XI. — Volume of pollen mother cells and microspores at various stages of
development .
Stage.
Number
Average
ured.
diameter.
M.
263
15-25
300
9.41
200
17.44
500
18.39
Volume.
Pollen mother cell at diakenesis
Liberated microspores
Microspores, division stage
Mature pollen
Cm. /I.
1,859.99
436. 28
2, 777. 40
3.255-39
Mar. II, 1918 Sterility in the Strawberry 653
Although the gametophytic generation must properly be considered
as beginning with the first appearance of the haploid chromosome
number, yet the liberation of the spores from the tetrad may be con-
sidered as marking the beginning of the independent growth period of
this generation and the rapid growth of the spore wall, division of the
microspore nucleus, the increase in cytoplasm, and finally the germi-
nation of the pollen grain and the production of sperms as develop-
mental stages in this very much foreshortened plant generation. The very
rapid wall growth, division of the nucleus, and finally the increase in
cytoplasm to 7.5 times its original volume, all within a relatively short
period, go to make this period of a plant's life history probably the
most finely adjusted and critical one which it is called upon to survive.
It is during the periods of rapid enlargement and of increasing cyto-
plasm that degeneration of the microspores takes place.
Although there is some evidence in the strawberry that degeneration
may begin before liberation of the microspores from the tetrad, so many
anthers have been examined which contain only normal microspores,
both in the tetrad stage and directly following their liberation, that
degeneration before this time can be considered as negligible. It is
probable that poor fixation may account for the few apparent instances
which have been found. Indeed, poor fixation constantly enters as a
disturbing factor in the interpretation of the condition of the supposedly
aborting grains, and it is only by finding the same types at later and
more advanced stages of degeneration when there can be no question as
to the condition of the microspore that the factor of poor fixation can
be eliminated. No aborted microspores have so far been found which
could be referred to as degeneration within the tetrad.
Directly following microspore liberation evidences of pollen abortion
may be noticed and from this period on through the various stages of de-
velopment, microspores and pollen grains are continually aborting.
Plate D, figures 3 and 4, shows microspores shortly after liberation, in
which the contents have completely broken down into a yellow oily mass
which turns black on exposure to osmic acid. The lighter areas are
globules evidently of a different substance.
Ordinarily during the growth period of the wall following liberation,
the cytoplasm becomes spread over the periphery in a thin layer. Plate
D, figures 5, 6, and 7, represent various conditions in which this normal
process has not occurred. Figures 7 and 1 1 are later stages of this type
of abortion in which degeneration of the nucleus and cytoplasm is grad-
ually taking place. Figure 15 apparently represents a further stage of
this series, in which the nucleus, although visible in outline, is, with the
cytoplasm, entirely functionless. Plate E, figure 11, shows the com-
pletely degenerate microspore of this type found among normal mature
pollen. The dark-staining mass is yellow and oily before killing.
654 Journal of Agricultural Research voi. xii, no. io
Although there has been a considerable amount of degeneration up to
the time of the completion of growth of the microspore wall, probably
more takes place between the time of the formation of the large vacuole
(Pi. C, 8) and the completion of the microspore division than at any
other period.
Plate D, figure 12, shows a full-grown microspore in which degeneration
is taking place, both in the cytoplasm and in the nucleus. This was found
in an anther containing full-grown i -nucleate microspores, and probably
is a case in which degeneration has begun during the period of enlarge-
ment.
Plate D, figure 13, and Plate E, figure 2, represent early stages in de-
generation of I -nucleate microspores which are at the stage at which
division should take place as they were found in anthers containing both
I- and 2-nucleate grains, as well as division stages. The microspore rep-
resented in Plate E, figure 7, was found among microspores of the stage
of development shown in Plate C, figure 15, while that of Plate E, figure
8, was found among mature pollen grains. Both are evidently later
stages of the condition represented in Plate D, figures 12 and 13, and Plate
E, figure 2.
During the period of division a few microspores contain a very scant
amount of cytoplasm. Such spores are seen in Plate D, figures 14 and 16,
and Plate E, figures i and 4. Eventually these completely degenerate.
Those of the type shown in Plate D, figure 16, and Plate E, figure i , prob-
ably are the forerunners of the completely degenerate microspore shown
in Plate E, figure 10, which is a common type in mature anthers.
Although degeneration takes place to a greater or less degree at all
stages, from microspore liberation to microspore nucleus division, no
instances of degeneration occurring during the process of division have
been noted. Directly following division, however, evidences of degener-
ation again become apparent, although much less numerous than in the
period just prior to division. Figure 3 of Plate E represents an early
stage of degeneration directly following microspore division. This is
probably an early stage of the more advanced degeneration stage shown
in Plate E, figure I. This young pollen grain was found among grains in
which the cytoplasm was increasing rapidly in amount. It will be noticed
that the generative cell is aborting, while the remainder of the grain
is normal. Apparentiy grains of this type continue to increase normally
with regard to the vegetative portion, while degeneration of the genera-
tive cell occurs. Figure 1 3 of Plate E represents such a grain found in a
mature anther. Figure 12 represents another type of degeneration
which takes place subsequent to division of the microspore nucleus. The
generative cell has completely degenerated. The vegetative nucleus,
although still present in outline, is f unctionless, while the cytoplasm bears
no resemblance to the normal. It still retains the property, however, of
Mar. ir. 1918 Sterility in the Strawberry 655
absorbing liquids when placed in them, and so the grains of the type
shown in Plate E, figure 13, are likely to cause inaccuracies when the ordi-
nary methods of determining the percentage of good and abortive pollen
are employed.
It seems clear thus far that degeneration of the microspores and pollen
grains is a phenomenon closely related to the very active metabolic proc-
esses which are taking place during this period of the plant's life history.
DISCUSSION OF RESULTS
T"hus far there has been shown to exist in the strawberry types of
sterility due to at least two distinct causes.
In some of the wild species, including F. elatior, F. platypeiala, F.
cuneifolia, and F. chiloensis, it seems highly probable that the species
are diecious, while F. virginiana is unquestionably so for the most part.
Dieciousness is expressed in the production of pistillate plants bearing
staminodia, which, so far as I have observed, never produce pollen.
The staminate plants bear normal stamens and pistils, which appear
normal but which seldom are fertile. Certain types are also found which
are intermediate between these two types. A few staminate plants may
bear fruit on one or more of the early flowers. These flowers may or
may not bear stamens, but both staminodia and intermediate anthers
are found on them. Other plants which are apparently staminate
develop only intermediate anthers in which abortion takes place at the
tetrad division or shortly after, resulting in a degenerate mass in the
anther. Another staminate type has recently been studied which bears
fruit on the primary and occasionally on some of the other flowers.
The anthers of the primary flowers are reduced to staminodia. The
secondary flowers carry pollen development through to the liberation
of the microspores when 40 to 80 per cent of them abort and degenerate
completely, forming a yellow oily mass. The remaining grains develop
normally and are fertile. These types are all found in plants of pure F.
■virginiana and represent varying degrees of expression of dieciousness.
The diecious condition of the wild species from which the cultivated
forms have been derived probably explains the greater sterility of the
later flowers of the cultivated hermaphroditic varieties, than is found
in those of the pistillates ; if we can accept the origin of the hermaphro-
dites as being from males which have developed partial fertility of the
female organs. The appearance of staminodia showing varying degrees
of development in the cultivated varieties is also the direct result of
dieciousness, while the intermediate types of anthers in the cultivated
forms are of the same nature as those found in wild staminate clones.
It is an interesting fact, in connection with the problem of sex deter-
mination and dieciousness, that where intermediate anthers or stami-
noids occur, either on wild clones or cultivated varieties which are able
656 Journal of Agricultural Research voi. xii, No. 10
to bear normal anthers, they are practically always borne on the primary
flowers, while the anthers produced later have a greater tendency toward
normal development. The tendency toward the production of stami-
nodia is much greater in the early spring than later. On the other hand,
pistil sterility is much more frequent on the later flowers of an inflores-
cence and when fruits set on wild staminate clones it is practically always
on the first flowers of a cluster which open.
A second type of sterility often associated with the above type but due
to a different cause is that which results in aborted microspores and
pollen grains in otherwise normal anthers. Aborted pollen has been
shown to be present in relatively small amounts in pure species of Fragaria,
but appears often in large quantities in many of the cultivated varieties.
This type of abortion has long been recognized in hybrids, and recently
Jeffrey and his students have gone so far as to consider any plant bearing
ever 15 or 20 per cent of this type of pollen a hybrid.
Selfed seedlings and F^ plants of crosses between varieties of cultivated
strawberries are so extremely variable for many factors that it seems
self-evident that they are of hybrid origin, and this is to a great extent
confirmed by what is known of the origin of the numerous cultivated
varieties, many of which are the result of variety crosses, while by far
the larger number are chance seedlings. It thus seems evident that
pollen abortion in the cultivated varieties is due to the same causes
which produce sterility in other hybrids.
As would be expected, there are varying degrees of sterility resulting
from hybridization and varying degrees of irregularities in the stages
which lead up to the final abortion of pollen. There appear in the
literature numerous instances of abortion in both male and female repro-
ductive organs following irregular reduction divisions. The irregular
divisions, especially in the pollen mother cells, result in the production
of more than four cells of unequal size in the tetrad. These produce
microspores of varying sizes, few of which ever come to maturity. Gates
(77, p. 98) pointed out that most of the forms studied by Wille (44)
showing supernumerary cells in the tetrad are either hybrids or have
been under cultivation for some time and are open to the suspicion of
being hybrids. Other plants, some of which are knowTi to be hybrids
while others which have been cultivated as horticultural varieties and
are under suspicion as hybrids, have been studied in more detail by
various workers. Tischler (40, 41, 42), who has done much work with
plants of this type, finds that in hybrids of Ribes spp. and in the sterile
hybrid Mirahilis jalapa X M. iubiflora pollen degeneration usually takes
place following normal divisions. In the hybrids Poientilla tabernae-
montani X P. rubens, Syringa chinensis, and Bryonia alba X B. dioica,
and in three varieties of banana {Musa paradisiaca) having different
chromosome numbers, irregular divisions are common and are always
followed by much pollen abortion. In these banana varieties, the
Mar. II. 1918 Sterility in the Strawberry 657
origins of which are unknown but which dififer cytologically in having
8, 16, and 24 chromosomes as the reduced number, it is significant that
the most frequent irregularities in cell division during reduction, and
most complete pollen sterility occurs in the two varieties having the
greater chromosome numbers. Thus, pollen abortion may or may not
be the result of irregularities at reduction division, but is apparently
related to hybridity and is associated with heterozygosity.
After working with several hybrid plants showing both normal divi-
sions and irregularities during reduction and formation of tetrads,
Tischler (41, p. 144) concluded that —
Die Sterilitat bei Hybriden hangt nicht von irgendwelcher Chromatin repulsion ab.
He concluded further that irregularities during tetrad division can not
be considered as characteristic only of hybrids. He thought that ste-
rility of hybrids was due to the coming together of two sex cells which
did not contain identical developmental tendencies and that these were
expressed at the critical time of the formation of the reproductive organs.
Actual abortion of the grains he thought was due to insufficiency of cyto-
plasm in the enlarged microspores.
One of the most striking cases of sterility following hybridization, the
cytological details of which have been worked out, is that of the hybrid
Drosera longifolia X D. rotundifolia, reported by Rosenberg (jj). The
striking feature of this hybrid is that it is between parents having differ-
ent chromosome numbers, the diploid number of D. longifolia being 40,
while that of D. rotundifolia is only 20. As a consequence the hybrid
contains 30. At reduction division in both the megaspore mother cells
and pollen mother cells there appeared 10 pairs of chromosomes and 10
single ones, the pairs supposedly being made of the 10 D. rotundifolia
chromosomes paired with lo from the D. longifolia parent, while the 10
single chromosomes were the remaining 10 D. longifolia chromosomes.
Reduction division resulted in the separation of the paired ones, these
being drawn regularly to opposite poles. The unpaired chromosomes,
on the other hand, were either drawn to one or the other pole or were
left in the cytoplasm to form another small nucleus. The homoeotypic
divisions took place normally. Following the organization of the micro-
spores within the tetrad, many proceeded to increase in size; in some,
division of the microspore nucleus proceeded normally, and then in
practically all cases abortion of the pollen took place. Following tetrad
formation in the feinale reproductive organs, three of the tetrads usually
aborted, as is common, while the other proceeded to form the egg sac.
Egg-sac formation was carried to various stages, but it was only very
rarely that a perfect egg sac, capable of further development, was formed.
Rosenberg concluded {p. 39) that because of the fact that the micro-
spore division was able to proceed normally, the degeneration of the
pollen grains was not the result of the irregular distribution of chromo-
658 Journal of Agricultural Research voi. xii. No. i©
somes during reduction division, but was due to a lack of cytoplasm.
The abortion of the egg sacs, he again concluded, was not due to irregular
divisions, as all of the divisions following reduction were normal., but was
due to poor nutrition.
Nakao (29) , working on the cytology of certain grain hybrids in which
very striking irregularities in reduction division took place, followed by
complete abortion of the microspore after liberation, concluded that
abortion in this case was due to an insufficiency of cytoplasm previous
to reduction division which resulted in abnormally early division and con-
sequent irregularities. He did not consider why these irregularities
should cause abortion of the microspores.
In view of the conditions existing within an anther at the time of
degeneration of microspores and because of certain genetic results which
can only be explained on the basis of selective elimination of certain
gametic combinations, it is difficult to agree with the conclusions of
Tischler and Rosenberg that degeneration has nothing to do either with
the irregularities or normal repulsion which occur at reduction division
in hybrids.
Although there are striking differences in the regularity with which
reduction takes place in sterile or partially sterile hybrids, there are
certain conditions, in both those which proceed normally and those
which show irregularities, which are alike and must be taken into account
in the consideration of the causes of pollen abortion. These conditions
are as follows :
(i) At reduction division there is a sorting out of the parent chromo-
somes, resulting in new combinations in the daughter cells, the mnuber
of which depends upon the degree of difference between the two parents.
If division proceeds normally, there is an equal number of chromosomes
in each daughter nucleus. If it takes place irregularly, imequal numbers
are found in the resulting daughter cells. In either. case the combina-
tions are new and may or may not contain all of the properties necessary
for perfect metabolism of the cell.
(2) If divisions take place regularly, there is an equal quantitative
and, as far as can be determined, also an equal qualitative division of
cytoplasm between the quadrants of a tetrad. If divisions have pro-
ceeded irregularly, the cytoplasm is divided between the members of the
tetrad in proportion to the amoimt of chromatin which they contain.
In either case at the time of liberation from the tetrad, or if liberation
does not take place, as in Drosera spp., at the period previous to enlarge-
ment, all of the microspores appear normal — that is, they contain an
organized nucleus and are filled with cytoplasm. In Minnesota 3
there is at this time as great uniformity in size and cytoplasmic content
of the individual microspores as is fotmd in entirely fertile plants of
F. virginiana.
Mar. II, 1918 Sterility in the Strawberry 659
(3) It is not until rapid growth of the microspores takes place and the
necessity of active cell metabolism appears that evidences of degenera-
tion appear. The necessity of active metabolism becomes apparent
when it is remembered that the microspores increase 7.5 times their
original volume during this growth period.
(4) There is no specific time at which degeneration of the grains
within a single anther takes place. In most of the sterile forms thus far
studied a series of degenerating stages appear from the first period of
growth of the microspores to the formation of nearly mature pollen.
It is becoming more and more evident that the growth and develop-
ment of plants and animals are directly dependent upon the chromosome
combination which they contain. Boveri {see Morgan, 28, p. 55), in
working with dispermic sea-urchin eggs, found that they very rarely
develop normal individuals, while if separated at the 4-celled stage
normal individuals often developed. This seems dependent upon the
fact that in the first four cells, which are the result of a single division,
the chances of one of the cells receiving at least one of each kind of
chromosome are relatively high and thus, when separated, some may
develop normally. On the other hand, the chances of each of the four
cells receiving one of each kind of chromosome necessary for perfect
development are small; and as a result the individual develops abnor-
mally. Bridges (5) has shown that in the fruit fly Drosophila ampelo-
phila certain variations from the normal chromosome combinations
have a definite effect upon the development of the zygote. Zygotes
containing 3X chromosomes die, while those containing 2X and a Y
chromosome develop normally. Male individuals may develop which
contain an X but no Y chromosome, but were found to be entirely sterile;
while those zygotes which received only a Y chromosome died as did
also those which received 2Y chromosomes, but no X. Zygotes con-
taining 2Y chromosomes plus an X, however, were able to develop into
normal males. Apparently the presence of an X chromosome is neces-
sary for the development of an individual, while the presence of a Y in
males is necessary if the male is to be fertile.
In plants the evidence for the dependence of development upon
chromosome combinations is becoming indisputable if the Mendelian
interpretation of the inheritance of factors is admitted. In F^ progeny
of hybrids, if the parents are homozygous, there is generally as much
uniformity as is shown by either parent, while the Fj progeny shows a
wide range of types, often overstepping the limits of the parents. If
such a variety of types with regard to hardiness, rust resistance, adapta-
bility to various regional and soil conditions, and vigor of the individual
plant are produced in the 2X generation as the result of new chromo-
some combinations, why is it not possible for a similar series to exist in
the iX generation with regard to the ability of the individuals to develop
in a given environment?
66o Journal of Agricultural Research voi. xii, no. lo
I have already pointed out that up to the time of liberation of the
microspores from the tetrad, in the strawberry, cell divisions have
resulted merely in an equal division of the cytoplasm of the original
mother cell between its four granddaughter cells with no evidence of
any metabolic changes resulting in an increase of cytoplasm. A similar
condition exists in those forms which show irregular divisions during
reduction with uneven distribution of cytoplasm between the resulting
cells. Liberation of the microspores from the tetrad marks the end of
the period of dependence of these cells upon the 2X generation, as far
as future growth and development is concerned. At this time there is
no difiference between the conditions surrounding the microspores of a
hybrid and those of a genetically pure individual. Both groups of
microspores are set free in a homogeneous anther sap to complete their
own further development.
The progress of the developmental stages in plants of pure F.
virginiana is characterized by great regularity of development of the
individual microspores within a loculus with regard to rate of enlarge-
ment, time of division of the microspore nucleus, and the subsequent
development of cytoplasm. In fact, there is as great uniformity shown
in these stages as was shown in the stages leading up to the mature
tetrad stage in either F. virginiana or Miimesota 3. Minnesota 3, on
the other hand, shows great irregularities during this developmental
period in rate of growth of the individual microspores, in the time
of division of the microspore nucleus, and in the rapidity of formation
of cytoplasm. This lack of uniformity is in striking contrast to the
uniformity shown in earlier stages of the same plant, while the cells
were dependent on the 2X individual. The liberated microspores of
Minnesota 3 are strikingly uniform in size and cell contents.
During any period, following liberation of the microspore to the
completion of development, microspores or pollen grains may be found
degenerating. As all of the grains within a loculus are free in a
homogeneous nutrient liquid, it seems difficult to believe that the varia-
tions in development can be due to anything but the individual
constitution of the microspores.
Indeed, there is constantly accumulating an increasing amount of
evidence which points to the continual elimination of gametes bearing
certain chromosome combinations. In 1894 Millardet (27) reported on
a series of hybrids between species of strawberry the progeny of which,
he said, formed an exception to the general rule of hybrids, as the specii&c
type of one or the other parent was always produced in the first and later
generations. The specific type shown in the second generation was,
with one exception, the same as that shown by its parent in the F^
generation. Millardet mentions complete sterility in one species of cross
and high percentages of sterility in the F^ generation of some of the other
combinations. Bellair (j) reports that in the tobacco cross Nicotiana
Mar. II. 1918 Sterility in the Strawberry 661
sylvesiris X N. tahacum the F^ generation resembled the A'', tabacum
parent and was partially fertile. From these he was able to obtain F3
plants apparently identical with the two parents and fully fertile. The
reappearance of types similar to the parents in large numbers in the
F3 generation suggests the elimination of gametes containing combinations
which would result in intermediate types.
Detlefsen {10), working with animals, reports results obtained from
cavy crosses which may readily be explained on the basis of the elimina-
tion of certain combinations in the gametes of the males. He crossed
.tame females to wild males. The F^ males were all sterile. The cross F^
female with wild male was not very successful and produced one sterile
male and a sterile female. The F^ females crossed to tame males gave
sterile males of which a few produced some nonfunctional sperms. The
females of this back-cross again crossed to tame males produced males
showing a low degree of fertility. As this process was continued, always
crossing back to tame males, the fertility of the male progeny increased
as they became more nearly homozygous for the tame condition. In the
sixth generation all of the males produced sperms, and 66.7 per cent of
these males were readily fertile. Apparently the more chromosomes of
one parent type which were present in the sperm, the greater its chance
of complete development.
East {13) in a short abstract gives the conditions which he found in
the progency of the partially sterile hybrid Nicotiana rustica humilis X iV.
paniculata. The Fj progeny of this hybrid were very uniform, but only
I to 6 per cent of the female gametes were functional, and 2 to 6 per
cent of the pollen grains were morphologically perfect. In the V^ gener-
ation some perfectly fertile plants were found, many possible Fg combina-
tions were omitted, many more homozygous combinations occurring
than should be expected, and the parent types appeared once in everv
100 to 200 plants, whereas if all of the possible Fj combinations appeared,
the parent types would be much more rare. East considered that the
results might be explained on the basis either of selective elimination of
F2 zygotes or selective elimination of F^ nonfunctional gametes. He con-
sidered further that the elimination of the nonfunctional gametes might
be due to irregularities of chromosome distribution, which scheme seemed
improbable; or the facts might be interpreted without this assumption if
certain conditions were met which are as follows :
If (i) there is a group of chromosomes in each parent that can not be replaced by-
chromosomes from the other parent; if (2) there is a group of chromosomes from each
parent, a percentage of which may be replaced by chromosomes from the other parent,
but where functional perfection of the gametes varies as their constitution approaches
that of the parental forms; if (3) there are other chromosomes that have no effect on
fertility and therefore can promote recombinations of characters in the progeny of
fertile Fg plants; if (4) a naked male nucleus entering the normal cytoplasm of the
egg in the immediate cross can cause changes in the cytoplasm that will affect future
reduction divisions; if (5) this abnormally formed cytoplasm is not equitably dis-
38325°— 18 4
562 Journal of Agricultural Research voi. xii, no. w
tributed in the dichotomies of gametogenesis in the Fj generation; if (6) it follows
from (4) and (5) that Fj zygotes may be formed which are less perfect in their gamete
forming mechanism than those of the Fj generation; and if (7) the heterotypic division
of gametogenesis does not necessarily form two cells alike in their viability.
In the strawberry, in which no irregular distribution of chromatin
occurs, certain of these assumptions — namely i, 2, 3, and 7 — would apply
in explaining partial male sterility in many partially sterile varieties,
but assumptions 4 and 5 and assumption 6, which is dependent upon
them, can hardly be considered applicable; assumption 4 because there
is no cytological evidence that there has been any disturbance of reduc-
tion division; 5, because there is apparently equal distribution of the
mother cell cytoplasm to each member of the tetrad, and 6, because
there is no cytological evidence which would lead one to believe that the
cytoplasm of the zygote had anything to do with the perfection of its
gamete forming mechanism. In the strawberry sterility seems to be
due to the inability of certain chromosome combinations to use the food
material in which they are embedded in the growth and development of
the liberated spore to a ripe pollen grain.
If we can accept the hypothesis of pollen abortion being due in hybrids
to certain chromosome combinations affecting the normal metabolism
of the microspore in its development, the question at once presents
itself: Is pollen abortion the result of the presence of one particular
chromosome or of certain combinations of two or more, or do all of the
chromosomes play some part in it ? This question can not be answered
from the facts so far obtained in the strawberry because of the heterozy-
gous condition of the material which has been used; but there is other
published evidence which throws some light on this question.
Belling (4) has made a very careful study of partial sterility of hybrids
between four species of the "bean" Stizolobium. He found that the F^
plants, of those crosses in which the velvet bean {S. deeringianum) was
used as one parent continually, aborted one-half of the pollen grains and
one-half of the egg sacs. Of the second generation plants one-half were
completely fertile and one-half partially sterile, as in Fj. The progeny
of fertile F2 plants continued to be fertile, while the progeny of the
partially sterile plants were one-half fertile and one-half partially sterile.
Belling explained these results on the basis of the presence in the velvet
bean of the factor K, which was not present in the other three varieties.
These three, however, contained the factor L not present in the velvet
bean. The presence of either K or L he assumed to be necessary for the
normal development of either egg sacs or pollen grains, the presence of
both K and L causing abortion. We may extend this working hypoth-
esis slightly and put it on a chromosome basis, in which case we must
consider that Belling's factors K and k form one allelomorphic pair,
being situated in a certain loculus of a specific chromosome in the velvet
and the other three varieties of beans, respectively, and that the factors
Mar. II, 1918 Sterility in the Strawberry 663
L and 1 form an allelomorphic pair located in a definite but different chro-
mosome of the three varieties and the velvet bean, respectively. Then,
to follow out Selling's scheme, the presence of the chromosomes bearing
the factors K and L in the same member of a tetrad causes abortion, and
likewise the presence of both chromosomes lacking the factors K and L
causes abortion. Belling states that abortion of the microspores takes
place in the vacuolate stage and that there are no intermediates between
the completely aborted and the most perfect grains, thus strengthening
the idea that in this case no more than two chromosomes have to do
with abortion.
Another set of studies which point to one instead of two chromosomes
being the cause of pollen abortion are those of Shull {35, 36) on the
inheritance of sex and of a sex-linked factor in Lychinis dioica. In
the first of these studies Shull showed that very probably L. dioica
9 is homozygous for the sex determining factors, while L. dioica $ is
heterozygous. In crossing these forms an approximate ratio of i pis-
tillate to I staminate usually resulted, but with nearly always a slight
excess of pistillate plants, suggesting, if the females are homozygous
for sex, an elimination of a portion of the male gametes bearing the
determiner for maleness. In a later study he was able to show that the
determiner for maleness was linked with a factor for narrow leaves while
in normal plants the determiners for femaleness were linked with broad
leaves. In a narrow-leaved mutant male found by Baur he showed
that the determiner for femaleness as well as maleness was linked with
the narrow-leaved determiner. It was as a result of the discovery of
this homozygous (for leaf width only) , narrow-leaf male that the factor
for narrow leaves linked to maleness was able to be discovered as, being
a recessive character and always in a heterozygous condition, it was
hidden in normal males.
When these homozygous narrow-leaved males were used in crosses
with either normal broad-leaved females or heterozygous females,
there was always produced a great excess of males, the females appear-
ing only in very small numbers. These results were apparently in con-
tradiction to those previously obtained in which females were more
abundant. Shull gave no explanation of these irregularities. They
suggest, however, that there is a fairly constant elimination of certain
gametes. A study of all of Shull's results, with this idea in mind, in-
dicates that an explanation based on the elimination of certain male
gametes will cover all cases of irregularity so far reported by him except
the nonappearance of homozygous hermaphrodites and of heterozygous
hermaphrodites containing male determiners. These two instances, if
we may draw analogies between plants and animals, are of the same
nature as the YY zygotes in species of Drosophila, and die (Bridges, 5).*
' Shull (55-56) has shown that the hermaphrodites have undoubtedly been derived from males; and
therefore the presence of two hermaphrodites or a male and an hermaphrodite determiner would be anal-
ogous to the presence of two male determiners.
664
Journal of Agricultural Research
Vol. Xri. No. 10
In every other instance in which irregularities in sex ratios occurred, a
male or hermaphroditic parent was used in which the condition for nar-
row leaves was linked with either a determiner for maleness, femaleness,
or hermaphroditeness. If the irregularities were relatively slight, as
was the case when normal broad-leaved males and females were crossed,
maleness and narrow leaf were linked. A partial elimination of these
male gametes would produce the actual results obtained. Hermaphro-
dites of Lychnis dioica acted in the same manner as the males. Hermaph-
rodites of Melandrium album, which we may assume to bear the narrow-
leaf and hermaphrodite determiners linked, as they have undoubtedly
been derived from males, acted in the same manner as the narrow-
leaved males — that is, they produced only females when crossed to
normal broad-leaved females, in place of a i to i ratio. These results
can be explained on the assumtion of complete elimination of the male
gametes of the M. album hermaphrodite, which carry the hermaphrodite
mutant and its linked factor, narrow leaf, and in the case of the narrow-
leaved L. dioica males of the nearly complete elimination of the male
gametes bearing the mutant factor for femaleness and narrow leaf.
In all cases it seems that the factor for narrow leaves has an inhibiting
action on the formation of the male gametes and results in the partial
or complete elimination of them. Elimination in the normal males is
not complete; otherwise this line would long ago have disappeared. In
the mutants in which narrow leaf is linked with femaleness, elimination
of male gametes bearing this mutant factor is nearly complete. Shull
has also given much evidence which shows that there is also some elim-
ination of female gametes bearing this mutant factor but to a less extent
than in the males. Such partial elimination of the female gametes was
shown in the cross heterozygous broad-leaved female (of the formula
FBFb) by a narrow-leaved male Fbfb, which produced two heterozygous
broad-leaved females, no homozygous narrow-leave'd females (the nearly
complete absence of both of these classes evidently being due to the
elimination of male gametes bearing the factor for narrow leaves and
femaleness), 630 broad-leaved males, and 463 narrow-leaved males. In
this cross all four types should have appeared in equal numbers.* The
1 The effect of the linked factors "narrow leaves" and "femaleness" on the productioa of male and
female gametes can be most readily seen by the use of'this simple diagram
(fFb
fb
?
Fb
fb
FB
FB
FB
3
6jo
Fb
fb
Fh
Fb
Fb
0
463
in which the male gametes are placed on the upper side of the square and the female gametes on the left
side, while the number of each of the types of progeny are placed within the small squares, with their
respective gametic combiaations.
Mar. II, 1918 Sterility in the Strawberry 665
inequality of the last two classes must be due to inequality in production
of the two kinds of female gametes FB and Fb ; the latter, which carries
femaleness and narrow leaf linked, appearing less frequently than FB.
Apparently with the suppression or loss of the determiner for broad
leaves in the sex chromosome, there has also been a partial suppression
of a factor necessary for the normal development of male and to a less
extent of female gametes. In personal correspondence with Dr. ShuU
he informs me that there is actually a great deal of pollen sterility in
the narrow-leaved Lychnis male.
CONCLUSIONS
(i) The flowers of Fragaria are pentamerous with regard to all parts
except pistils. The stamens are arranged in three whorls; the outer
parapetalous series of 10 stamens, the middle antipetalous, short fila-
mented series of 5, and the inner antisepalous series of 5. Increases in
stamen number are due to the addition of 5, or a multiple of it, to either
the antipetalous or the antisepalous series. Decreases in stamen number
are due to the loss of first the antipetalous and next the antisepalous
series. Apparently the parapetalous series are permanent. Decrease
in stamen number is in no way related to dieciousness.
(2) There is a positive correlation between flower position, flower part
number, and size of fruit in the strawberry.
(3) The wild American species of strawberry, from which the cultivated
varieties have been derived, are for the most part diecious. The pistillate
plants bear staminodia, which rarely develop as far as the pollen mother
cell stage, and the staminate plants bear pistils which superficially ap-
pear to be perfect but which are only occasionally functional. In a few
wild clones of F. virginiana, which appear to be sterile, pollen develop-
ment is carried as far as the tetrad division or slightly beyond this to
the liberation of the microspores, when complete disintegration of the
anther contents to an oily mass takes place. In other instances a por-
tion of the microspores develop normally while the remainder within the
same anther disintegrate, while in other clones shortly after liberation,
and following a slight growth of the microspores, complete abortion of
the same type as that found in hybrids takes place. These anther types,
in wild clones, all appear to be various expressions of a tendency toward
dieciousness and are not the result of hybridization. Similar anther
types are common in certain cultivated varieties, on the early flowers of
an inflorescence, and especially on those appearing early in the season.
(4) There is a correlation between flower position and fertility of
pistils, fertility decreasing in the later flowers of an inflorescence. Pistil
sterility is expressed in the production of irregularly shaped berries or
entirely sterile flowers. Sterility of the later flowers of an inflorescence
is more general in hermaphrodites than in pistillates, suggesting that
the hermaphrodites have been derived from staminates of the diecious
wild forms.
666 Journal of Agricultural Research voi. xii, no. io
(5) The appearance of considerable amounts of aborted pollen in
wild plants of F. virginiana and F. americana is rare except in anthers
of the intermediate type. Most cultivated varieties produce considerable
amounts of aborted pollen of the type common in hybrids. The percentage
of aborted grains is not constant in the individual flowers of a variety,
and neither is it constant in the individual anthers of a single flower, as
just as great variations appear within the anthers of a flower as are
shown by composite pollen samples of individual flowers,
(6) In those varieties producing high percentages of aborted grains
a portion, at least, of the morphologically normal pollen grains are
functional, as shown by germination and bagging tests. There is no
evidence of a physiological self-sterility in the strawberry.
(7) In the partially sterile variety Minnesota 3 pollen development
is carried on normally up to the liberation of the microspores from the
tetrad. At this time all of the microspores appear normal and alike.
Following liberation, variations in rate of growth, time of division of the
microspore nucleus, and ability of the individual microspores to develop
normally are shown. At all stages, during this growth period micro-
spores were found in various stages of abortion. F. virginiana exhibits
as great regularity during this, growth period as is shown in the stages
leading up to liberation of the microspores.
(8) Liberation of the microspores from the tetrad marks the beginning
of an independent gametophytic generation, so far as the metabolic
processes of growth are concerned. The individual microspores float in
a homogeneous nourishing medium provided by the sporophyte, but the
use of this food material in cell metabolism depends entirely upon the
individual organization of the microspores.
(9) Specific chromosome combinations have been shown by various
investigators to be a potent factor in the developm^t or lack of develop-
ment of individual plants or animals. In plants heterozygous for a
number of factors, as are the varieties of strawberries, numerous new
chromosome combinations occur for the first time in the microspores.
The varying rates of growth, time of microspore division, ability to
increase the cytoplasm, and inability in many cases to develop normally
seem to be the outward expression of the differential ability of these
new chromosome combinations to carry on cell metabolism.
LITERATURE CITED
(i) Allen, C. E.
1905. NUCLEAR DIVISION IN THE POLLEN MOTHER-CELLS Olf LILIUM CANADENSE.
/n Ann. Bot., v. 19, no; 74, p. 189-258, pi. 6-9. Literature cited,
p. 252-256.
(2)
1905. DAS VERHALTEN DER KERNSUBSTANZEN WAHREND DER SYNAPSIS IN
DEN POLLENMUTTERZELLEN VON LILIUM CANADENSE. In Jahtb.
Wiss. Bot. [Pringsheim], Bd. 42, Heft i, p. 72-82, pi. 2.
Mar. II, 1918 Sterility in the Strawberry 667
(3) Bellair, Georges.
1913. RECROIS^ES ENTRE ELLES DEUX ESP^CES QUI SE SONT D^GAG^ES d'uN
HYBRIDE n'OB^ISSENT PLUS A LA LOI MENd6lIENNE DE LA DOMINANCE.
In Compt. Rend. 4th Conf. Intern. Genetique, 1911, p. 201-203.
(4) Bellinc, John.
1914. THE MODE OP INHERITANCE OP SEMI-STERILITY IN THE OFFSPRING OP
CERTAIN HYBRID PLANTS. In Ztschr. Induk. Abstam. u. Vererbungsl.,
Ed. 12, Heft 5, p. 303-342, 17 fig. Literature cited, p. 341-342.
(s) Bridges, C. B.
1916. NON-DISJUNCTION AS PROOF OP THE CHROMOSOME THEORY OP HEREDITY.
In Genetics, v. i, no. i, p. 1-52; no. 2, p. 107-163, 9 fig., i pi. Bib-
liography, p. 162-163.
(6) BUNYARD, E. A.
1914. THE HISTORY AND DEVELOPMENT OP THE STRAWBERRY. In Jour Roy.
Hort. Soc. [London], v. 39, pt. 3, p. 541-552, fig. 164-168. Bibfiog-
raphy, p. 551-552-
(7) Castle, R. L.
1904. POMOLOGY AS A STUDY. In Jouf. Roy. Hort. Soc, V. 39, pt. 1/3, p.
146-160.
(8) Cole, Ruth D.
191 7. IMPERFECTION OP POLLEN AND MUTABILITY IN THE GENUS ROSA. In
Bot. Gaz., V. 63, no. 2, p. 110-123, pL 4-6.
(9) Darrow, G. M.
1916. SOUTHERN STRAWBERRIES. In Jour. Heredity, v. 7, no. 12, p. 531-
540, 6 fig.
(10) Detlepsen, J. A.
1914. GENETIC STUDIES ON ACAVY SPECIES CROSS. Carnegie Inst. Washington
Pub. 205, 134 p., 2 fig., 10 pi. Bibliography, p. 129-132.
(11) Digby, L.
I914. A CRITICAL STUDY OF THE CYTOLOGY OF CREPIS V^RENS. In Arch. Zell-
forsch., Bd. 12, Heft i, p. 97-146, pi. 8-10. Bibliography, p. 138-141.
(12) DORSEY, M. J.
1914. POLLEN DEVELOPMENT IN THE GRAPE WITH SPECIAL REFERENCE TO
STERILITY. Minn. Agr. Exp. Sta. Bui. 144, 60 p., 4 pi. Bibliog-
raphy, p. 50-60.
(13) East, E. M.
1915. AN INTERPRETATION OP STERILITY IN CERTAIN PLANTS. In Proc. Amet.
Phil. Soc, V. 54, no. 216, p. 70-72.
(14) Fletcher, S. W.
191 5. fragraria virginian a in the evolution op the garden straw-
BERRY IN NORTH AMERICA. In Proc. Soc Hort. Sci., 1915, p. 125-137.
(15)
1917. STRAWBERRY-GROWING. 325 p., 23 fig., 24 pi. New York.
(16)
I917. THE STRAWBERRY IN NORTH AMERICA. 234 p., 26 fig. NeW York.
(17) Gates, R. R.
1907. pollen development in hybrids op oenothera lata x o. lamarck-
lANA, AND ITS RELATION TO MUTATION. In Bot. GaZ., V. 43, nO. 2, p.
81-115, pi. 2-4. Literature cited, p. 110-112.
(18)
1915. THE MUTATION FACTOR IN EVOLUTION. 353 p., 114 fig. Bibliography,
p. 323-342.
668 Journal of Agricultural Research vd.xii, no. lo
(19) Georgeson, C. C.
1907. BRIEF SUMMARY OF WORK. SITKA STATION. In Alaska AgT. Exp. Sta.
Ann. Rpt., 1906, p. 9-15, 2 pi.
(20)
1911. WORK AT SITKA STATION. In Alaska AgT. Exp. Sta. Ann. Rpt., 1910,
p. 10-29, 4 pi.
(21) GooDSPEED, T. H.
1913. ON THE PARTIAL STERILITY OF NICOTIANA HYBRIDS MADE WITH N. SYL-
VESTRIS AS A PARENT. I. In TJniv. Cal. Pub. Bot., v. 5, no. 4, p.
189-198.
(22) and Kendall, J. N.
I916. ON THE PARTIAL STERILITY OF NICOTIANA HYBRIDS MADE WITH N. SYL-
VESTRIS AS A PARENT. III. AN ACCOUNT OF THE MODE OP FLORAL
ABSCISSION IN THE Fj SPECIES HYBRIDS. In Univ. Cal. Pub. Bot.,
V. 5, no. 10, p. 293-299.
(23) Hoar, C. S.
1916. sterility as the result of hybridization and the condition of
POLLEN IN RUBUS. In Bot. Gaz., v. 62, no. 5, p. 370-388, pi. 10-12.
^ Literature cited, p. 386-387.
(24) JanczEwski, Ed.
1908. SUR LES ANTH^RES ST^RILES DES GROSEILLIERS. In Bul. Acad. Sci«
Cracovie, 1908, p. 587-596, i pi.
(25) Jeffrey, E. C.
1914. spore conditions in hybrids and the mutation hypothesis op de
VRiES. In Bot. Gaz., v. 58, no. 4, p. 322-336, pi. 22-25.
(26) JESENKO, F.
1913. SUR UN HYBRIDE FERTILE ENTRE TRITICUM SATIVUM 9 9 (blE MOLD-
squarehead) ETSECALECEREALE^ (SEIGLE DE PETKUS). /w Compt.
Rend. 4tli Conf. Intern. Genetique, 1911, p. 301-311.
(27) Millardet, a.
1894. NOTE SUR l'HYBRIDATION SANS CROISEMENT OU FAUSSE HYBRIDATION.
In Soc. Sci. Phys. et. Nat. Bordeaux, s. 4, t. 4, cahier 2, p. 347-372*
I fig.
(28) Morgan, T. H.
1914. HEREDITY AND SEX. ed. 2, 248 p., 121 fig. N^sv York.
(29) Nakao, M.
191 1. cytological studies on the nuclear di\^sion of the pollen mother-
CELLS OF SOME CEREALS AND THEIR HYBRIDS. In JotUT. Col. Agr.
Tohoku Imp. Univ., v. 4, no. 3, p. 173-190, 4 pi.
(30) Overton, J. B.
1905. tJBER REDUKTIONSTEILLUNG IN DEN POLLENMUTTERZELLEN EINIGER
DiKOTYLEN. In Jahrb. Wiss. Bot. [Pringsheim], Bd. 42, Heft i, p.
121-153, pi. 6-7.
(31) Richardson, C. W.
1914. A preliminary NOTE ON THE GENETICS OF FRAGARiA. In Jout. Gene-
tics, V. 3, no. 3, p. 171-177, 4 fig-, pl- 7-
(32) Rimpau, W.
1891. KREUZUNGSPRODUKTE LANDWIRTSCHAFTLICHER KULTURPFLANZEN. In
Landw. Jahrb., Bd. 20, p. 335-339-
(33) Rosenberg, O.
1909. cytologische und morphologische studien an drosera longlfolia
X ROTUNDIFOLIA. K. Svcnsk. Vetensk. Akad. Handl., Bd. 43, No.
II, 64 p., 33 fig., 4 pi. Literaturverzeichnis, p. 61-62.
Mar. II, 1918 Sterility in the Strawberry 669
(34) Rydberg, p. a.
1898. ▲ MONOGRAPH OF THE NORTH AMERICAN POTENTILLEAE. Mem. Dept.
Bot. Columbia Univ., v. 2, 223 p., 112 pi.
(35) Shull, G. H.
I910. INHERITANCE OP SEX IN LYCHNIS. In Bot. Gaz., V. 49, HO. 2, p. IIO-
125, 2 fig. Literature cited, p. 125.
(36)
1915. SEX-LIMITED INHElilTANCE IN LYCHNIS DIOICA L. In Ztschr. Induk.
Abstam. u. Vererbimgsl., Bd. 12, Heft 5, p. 265-302, 5 fig. Literature
cited, p. 299-302.
(37) Standish, L. M.
1916. what is happening to the hawthorns? In Jour. Heredity, v. 7, no.
6, p. 266-279, fig. 7-17. Literature cited, p. 279.
(38) Stevens, N. E.
1912. observations on heterostylous plants. In Bot. Gaz., v. 53, no. 4,
p. 277-308, pi. 21-23. Literature cited, p. 305-307.
(39) Strauburger, Eduard.
1909. ZEITPUNKT DER BESTIMMUNG DES GESCHLECHTS, APOGAMIE, PARTHENO-
GENESIS UND REDUKTiONSTEiLUNG. In his Histologische Beitrage,
Heft 7, 124 p., 3 pi. Jena.
(40) TiSCHLER, G.
1906. tJBER DIE ENTWICKLUNG DES POLLENS UND DER TAPETENZELLEN BEl
RiBESHYBRiDEN. In Jahrb. Wiss. Bot. [Pringsheim], Bd. 42, Heft 4,
P- 545-578, pi. 15- Literatur-Verzeichnis, p. 575-577-
(41)
1908. ZELLSTUDIEN AN STERILEN BASTARDPFLANZEN. In Aich. Zellforsch.,
Bd. I, Heft I, p. 33-151, 120 fig. Literatur, p. 104-106, 147-151.
42)
1910. UNTERSUCHUNGEN tJBER DIE ENTWICKLUNG DES BANANEN-POLLENS. I.
In Arch. Zellforsch., Bd. 5, Heft 4, p. 622-670, 4 fig., pi. 30-31. Ver*
zeichnis der zitierten Literatur, p. 67.
(43) VallEau, W. D.
1916. INHERITANCE OF SEX IN THE GRAPE. In AmcT. Nat., V. 50, no. 597,
P- 554-564- Literature cited, p. 563-564.
(44) WiLLE, N.
1886. UEBER DIE ENTWICKLUNGSGESCHICHTE DER POLLENKORNER DER ANGIO-
SPERMEN UND DAS WACHSTHUM DER MAMBRANEN DURCH INTUSSUS-
CEPTION. Forhandl. Vidensk. Selsk. Christiania, no. 5, 71 p., 3 pi.
Original not seen.
PLATE B
Minnesota 3 :
1. — Pollen mother cell previous to synapsis. X 2,000.
2. — Presynapsis in the pollen mother cell showing loops extending out from
synaptic mass. X 2,000.
3. — ^Two loops and portion of a loop extending from the presynaptic mass. Same
stage as figure 2. X 3,280.
4. — Synapsis in a pollen mother cell. X 2,000.
5. — A postsynaptic stage. The synaptic mass is unraveling into bivalent loops.
X 2,000.
6. — Open spireme stage. X 2,000.
7. — A presegmentation stage of the spireme. X 2,000.
8. — A portion of a bivalent spireme thread of the same stage as figure 7. X 3)28o.
9. — Segmentation of the bivalent spireme into chromosome pairs. The cytoplasm
is roxmding up and is partially surrounded by a gelatin-like sheath. X 2,000.
10. — A portion of the bivalent spireme during segmentation. X 2,340.
II, 12. — Chromosome pairs during the contraction period following segmentation.
X 2,000.
13. — Individual chromosome pairs showing various figures commonly formed during
contraction. X 3.280.
14. — Diakenesis in the pollen mother cell. X 2,000.
15. — Multipolar spindle stage of pollen mother cell. X 2,000.
16. — Early anaphase of the heterotypic division. X 3,280.
(670)
Strawberry
Plate B
\
A'^'
\.i
' !^*.»^C
4
13
12
I?
^
15
^
8
16
Journal of Agricultural Research
Vol. XII, No. 10
I B-einHun CO wash' d c
PLATE C
I. — Late anaphase of the heterotypic division. The cytoplasm is rounding up
from the mother cell wall and is partially surroimded by a thick gelatin-like sheath.
X 2,000.
2. — Chromosomes on the equatorial plate of the homeotypic division.
3. — A portion of an anther in the tetrad stage, showing the microspores embedded
in the gelatin-like sheath. The original mother cell walls are still present.
4. — A tetrad at the same stage as those shown in figure 3. The mother cell wall
is not shown. X 2,000.
5. — A microspore shortly after liberation from the tetrad. X 2,000.
6. — A liberated microspore in which growth has commenced. X 2,000.
7. — A later stage than figure 6, showing the slight thickening of the wall and
the irregularities due to growth of the wall. X 2,00c.
8. — Microspore growth completed previous to division of the microspore nucleus.
X 2,000.
9. — A section through a microspore nucleus in prophase showing the continuous
univalent spireme. X 3,280.
. 10. — Another section of the same nucleus, showing the first stages of the disappear-
ance of the nucleolus. The nucleolar strands are attached to the spireme. X 3,280.
II. — Metaphase of the division of the microspore nucleus. A spindle in this posi-
tion results in the nuclear arrangement shown in figure 15. The thickened extine is
shown. X 2,000.
12. — F. virginiana. Anaphase in the division of the microspore nucleus. The
spindle lying parallel to the wall results in the nuclear arrangement shown in figiu-e 14.
X 2,000.
13. — Telophase of the division of the microspore nucleus. The wall which eventu-
ally surrounds the generative nucleus is not always apparent at this time.
14. — A later stage than figure 13 in which the generative cell has been definitely
cut off. X 2,000.
15. — A yotmg pollen grain shortly after division, showing an increase in cytoplasm
content. The thickened extine is shown. X 2,000.
16. — End view of a pollen grain showing the pattern of the laminate layers shown
in figure 15 and Plate D, figures i, 6, and 15. The arrows mark the ends of the three
sutures which bear the germ pores.
Sterility in tlie Strawberry
Plate C
:'?^
c
'^«^
10
m
II
12
14
15
16
#
Journal of Agricultural Research
Vol. XII. No. 10
A B eKAHAM CO. WASHf DC.
PLATE D
I. — Nearly mature pollen grain. The central body is the vegetative nucleus
while the other is the generative cell. X 2,000.
2. — Mature pollen grain. The extine is not shown. The killing fluid causes the
dry folded grains to become spherical. X 2,000.
3. 4> 5. 7- — Various types of degenerate microspores from anthers bearing
microspores of the stage shown in Plate C, figure 6. X 2,000.
6. — An aborting microspore from an anther containing half-grown microspores.
X 2,000.
8. — An aborting microspore of the same type as that shown in figure 6 from an
anther containing nearly full-grown microspores as in Plate C, figure 8. As ia Plate
D, figiu-e 6, the cytoplasm and nucleus still appear normal. X 2,000.
9, II. — Microspores of the same types and same age as figures 6 and 8, in which
degeneration has proceeded farther X 2,000.
10. — An aborted microspore from an anther containing microspores of the stage
shown in Plate C, figure 8. Apparently abortion took place shortly following libera-
tion from the tetrad.
12. — An early stage of degeneration in a full-grown i-nucleate microspore. X 2,000.
13. — An early stage of degeneration in a full-grown i-nucleate microspore. A
ntmiber of normal microspores in this anther are already dividing.
14. — An aborting microspore containing an. abnormally small amoimt of light
staining cytoplasm; from an anther containing i- and 2-nucleate microspores. X 2 ,000.
15. — An aborted microspore from an anther containing i- and 2-nucleate micro-
spores. Apparently this is a late stage of the type of degeneration shown in figures
6 and 8. X 2,000.
16. — An aborted microspore containing very scant cytoplasm. The nucleus has
completely degenerated and degeneration of the cytoplasm has begxm.
Sterility in the Strawberry
PLATE D
t;--,--:i.u6. j-- _
14
15 ""'^V^Wv./
16
Jniirnal of Atrriculturai Research
Vol. XII, No. 10
A a euHAH CO. wiLSHf o c.
PLATE E
I, — A slightly more advanced stage of the condition shown in Plate D, figure i6.
2. — An early stage in the abortion of a full-grown i-nucleate microspore. X 2,000.
3. — An early stage of abortion directly following microspore division. X 2,000.
4. — A full-grown i-nucleate microspore containing very scant light-staining cyto-
plasm; from an anther containing i- and 2-nucleate microspores. X 2,000.
5. — Another type of degeneration of a full-grown I-nucleate microspore. X 2,000.
6. — An aborted microspore fotmd among i- and 2-nucleate microspores. X 2,000.
7. — A later stage of the type of degeneration shown in Plate D, figure 13; from
an anther containing microspores of the stage of development shown by Plate C,
figure 15. X 2,000.
8. — Degeneration of the generative cell shortly after division. The vegetative
nucleus and cytoplasm are still normal. X 2,000.
9, 10. — Common types of aborted microspores found with mature pollen. Evidently
abortion took place before the division of the microspore nucleus. X2,ooo.
II. — An aborted microspore, of the same type as that shown in figure 7. Found
with mattu"e pollen. X 2,000.
12. — A pollen grain showing abortion of the generative cell and an abnormal vacuo-
late condition of the cytoplasm. X 2,000.
13. — A later stage of the type of degeneration shown in figure 8. The vegetative
nucleus and cytoplasm are normal. X 2,000.
Sterility in the Strawberry
Plate E
A
(a
-^
13
10
Journal of Agricultural Research
Vol. XII. No. 10
1 a SHkHt 1 CO . WASH " 0 c
38325"— 18 5
PLATE 35
A. — ^Tertiary flower of the pistillate variety, Minnesota roiyX Progressive — 13-40,
showing prominent staminodia.
B, C. — Primary and secondary flowers of the perfect variety, Minnesota loiyXPro-
gressive~9-24; B showing intermediate and C perfect anthers.
D, E, F. — Two primary and a secondary flower of the perfect variety, Minnesota
loiyXProgressive— 2-55, showing pistillate, intermediate, and perfect types of
flowers.
G, H, I, J. — Flowers from the perfect variety, Minnesota 1017X Progressive— 32-1.
G and H are primary and secondary flowers, respectively, and are pistillate; I a
secondary imperfect flower with a few normally developed stamens and J a tertiary
perfect flower.
sterility in the Strawberry
Plate 35
Journal of Agricultural Research
Vol. XII, No. 10
sterility in tlie Strawberry
Plate 36
Journal of Agricultural Research
Vol. XII, No. 10
PLATE 36
A, B, C, D. — Cross sections of two loculi of staminodia of the pistillate varie-
ties, Crescent, Columbia, Minnesota loi 7 X Progressive — 11-59, and Seedling 140,
respectively.
E. — Degeneration of the tetrads in an intermediate anther of Fragaria virginiana.
F, G. — Later stages of the condition shown in figure E.
H. — A portion of an intermediate anther from the first flower of Minnesota 3.
EFFECT OF NITRIFYING BACTERIA ON THE SOLU-
BILITY OF TRICALCIUM PHOSPHATE^
By W. P. Kblley^
Professor of Agricultural Chemistry, Graduate School of Tropical Agriculture and
Citrus Experiment Station, University of California
INTRODUCTION
The solution of tricalciurn phosphate and the chemical changes through
which it passes in soils are subjects of special interest. The phosphorus
compounds of soils have been mainly derived from tricalcium phosphate,
and are relatively insoluble in water. A considerable part of the phos-
phorus of soils probably actually occurs as tricalcium phosphate. The
soluble phosphates of processed fertilizer also become relatively insoluble
soon after being mixed with soil, some of which may be converted into
tricalcium phosphate. In addition, tricalcium phosphate, in the form
of untreated rock phosphate, has been recommended as a fertilizer at
various times in the past and at present is being applied in considerable
amounts in different parts of America.
While tricalcium phosphate is relatively insoluble in water, it is well
known that this substance is notably soluble in water saturated with
carbon dioxid. For this reason it has long been suspected that the
carbonic acid of soils promotes the solution of the phosphates present.
Likewise the increased fertilizing effect resulting from the application
of rock phosphate in conjunction with decaying organic matter has
commonly been assumed to be due to the solvent effects of the carbonic
acid, and possibly other organic acids, that are formed in the decom-
position of the organic matter.
As a rule, however, investigators have not been able to detect any
increase in the solubility of rock phosphate when left in contact with
decaying organic matter. For example, Lupton ((5)^ McDowell (7),
Truog (9),Tottinghara and Hoffmann (8), and various European workers*
found, as a result of composting rock phosphate with various fermenting
mixtures, that in no case more than slight increases in the solubility of
the phosphate took place and in certain instances decreases in solubiUty
were noted.
> Paper No. 45, University of California Graduate School of Tropical Agriculture and Citrus Experiment
Station, Riverside, Cal.
» Credit is due Mr. A. B. Cummins for assistance in this investigation.
' Reference is made by number (italic) to " Literature cited," p. 685.
* A very complete bibliography of this subject is given by Lipman, McLean, and Lint (5).
Journal of Agricultural Research, Vol. Xil, No. 10
Washington. D. C. March n,' 1918
n* ,, , KeyNo. Cal. 18
(671)
672 Journal of Agricultural Research voi. xii. No. 10
A number of investigators have shown that the solution of tricalcium
phosphate may be efifected by the biochemical oxidation of sulphur.
Recently Lipman, McLean, and Lint (5) found that large amounts of
rock phosphate may be made soluble by the sulphur bacteria in fer-
menting mixtures containing elemental sulphur.
The effects of the nitrifying bacteria on the solubility of tricalcium
phosphate have recently been investigated by Hopkins and Whiting (j).
They found that the nitrite bacteria (Niirosomonas spp.) have the power
of oxidizing ammonium sulphate in solution cultures containing trical-
cium phosphate but no carbonate or free base, and that under these con-
ditions the nitrous acid and sulphuric acid, formed from the ammonium
sulphate, attacked the tricalcium phosphate and rendered notable
amounts of phosphorus and calcium soluble in water. Similar effects
were found when ammonium nitrate was substituted for ammonium sul-
phate. By calculation they found, as an average of 13 tests, that for
every 56 pounds of nitrogen oxidized, 115 pounds of phosphorus and 211
pounds of calcium were made soluble in water. They also found that no
change in the solubility of tricalcium phosphate takes place as a result
of the action of the nitrate bacteria (Nitrobacter spp.). As pointed out
by them, the oxidation of nitrite to nitrate does not necessitate an in-
crease in acidity, but is merely a matter of adding an atom of oxygen to
the nitrite.
Hopkins and Whiting have discussed their views regarding the prac-
tical bearing of these experiments at considerable length, and have as-
signed special importance to the nitrite bacteria as agents in promoting
the solution of rock phosphate in the field. In commenting on these
investigations Davenport ^ even suggested that the importance of the
nitrite bacteria as agents in the solution of rock phosphate, is on a par
with that of the legume bacteria in nitrogen fixation.
Since the formation of nitrous acid is commonly considered to be an
essential step in the nitrification process and as nitrification is generally
active in productive soils, it is at once apparent that relatively large
amounts of phosphate will be made soluble by this group of bacteria,
provided the reactions that take place in soils be similar to those found in
solution cultures.
The practical importance of phosphates in agriculture and the general
interest in the several phases of the nitrification process justify further
investigation of this problem. Accordingly, the writer has made some
studies on it at the University of California Citrus Experiment Station.
In these studies both soil and sand cultures have been employed. The
formation of nitrate and the solubility of calcium and phosphoric acid in
water were used in this investigation as measures of the biochemical
action of the nitrifying organisms.
• Foreword to the publication by Hopkins and Whiting (3).
Mar. II, 1918 Nitrifying Bacteria and Tricalcium Phosphate 673
EXPERIMENTAL RESULTS
The soil used in this investigation was drawn from one of the plots
(F) now being used in a fertilizer experiment with Citrus trees. This
plot has been treated annually for 10 years with light applications of
stable manure, but no commercial fertilizer or lime has been applied
to it. The soil is a light sandy loam of granitic origin, the coarser par-
ticles of which are composed largely of granite. It contained 8.5 p. p. m.
of nitric nitrogen, and 0.188 per cent of total phosphorus pentoxid
(P2O5), of which 17.4 p. p. m. were soluble in water when the experiment
was begun. The total carbonate (CO3), as determined by the Gaither
(2) method, amounted to 0.03 per cent, but the sample was free from
water-soluble carbonate (CO3).
Portions of 2,000 gm. each of fresh soil were weighed into half -gall on
(1.89 Hters) fruit jars. A solution of ammonium sulphate (c. p.) was
added to certain portions at rates supplying o.oi gm. of nitrogen per
100 gm. of dry soil. To other portions an equal quantity of nitrogen was
added in the fonr) of dried blood. Still other portions were employed with-
out the addition of any nitrogenous substance. Baker's analyzed trical-
cium phosphate was added in certain cases at the rate of o.io gm. per 100
gm. of soil and calcium carbonate (c. p.) at the rate of 0.25 gm. per 100
gm. of soil. The experiments were made in duplicate. After a thorough
mixing, adding suitable amounts of water and mixing again, the jars
were loosely covered and incubated at room temperature.
It is, of course, well known that the purest tricalcium phosphate is
somewhat soluble in water and that ammonium sulphate affects the
solubility of certain soil constituents, notably calcium, without the inter-
vention of bacteria. Consequently it was deemed necessary to deter-
mine the solubility of calcium and phosphoric acid after the above-
named substances had been mixed with the soil, but before sufficient time
had elapsed to permit measurable bacterial action. It is obvious that
the amounts of soluble calcium and phosphoric acid present in the soil
at the beginning of the experiment should not be considered as having
been dissolved by subsequent bacterial action. Accordingly portions of
soil each containing 200 gm. were placed in flasks, the same relative
proportions of tricalcium phosphate, calcium carbonate, and ammonium
sulphate added as in the incubated series, and the contents thoroughly
mixed. Distilled water was added at the rate of 250 parts per 100
parts of dry soil, and the contents were vigorously shaken once every
10 minutes during an hour, and were then filtered through Chamberland-
Pasteur filters. Calcium was determined in the filtrates by the volu-
metric permanganate method and phosphoric acid by the Pemberton
volumetric method. The average results obtained from closely agreeing
duplicate solutions, expressed in parts per million of dry soil, are sub-
mitted in Table I.
674
Journal of Agricultural Research
Vol. XII, No. lo
Table I. — Solubility of calcium and phosphate in soil immediately after adding calcium
carbonate, tricalcium phosphate, and ammonium sulphate
Treatment.
Soil only
Soil and calcium carbonate
Soil and tricalcium phosphate
Soil, tricalcium phosphate, and calcium carbonate. .
Soil and ammonium sulphate
Soil, ammonium sulphate, and tricalcium phosphate
Soluble
calcium
(Ca).
Soluble
phosphoric
acid (P2O5).
P. p. TO.
P. p. m.
27-5
26.8
17.4
13.8
31- I
28.6
33-5
90. I
27. I
16.6
92. 6
25. I
These data show that the addition of calcium carbonate produced no
effect on the immediate solubility of the calcium already in the soil or
that added as tricalcium phosphate, but the addition of tricalcium phos-
phate produced an increase of about 5 p. p. m. of soluble calcium and
II. 2 p. p. m. of soluble phosphoric acid. The most notable effect was
produced by ammonium sulphate, which caused an increase in water-
soluble calcium from 27.5 to 90.1 p. p. m.
The data submitted in Table I should not be considered as represent-
ing true solubility determinations, for it is not certain that equilibrium
was completely established, either between the various solids present
and the solvent (water), or between the constituents of the soil and the
chemical substances that were added to it. A longer period of contact
might have yielded extracts either more or less concentrated, depending
on whether or not the rate of solution was greater or less than the rate
of fixation. The same procedure was followed in making these deter-
minations, however, as was used at the close of the incubation periods,
and, although the results are not strict solubility determinations, they
are believed to be comparable, and that any difference between the
amounts of calcium and phosphoric acid found at the beginning and the
close of the periods of incubation may be assumed to have arisen mainly
through the action of biochemical agents.^
After incubation periods of 28, 57, and 157 days, quantities containing
200 gm. of dry soil were transferred from the incubation jars to flasks,
distilled water was added at the rate of 250 parts per 100 parts of dry
soil, was shaken vigorously once every 10 minutes during an hour, and
was then filtered through Chamberland-Pasteur filters as in the pre-
ceding series. Soluble calcium and phosphoric acid were determined
in the filtrates by the methods already referred to, and nitric nitrogen
by the phenoldisulphonic-acid method. The filtrates were also tested
for nitrite, but not more than 0.5 p. p. m. was found in any case. The
average results of closely agreeing duplicates are recorded in Table II.
*It was not deemed advisable to maiataia separate portions in a sterile condition, owing to the fact
that the various methods now in use for bringing about complete sterilization in soils probably affect the
solubility of the various constituents.
Mar. II. 1918 Nitrifying Bacteria and Tricalcium Phosphate 675
Table II. — Effects of nitrification on the solubility of tricalcium phosphate in soil
Materials added.
Control
Calcium carbonate
Tricalcium phosphate. .
Calcium carbonate and
tricalcium phosphate.
Ammonium sulphate . .
Ammonium sulphate
and calcium car-
bonate
Ammonium sulphate
and tricalcium phos-
phate
Ammonium sulphate,
calcium carbonate,
and tricalcium phos-
phate
Dried blood
Dried blood and cal-
cium carbonate
Dried blood and trical-
cium phosphate
Dried, blood calcium
carbonate , and trical-
cium phosphate
After 2S days.
Nitric
nitro-
gen.
P. p. m.
20. O
22. o
21. o
22. o
98.0
Cal-
cium.
P. p. m.
45- o
56.5
53-5
59-1
219.4
97.0 254.4
99.0 217. 7
100. o
91. o
89.0
82.0
253-4
107.7
107. 2
Phos-
phoric
acid.
P. p. m.
13- I
II. 9
24. 2
17-3
18.5
18.5
52.1
26.6
9-7
9.8
After s7 days.
Nitric
nifiro-
gen.
111.71 24.3
118. 2I 19. 5
P. p.m.
25-5
29. o
28.0
28.0
99.0
99.0
lOI. o
90. o
90. o
88.0
87.5
Cal-
cium.
P. p. m.
50. 6
70.8
58.8
70. I
225.4
Phos-
phoric
acid.
P. p. m.
13.2
25.0
22. 4
19.4
270. 5! 7- 4
229. 6
230.4
113-9
140. 2
117. 7
38-0
13-9
10. o
II- 5
22. 2
18.3
After IS7 days.
Nitric
nitro-
gen.
P. p.m.
114. O
94.0
Cal-
cium.
Phos-
phoric
acid.
P. p. fn. \P, p. m.
232. I
218.4 30.0
116. 4
5-7
A series of experiments with the use of silica sand corresponding
closely with the preceding soil series was conducted at the same time.
The silica sand was obtained from Monterey, Cal., and was free from
carbonate, but contained small amounts of feldspar, hornblende, and
mica particles, and possibly traces of other minerals. Portions of 1,000
gm. each were placed in fruit jars, and quantities of ammonium sulphate,
dried blood, calcium carbonate, and tricalcium phosphate were added
in duplicate at the same rates and arranged after the same plan as in the
preceding soil series.
Mixed cultures of bacteria were supplied by adding 150 c. c. of an
ordinary soil infusion obtained from the soil used in the preceding series.
The infusions were quite clouded with suspended matter, which probably
included small amounts of various soil constituents. In addition, 50 c. c.
of a nutrient solution, composed of 2 gm. of sodium chlorid, 0.2 gm. of
magnesium sulphate, 0.5 gm. of potassium sulphate, and 6 drops of a 10
per cent solution of ferric chlorid per liter, were thoroughly mixed with
the sand in each jar. The jars were loosely covered and incubated at
room temperature.
After periods of 28, 56, 98, and 157 days, quantities containing 200 gm.
of dry sand were withdrawn, 500 c. c. of distilled water added, and after
shaking vigorously as in the preceding series, were filtered through
676
Journal of Agricultural Research
Vol. XII, No. 10
Chaniberland-Pasteur filters. Nitrate, calcium, and phosphoric acid
were determined in the filtrates by the methods already referred to.
Nitrite, when present, was also determined by the Greiss-Ilosovay method.
The calcium determinations for the 98-day period have been omitted
from the table, owing to an error in the analytical procedure. The
average results from duplicate incubations are recorded in Table III.
Table III. — Effects of nitrification on the solubility of fricalcium phosphate in sand
cultures
[Results expressed in parts per million)
Aiter 28 days.
After 56 days.
After 98
days.
After IS 7 days.
Materials added.
6
a
d
3 M
0 0
i
0
0 .
1
a a
0 0)
•a
•n
0
0 .
0 (U
'3 .
ss
0 .
0 «
6
■3
1°
a
3
•0
0 .
0 "
S
iS
6
iz;
6
p4
s
g
g
S
2
c3
S
Calcium carbonate
7-S
0.0
72-5
0.0
g.o
0.0
77-6
0.0
12. 2
0.0
0.0
n.o
0.0
77.6
0.0
Tricalcium phosphate
7.2
.0
52-1
67. S
8.1
.0
SI. 8
S2.3
ICO
.0
ss- s
10. 2
.0
44.6
4S-7
Calcium carbonate and tricalcium
phosphate
7-3
.0
75-1
23-7
8.3
.0
61.3
18.0
"•5
.0
21.6
10.5
.0
64-3
19.8
Ammonium sulphate
24-5
I- 5
6.9
28. 7
Ammonium sulphate and calcium
carbonate
92.0
.0
308.0
.0
79- 0
.0
298.0
.0
82. s
.0
.0
92.0
.0
345-1
■ 0
Ammonixmi sulphate and trical-
cium phosphate
2. 1
.0
62.6
77-9
7-S
.0
87.1
79-9
14.0
.0
83.9
33- 0
.0
120.4
8s. I
Ammonium sulphate, calcium car-
bonate, and tricalcium phos-
phate
33-5
8.2
207.8
17.7
80.0
.0
269.2
12.5
80.0
.0
X5-4
92.0
.0
304- 0
15-8
Dried blood
1. 1
41.2
23- 0
.0
3.0
^0.0
20.5
.0
7-2
23.7
.0
29.0
S.O
35-5
.0
Dried blood and calcium carbonate.
4S-0
IS. 6
150-3
.0
7S-0
.0
191-5
• 0
81.0
.0
.0
91.0
.0
210.6
.0
Dried blood and tricalcium phos-
phate
•S
2S.0
^S.'!
■I?-!
I.O
Si.o
40.3
"^S-?
2.0
2S.O
S2. 2
12.2
22. S
4S-S
46. s
Dried blood, calcium carbonate.
and tricalcitun phosphate
1-3
62.5
100.2
15-1
2.9
S2.0
115- 7
12.0
2.6
50.0
18.3
II. 0
75- 0
162.0
12.7
DISCUSSION OF EXPERIMENTAL RESULTS
Soil series. — The data submitted in Table 11 show that active nitri-
fication took place in the soil series. For example, the nitric nitrogen
increased in the control portions from 8.5 p. p. m., originally present, to
20 p. p. m. in 28 days. Where ammonium sulphate was added, the con-
centration increased to 98.5 p. p. m., while dried blood yielded 91.0
p. p. m. of nitric nitrogen. The addition of calcium carbonate and tri-
calcium phosphate either singly or together produced very slight, if any,
effects on nitrification in this series.^ After subtracting the amounts of
nitric nitrogen in the controls, it is found that 78 per cent of the am-
monium sulphate and 71 per cent of the dried-blood nitrogen were
oxidized in 28 days.
When allowance is made for the soluble calcium found at the beginning
of the experiment (Table I), the data show that in every case nitrification
1 In other experiments with this soil the addition of calcium carbonate has slightly stimulated the nitri-
fication of ammonium sulphate, but not of dried blood.
Mar. II, 1918 Nitrifying Bacteria and Tricalcium Phosphate 677
was accompanied by increases in the solubility of the calcium present.*
This was noted to some extent in the portions to which no nitrogenous
additions were made ; was considerably greater when dried blood was sup-
plied; and was greatest with the addition of ammonium sulphate. The
data show, however, that the concentration of soluble calcium was not
increased as a result of adding tricalcium phosphate. On the other hand,
soluble calcium was considerably increased in a number of cases by the
addition of calcium carbonate. It would seem, therefore, that calcium,
in the form either of the carbonate or of such silicates as occur in this
soil, will be dissolved by the biochemical oxidation products in preference
to tricalcium phosphate.
In contrast to the effects on the solubility of calcium, the data show
that nitrification of the soil nitrogen and that added as dried blood was
accompanied in each case by a well-defined decrease in soluble phos-
phoric acid. Where ammonium sulphate was added alone, the amounts
of soluble phosphoric acid found at the 28- and 57-day periods were
approximately the same as found at the beginning of the experiment.
It would appear, therefore, that the solvent action of the bacteria in
this case was almost exactly equal to the precipitating action that evi-
dently took place in the control and dried-blood portions. After 157
days, however, more than half of the soluble phosphoric acid originally
present in this portion had disappeared. It is also shown that, while
larger amounts of soluble phosphoric acid were found where tricalcium
phosphate had been added than in the control portions, the increases
can not be definitely ascribed to the action of bacteria in any case except
where ammonium sulphate was also added, and then only without the
addition of calcium carbonate. For in all other cases the solubility
immediately after adding tricalcium phosphate was equal to or greater
than that at the end of the incubation periods.
In the absence of calcium carbonate, however, the oxidation of am-
monium sulphate dissolved tricalcium phosphate, as shown by the fact
that the concentration of soluble phosphoric acid was increased in 28
days from 24.2 to 52.1 p. p. m. Later the solubility steadily declined
until at the end of 157 days the concentration had been reduced to
30 p„ p. m.
Assuming in this case that the increase in nitric nitrogen over the
amounts found in the controls was due to the oxidation of ammonium
sulphate, we find that the oxidation of 78.0 p. p. m. of nitrogen resulted
' It should be clearly understood that the results obtained in this investigation represent the algebraic
sumanddifferenceoi the results of a number offerees. In the first placeit is highly probable that biochem-
ical agent"! other than the nitrif jTng organisms are capable of affecting the solubility of calcium and phos-
phoric add in soils. A part of this effect may be referred to as positive and a part as negative, since, on the
one hand, carbonic acid, formed in the life process of bacteria, is a solvent for calcium and phosolioric acid,
andon the other hand, the organisms themselves absorb phosphoric acid {8). in the second place diffusion
tends to bring about more or les^ fixation in soils. The concentration at a given moment, therefore, is
really dependent on the Interaction of a number of forces. Consequently a full explanation of the results
obtained is not possible at present.
678 Journal of Agricultural Research voi. xii. no. 10
in the solution of 27.9 p. p. m. of phosphoric acid, or 12.2 p. p. m. phos-
phorus. By comparing these data with the rate of solution reported by
Hopkins and Whiting from solution cultures, it will be seen that, while
the oxidation of i pound (454 gm.) of nitrogen was accompanied by the
solution of 2.033 pounds (922 gm.) of phosphorus in their experiments;^
in these experiments with soil cultures the maximum amount of phos-
phorus made soluble was only 0.156 pound (70.8 gm.) per pound of
nitrogen oxidized. Therefore, the oxidation products of ammonium
sulphate were approximately 13 times as effective in dissolving trical-
cium phosphate in solution cultures as in this soil.
It is interesting to note that the addition of calcium carbonate tended
to lower the solubility of tricalcium phosphate wherever applied.
Sand series. — Ammonium sulphate, when added alone, underwent
almost no nitrification in the sand series (Table III) until the last period
of the experiment, during which small amounts of nitrate were formed.^
The presence of calcium carbonate, however, promoted very active
nitrification of ammonium sulphate. In this case the concentration of
nitric nitrogen reached its maximum (92 p. p. m.) in 28 days. On the
other hand, the effects resulting from the addition of tricalcium phos-
phate only began to be manifested by the fifty-sixth day. Later the
nitrate content slowly increased until the close of the experiment, when
33 p. p. m. were found. Nitrification of ammonium sulphate in the
proportions containing both calcium carbonate and tricalcium phosphate
was not so pronounced during the first 28 days as with calcium carbon-
ate only, but later the effects were almost identical.
The nitrification of dried blood in sand proved to be especially inter-
esting in that the intermediate formation of nitrite proceeded much
more rapidly than the oxidation to nitrate.^ When dried blood alone
was added, 42.1 p. p. m. of nitrite nitrogen were found after 28 days and
only I.I p. p. m. of nitric nitrogen. Later no further accumulation of
nitrite took place, but the formation of nitrate set in slowly with the
result that 29 p. p. m. had been formed by the close of the experiment,
but even then 5 p. p. m. of nitrite still remained.
> It should not be inferred that Hopkins and Whiting claim that the products of nitrification will dis-
solve rock phosphate at the same rate in soil as in solution cultures. They pointed out (j. />. 405), for
example, that nitrous acid may combine with calcium sihcate, calcium carbonate, and other compounds
in soils as well as with tricalcium phosphate. Nevertheless they hold that the nitrite bacteria are important
agents in bringing about the solution of rock phosphate in field soils.
2 Appreciable amounts of nitrate were formed in the portions which were intended to be free from com-
Kined nitrogen. The nitrate in these instances probably originated from organic matter held in suspen-
sion in the soil infusions that were added. The amounts formed, however, were small and consistent ia
every case, increasing steadily from an average of 7.3 p. p. m. at the 28-day period to a maximum of ii.i
p. p. m. at 98 days.
5 Data showing that nitrites may accumulate in nitrification experiments have previously been reported
(4), but this is a condition not commonly met in the field. The accumulation of nitrites indicates, of
course, that some factor in the medium was abnormal, but more favorable for the nitrite bacteria than
for the nitrate bacteria. It is known that the nitrate bacteria are more sensitive to adverse conditions
than the nitrite bacteria.
Mar. u, 1918 Nitrifying Bacteria and Tricalcium Phosphate 679
The application of calcium carbonate notably stimulated the nitrifi-
cation of dried blood, and after 56 days the yield was approximately the
same as from ammonium sulphate. On the other hand, tricalcium
phosphate produced no stimulation in the nitrification of dried blood at
any period, but the application of both calcium carbonate and tricalcium
phosphate promoted active nitrite formation, which during the last
period of the experiment resulted in the oxidation of approximately as
much nitrogen as in any other case in the experiment. The final oxida-
tion to nitrate in this case, however, was very feeble throughout the
entire experimental period.
It is evident from the above results, therefore, that, while tricalcium
phosphate may promote nitrification in the absence of carbonate, more
favorable conditions for nitrification were produced by calcium carbonate
than by tricalcium phosphate.
Large amounts of calcium carbonate were made soluble by the nitrifi-
cation of ammonium sulphate in the sand series, but since enfeebled
nitrification of ammonium sulphate took place when tricalcium phos-
phate only was added, relatively small increases in soluble calcium were
produced. At the final period of the experiment, however, the increase
in soluble calcium resulting from tricalcium phosphate and the small
amounts of nitrogen that had been oxidized agree closely with theo-
retical calculation.
Again, considerably less calcium was dissolved in the nitrification of
dried blood than in that of ammonium sulphate, a result which is in
harmony with generally accepted vievv's regarding the nature of the
oxidation products formed in the two cases. With the latter sulphuric
acid is formed in addition to nitrous acid, while with the former carbonic
acid is probably one of the end products.
The limited nitrification of dried blood, found where tricalcium phos-
phate had been added, was associated with a lower soluble-calcium
content than occurred where tricalcium was added alone. But in view
of the fact that not more than 30 per cent of the nitrogen was oxidized,
it is probable that the medium remained alkaline as a result of the
ammonification of the dried blood, and, consequently, the lower yields
of soluble calcium may have been due to slight precipitarion of calcium
as calcium carbonate.
It is especially interesting that smaller amounts of calcium were
dissolved by the nitrification of ammonium sulphate and dried blood
in the presence of both calcium carbonate and tricalcium phosphate
than with calcium carbonate alone.
Considering the phosphoric-acid determinations, it is at once apparent
that nitrification did not produce an increase in the solubility of tri-
calcium phosphate when carbonate was present. In the absence of
carbonate, however, an increase in solubility resulted from the nitrifica-
tion of ammonium sulphate. In this case, although nitrification was
68o Journal of Agricultural Research vot. xii. no. w
less active than elsewhere, the data show that the small amounts of
oxidation products formed dissolved tricalcium phosphate. On the
other hand, wherever calcium carbonate was also present, the oxidation
products not only combined with it, but the initial solubility of the
tricalcium phosphate was very materially lowered as well. This was
true in the nitrification of both ammonium sulphate and dried blood.^
INTERPRETATION OF RESULTS
In making a practical interpretation of these investigations it should be
borne in mind that fertile soils commonly contain at least small amounts
of carbonate, that even the so-called acid soils frequently contain con-
siderable amounts of bicarbonate, and that the presence of calcium
carbonate in soils is generally considered to promote conditions that are
favorable for the growth of most crops. Furthermore, large amounts
of calcium carbonate are being applied to soils in many localities,
especially in the humid sections. In the Central West, for example,
ground limestone is being applied on a large scale, and generally it is
recommended that the application be repeated every few years.
Under the conditions that result, chemical reasoning (j) and the experi-
mental results reported above agree in suggesting that the action of the
biochemical oxidation products, formed in the nitrification of organic
nitrogen, would be spent on the carbonate and not on tricalcium phos-
phate. Furthermore, it seems doubtful whether this could be avoided
by the application of limestone and rock phosphate at different times
in a rotation, as was suggested by Hopkins and Whiting. Although
it is possible that under this condition the particles of rock phosphate
may chance to occur in local centers that are somewhat removed from
solid particles of calcium carbonate, and nitrification may happen to
take place in these centers, it does not even then necessarily follow
that the phosphate would be dissolved. For such centers would probably
always be in contact with soil silicates, and the above data indicate
that at least some soil silicates may be attacked by the products of
nitrification in preference to tricalcium phosphate. When the con-
ditions permit the accumulation of considerable acidity, however, such
as probably obtain when ammonium sulphate is applied to a soU low
in carbonate, it was found that small amounts of the phosphate were
dissolved. But then a degree of acidity that is distinctly injurious to
crops may soon develop, as has been found by field trials in a number
of localities. Furthermore, bicarbonate, arising from the action of
carbonic acid on the solid particles of calcium carbonate, would certainly
tend to diffuse toward the supposed centers of acidity, thus precipitating
• Soluble phosphoric acid may have been utilized to some extent in the life processes of the bacteria
present, as was found by Tottingham and Hofimann (,8). But the simultaneous losses m soluble calcium
also suggest the precipitation of phosphoric acid, a view that is in harmony with the well-known fact that
calcium carbonate may precipitate phosphoric acid irom solution.
Mar. II, i9i8 Nitrifying Bactevla and Triccdcium Phosphate 68 1
the phosphoric acid. The results of many studies on the concentration
of nitric nitrogen in soils supporting growing crops show, for example
that diffusion is an important force in maintaining equilibrium in soils.
Tricalcium phosphate can not be converted into monocalcium phos-
phate (the water-soluble form) without active acidity being developed.^
Nor are acids neutralized by converting tricalcium phosphate into mono-
calcium phosphate, since the latter is an acid compound. But when the
acidity necessary to the solution of tricalcium phosphate is neutralized,
the phosphoric acid will be precipitated. If it be desirable that a con-
dition either of neutrality or slight alkalinity obtain in soils, as has been
widely taught, it is difficult to see how more than traces of monocalcium
phosphate can exist therein at the same time.
It should also be recalled that untreated rock phosphate generally
contains considerable amounts of calcium carbonate intimately com-
mingled with the phosphate. Before dilute acids, formed by the action
of bacteria or otherwise, can dissolve the phosphate, the carbonate must
first be neutralized, as is commonly recognized in the commercial pro-
cesses employed in the manufacture of acid phosphate.
It seems possible, however, that the nitrifying bacteria may dissolve
limited amounts of rock phosphate in acid soils. The soil used in this
investigation, although low in carbonate, was not acidic, and more active
solution of tricalcium phosphate would probably take place in an acid
soil. But, in any case, with the possible exception of very sandy types
of soil, it is probable that phosphoric acid, made soluble by the nitrifying
bacteria, would tend to become fixed through being brought into con-
tact with other soil constituents by diffusion. Many investigations have
shown, for example, that acid phosphate soon becomes fixed, even in
acid soils.
From these investigations it seems, therefore, that while the nitrite
bacteria are capable of effecting the solution of tricalcium phosphate
under restricted conditions, they are not the potent agents in the solu-
tion of rock phosphate in the field that Hopkins and Whiting were led to
infer from their experiments with solution cultures.
In the opinion of the author the results obtained in this investigation
should not be interpreted as being definitely opposed to the use of un-
treated rock phosphate as a fertilizer. It is true that the means by which
tricalcium phosphate is made soluble in soil have not been definitely
determined, but the important fact remains that in various localities
beneficial effects on the growth of crops have frequently been produced
by rock phosphate. It therefore remains for further investigation to
lay bare the reasons. The suggestions offered by Truog (9) in this con-
nection seem to be deserving of special consideration.
• It is, of course, understood that hydrolysis is excepted in this case. The absolute increases in soluble
phosphoric acid in soib resulting from the hydrolysis of tricaldum phosphate are probably quite small.
682 Journal of Agricultural Research voi. xii, no. lo
SUMMARY
The investigations reported in this paper include a study of (i) the
effects of adding calcium carbonate, tricalcium phosphate, and ammo-
nium sulphate on the immediate solubility of calcium and phosphoric
acid in a light sandy loam soil; (2) the effects of nitrification of the soil
nitrogen, ammonium sulphate, and dried blood on the solubility of the
naturally occurring calcium and phosphoric acid; (3) the effects of nitri-
fication in soil and sand cultures on the solubility of tricalcium phosphate
both with and without the application of calcium carbonate.
The following results were obtained:
(i) The addition of calcium carbonate produced no effect on the imme-
diate solubility of the soil calcium or that added as tricalcium phosphate.
The addition of tricalcium phosphate produced an increase of about 5
p. p. m. of soluble calcium and 11.2 p. p. m. of soluble phosphoric acid,
while the addition of ammonium sulphate brought about an increase in
water-soluble calcium from 27.5 to 90.1 p. p. m.
(a) Active nitrification of ammonium sulphate and dried blood took
place in the soil series, and at the same time notable increases in soluble
calcium were produced.
(3) No increase in the solubility in water of the soil phosphates or of
tricalcium phosphate was produced by bacterial action except in the
nitrification of ammonium sulphate when added wdthout calcium car-
bonate. In this case 0.156 pounds (70.8 gm.) of phosphorous were dis-
solved for every pound of nitrogen oxidized, whereas Hopkins and Whit-
ing found from solution cultures that 2.033 pounds (922 gm.) were dis-
solved for every pound of nitrogen oxidized.
(4) The addition of calcium carbonate brought about an increase in
soluble calcium but tended to lower the solubility of tricalcium phosphate.
(5) In the absence of calcium carbonate the nitrification of ammonium
sulphate in sand cultures was accompanied by the solution of theoretical
amounts of tricalcium phosphate. When calcium carbonate was present,
however, the solubility of tricalcium phosphate was not increased by
nitrification.
(6) The formation of nitrite from dried blood took place more rapidly
in the sand cultures than the formation of nitrate.
(7) Tricalcium phosphate was not dissolved by the nitrification of dried
blood in the sand series.
(8) It was found that calcium carbonate promoted more active nitri-
fication than tricalcium phosphate.
(9) The experimental results indicate that the nitrification of organic
forms of nitrogen does not increase the solubility of rock phosphate under
field conditions that are favorable to crop growth. It is possible, how-
ever, that the nitrification of ammonium sulphate may result in the solu-
tion of small amounts of tricalcium phosphate in soil low in carbonate.
Mar. ij, is,i8 Nitrifying Bacteria and Tricalcium Phosphate 683
LITERATURE CITED
(i) Cameron, F. K., and Bell, J. M.
1907. THE ACTION OF WATER AND AQUEOUS SOLUTIONS UPON SOIL PHOSPHATES."
U. S. Dept. Agr. Bur. Soils Bui. 41, 58 p., 5 fig.
(2) Gaither, E. W.
191 2. a new apparatus for the determination op carbon dioxide. in
Jotir. Indus, and Engin. Chem., v. 4, no. 8, p. 611-613, 2 fig.
(3) Hopkins, C. G., and Whiting, A. L.
1916. soil BACTERIA AND PHOSPHATES. 111. Agr. Exp. Sta. Bui. 190, p. 395-406.
Foreword by E. Davenport.
(4) Kellev, W. p.
1916. NITRIFICATION IN SEMIARID SOILS, I. In JouT. Agr. Research, v. 7, no.
10, p. 417-437. Literature cited, p. 436-437.
(5) LiFMAN, J. G., McLean, H. C, and Lint, H. C.
I916. SULFUR OXIDATION IN SOILS AND ITS EFFECT ON THE AVAILABILITY OK
MINERAL PHOSPHATES. In Soil Sci., V. 2, no. 6, p. 499-538, 5 fig.
Literature cited, p. 535-538.
(6) LuPTON, N. T.
1893. THE EFFECT OF DECOMPOSING ORGANIC MATTER ON NATURAL PHOS-
PHATES. In Ala. Agr. Exp. Sta. Bui. 48, p. i-io.
(7) McDowell, M. S.
1908. is the phosphoric acid op floats made soluble by rotting manure?
In Pa. Agr. Exp. Sta., Ann. Rpt., 1907-08, p. 175-178.
(8) ToTTiNGHAM, W. E., and Hoffmann, C.
1913. NATURE of the CHANGES IN THE SOLUBILITY AND AVAILABILITY OF
PHOSPHORUS IN FERMENTING MIXTURES. Wis. AgT. Exp. Stfi. Research
Bui. 29, p. 275-321, 3 fig.
(9) Truog, E.
1912. factors influencing the availability of rock phosphate. wjs.
Agr. Exp. Sta. Research Bui. 2c, p. 17-51, 4 fig.
38325°— 18 6
ADDITIONAL COPIES
OP THLS PtJBLICATION MAY BE PKOCUBED FROM
THE SUPERINTENDENT OF DOCUMENTS
GOVERNMENT PRINTING OFFICE
WASfflNOTON, D. C.
AT
30 CENTS PER COPY
SoBSCEiPTiON Price, J3.00 Peb Year
V
Vol. XII NIARCH IS, 1918 No. 11
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTKNTS
Page
Respiration of Stored Wheat - - - - - - 685
C. H. BAILEY and A. M. GURJAR
( Contribution Irom Minnesota Agricultural Experiment Station )
E£fects of Mistletoe on Young Conifers - - - - 715
JAMES R. WEIR
(Contribution from Bureau of Plant Industry )
Determination of Fatty Acids in Butter Fat - - - 719
E. B. HOLLAND and J. P. BUCKLEY, Jr.
(Contribution from Massachusetts Agricultural Experiment Station)
PUBLISHED BY AUTHORITY OF THE SECRETARY OF AGRICULTURE,
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COLLEGES AND EXPERIMENT STATIONS
\VASMINGTON, D. C.
WAtHINOTON: OOVERNMENT PRINTINO OFFICE : Itll
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
FOR THE ASSOCIATION
KARLF.KELLERMAN. Chairman RAYMOND PEARL
Physiologist and Associate Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARI,EvS L. MARLATT
Eniomologisl and Assislajit Chief, Bureau
of Enionwloay
Biolpqisl. Maine Agriculliiral Experiment
Station
H. P. ARMSBV
Director . Institute of Animal A'utriliim, The
Pennsyhania State College
E. M. FREEMAN
Botanist, Plant Pailwlot/ist 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.
* Dr. Pearl has undertaken special work in connection with the war emergency ;
therefore, imtil further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P. Armsby, Institute of Animal Nutrition,
State College, Pa.
mm
Vol.. XII
Washington, D. C, March i8, 1918
No. II
RESPIRATION OF STORED WHEAT ^
By C. H. Bailey, Cereal Technologist, and A. M. Gurjar, Assistant in Agrkultural
Biochemistry, Division of Agricultural Biochemistry, Minnesota Agricultural Experi-
ment Station
INTRODUCTION
The preservation in storage of large quantities of thrashed grain in-
volves certain difficulties. In addition to protection from vermin, it has
long been known that the grain must be dry when stored, and the ingress
of moisture prevented. The ancients were aware that damp grain, or
similar vegetable material, will heat and become decomposed when
stored in quantities. It is evident that in prehistoric times carefully
constructed receptacles were employed for the protection of the reserve
supplies of cereals. Those ancient people of India, the Hindus, some-
times resort to the use of receptacles which are submerged in cool water,
reducing the temperature of the stored grain and also the supply of air.
In modern practice carefully constructed tanks or silos, frequently of
large capacity, are chiefly used for the storage of grain.
CAUSE OF THE HEATING OF GRAIN
The cause of heating of damp vegetable matter was not known, how-
ever, until comparatively recent times, and it is only mthin the last
decade or two that any data have been accumulated which indicate the
exact effect of various factors on the rate of heating of grain and similar
material. That the phenomenon known as respiration is responsible for
the heat energy released in a mass of damp grain is shown by modern
research. Loew (i8pc>y, in discussing the fermentation and heating of leaf
tobacco, maintained that the release of energy and rise in temperature
was occasioned by the activity of the oxidizing enzyms of the leaf cells.
The microbial flora were not believed to play any considerable part in
these changes. In advancing this view he opposed vigorously the bac-
terial-fermentarion theory of Suchsland {i8gi).
Rahn {1910) states that the curve of the process of spontaneous heat-
ing of organic matter, including grain, would not in itself indicate whether
00 ' Published with the approval of the Director, as Paper 97, Journal Series, Minnesota Agricultural Experi-
^ir jnent Station.
_ 2 Bibliographic citations in parentheses (italic) refer to " Literature cited," pp. 710-713.
. -J Journal of Agricultural Research,
Washington, D. C.
(685)
Vol. XII, No. n
Mar. 18, 1918
Key No. Minn. 26
686 Journal of Agricultural Research voi. xii. No. n
the heat was produced by chemical or microbial causes. Gore {191 1,
P- 33) concluded that in self-heating, physiological processes are probably
the first to operate. Oxidizing enzyms are the active agents in many
cases. Chemical oxidation would intervene only when the temperature
had been raised to the combustion point of the substances present. Gore
presents formulas and graphs illustrating the theoretical progress of the
accumulation of heat.
The studies of E. M. Bailey {1912) on the ripening of bananas indicate
that bacterial activity is not responsible for the heat produced or other
changes resulting during the ripening of the fruit.
Nabokich {1903) found that seeds which had been sterilized by chemi-
cals such as mercuric chlorid respired during the first few days as much
carbon dioxid as did the controls. In some instances the respiration
of the sterilized seeds exceeds that of the controls.
Respiration may be briefly defined as the release of energy through
the biochemical oxidation of organic compounds as accelerated by certain
enzyms. Carbon dioxid and water are the characteristic chemical end
products. It is shown by De Saussure {1804) that respiration was ac-
companied by the disappearance of oxygen. Pfliiger {1875) maintained
that the inspired oxygen combines in some manner with cyanogen radi-
cals of the living protoplasm. This effects a readjustment, as the result
of which carbon dioxid and water are eliminated. The decomposition
is of an explosive character, and the reaction liberates heat energy.
Verworn {1899, p. 483) evolved the " biogen" theory, according to which
the oxygen enters the "biogen molecules," which are thus rendered less
stable. Slight impulses are then required to bring about a chemical
union of this oxygen with the carbon in the cyanogen group.
Since the grain itself is a poor conductor of heat, it follows that the heat
energy released through respiration accumulates in the mass in propor-
tion to its bulk so that the increase in temperature may in time become
very marked.
MATERIAL OXIDIZED IN RESPIRATION
In the case of resting tissues and storage structures, such as the grain
kernel or caryopsis, the exact character of the substrate or material
oxidized in respiration is of significance. Wehmer {1892) fed Aspergillus
on peptone, and found it capable of satisfying its requirements from
this source. Gore (19 14) found that in ripening bananas the rate of
starch hydrolysis paralleled the rate of respiration. There was slight
change in the quantity of protein and fats in the fruit. Maige and
Nicolas {1910) found that the immersion of etiolated leaves, shoots, or
seedlings in sugar solutions resulted in an increased rate of respiration.
Langworthy and Milner (1913) observed that in ripening bananas the
"thermal quotient," or quantity of heat produced per unit of carbon
dioxid respired, indicated the combustion of carbohydrate. The thermal
Mar. is, lyib Respuation of Stored Wheat 687
quotient found for one period of their studies, when conditions for
accurate measurement were at an optimum, was 2.6. This is so nearly
the theoretical for the combustion of carbohydrates as to indicate that
little else was involved. As pointed out by Dr. Milner in a private
communication, the thermal quotient when fat is burned is 3.4, while
that when protein is burned is 2.9.
Hasselbring and Hawkins (1913) found no general correlation between
the total sugar content of the sweet potato and its respiratory activity.
Cane sugar is relatively stable and does not appear to be used in the
process of respiration. A simultaneous decrease in the reducing sugars
and the respiratory activity was observed. The reducing sugars are,
in their opinion, the immediate source of respiration material.
That the fat or ether extract of the wheat embryo is not the principal
substrate upon which the respiratory enzyms act, and that it is not
burned or destroyed during germination is indicated by the experiments
of Le Clerc and Breazeale (1911, p. 12). Their data show that there w^as
more fat in the total plant (seed residues plus axes) at practically all
stages of germination than there was in the original seeds. Accordingly
little energy could have been derived from the oxidation of the fats; on
the contrary, energy derived from some other source must have been
utilized in the synthesis of the additional fat produced in the seedling.
All available evidence therefore seems to indicate that the heat of respira-
tion is produced by the oxidation of reducing sugars.
SEAT OF RESPIRATION IN THE WHEAT KERNEL
There are a number of reasons for believing that the germ or embryo
of the wheat kernel is the location of the larger part of the biological
oxidation that occurs incidental to respiration. The embryo is, in a
general w^ay, decidedly richer in enzyms than is the endosperm or any
other kernel structure. Brown and Morris {i8go) consider that the
endosperm of the ripened kernel is no longer a vital tissue. The secretion
of diastase and other enzyms is assigned to the scutellum, an organ of the
embryo. Maim and Harlan (1915) concluded that in germinating
barley the conversion of the endosperm is effected by enzyms secreted
by the epithelial layer of the scutellum.
Karchevski (1903) found the energy of carbon-dioxid respiration
to be 12 times as great in the wheat embryos as in the seeds themselves.
Burlakow {1898) states that the respiratory activity of the germ is 20
times greater than is that of the endosperm. Wender {^1905) called
attention to the pronounced catalase activity of the germ structures,
as contrasted with mill products derived almost exclusively from the
endosperm. This has also been observed in an unpublished study of
catalases which was made in this laboratory. This fact is more signifi-
cant, in view of Appleman's {1915) discovery that catalase activity in
potato juice shows a striking correlation with respiratory activity in the
688 Journal of Agricultural Research voi. xu.no. n
tubers. If catalase activity parallels respiratory activity, it may be
reasoned by analogy that those structures of the wheat kernel which
exhibit the greatest catalase activity are the seat of the larger part of
respiration.
Barnes and Grove {1916) lend further support to the hypothesis that
the seat of oxidation activity is in the embryo by their obsenl^ation that
in air-dry wheat the embryo becomes shrunken after a time, while the
food materials of the endosperm are unimpaired. This is interpreted
by them to indicate a destruction of the material of the embryo itself
as the result of respiration when the dryness of the kernel suppresses
diffusion. Were respiration equally pronounced in the endosperm it
too should exhibit a similar loss of material.
Osterhout's (1917) observation that oxidation is more rapid in the
nucleus than in the cytoplasm might lead to the deduction that, since
the embryo cells of the wheat caryopsis have a much larger proportion
of nuclei than the endosperm cells, oxidation should proceed more rapidly
in the embryo tissues.
All of the above facts are in harmony with what might logically be
expected. The principal release of energy as the result of biological
combustion should occur in the structure where such energy is required
for the synthesis of new organic compounds. Since the embryo is
endowed particularly with that function, respiration must of necessity
be most pronounced in it, if not confined to it.
MEASUREMENT OF THE RATE OF RESPIRATION
There are two general methods which may be used in the quantita-
tive estimation of the rate of respiration of vegetable material. One
that is employed where faciHties are available is to measure in terms
of Calories the heat energy released per unit of time and material. The
elaborate device for this purpose is known as the respiration calorimeter.
The second general method includes the determination of one of the end
products of the reaction, carbon dioxid. This method may easily be
made decidedly accurate without entailing the assembling of a calori-
meter. In using the calorimeter both carbon dioxid respired and heat
evolved may, if desired, be determined simultaneously.
Inasmuch as the writers were not provided with a respiration calori-
meter suited to this purpose, the carbon dioxid evolved b}^ stored wheat
was measured, and from this data the rate of respiration was calculated.
Truog's {1915) method and absorption tower was used for this purpose,
the tower being slightly modified, or rather, added to, in order to adapt
it to the present work. The procedure followed and a description of
the apparatus has been published by the junior author (Gurjar, 19 17).
To compute the Calories of heat evolved, the factor found by Langworthy
and Milner may be employed; i gm. of respired carbon dioxid equals
2.6 Calories of heat.
Mar. 18, 1918 Respiration of Stored Wheat 689
Calcium-chlorid towers were used as respiration chambers. Paraffined
wire netting was fitted into the constriction near the base, and on this
the grain rested. Rubber connections were made as short as possible,
and all stoppers and tubing were thoroughly paraffined to prevent
selective absorption of carbon dioxid. In all instances where the same
lot of grain was worked with at different moisture contents, the several
portions were brought to approximately the desired percentage of
moisture by adding water from a burette, at the same time stirring
thoroughly. The samples were allowed to stand in sealed jars for three
days, in order to insure uniform distribution of the water through the
kernels. It had been found by determining the rate of swelling of the
kernels that they reached their maximum size in considerably less than
three days, and from this it was concluded that the moisture distribution
would be complete in the 3-day interval. When the grain was ready
to work with, a weighed quantity v^^as sealed into the tower. Samples
were taken at this time for the determination of moisture. The amount
of grain placed in the tower varied with the moisture content, about 500
gm. being employed in the case of the lower moisture limits, while
about 300 gm. were used when the percentage of moisture exceeded
15.5 per cent. In this manner convenient and accurately determinable
quantities of carbon dioxid were obtained. After sealing the grain into
the glass cylinders, the air was removed and replaced by carbon-dioxid-
free air. The towers were then placed in the thermostat, which, except
when temperature was the variable, was maintained at 37.8° C. (100° F.).
The period of incubation was fixed at four days, the exact number
of hours being noted at the time of removing the respiration chambers
from the thermostat. The accumulated carbon dioxid v/as then re-
moved through the tubulure at the bottom of the tower, carbon-dioxid-
free air being simultaneously admitted through the top. The carbon
dioxid was absorbed in N/4 barium hydroxid [Ba(0H)2] solution in
the special absorption tower, as described by Gurjar in the above-
mentioned paper. The respiration data given in the tables are stated
in terms of milligrams of carbon dioxid respired per 24 hours by each
100 gm. of dry matter.
RELATION OF THE MOISTURE CONTENT OF WHEAT TO THE RATE OF
RESPIRATION
The observation of Bonnier and Mangin {1885) that respiration of
living plants varies directly with the humidity of the air might be
interpreted to mean that the moisture content of the tissues increased
in a humid atmosphere. This increase in turn may have occasioned
the rise in the rate of respiration. Maquenne {1900) concluded that
a reduction in the moisture content of seeds is accompanied by a reduc-
tion in the rate of respiration, and Lund (1894) discovered that the
desiccation of roots and tubers reduced their rate of respiration.
690 Journal of Agricultural Research \'o;. xn, no. h
Kolkwitz (1901) found that barley grains containing 19 to 20 per
cent of moisture respired 3.69 mgm. of carbon dioxid per kilo in 24
hours at summer temperature, while at the same temperature barley
containing 14 to 15 per cent of moisture respired 1.4 mgm. per kilo,
and 0.35 mgm. per kilo when containing 10 to 12 per cent of moisture.
White (1909) found that all cereals gave off appreciable quantities
of carbon dioxid when stored in an air-dried condition, the respiration
of wheat containing 11.9 per cent of moisture being especially pro-
nounced. Wheat dried for eight days at 45° C. did not respire a determin-
able quantity of carbon dioxid.
Babcock (1912) states that respiration is practically suspended in
dry seeds and spores, and is most pronounced when vital processes
are most active, as during the germination of seeds. The metabolic
water produced as respiration proceeds is believed to play an important
part in the vital phenomena of the cells.
Duvel (1904), in studying the vitality of stored seeds, observed that
the rate of respiration, as indicated by the carbon-dioxid content of
the air in the closed container, was increased on increasing the moisture
content of the seed. At the same time there was a marked diminution
in the percentage of viable seed.
Qvam {1906) observed an increased rate of respiration in barley as
the percentage of moisture was increased.
Duvel (1909) held a lot of corn in storage in an elevator bin. The
moisture content averaged 17.8 per cent, and the initial temperature
on February 17, 1909, was 36° to 40° F., which increased near the surface
of the grain to 133° F. on April 27, 1909. The temperature from the
middle to the bottom of the bin was only about 40° F. A portion of
the hot corn from the top of the bin was artificially dried to an average
of 14.57 per cent of moisture. This was put in a car, and as a control
a lot of the cool, undried corn from the same bin was put in another
car. The latter had an average moisture content of 17.5 per cent.
The dried corn remained for 37 days in as good condition as when put
in the car, its temperature rising from 57° to 67° F., or a total increase o£
10 degrees. The cool, undried corn began to show signs of deterioration
in 23 days, and five days later a point near the surface reached a tem-
perature of 122° F. This indicates the increased tendency of the damp
grain to heat in storage.
Shanahan, Leighty, and Boerner {1910) examined cargoes of American
corn on arrival at European ports and observed an increased tendency
to heat and "go out of condition" as the moisture content increased.
Duvel and Duval (1913) Studied the temperature changes in carloads
of corn containing different percentages of moisture. In one experiment
Mar. i8, 1918
Respiration of Stored Wheat
691
running from March 2 to March 29, 191 1, the following temperature
changes were observed in cars held on track at Baltimore, Maryland:
Moisture.
Average temperature of corn when —
Loaded.
Unloaded.
Per cent.
21. 6
19.9
17.4
14. I
"F.
40. 0
40. 0
40. 0
40. 0
•F.
109.7
41-5
40.5
41-3
In a similar experiment, running from May 11, 191 1, to June i and 3,
191 1, corn containing 16.9 per cent of moisture or more was heating
when unloaded, while that which contained 13.9 per cent of moisture
was still cool. The extent of heating bore a fairly definite relation to
the shrinkage or loss in weight of the grain.
Bailey (1917a) reported to the Second Interstate Cereal Conference in
1916 the results of storage experiments with wheat at Duluth, Minn.
In the cool climate of that city it was found that wheat containing 15.5
per cent of moisture when put in a bin in the fall kept 333 days before
it developed a sufficiently high temperature to necessitate turning it,
while wheat containing 16.5 per cent of moisture was actively heating
in 49 days.
The exact reason for an accelerated respiration with an increased
moisture content had not been adequately explained. In the discussion
referred to in the paper mentioned in the preceding paragraph an hy-
pothesis was advanced to account for this relation. Moisture in grain
may, in the light of recent discoveries in the field of physical chemistry,
be assumed to exist as imbibed water in loose combination with the
organic colloids. The organic colloids which form the principal con-
stituents of the wheat kernel have the property of imbibing considerable
quantities of water and forming elastic gels. The gel swells as the water ^
is increased, although the total volume of the dry colloid plus the added
water is diminished. The water-imbibing capacity of the several colloids
varies widely, starch having an imbibing capacity materially lower than
that of wheat gluten. There is no fixed amount which a given dry
colloid will imbibe; thus, gels of varying viscosity can be produced,
depending upon the proportion of water present, and other variables,
such as temperature, mineral salts, and other substances. The rate of
diffusion in a gel varies with the viscosity, as pointed out by Plimmer ■
(1975, p. 386). In dilute gels diffusion takes place as in water, while in
strong gels the rate is slower. It is probable that in very dry grain the
imbibed water is not sufficient to produce a gel in the endosperm struc-
tures. The colloidal material there located accordingly does not have a
692
Journal of Agricultural Research
Vol. XII, No. n
continuous structure, and the possibilities of diffusion are decidedly
reduced under such conditions. The exact percentage of moisture below
which this discontinuous structure exists in a normal wheat kernel is
not known; it probably varies with the percentage of gluten in the grain
since gluten possesses a greater water-imbibing capacity than starch.
Increasing the moisture content above the maximum at which discon-
tinuity exists results in the formation of a gel through which diffusion
AS
OF /'TO/SrU/PS
Fig. I. — Graph showing the relation of the moisture content of wheat to the rate of respiration.
can occur. Further increases in moisture content up to maximum imbi-
bition produce progressively less viscous gels, and correspondingly in-
crease the possible rate of diffusion. Since the rate of respiration in
grain doubtless depends in part upon the rate of diffusion between the
various kernel structures, it follows that the less viscous the gelatinous
material of which the cell contents are composed, the more rapid the
production of heat through respiration. To restate, the production of
heat is dependent upon the activity of the oxidizing enzyms of the
Mar. i8, 1918
Respiration of Stored Wheat
693
kernel, the complex phenomenon being known as respiration. The latter
is accelerated by an increase in the rate of diffusion, which in turn is
dependent upon the existence of a gel, and the relative viscosity of that
gel. For these reasons the moisture content of sound grain determines
to a considerable extent the rate of respiration and consequent liability
of heating when bulk grain is stored.
To determine the relation of moisture content to respiration in stored
wheat, a large sample of spring-sown Haynes Bluestem wheat known as
Minnesota 169 was obtained. The weight per bushel of the sample was
57K pounds (26.08 kgm.) ; the w^eight per i ,000 average kernels, 24.62 gm. ;
and it contained 2.21 per cent of nitrogen on the dry basis. It was then
divided into several portions, and each portion was brought to a different
moisture content, the percentages of moisture ranging from 12.50 to
17.07 per cent. The quantity of carbon dioxid respired per 24 hours by
each 100 gm. of dry matter is given in Table I and is shown graphically
in figure i. The rate of increase in respiration is fairly gradual from
12.50 to 14.78 per cent, but after the latter percentage is exceeded the
rate is markedly accelerated. The break in the curve occurs when the
moisture content slightly exceeds 14.5 per cent, and it is probable that
this represents about the maximum percentage of moisture that this class
of wheat may safely contain without danger of heating when stored in
bulk.
Table I. — Respiration of Haynes Bluestem (Minnesota i6g) -wheat, <^ incubated at j^ -8° C
for four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Per cent.
12. 50
13-93
14.78
15.42
\IC7n.
0-54
.65
.86
I. 62
Per cent.
16.08
16.65
17.07
Mgm.
2.88
6.86
II. 72
"Weight per bushel of sample, 57^4 pounds,
basis, 2.21 percent.
Weight per 1,000 kernels, 24.62 gm. Nitrogen on dry
The acceleration of respiration as the moisture content increases is
shown in Table II. These data are based upon the estimated respiration
values at even percentages of moisture, and the computed increase in
respired carbon dioxid for each increase of i per cent of moisture. In
computing these data the following formula was employed : —^ — ^^^
in which Km represents the respiration value at a particular percentage
of moisture, and Km _ i represents the respiration value for the same wheat
containing i per cent of moisture less than Km. It is evident that the
acceleration between 12 and 14 per cent of moisture is very gradual, while
it increases markedly after 14 per cent is exceeded.
694
Journal of Ag
ricultural Research
Vol. XII, No. II
Table II.
— Acceleration of the rate of respiration of hard spring wheat with increasing
moisture content
Formula.
Acceleration at following percentages of moisture.
12 to 13
13 to 14
14 to IS
IS to 16
16 to 17
Km — Km —
t
o. i6
0.17
0.66
I. 41
Km— 1
3. 02
/^(?/^.
24
RELATION OF THE CON-
SISTENCY OF THE
WHEAT KERNEL TO
THE RATE OF RESPI-
RATION
Reference has already
been made to the differ-
ence in the relative
water-imbibing capacity
of the various organic col-
loids of the wheat kernel.
Since starch and gluten
constitute a large propor-
tion of the endosperm,
their differences in this
regard are of principal
interest. Simple tests in-
dicate that the water-im-
bibing capacity of gluten
is materially greater than
that of starch. In con-
sequence it follov\rs that,
as a general rule, those
kernels which contain a
high percentage of gluten
will, at any particular
moisture content, be more
viscous than will kernels
containing a lower per-
centage of gluten.
The gluten content is also related to the relative consistency or hard-
ness of the wheat berry. In general, the hard, vitreous grains contain
a higher percentage of gluten than do the soft, starchy grains. Accord-
ingly we may expect exactly what we find — viz, that the soft wheats
are "tougher" and of a lower viscosity at any given moisture content
(within the limits found in ordinary commercial grain) than the hard,
vitreous wheats.
/2
/3 Af /s /e /7
Fig. 2.— Graphs showing the comparative rate of respiration
hard spring, soft red winter, and soft white winter wheat.
Mar. i8, 19 1 £
Respiration of Stored Wheat
695
It is commonly recognized in the grain trade that the keeping quali-
ties of soft wheats are inferior to those of hard wheat. Because of the
relation of respiratory activity to rate of diffusion, it should follow that
with the same moisture content respiration would proceed more rapidly
in a soft than in a hard or vitreous kernel. A sample of soft red winter
wheat of the Fultz variety was obtained from the Experiment Station
at Columbia, Mo., and another of white winter wheat from Grand Blanc,
Mich. The rate of respiration in these soft wheats at different moisture
contents was studied, and it was found that, except at the lower per-
centages of moisture, the rate of respiration was higher in the soft red
winter wheat than it was in the hard spring wheat, and still higher in
the white winter wheat. These data are given in Tables, III, IV, and
V, and graphically in figure 2. As shown by the graph, the curves tend
to converge at about 1 2 per cent of moisture, indicating that at less than
this moisture content the discontinuity of endosperm structure referred
to above may exist in sound wheats and respiration proceed at the ex-
pense of substances in the germ rather than by oxidation of materials
which diffuse to it from the endosperm.
It may further be seen that the quantity of heat evolved by hard spring
wheat containing 14.5 per cent of moisture, as evidenced by the rate of
respiration at that moisture content, was evolved by these soft wheat
samples when they contained about 13.6 to 13.8 per cent of moisture.
This is of interest, in view of the moisture limits prescribed in the United
States Grain Standards for wheat, which are I4>^ per cent for No. 2 hard
spring, and 1 3 per cent for No. 2 soft red winter, and the same for com-
mon white and white club wheat.
Table III. — Respiration of soft red winter wheat (^ from Missouri, incubated at j/.8'°C.
for four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. 01
dry Eoatter.
Per cent.
13-07
14.70
Mgm.
0. 65
.80
•95
Per cent.
15-45
16.37
17. 40
Mgm.
2. 00
5.06
22.03
o Weight per bushel of sample, 61 pouadi. Weight per 1,000 Iceruels, 29.97 gm- Nitrogen on dry basis.
1.54 per cent.
696
Journal of Agricultural Research
Vol. XII, Mo. II
Table IV. — Respiration of white winter wheat "■ from Michigan, incubated at 27-8°C.
for four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 jrm. of
dry matter.
Per cent.
11.94
13.04
14.32
Mgm.
0.48
.60
.89
Per cent.
15-57
16.83
Mgm.
3. 20
22. 77
<J Weight per bushel of sample, 59 pounds. Weight per 1,000 kernels, 37.73 gm. Nitrogen on dry basis,
1.53 per cent.
Table V. — Interpolated quantity of carbon dioxid respired per unit of time and material,
at even percentages of m,oisture
Class of wheat.
Hard spring . . . .
Soft red winter.
White winter. . .
Carbon dioxid respired per 24 hours for each 100 gm. of dry matter.
per cent
moisture.
Mgm.
O. CO
49
13
per cent
mojsture.
Mgm.
C. K&
■ h
. 60
14
per cent
moisture.
Mgm.
0.68
.81
■83
IS
per cent
moisture.
Mgm.
I- 13
1-37
4- 15
16
per cent
moisture.
Algm..
2. 72
3-84
9-85
17
per cent
moisture.
Mgm,.
IO-73
15-51
25.18
RELATION OF THE RELATIVE PLUMPNESS OF THE WHEAT KERNEL
TO THE RATE OF RESPIRATION
It is generally recognized that the velocity of enzym action conforms
quite closely to the law of mass action. Thus any condition which
affects the quantity of either the substrate or the enzym will cause
variations in the rate of the reaction. Since respiration is occasioned
by enzyms, the rate of respiration of the wheat kernel should vary with
these conditions.
A shriveled condition of the wheat kernel is due generally to factors
operating during the later stages of kernel development. The trans-
location of reserves to the kernel is interfered with by rust, drouth,
desiccation by hot winds, or some other agency; and an incomplete
filling of the endosperm results. According to Brenchley {igog), the
germ portion of the kernel is developed earlier than the endosperm,
and tends to escape injury from the agency causing shrunkenness of
the endosperm more than does the latter. The diminished size and
weight of the shriveled kernel is therefore due principally to the decreased
quantity of endosperm.
The enzymic activities of the kernel seem to be mainly invested in the
embryo. This was discussed in one of the foregoing paragraphs. In
the embryo of the shriveled wheat berry the enzyms are probably repre-
sented practically as they are in the normal kernel. In instances where
the average weight of the individual kernel is only about half the normal,
Mar. i8, 1918
Respiration of Stored Wheat
697
it follows that there is approximately twice the enzymic activity per
unit of mass than that shown by normal wheat. The normal spring
wheat used by the writers weighed 24.62 gm, per 1,000 average kernels.
A shriveled sample of the same type of wheat was obtained which weighed
11.73 g™- per 1,000 kernels, or less than half the weight of the normal.
The rate of respiration of these two lots was compared, and, as shown
in figure 3, the respiratory activity of the shriveled sample decidedly
exceeded that of the normal, or plump wheat. Thus the quantity of
carbon dioxid respired by the latter when it contained 14.5 per cent of
moisture was respired by the shriveled wheat used when it contained
only 12.8 per cent of moisture. The curves tend to converge at moisture
contents slightly below 12 per cent.
The respiration data for the shriveled sample are given in Table VI,
while in Table VII is shown the interpolated quantity of carbon dioxid
respired by the normal and shriveled wheats at even percentages of
moisture.
Table VI. — Respiration of shriveled spring wheat, '^ incubated at 27-8° C . for four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Per cent.
12.68
13- 19
14. 29
15-30
Mgm.
0.75
.94
1.38
3.02
Per cent.
15.68
16. 09
16.44
16.80
Mgm.
4- 50
10. 51
16. 92
21. 65
a Weight per bushel of sample, 47M pounds. Weight per 1,000 kernels, 11.73 gm. Nitrogen on dry
basis, 2.03 per cent.
Table VII. — Interpolated quantity of carbon dioxid respired per unit of time and material
at even percentages of moisture
Carbon dioxid respired per 24 hours for each 100 gm. of dry matter.
Class of wheat.
12
per cent
moisture.
13
per cent
moisture.
14
per cent
moisture.
IS
per cent
moisture.
16
per cent
moisture.
17
per cent
moisture.
Plump spring wheat
Mgm.
0. 50
•65
Mgm.
0.58
.88
Mgm.
0. 68
1. 26
Mgm..
I- 13
2.54
Mgm.
2. 72
9.41
Mgm.
10.93
22. 65
Shriveled spring wheat
SOUNDNESS OF THE WHEAT KERNEL IN ITS RELATION TO THE RATE
OF RESPIRATION.
A form of unsoundness recognized as such by the grain trade and
frequently occurring in spring wheat is the frosted condition. This
results from the freezing of the plants before the grain is matured and
desiccated. The plants usually thaw later, and, vv^hile the protoplasm
698
Journal of Agricultural Research
Vol. XII. No. II
is killed or disorganized and its synthetic activities reduced or destroyed,
certain hydrolytic enzyms are activated, and hydrolysis or splitting of
certain of the kernel constituents ensues. As a result, there is an accumu-
lation of the split products of starch and proteins, particularly dextrose
and amino acids. It is probable that the extent of starch hydrolysis
by amylases depends in large part upon the percentage of moisture in the
grain after it thaws out. If the kernels are nearly dry, less change \vill
occur than if the kernels contain considerable moisture.
The dextrose which thus ac-
cumulates in the kernel is pre-
sumably available as substrate
for the respiratory enzyms.
In accordance with the law
of mass action, a greater con-
centration of substrate should
accelerate the rate of respira-
tion. There is an additional
factor in the case of frosted
wheat that would also tend to
result in an increased rate of
respiration at any particular
percentage of moisture. The
hydrolysis of the gluten sub-
sequent to thawing results in
products having a materially
lower water-imbibing capacity
than the normal gluten. In
fact, the amino acids formed
are not colloids, and form true
solutions. Consequently the
relative viscosity of frosted
grain at any moisture content
will be less than in normal
grain, the difiference depending
upon the extent of hydrolysis.
To ascertain the efifect of
frosting upon the rate of respiration, two samples of commercial
wheat containing frosted kernels were secured. These were marked
"moderately frosted" and "badly frosted," respectively. The
respiratory activity of these frosted samples was determined with
five different percentages of moisture present. Tables VIII and IX
show the respiration data of these two lots, while in Table X are
given the interpolated values at even percentages of moisture in com-
parison with sound spring wheat. The same data are shown graph-
ically in figure 4. There is a slight overlapping of the curves for the
I
I
\
to
Q /i? /S /^ /S /& /y'
Fig. 3. — Graphs showing the rate of respiration of shriveled
■wheat and of plump wheat of the same class.
[■
1
J
1
^1
r.l
if
fi
1
' 3
/
1
/
y
"^y
A
y
Mar. i8, 1918
Respiration of Stored Wheat
699
moderately and badly frosted samples between 14.5 and 15.5 per cent of
moisture, the reason for which is not clear. The discrepancy is not
great, and the similarity of the curves indicates that possibly our judg-
ment was at fault in describing the two lots of frosted wheat as "moder-
ately" and "badly" damaged. The decided differences in the rate of
/P
/O
\
w
•5^
li
/
^1
/
if.
1
ii
1 i
»/
/
J
1
,(C
f/
/
/
1
V
/
.y^j
/
/
^
'^
y
r
**'
—
y4?
/C?
/-S* /^ /6-
/T"
Fig. 4. — Graphs showing the rate of respiration of frosted wheat and sound wheat of the same class.
respiration of frosted and of normal wheat containing up to 16.5 per
cent of moisture substantiates what had previously been empirically
observed by the senior author (Bailey, 1917b) — viz, that frosted wheat
tends to heat more readily when stored than does sound wheat contain-
ing the same percentage of moisture.
700
Journal of Agricultural Research
Vol. XII, No. ir
Table VIII. — Respiration of moderately frosted wlieatfl incubated at J/.8° C. for
four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Per cent.
12.32
14.44
Mgm.
0.74
I. 04
1.89
Per cent.
14-95
15-42
Mgm.
3-75
5-21
oWeight per bushel of sample, 58 pounds. Weight per i,ooo kernels, 36.94 gm. Nitrogen on dry basis,
2.02 per cent.
Table IX. — Respiration of badly frosted wheatfl incubated at 37.8° C. for four days
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Moisture.
Carbon dioxid
respired per 24 hours
for each 100 gm. of
dry matter.
Per cent.
13-79
14.67
15-74
Mgm.
1.63
2. 64
5-24
Per cent.
16.81
16.95
Mgm.
9.82
II. 40
o Weight per bushel of sample, 53 pounds. Weight per 1,000 kernels, 22.52 gm. Nitrogen on dry basis,
2.10 per cent.
Table X. — Interpolated quantity of carbon dioxid respired per unit of time and m,ate-
rial, at even percentages of moisture
Class of wheat.
Carbon dioxid respired per 24 hours for each 100 gm. of dry matter.
per cent
moisture.
13
per cent
moisture.
14
per cent
moisture.
per cent
moisture.
16
per cent
moisture.
17
per cent
moisture.
Sound spring wheat
Moderately frosted. ,
Badly frosted
Mgm.
0. 50
-65
1. 12
Mgm,.
0.58
•94
I. 20
Mgm.
0.68
1.52
1.87
Mgm..
1-13
3-90
3-46
Mgm.
2. 72
Mgm.
10.93
6-35
11.97
Table XI. — Respiration of several samples of frosted wheat from, car lots, and containing
varying percentages of moisture. Incubated at 37.8° C. for four days
Laboratory No.
Moisture.
Weight per
bushel.
Weight per
1,000
kernels.
Nitrogen on
dry basis.
Carbon
dioxid re-
spired per
24 hours for
each 100 gm.
of dry
matter.
Interpolated
respiration
of sound
wheat of
same water
content.
G122
Per cent.
14.30
14.82
16.16
16. 19
Pounds.
60K
61
56K
57
Gm.
28.36
28. 64
23-56
28. 88
Per cent.
2. 23
2. 20
2. 27
2. 00
Mgm.
I. 71
2-34
8.71
8.24
Mgm.
1.30
1-73
6 86
Gl2^
GII8
GI02
7-15
Several wheat samples taken from car lots by samplers of the State
Grain Inspection Department were frosted, and the respiratory activity
Mar. i8, 1918
Respiration of Stored Wheat
701
of four of these was measured. In every case the rate of respiration
was greater than in sound wheat of the same moisture content which
was taken from car lots at the same time. The tests of these frosted
samples are shown in Table XI, the last column of which shows the
interpolated respiration data of sound wheat.
THE PERIOD OF DAMPNESS AS INFLUENCING THE RATE OF RESPI-
RATION OF WHEAT
When damp wheats are stored, the excess moisture which is responsible
for their dampness has been present for varying lengths of time. This
is called for the purposes of this discussion the "period of dampness."
In connection with these studies a series of wheat samples was taken
from car lots by the State Grain Inspection Department between the 15th
and 27th of March, which contained varying percentages of moisture.
The respiratory activity of each of these was determined, and on plot-
ting a curve it was observed that these samples gave materially higher
values than the samples which had been dampened and allowed to
stand for three days before measuring the rate of respiration. The
curves tend toward convergence at a moisture content of 12 per cent.
These data are given in Tables XII and XIII and are shown graphically
in figure 5.
Table XII. — Respiration of natural hard spring wheat collected from car lots, and
incubated at ^7.8° C. for four days
Laboratory No.
G128
G127
G124
G120
G104
G103
G107
G109
Moisture.
Per cent.
12.47
13. II
14.70
15-51
15-73
16. 00
16.53
16. 90
Weight per
bushel.
Pounds.
62
63
58i
52
59
59
56
51
Weight per
1,000 kernels.
Gm.
30-56
32-94
24. 60
26. 92
28.64
25.72
22.88
Nitrogen in
dry matter.
Per cent.
2. 14
2. 16
2.
2.
2.
2.
05
31
47
37
2.25
Carbon
dioxid re-
spired per
24 hours for
each 100 gm.
of dry matter.
Mgm.
O. 61
•75
1.49
3.26
3-94
5- 69
10. 65
13.86
Table XIII. — Interpolated quantity of carbon dioxid respired per unit of time and mate-
rial, at even percentages of moisture
Carbon dioxid respired per 24 hours for each 100 gm. of dry matter.
Class of wheat.
13 13
per cent per cent
moistiu-e. moisture.
14
per cent
moisture.
IS
per cent
moisture.
16
per cent
moisture.
17
per cent
moisture.
Freshly dampened wheat ....
Natural wheat
Mgm.
0. 50
•51
Mgm.
0.58
•73
Mgm.
0.68
I- 15
Mgm.
I- 13
2. 14
Mgm.
2. 72
5-69
Mgm.
IO-73
15-03
3832G°— 18 2
702
Journal of Agricultural Research
Vol. XII, No. II
Since the principal difference between the samples dampened by the
writers and the damp grain obtained from the freight cars was the length
of the period of dampness, an experiment was conducted to ascertain
the extent to which that variable affected the rate of respiration. Two
samples of Bluestem wheat that had been dampened some time previous
Af&/*f.
/v? M /s /e
/7
Fig. s.— Graphs showing the comparative respiratory activity of naturally damp wheats and of wheats
dampened in the laboratory three days before they were incubated.
and stored at a temperature of about 25° C. in the laboratory vault
were incubated in the usual manner at 37.8° C. with the results shown
in Table XIV. The quantity of carbon dioxid respired by the sample
containing 15.21 per cent of moisture, which was stored for 55 days after
it was dampened, was about four times as great as that respired by
freshly dampened wheat of the same moisture content, while that from
Mar. 18, 1918
Respiration of Stored Wheat
703
the sample containing 15.71 per cent of moisture, which was stored for
108 days, was about eight times as great as for freshly dampened wheat
containing that percentage of moisture.
Table XIV. — Respiration of dampened wheat after storage at about 25° C. Incubated
at 37.8° C. for four days
Moisture per cent. .
Number of days stored
Carbon dioxid respired per 100 gra. of dry matter in each 24
hours mgm . .
Carbon dioxid respired per 100 gm. of dry matter in each 24 hours,
of same lot of wheat 4 days after dampening mgra . .
Carbon dioxid respired by naturally damp wheat from car lots,
of same moisture content mgm . .
Lot B.
15-72
108
17. 00
2. 17
3-85
Table XIV also shows a decided difference in the comparative rate of
respiration of wheat stored for a time in a warm room and that obtained
from freight cars. The latter had no doubt been damp for a longer time
than that which was dampened and held in the room. These data
suggest that not only does the period of dampness affect respiration,
but the conditions of storage may have an equally important effect.
The grain taken from cars was cold at the time, and had probably been
exposed to the cold atmospheric conditions of the preceding winter
months. If, as postulated by the writers, dextrose tends to accumulate
in the stored damp grain the rate of accumulation would depend upon the
temperature of the grain as well as upon the time elapsed. Accord-
ingly there may have been more substrate (dextrose) for the respiratory
enzyms in the grain which had been stored in the warm room than
there was in that stored out of doors during the winter months. A
series of experiments with both the period of dampness and the tempera-
ture of storage as variables have accordingly been begun, and the results
will be reported later in another publication.
INFLUENCE OF TEMPERATURE ON THE RESPIRATION OF STORED
WHEAT
Pfeffer (1878) observed that the intensity of respiration increases
with the temperature until the latter begins to injure all the vital proc-
esses. Hoff {1896, p, 12^) stated that the rate of respiration increases
two or three times for each lo-degree rise in temperature in accordance
•vvith the usual rule of chemical reactions. Ziegenbein {i8gj) found
that temperatures above 45° C. were injurious
Clausen {i8po) studied the respiration of germinating wheat at differ-
ent temperatures and found the optimum to be about 40° C. The rate of
respiration was 2.86 times as great at 10° as at 0° and 1.09 times as great
704 Journal of Agricultural Research \o\. xii, no. h
at 40° as at 30°. The average increase between 0° and 40° was 2.71
times for each lo-degree increase in temperature.
Qvam (1906) found the rate of respiration increased up to at least 45° C,
which was apparently the highest temperature at which observations
were made. The grain was very moist, 100 gm. of water having been
added to 200 gm. of grain.
Matthaei (1905) investigated the respiration of cherry-laurel leaves at
different temperatures and reported an increasing rate of respiration
between 5.8° and 33.1° C. At 5.8° 2 gm. of green leaves respired o.i
mgm. of carbon dioxid per hour, and at 33.1° the rate was 1.35 mgm. per
hour.
The experiments of Duvel and Duval (1913) with shelled corn indicate
the relation of air temperatures to the heating of such material. Shelled
corn in transit and on track containing 16.9 per cent of moisture began
to heat and go out of condition between May 1 1 and June 3, 191 1, while
in the period from December 24, 1910, to January 20, 191 1, corn con-
taining 22 per cent of moisture gained only a few degrees in temperature.
The purely physical factor of heat loss into the cold winter atmosphere,
of course, served to reduce the rate of rise in temperature, but there was
probably a diminished rate of evolution of heat as well.
Attention has been called by Bailey {1917a), to the influence of at-
mospheric temperatures upon the rate of heating of damp wheat. A
lot of wheat containing 16.5 per cent of moisture required but 11 days
to increase in temperature from 70° to 80° F., when the mean air temper-
ature was 62.1° F., while later in the year, when the mean air tempera-
ture was 44.3° F., another lot of wheat containing the same percentage
of moisture was stored 49 days before its temperature increased to the
same extent.
Another comparison of the influence of temperature on the rate of
heating in storage was afforded by two lots of wheat put into bins at about
the same time and containing nearly the same percentage of moisture.
The initial temperature of one lot was 74° F., and of the other 70°. The
latter required over five times as long to reach a temperature of 80° as
did the former, owing to the slow increase in temperature at the outset
as contrasted with the rapid rate of increase as the temperature mounted
higher.
To ascertain the relation of temperature to the rate of respiration in
stored grain, a large sample of Minnesota i69Bluestem wheat was damp-
ened until it contained 14.96 per cent of moisture. Aliquots of this
sample were sealed in glass jars and kept in a refrigerator until they were
used. This was done to minimize enzymic changes in the grain. The
necessary quantities were drawn from the refrigerator for incubation at
the several temperatures.
The lowest temperature at which observations were made was 4°C.,
and since the increase in rate of respiration between 4° and 25° was rela-
Mar. i8, 1918
Respiration of Stored Wheat
705
lively small, no intermediate temperatures were employed. Regular
increases in the temperature of the thermostat by lo-degree intervals
were then made until the respired carbon dioxid showed a marked
diminution.
The data in Table XV and the graph in figure 6 show that the rate of
respiration increased to a maximum at 55° C. This is therefore the tem-
perature at which the most rapid evolution of heat would occur. A dis-
coloration of the seed coat of the wheat kernel begins to show on some
kernels at about 35° C, while at 55° the whole mass is of a mahogany
color. At 65° the respiratory enzyms have been ' partially but not
J
\
/
\
/
\
/
\
/
\
\
/
f
0
.. . «
1
/s
2'S 35 '^S .5i5
es
7^
Fig. 6. — Graph showing the relation of temperature to the rate of respiration.
wholly inactivated, while at 75° this inactivation has proceeded still
further, and some roasting of the grain has occurred.
Table XV. — Respiration of hard spring wheat at different temperatures
■ Carbon dioxid
Carbon dioxid
Temperature.
respired per 24 hours
Temperature.
respired per 24 hours
for each 100 gm. of
for each icxj gm. of
dry matter.
dry matter.
•c.
Mgm.
"C.
Mgni.
4
0. 24
55
31-73
25
•45
65
15-71
35
1.30
75
a 10. 28
45
6.61
"A part of this carbon dioxid may have resulted from roasting the grain.
7o6
Journal of Agricultural Research
Vol. XII, No. II
The proportional change in respiration for each lo-degree rise in tem-
perature is shown in Table XVI. The data in this table were obtained
by employing the conventional formula -^ in which Vt represents the
rate of respiration at the specified temperature, and F^+^o represents
the rate at a temperature lo degrees higher. The values at 5° and
15° were computed by integrating the actual data obtained at 4° to
25° and 35° C.
Table XVI. — Acceleration of respiration for each lo-degree increase in temperature
t.
Vt
■
t.
Vt
° c.
5
25
I. 16
1-55
2.89
° C.
35
45
55
S.oS
4. So
•49
INFLUENCE OF ACCUMULATED CARBON DIOXID UPON RESPIRATION
Muntz {1881) observed that ten times as much carbon dioxid was
respired when grain had access to free air as when sealed air-tight and
that the yield of carbon dioxid was much greater after the moisture
exceeded 13 to 14 per cent.
Mangin (i8g6) found the evolution of carbon dioxid and absorption
of O2 to be reduced when germinating seeds were put in air containing
CO
up to 5 per cent of carbon dioxid. The -^r ratio was increased, indi-
eating that absorption of oxygen was diminished more than the evolu-
tion of carbon dioxid.
Duvel {1904), Babcock (191 2), Barnes and Grove (19 16), and others
have called attention to the reduced vitality and germination of seeds
stored in carbon dioxid, or in tight containers in which the respired car-
bon dioxid accumulated.
In the case of grain stored under ordinary commercial conditions it
follows that the oxygen in the space surrounding the kernels must be
replaced with respired carbon dioxid. The rate of such replacement
will, of course, hinge upon the factors influencing the rate of respira-
tion. To determine the relative change in the respiration of grain stored
in a tight container in which the respired carbon dioxid must accumu-
late, the following experiment was conducted: A sample of wheat
containing 15.05 per cent of moisture was divided into four portions,
which were incubated in the usual manner at 37.8° C. At the end of i,
4, 8, and 12 days a cylinder was removed and the respired carbon dioxid
determined. The rate of respiration for the first day and the average
rate for each of the 4-day periods are shown in Table XVII and graph-
Mar. iS, 1918
Respiration of Stored Wheat
707
ically in figure 7. Both table and graph show plainly that the rate of
respiration is reduced by the accumulated carbon dioxid, and it is prob-
able that a further reduction in respiratory activity would result if the
proportion of carbon dioxid were increased. This fact has an important
bearing on commercial practices. It indicates that stored grain should
not be disturbed so long as its temperature does not exceed that of the
atmosphere, since exposing it would serve to ventilate the grain and thus
P2
Af(P/V.
N
^
.^
^\
y
A
y
/
>
y
/
y
t
r
/
/
J
/
/
/
/
3 ^ S £^ 7- <9 ^
/iP //
/^
Fig. 7. — Graph showing the rate of respiration during successive intervals when the respired carbon
dioxid was permitted to accumulate in the mass of grain.
remove the carbon dioxid. On returning such grain to the bin after
aerating it the rate of respiration would be increased over that of the
grain which was not handled and exposed, since the rate of respira-
tion of the unaerated grain has been depressed, owing to the accumulated
carbon dioxid. The aerated grain will accordingly heat more rapidly
than before aerating, unless its temperature has been materially reduced
by exposing it to the air.
7o8
Journal of Agricultural Research voi. xii, no. n
Table XVII. — Rate of respiration per day for several successive periods
Period.
Carbon dioxid
respired per
loo gm. of dry
matter.
First day
Average rate per day for first 4-day period. . .
Average rate per day for second 4-day period
Average rate per day for third 4-day period. .
Mgm.
4. II
2.68
1.49
I. II
RESPIRATION IN OXYGEN-FREE ATMOSPHERE
That respiration may occur in the absence of oxygen was first dis-
covered by Rollo in 1798 {Hill, 191 3) in vi^orking with barley grains.
Other investigators confirmed this observation, and Pfefifer (1S78)
suggested the term "intramolecular respiration" for this class of phe-
nomena. Takahashi (1903) reported that rice can germinate in water
without the presence of sugar and in the entire absence of any air.
Hill (1913) determined the rate of respiration of water-soaked and
sterilized wheat in air, nitrogen, and hydrogen. The decrease in respira-
tion in a continuous current of hydrogen and nitrogen below that in a
continuous current of air was about 50 per cent in seeds sterilized in
alcohol and about 80 per cent in seeds sterilized in formalin.
It appeared desirable to determine the effect of the elimination of
oxygen upon the respiration of stored wheat, and to this end the follow-
ing experiment was conducted: Two lots of wheat containing 15.6 per
cent and 17.6 per cent of moisture respectively were secured. A por-
tion of each lot was sealed in cylinders, the air was removed and replaced
by nitrogen. These lots and controls in ordinary atmospheric air freed
from carbon dioxid were then incubated for four days at a temperature
of 23.9° C. The data in Table XVIII show the rate of respiration in
the oxygen-free atmosphere to have been reduced to about two-fifths
of that in a normal atmosphere.
Table XVIII. — Comparative rate of respiration in oxygen-free and normal atmosphere
Moisture.
Carbon dioxid respired per 100 gm.
of dry matter in eacli 24 hours.
Oxygen-free
atmosphere.
Normal
atmosphere.
Per cent.
15.6
17.6
Mgm.
0.43
2.80
Mgm.
I. 10
6.80
Mar. i8. 1918 Respiratiou of Stored Wheat 709
CONCLUSIONS
(i) Deductions from these investigations support the findings of
earlier investigators that spontaneous heating in damp grain is occa-
sioned by the biological oxidation of dextrose and similar sugars, chiefly
in the germ or embryo of the kernel.
(2) Moisture is one of the determining factors in respiration. It
establishes the comparative rate of diffusion between the several kernel
structures. Any gain in the moisture content of the kernel accord-
ingly increases the rate of diffusion and, simultaneously, the rate of
respiration. The increase is gradual and fairly uniform until the mois-
ture exceeds 14.5 per cent, in the case of plump spring wheat, when it
is markedly accelerated.
(3) Density of the wheat Kernel generally parallels the gluten con-
tent. Gluten possesses the property of imbibing more water than starch,
and thus varying percentages of gluten result in varying degrees of
viscosity at the same moisture content. The relative viscosity affects
the rate of diffusion and this in turn directly affects the rate of respira-
tion. The soft, starchy wheats thus respire more rapidly than hard,
vitreous wheats containing the same percentage of moisture.
(4) Plumpness of the wheat kernel affects the rate of respiration, as
shown by contrasting plump and shriveled grain. The shriveled wheat
respired two to three times as much as did the plump wheat at moisture
contents above 14 per cent. At percentages of moisture below 14 per
cent the difference is not very marked. The high acceleration of respira-
tion in shriveled wheat containing more than 14 per cent of moisture
is attributed to the higher ratio of germ to endosperm and hence the
larger percentage of enzym to substrate as compared with plump wheat.
(5) The period of dampness — that is, the length of time the excess
moisture has been present in the wheat — bears a relation to the rate of
respiration. This is shown by comparing the respiration of freshly
dampened wheat with that of naturally damp grain and with grain
that had been dampened and stored for varying lengths of time. The
curve of respiration diverges from that of freshly dampened wheat when
the moisture content exceeds 12 per cent, and this divergence is more
marked after 13 per cent of moisture is reached. In the case of wheat
dampened and stored, the quantity of carbon dioxid respired varies
directly with the number of days the wheat remained in storage. The
temperature at which the grain is stored affects the rate of diastatic
action, thus increasing the quantity of substrate available to the respira-
tory enzyms. This is indicated by the greater rate of respiration of
wheat stored at room temperature than that stored at the outdoor
temperature during the winter months.
(6) Unsoundness of wheat caused by the freezing of the unripe plant
results in higher respiratory activity in the threshed grain. This was
7IO Journal of Agricultural Research. voi. xii. no. n
shown by comparing moderately and badly frosted wheats with sound
wheat. The frosted wheat respired more vigorously than the sound
wheat. This was attributed to the arresting of the synthetic processes
on freezing, and subsequent activities of the hydrolytic enzyms on thaw-
ing of the frozen wheat. The accumulation of glucose as the result
of starch hydrolysis furnishes larger quantities of substrate to the
respiratory enzyms.
(7) Increasing temperatures accelerate the rate of respiration until
55° C. is reached. As the temperature rises the diastatic action upon
starch increases. A point is reached, however, at which the enzym
activity diminishes.
(8) Accumulation of carbon dioxid in the respiration chamber
decreases the rate of respiration. The mean rate by four-day intervals
is highest for the first four days and diminishes materially in successive
periods.
(9) Respiration is reduced in an oxygen-free atmosphere, the ratio
to that occurring in a normal atmosphere being about i to 2.5.
LITERATURE CITED
ApplEman, CO.
1915. RELATION OP CATALASE AND OXIDASES TO RESPIRATION IN PLANTS. Md. AgT.
Exp. Sta. Bul. 191, 16 p., 2 fig. Literature cited, p. 16.
Babcock, S. M.
I912. METABOLIC water: ITS PRODUCTION AND r6lE IN VITAL PHENOMENA. Wis.
Agr. Exp. Sta. Research Bul. 22, p. 87-181.
Bailey, C. H.
1917a. THE HANDLING AND STORAGE OF SPRING WHEAT. hi Jour.- Amer. SoC.
Agron., V. 9, no. 6, p. 275-281, fig. 16-20.
1917b. THE MOISTURE CONTENT OF HEATING WHEAT. In Jour. Amer. Soc. Agron.,
V. 9, no. 5, p. 248-251.
Bailey, E. M.
1912. biochemical and bacteriological STUDIES ON THE BANANA. In JoUr.
Amer. Chem. Soc, v. 34, no. 12, p. 1706-1730.
Barnes, J. H., and Grove, A. J.
I916. THE INSECTS ATTACKING STORED WHEAT IN THE PUNJAB, AND THE METHODS
OF COMBATING THEM, INCLUDING A CHAPTER ON THE CHEMISTRY OF RESPI-
RATION. In Mem. Dept. Agr. India, Chem. s., v. 4, no. 6, p. i65-28od,
12 fig., 4 col. pi.
BoNNiER, Gaston, and Mangin, Louis.
1885. LA FONCTiON RESPiRATOiRE CHEZ LES v^G^TAUX. In Ann. Sci. Nat. Bot.,
s. 7, t. 2, p. 365-380.
Brenchley, W. E.
1909. ON THE STRENGTH AND DEVELOPMENT OP THE GRAIN OP WHEAT (TRITICUM
vulgare). In Ann. Bot., v. 23, no. 89, p. 117-139, 5 fig., pi. 8-9. Litera-
ture cited, p. 138.
Mar. i8, 1918 Respiratiofi of Stored Wheat 711
Brown, H. T., and Morris, G. H.
1890. RESEARCHES ON THE GERMINATION OP SOME OP THE GRAMINEAE. I. In
Jour. Chem. Soc. [London,] v. 57, p. 458-528, 2 pi.
BURLAKOW, G.
1897. UEBER ATHMUNG DES KEIMES DBS WEIZENS, TRITICUM VULGARE. (Abstract.)
In Bot. Centbl., Bd. 74, No. 11, p. 323-324. 1898. (Original article in
Arb. Naturf. Gesell. K. Univ. Charkow, Bd. 31, Beilage, p. i-xv. 1897.
Not seen.)
Clausen, H.
1890. BEITRAGE ZUR KENNTNISS DER ATHMUNG DER GEWACHSE. In Landw. Jahrb.,
Bd. 19, p. 893-930, pi. 19-20.
DuvEL, J. W. T.
1904. THE VITALITY AND GERMINATION OP SEEDS. U. S. Dcpt. Agr. BUT. Plant
Indus. Bui. 58, 96 p., I fig. Literature cited, p. 90.
1909. THE DETERIORATION OP CORN IN STORAGE. U. S. Dept. Agr. Bur. Plant
Indus. Circ. 43, 12 p., 3 fig.
, and Duval, Laurel.
I913. THE SHRINKAGE OF SHELLED CORN WHILE IN CARS IN TRANSIT. U. S. Dcpt.
Agr. Bui. 48, 21 p., 8 fig.
Gore, H. C
1911. STUDIES ON FRUIT RESPIRATION. U. S. Dept. Agr. Biu". Chem. Bui. 142, 40 p.,
17 fig-
1914. CHANGES IN COMPOSITION OP PEEL AND PULP OP RIPENING BANANAS. In JoUT.
Agr. Research, v. 3, no. 3, p. 187-203, i fig. Literature cited, p. 202-203.
GURJAR, A. M.
I917. THE ADAPTATION OP TRUOG'S METHOD FOR THE DETERMINATION OP CARBON
DioxiD TO PLANT RESPIRATION STUDIES. In Plant World, V. 20, no. 9,
p. 288-293, 1 fig-
Hasselbring, Heinrich, and Hawkins, L. A.
1915. RESPIRATION EXPERIMENTS WITH SWEET POTATOES. In Jour. Agr. Research,
V. 5, no. 12, p. 509-517. Literature cited, p. 517.
HiLL, G. R., Jr.
I913. RESPIRATION OP FRUITS AND GROWING PLANT TISSUES IN CERTAIN GASES,
WITH REFERENCE TO VENTILATION AND FRUIT STORAGE. N. Y. Cornell Agr.
Exp. Sta. Bui. 330, p. 373-408. Bibliography, p. 407-408.
Cites Rollo, p. 379-
HoPP, J. H. van't.
1896. STUDIES IN CHEMICAL DYNAMICS. Translated by Thomas Ewan. 286 p., 49
fig. Amsterdam, London.
KarchEvski.
1901. influence of the fluctuation op temperature on THE respiration OP
SEEDS AND EMBRYOS OP WHEAT. (Abstract.) In Exp. Sta. Rec, v. 14,
no. 9, p. 839. 1903. (Original article in Mem. Warsaw Univ., v. 9, 114 p.
1901. Not seen.)
KOLKWITZ, R.
1901. UEBER DIE ATHMUNG RUHENDER SAMEN. In Ber. Dcut. Bot. Gesell., Bd. 19,
Heft 4, p. 285-287.
712 Journal of Agricultural Research voi. xii. No. n
Langworthy, C. F., and Milner, R. D.
I913. SOME RESULTS OBTAINED IN STUDYING RIPENING BANANAS WITH THE RESPIRA-
TION CALORIMETER. In U. S. Dept. Agr. Yearbook, 1912, p. 293-308.
LE Clerc, J. A., and Breazeale, J. F.
191 1. TRANSLOCATION OF PLANT FOOD AND ELABORATION OP ORGANIC PLANT MATE-
RIAL IN WHEAT SEEDLINGS. U. S. Dept. Agr. Bur. Chem. Bui. 138, 32 p.
LoEw, Oscar.
1899. CURING AND FERMENTATION OF CIGAR LEAF TOB.'VCCO. U. S. Dept. Agr. Dept.
Rpt. 59, 34 p.
Lund, J. F.
1894. NOTE SUR L'iNFLUENCE DE LA DESSICCATION SUR LA RESPIRATION DBS TUBER-
CULES. In Rev. Gen. Bot., t. 6, p. 353-355.
Maige, a., and Nicolas, G.
1910. RECHERCHES SUR L 'INFLUENCE DES SOLUTIONS SUCR^ES DE DIVERS DEGR^S
DE CONCENTRATION SUR LA RESPIRATION , LA TURGESCENCE ET LA CROISSANCE
DE LA CELLULE. In Ann. Sci. Nat. Bot., s. 9, t. 12, no. 2/6, p. 315-368.
Mangin, Louis.
1896. SUR LA V]6g6TATION DANS UNE ATMOSPHl^RE VIClfiS PAR LA RESPIRATION.
In Compt. Rend. Acad. Sci. [Paris], t. 122, no. 12, p. 747-749.
Mann, Albert, and Harlan, H. V.
191 5. MORPHOLOGY OF THE BARLEY GRAIN WITH REFERENCE TO ITS ENZYM-SECRETING
AREAS. U. S. Dept. Agr. BuL 183, 32 p., 7 fig., 8 pi.
Maquenne, L.
1900. RECHERCHES SUR LA GERMINATION. In Ann. AgTon., t. 26, no. 7, p. 321-332.
Matthaei, G. L. C.
1905. EXPERIMENTAL researches ON VEGETABLE ASSIMILATION AND RESPIRATION.
III. ON THE EFFECT OF TEMPERATURE ON CARBON-DIOXIDE ASSIMILATION.
In Phil. Trans. Roy. Soc. [London], s. B, v. 197, no. 228, p. 47-86, 6 fig.
Bibliography, p. 85-86.
MtJNTZ, A.
1881. SUR LA CONSERVATION DES GRAINS PAR l'ensilage. In Compt. Rend.
Acad. Sci. [Paris], t. 92, no. 3, p. 137-139.
Nabokich, a. J.
1903. UBER DEN EINFLUSS DER STERILISATION DER SAMEN AUP DIE ATMUNG. In
Ber. Deut. Bot. Gesell., Bd. 21, Heft 5, p. 279-291.
OSTERHOUT, W. J. V.
1917. THE r6le of the nucleus IN OXIDATION. In Science, n. s., v. 46, no. 1189,
P- 367-369-
Pfeffer, W.
1878. DAS WESEN UND DIE BEDEUTUNG DER ATHMUNG IN DER PFLANZE. In Landw.
Jahrb., Bd. 7, p. 805-834.
PFLtJGER, E.
1875. UEBER DAS PHYSIOLOGISCHE VERBRENNUNG in DEN LEBENDIGEN ORGANISMEN.
In Arch. Physiol. [Pfluger], Bd. 10, p. 251-367.
Plimmer, R. H. a.
1915. PRACTICAL ORGANIC AND bio-chemistry. 635 p., 86 fig., I col. pi. London,
New York.
Mar. i8, igis RespiratioYi of Stored Wheat 713
QvAM, Olaf.
1906. KORNETS AANDING. EN RELATION MELLEM SPIREEVNE OG AANDINGSINTEN-
siTET. Tidsskr. Norske Landbr., Aarg. 13, p. 263-284, 14 fig.
Rahn, Otto.
1910. THE usefulness of curves in the interpretation op microbial and
BIOCHEMICAL PROCESSES. Mich. AgT. Exp. Sta. Tech. Bui. 5, p. 347-372,
18 fig.
SaussurE, Theodorde.
1804. recherches chimiques sur la v6g6tation. 327 p., i pi. Paris.
Shanahan, J. D., Leighty, C. E., and Boerner, E. G.
1910. AMERICAN EXPORT CORN (maize^ IN EUROPE. U. S. Dcpt. AgT. BuF. Plant
Indus. Circ. 55, 42 p., 7 fig.
Suchsland, Emil.
1891. UEBER TABAKSFERMENTATiON. In Bet. Deut. Bot. Gesell., Bd. 9, Heft i,
p. 79-81.
Takahashi, T.
1905. is GERMINATION POSSIBLE IN THE ABSENCE OF AIR? (Abstract.) In Exp.
Sta. Rec, v. 16, no. 11, p. 1090. 1905. (Original article in Bui. Col. Agr.
Tokyo Imp. Univ., v. 6, no. 4, p. 439-442. 1905. Not seen.)
Troug, E.
I915. METHODS for the DETERMINATION OF CARBON DIOXIDE AND A NEW FORM OF
ABSORPTION TOWER ADAPTED TO THE TITRIMETRIC METHOD. In Jour.
Indus, and Engin. Chem., v. 7, no. 12, p. 1045-1049, i fig.
Verworn, Max.
1899. GENERAL PHYSIOLOGY. Translated by F. S. Lee. 632 p., 285 fig. London.
Wehmer, C.
1892. oxalsaures ammon als pilzliches stoffwechselproduct die ernahrung
durch eiweiss. (Abstract.) In Bot. Centbl., Bd. 51, No. 11, p. 337.
1892. (Original article in Jahresber. Natiu-h. Gesell. Hannover, 1892,
p. 99-106. Not seen.)
Wender, Neumann.
1905. peinheitsbestimmung der mehle. In Ztschr. Untersuch. Nahr. u. Gen-
ussmtl., Bd. 10, Heft 12, p. 747-756.
White, Jean.
1909. the ferments and latent life of resting seeds. In Proc. Roy. Soc.
[London], s. B, v. 81, no. 550, p. 417-442. Special bibliography, p. 441-442.
Ziegenbein, Ernst.
1893. UNTERSUCHUNGEN USER DEN STOFPWECHSEL UNO DIE ATHMUNG KEIMENDER
kartoffelknollen sowie anderer pflanzen. In Jahrb. Wiss. Bot.
[Pringsheim], Bd. 25, Heft 4, p. 563-606.
EFFECTS OF MISTLETOE ON YOUNG CONIFERS
By James R. Weir
Forest Pathologist, Investigations in Forest Pathology, Bureau of Plant Industry,
United States Department of Agriculture
During the years 191 1 to 1917 the writer has been engaged in the
investigation of the injurious effects of mistletoes of the genus Razou-
mofskya (Arceuthobium) on conifers. A part of these investigations
have already been published.* During the course of the work, a number
of studies were made which have not yet been reported. It seems that
these studies are of sufficient importance to be presented at this time.
The fact that these parasites are a cause of suppression ^ in forest
trees is readily appreciated after the injury has become acute. In
middle-aged and older trees where accumulated injury has resulted in
small diameters and broomed scraggly crowns with reduced leaf surface,
the effect of the parasites is quite evident (PI. 37, A). The effects of
mistletoe on its host is by no means so apparent on trees ranging in age
from 4 to 10 years. Trees which have become infected early in life
may begin to react in marked degree, but the extent of the suppression
is not always apparent to the eye. It may be noticed that the branches
of young trees are broomed along with other types of infection of a more
general nature, but the retardation of the excurrent or elongated growth
of the main stem may not be readily recognized. This is where detailed
measurements are of value and prove or disprove the early suppression
of trees of the younger age classes by mistletoe.
In' order to demonstrate the suppression in young trees, a series of
measurements of the height growth of w^estern yellow pine (Pinus ponde-
rosa) were made in Spokane County, Washington. Two representative
plots of I acre each, consisting of infected and uninfected reproduction
and representing all age classes, were selected. The site was level bench
land with a sparse stand of merchantable-sized trees. Some of these trees
were severely infected with mistletoe {Razoumofskya campy lopoda
' Weir, J. R. larch mistletoe: some economic considerations of its injurious effects. U. S.
Dept. Agr. Bui. 317, 25 p., 13 fig. 1916.
mistletoe injury to conifers in the northwest. U. S. Dept. Agr. Bui. 360, 39 p., 27 fig.
X916. literature cited, p. 39.'
some suggestions on the control of mistletoe in the N.^TION.^L forests op the northwest.
In Forestry Quart., v. 14, no. 4, p. 567-577. 1916.
' The term "suppression" is not used here in the ordinary forestry sense, but refers to a retardation of
growth induced by parasitic organisms.
Journal of Agricultural Research, Vol. XII, No. n
Washington, D. C. (/^S) VltiT. 18, 1918
ml Key No. G— 138
7i6
Journal of Agricultural Research
Vol. XII, No. ir
(Engelm.) Piper] and were the source of infection of the reproduction.
The young trees were well distributed over the area and were dominant.
The results of this study are summarized in Table I and demonstrate
very clearly the damage which may result from attack by this parasite
on yellow pine.
Table I. — Effect of mistletoe on yellow pine
Age of tree.
Basis No.
Infected.
Years.
4 ' 8
5 lo
6 14
7 ! 9
8 j 19
9 1 10
10 1 7
I
Total 77
Unin-
fected.
50
27
42
35
20
23
16
213
Average height of trees.
Infected.
Cm.
15. 22
22. 84
35-53
48. 22
53-29
63-45
73.60
44,72
Uninfected.
Cm
25
35'
50'
71
78.
93
109
■38
•53
.76
.06
.68
.90
- 13
57-87
DiEference in
height be-
tween infect-
ed and un-
infected
trees.
Cm..
10. 15
12. 69
15.22
22. 84
25-38
30-45
35- 53
13-15
During the above study it was plainly observed that the length of the
intemodes of the infected trees, also the length of the terminal and lateral
buds of the main shoot, were much shorter than those of the uninfected,
and led to the following study on Douglas fir {Pseudotsuga taxijolia) in
the Missoula River region near Missoula, Mont. The trees were infected
with Razoumofskya douglasii Engelm. For this study trees were selected
over an area without recourse to sample plots. The only point adhered
to was the selection of trees on same type of site, condition of growth,
and an average age of 18 years. The results are given in Table II
Table II. — Effect of mistletoe on Douglas fir
Trees seriously infected.
Trees vigorous, not infected.
Average
Average di-
Average
Average di-
V,
Average height
dimen-
mension and
u
Average height
dimen-
mension and
■w
growth of last four
sion ter-
number of
4-»
growth of last four
sion ter-
number of
■3
intemodes.
minal
lateral termi-
a
intemodes.
minal
lateral termi-
1->
St
bud.
nal buds.
a
bud.
nal buds.
a
J3
.a
A
5
ja
.a
A
5
•a
n
I
3
3
4
1
n
No.
bo
%
P3
.52
I
"
3
4
ti
S
No.
ti
1
Cm.
Cm.
Cm.
Cm.
Mm.
Mm.
Mm.
Mm.
Cm.
Cm.
Cm,.
Cm.
Mm.
Mm.
Mm.
Mm.
so
28
23
33
18
8
3
loS
S
3
so
36
33
36
35
13
4
157
7
3 +
The foregoing results clearly demonstrate the eflFect of the formation
of brooms and burls on the storage of food materials in the terminal buds
and shoots. It is a well-known fact that in the terminal bud are stored
Mar. i8. i9is Effects of Misiletoe Oil Young Couifers 717
the elaborated food materials for its early development the following
season. If this food material is reduced in amount b}' its becoming
localized in other parts of the tree, the growth of the main shoot must be
necessarily retarded (PI. 37, B), and the bud itself will form earher in
the season and be much reduced in size. Two yellow-pine trees, each
8 years of age, one with a conspicuous infection with broom formation,
the other entirely free from infection, were carefully observed to deter-
mine this point. The former not only started the elongation of the main
shoot nine days later than the other tree, but ceased to develop altogether
at the end of the first month. The shoot of the uninfected tree continued
to elongate for two months and showed a gain of 1 1 inches over that of the
infected tree. The trees grew under exactl}' similar conditions, and their
root systems were practically equal in extent. Kirkwood ^ states that —
It is improbable that the whole growth of the new leader is at the expense of the food
stored in the bud alone. That from other parts also doubtless contributes, but the
tendency is to crowd the formative materials toward the extrmities of the main shoot
and the branch. In the sharing of these materials the main shoot leads and the
branches follow in order of their importance. The principal growth, however, is
undoubtedly at the expense of the locally stored materials, the substances stored else-
where having their part to play in the development of the tissues in their immediate
proximity.
The storage of food in the shoot and branches is exactly what the
formation of brooms and burls prevents in a large measure in all parts of
the tree above the seat of infection, and eveaatually results in retardation
and the appearance of spike top or staghead. The writer has repeatedly
called attention to this in previous publications. In order to demonstrate
that there is an actual storage of food materials in mistletoe brooms greater
than that of normal branches the results of an experiment may be given.
Late in the m.onth of October, 191 4, after the leaves had fallen, 10 mistle-
toe brooms and 10 uninfected branches from points ranging from 5 to 15
feet above the fomier were cut from a western larch and thrown in a damp,
shady ravine. In June of the following year the brooms and branches
were examined with the following result: Practically all foliar spurs of
the brooms developed needles in proportion of about one-third of the
normal length. A fev/ of the foliar spurs of the uninfected branches pro-
duced needles barely protruding from the bud scales, and in most cases
there was no leaf production whatever. Observation of brooms in the
crowns of larch cut late in the fall during logging operations showed in
the production of needles in the spring that there must be a great localiza-
tion of elaborated food materials in the branches of the brooms over that
of the normal branches.'* The latter showed no signs of foliation.
' Kirkwood, J. E- the influence op preceeding seasons on the growth of yellow pine. In
Torreya, v. 14, no. 7, p. 118. 1914.
* Certain parasitic fungi also cause a flow of building materials to the place of infection. See GoebEL,
K . E. EINLEITUNG IN DIE EXPEKIMKNTELLE MOKPHOLO'^IE DER PFLANZEN. p. 75. Leipzig uud Berlin.
1908.
3832G°— 18 3
71 8 Journal of Agricultural Research voi. xn. xo.
-^v-
From the foregoing studies it may be concluded that the false mistle-
toes are serious agents in the suppression of young forest growth.
Moreover, young growth once infected on the main stem (and it is
usually so infected) can not recover and produce merchantable material.
Suppression in young yellow pine even up to the sapling stage is of
serious consequence. Very seldom do such trees ever overcome the
early influence. This is also noticeable in trees of dense stands or
when overtopped by older classes. The ultimate effect on the growth
of the tree is exactly equivalent to lack of light, only worse. In the
former case the excurrent growth is arrested, and the tree either
develops into one continuous broom or dies; in the latter, a sapling
of some merchantable value may result. It is clear that every effort
should be made in regions of heavy mistletoe infection ro reduce the
infection of reproduction by cutting all infected overtopping trees.
Care should be taken to prevent the introduction of mistletoe-in-
fected transplants in regions where the parasites do not occur/' Such
resfions, for examole, are the Black Hills of South Dakota and several
of the southeastern Montana forests.
SUMMARY
The height growth of young trees is greatly retarded by mistletoe.
The effects to be observed are reduction in the length of the internodes,
small dimensions of the terminal buds, and reduction in number.
This result is caused by the localization of food materials at the seat
of infection.
To reduce the chances of infecting young growth, all overtopping
infected trees should be killed. Infected trees of any age should be
killed if possible. Care should be taken that infected trees are not
planted in regions w^here mistletoe does not occur.
PLATE 37
A. — Pseudotsuga taxifolia infected with Razoumofskya douglasii.
B. — Effect of an inoculation with Razoumofskya campylopoda on the height growth
of 6-year-old Pimts jeffreyi. Culture made in a greenhouse at Missoula, Mont.
5 Weir, J. R. mistletoe; injury to conifers in the northwest. U. S. Dept. Agr. Bui. 360, 39 p.
27 fig. 1916. Literature cited, p. 39.
Effects of Mistletoe on Young Conifers
Plate 37
Journal of Agricultural Research
Vol. XII, No. II
DETERMINATION OF FATTY ACIDS IN BUTTER FAT: V
By E. B. Holland, Associate Chemist, and J. P. Buckley, Jr., Assistant Chemist,
Massachusetts Agricultural Experiment Station .
INTRODUCTION
Since reporting a method (14) ^ for the determination of stearic acid
in butter fat, work has been continued with a \'iew of evolving a process
for determining some of the remaining fatty acids. The object of the
investigation was to deduce practical analytical methods which might
serve to measure the effect of feed upon the composition of the resulting
butter fat. Such methods, however, if quantitative and reasonably
workable, would be applicable to other fixed oils and fats and play a part
in the so-called technical examination of such products.
There are several distinct lines of procedure upon which methods
for determining different fatty acids in a mixture might be based :
(i) Crystallization of the acids ;
(2) Solubility of various salts;
(3) Fractionation of the acids in vacuo;
(4) Fractionation of methyl or ethyl esters.
The iodin absorption and the acetyl substitution are measures of
unsaturated and of hydroxy acids, respectively. They are valuable
adjuncts, but of limited rather than of general application.
Attention has been called in several instances to some of the inherent
faults (jj, 14) of the different schemes of separation, and a discussion
of their relative merits appears unnecessary at this time. It will suffice
to summarize the present application and limitation of the schemes.
Crystallization methods have been employed for the quantitative
separation of stearic and of arachic (27) acids but seldom for other acids.
The lead-salt-ether method, or Gusserow (7)-Varrentrapp (29) process,
for the separation of liquid from solid acids is the most prominent illus-
tration of the salt-soluble methods which, as a rule, have not proved
sufficiently discriminative for quantitative use.
Fractionation of the acids in vactw has failed as an analytical process.
Fractionation of ethyl esters appeared applicable to many fatty acids,
although chemists generally consider the process as having little quanti-
tative significance. The experience gained in purifying fatty acids indi-
cated that the method was practicable and at least worthy of additional
study.
> From the Department of Chemistry, Massachusetts Agricultural Experiment Station. Printed with
the permission of the Director of the Station.
2 Reference is made by number (italic) to " Literature cited," pp. 73i-7,i2-
Journal of Agricultural Research, Vol. XII, No. ii
Washington, D. C. Mar. i8. igiS
mn (719) Key No. Mass. 4
-20 Journal of Agricultural Research voi. xii, No. n
EARLIER INVESTIGATIONS
At the present writing esterification methods have so wide an appli-
cation and contributors to the literature on the subject are so numerous
that only references bearing directly on the analysis of oils and fats
will be cited.
Rochleder (28) studied the action of absolute alcohol and dry hydro-
chloric acid gas upon castor oil in the separation of glycerol.
Berthelot (2, p. 311-312) applied the reaction to a number of glycerides
and showed that esters of the fatty acids were formed in addition to
glycerol.
Juillard {15, p. 239) prepared methyl and ethyl esters of dihydroxy-
stearic acid by boiling the acid with 10 times its weight of alcohol and
2 or 3 drops of sulphuric acid.
Fischer and Speier (5) esterified various organic acids with different
amounts of absolute ethyl alcohol and dry hydrochloric or concentrated
sulphuric acid, also of methyl alcohol and hydrochloric acid and noted
the yield. They did not deduce a general method.
Haller {8-1 1) heated a variety of oils and fats with twice their weight
of absolute methyl alcohol containing i to 2 per cent of hydrochloric
acid, removed the glycerol and excess alcohol with water, or preferably
brine, and fractionated the esters up to 194° C. (methyl caprylate)
at atmospheric pressure and the residual esters in vacuo. The methyl
oleate present in the myristic and higher fractions was removed by
chilling and absorption on porous plates. When complete alcoholysis
was not obtained, he recommended a second treatment with a new por-
tion of acidulated alcohol, the employment of a larger amount of alcohol
at the outset, or the addition of an inert solvent such as ether to facilitate
the reaction, particularly in the case of butter fat and of drying oils which
readily oxidize and polymerize. The production of a small amount of
aldehyde was noted in some instances. Haller neutralized the esters
wAth. barium carbonate or a solution of sodium carbonate and dried over
calcium chloride or anhydrous sodium sulphate.
Phelps and Hubbard {18) esterified succinic acid with ethyl alcohol
and hydrochloric acid, and secured the greatest yield with maximum
dehydration. In other experiments {ig) the addition of anhydrous zinc
chlorid enhanced the reaction.
Complete esterification of 50 gm. of benzoic acid {26) was obtained
by treating for 4 hours with 400 c. c. of absolute alcohol containing 1.25
per cent of dry hydrochloric acid and 10 gm. of anhydrous zinc chlorid,
or for 3 hours with 200 c. c. of absolute alcohol and 2 gm. of sulphuric
acid.
Several other chlorids {24, p. 296-297; 20-25) proved nearly as effi-
cient as zinc chlorid under like conditions of operation.
Mar. i8, 1918 Fatty Acids ifi B litter Fat 721
Meyer (17) employed substantially the Haller method with cotton-
seed oil, but increased the amount of methyl alcohol to four times the
weight of the oil, and obtained a yield of about 90 per cent.
Elsdon {3, 4) employed the Haller method with coconut oil and palm-
kernel oil, fractionated the resulting methyl esters in vacuo, and refrac-
tionated to constant boiling point. He stated that the process had
qualitative and a considerable amount of quantitati\"e value, but was
too lengthy for ordinary use.
Kailan {16) found that ether, benzene, and carbon tetrachlorid did
not accelerate the esterification of benzoic acid with absolute alcohol
and hydrochloric acid or with dilute alcohol and acid.
Wolff and Scholze (50) used a dilute sodium-bicarbonate solution to
purify the esters when shaking out with ether.
Abderhalden and Kautzsch (j) esterified the silver salt of an amino
acid by boiling with an excess of ethyl iodide.
Grandmougin, Havas, and Guyot (6) showed how an organic acid after
treatment with sodium methylate might be converted by means of an
excess of dimethyl sulphate into the methyl ester of the organic acid and
sodium methyl sulphate.
Possibly methyl sulfonic acid might be substituted in some instances
for dimethyl sulphate or methyl halide as indicated by an English patent,
No. 9359-^
Hauser and Klotz {12) esterified organic acids with alcohols by pass-
ing the vapors over glucinum oxide heated to 310° C.
Permissible space does not allow one to do justice to the articles cited.
PRELIMINARY WORK
Further study of esterification was undertaken with a view of secur-
ing a method for determining the percentage of another insoluble acid
besides stearic and oleic in butter fat. Laurie acid was naturally the
most promising, on account of the lower boiling point of its esters,
although myristic acid was also considered a possibility. The per-
centage of lauric acid, together with ordinary analytical data and the
amount of stearic acid determined by crystallization, would permit a
satisfactory calculation of the remaining insoluble acids.
Material. — At the outset the insoluble acids were employed for
esterification with an idea that the previous elimination of soluble acids
and of glycerol would be an advantage. In reality such did not prove
the case, as water, a limiting factor in esterification, was produced in
the reaction between fatty acids and alcohols, while glycerol was pro-
duced in the case of fats (glyceryl esters) and alcohols as shown bv the
following equations :
RCOOH -f- RiOH -^catalyzer=RCOOR^-^ H.,0
fatty acid alcohol ester water
(RCOO),C3H54-3RiOH-f-catalyzer=3RCOOR,-|-C3H5(OH),
fat alcohol ester glycerol
' Cliem. iVhs.. v. 6. no. 4, p. 5.55. lorj.
722
Journal of Agricultural Research
Vol. XII, Ne. II
F'urthennore, the employment of the original product was preferable
from an analytical standpoint to say nothing of the time and labor
Fig. 1. — Apparatus employed in esterification.
involved in the preparation of a large stock of insoluble acids. Butter
fat was used in all subsequent work.
Mar. 38. igts Fatty Acids in Butter Fat 723
EsTERiFiCATiON (fig. i). — Earlier investigators have indicated various
methods of esterification. Absolute alcohol with a mineral acid catalyzer
appeared the most practicable for analytical purposes and was adopted
tentatively. Between methyl and ethyl alcohols there was little choice,
except as to cost and convenience, although the methyl esters have a
somewhat lower boiling point. Ethyl alcohol was used exclusively.
The preparation of dry ethyl alcohol substantially free from aldehydes
was found considerable of a problem. For dehydration neither metallic
calcium nor any quicklime from the usual sources proved efficient. A
granulated caustic lime containing about 95 per cent of calcium oxid was
eventually obtained from the manufacturers which would produce a dry
alcohol on the third distillation.
Different schemes have been suggested for the removal of aldehydes,
such as oxidation with silver nitrate, potassium permanganate, or
potassium bichromate, polymerization with caustic alkali or fractional
distillation, and all were tried in some form or other. The following
process finally proved satisfactory and w^as adopted :
Approximately 2 liters of alcohol w^ere fractionated in a water bath
over 600 to 700 gm. of caustic lime and 2 to 3 gm. of caustic soda. The
main portion of the distillate, the first and last being rejected, w^as refrac-
tionated twice in a similar manner over fresh lime and soda. The
rejected portions were united and retreated.
Dry hydrochloric acid or concentrated sulphuric acid has been the
catalyzer almost invariably employed by different workers for esterifica-
tion. The former with a greater hydrogen-ion concentration is apparently
less efficient, gram for gram, than the latter, and is generally used in
larger amounts. This may be due in part to loss of hydrochloric acid
as ethyl chlorid by volatilization, but more likely to the dehydrating
effect of sulphuric acid. By using a larger quantity of alcohol, together
with certain chlorids, the yield of esters with hydrochloric acid may be
increased, as shown by Phelps.
A considerable excess of alcohol is also required to insure the necessary
mass action irrespective of its dehydrating action. For esterif\ing 150
gm. of butter fat, 400 c. c. of absolute ethyl alcohol containing 8 gm. of
dry hydrochloric acid or 4 c. c. of concentrated sulphuric acid were
employed. This amount of alcohol furnished about 1 1 times that needed
for combination with the fatty acids.
The use of a neutral solvent, such as ether, did not appear to accelerate
esterification and w^as omitted after a few trials. Butter fat diffused
rapidly through the acid alcohol on boiling, and the solution generally
cleared in a few minutes. Short boiling periods were tried, but 24 hours
were considered more reliable and adopted.
Purification of esters. — After completing the esterification and
cooling the solution, the esters must be precipitated and freed from
mineral acid, glycerol, and excess alcohol.
724
Journal of Agricultural Research
Vol. Xlf, No. II
Attention has already been called to various methods of neutralizing,
purifying, and drying the esters, as described by other workers. Most
of these processes and innumerable modifications were given careful trial,
but the readiness with which at least a portion of the butter-fat esters
hydrolyzed precluded the use of water in their purification. This was an
extremely exacting condition and called for a salt soluble in alcohol,
neutral in reaction, and with dehydrating properties. Dry magnesium
Fig. 3. — Apparatus employed in fractionation.
chlorid satisfactorily met these requirements, and furthermore was cheap
and easily procured. On the addition of the dry salt in the presence of
ether, a rapid separation of esters was obtained. The underlying solu-
tion was drawn off by means of a separate ry funnel, and the esters were
purified by "shaking out" several times with ether and a saturated
alcoholic solution of magnesium chlorid. The ether facilitates the
separation and shouid always be added before the magnesium chlorid.
Mar. 1 8, 1918
Faity Acids in Butter Fat
725
All the agitation necessary can be obtained by reversing the separatory
funnel several times, which gives a clear separation much quicker than a
more violent shaking. The resulting esters contain ether and probably
some alcohol, but do not require neutralizing or drying.
Fractionation (fig. 2). — Fractionation of the esters in vacuo by means
of a Bruehl or other type apparatus was found impracticable as a quanti-
tative process for the reason that a constant level of the liquid in the dis-
tilling flask and a constant pressure were prime requisites for a definite
fraction, neither of which co^ld be successfully maintained with the
facilities at hand. Furthermore, the use of such apparatus necessitates
a certain aptitude or technic not possessed by all analysts.
Fractionation at atmospheric pressure required a high temperature,
but proved feasible. Gas could not be used as a direct source of heat
owing to fluctuations in pressure, influence of air currents, tendency to
decompose the esters, etc., but when applied to a bath of superheated
valve oil proved entirely satisfactory. This oil is a cheap commercial
product that will safely withstand a temperature of OA^er 400° C. when
covered and may be used three times, possibly more, without appreci-
able loss of efficiency. All the exposed surface of the side-neck flask
should be wound with asbestos paper to prevent chilling the vapors and
breakage of the flask.
For some reason fractionation of the purified esters proved imprac-
ticable, possibly due to the influence of a relatively large amount of the
higher esters particularly oleic, but was readily accomplished after a pre-
liminary distillation. Accurate results are dependent in large measure
on a slow, steady rise in temperature during the first distillation and
subsequent fractionation. Glass beads were found helpful in boiling.
T.'^BLE I. — Boilinq point and range of fractions {uncorrected) of esters of butter fat
Ester.
Boiling point.
No.
I
2
3
4
5
6
Esters.
Range of
I'ractioiij.
Ethyl butyrate ....
"C.
119. 9-1 2 1
165 -167
205 -208
243 -245
269
Ethyl butyrate, caproate, and
oleate
'C.
Ethyl caproate
I25-IS0
180—221;
Ethyl caprylate
Ethyl caprate
Ethyl caproate, caprylate, and
oleate
Ethyl laurate
Ethyl caprylate, caprate, and
oleate
Ethyl myristate
Ethyl palmitate
Ethyl stearate
22C-270
Ethyl caprate, laurate, and
oleate
270-300
Ethyl oleate ....
Ethyl laurate, myristate, and
oleate
•^oo— •;2';
Ethyl myristate, palmitate, and
oleate
325-365
The range of the several fractions is more or less arbitrary, being de-
pendent on the speed of distillation and the distance the vapors have to
rise. The object was to secure fractions that did not contain more than
two esters in addition to oleic ester, and, furthermore, adjacent fractions
should contain approximately one-half of the saturated ester appearing
726
J Of. r rial of AgricuJiiiYal Research
Vol. Xn, No. II
in each. The reported boiling point of the ethyl ester of a number of
fatty acids that occur in butter fat, together with the range of the sev-
eral fractions as determined by analysis for a "high" side-tube 500-c. c.
distillation flask, may be noted in Table I.
The apparent lack of agreement is probably due to the greater distance
the vapors have to rise in practical fractionation and to the influence of a
gradually increasing amount of ethyl oleate in the several fractions.
Analysis seemed to be the only method for accurately establishing the
required range. Hempel tubes or similar apparatus could not be em-
ployed to break up the distillate.
As the preliminary work advanced, the results became more concord-
ant and indicated that it was possible to determine not only lauric and
myristic acids but also caproic, caprylic, and capric acids. A part of
the butyric acid was recovered, but the main portion was evidently lost
during the purification of the esters or distilled over with the ether, owing
to the greater solubility and volatility of this ester.
Analysis of fractions. — Only the determinations of saponification
and of iodin numbers by the usual methods employed wth oils and fats
were required.
Calculation of results. — Having determined the weight, saponifi-
cation and iodin numbers of the several fractions, the analyst must
ascertain whether the range of the fractions had been accurately estab-
lished. If correctly fractionated, the percentage of the different ethyl
esters in the several fractions can be calculated algebraically and their
weight computed, from which the percentage of the corresponding acids
in the butter fat can be determined.
As a matter of convenience, the data necessary for these calculations
have been compiled in Table II.
Table II. — Fatty acids and their ethyl esters
[C, 12.005; H, 1.008; 0, 16.000; K, 39.10; 1, 1^6.92]
Add.
Molecular
weight of
acid.
Molecular
•weight of
ester.
Saponifica-
tion No.
of ester.
Iodin No.
of ester.
Recipro-
cal.
Conver-
sion fac-
tor, ester
to acid.
Acetic ' 60. 042
Butyric [ 8S. 084
Valeric ; 102. 105
Caproic 116. 126
Caprylic 1 144- 168
Capric .
Lauric
^lyristic
Palmitic
Stearic
Arachic
Oleic
Erucic
Linolic
Linolenic
Clupanodonic . . .
Ricinoleic
Dihydroxystearic .
172. 210
200. 252
228. 294
256- 2,3^
284. 378
312. 420
282. 362
338. 446
280. 346
278. 330
276.314
298. 362
316.378
88.084
116. 126
130. 147
144. 168
172. 210
200. 252
228. 294
256- z?,(>
284. 378
312. 420
340. 462
310. 404
366. 488
308. 388
306. 372
304- 356
326. 404
344. 420
Mgm.
636. 983
483. 165
431- "3
589. 185
325.812
280. 187
245. 771
218. 885
197. 301
179. 592
164. 800
180. 758
153.096
181. 940
183. 137
184. 350
171.897
162. 906
0.81777
. 69263
I. 64624
2. 48561
3- 33609
• 77769
I. 22284
I- 44377
.60744
. 40232
• 29975
I.
68164
75852
78454
80540
83716
85997
87717
, 89060
90139
, 91024
, 91764
, 90966
, 92348
, 90907
,90847
. 90786
,91409
, 91858
Mar. is, 191S Fatty A cids in Butter I- at 727
As an illustration of the method of calculation, take the fifth fraction
of a sample of butter fat (Table III) vv^hich weighed 50.69 gm., had a
saponification number of 219.280 and an iodin number of 14.055, and
was intended to contain ethyl laurate, myristate and oleate.
The iodin number, convertible to ethyl oleate by the factor 1.22284, is
equivalent to 17.187 per cent.
Then i.oo— 0.17187, or 0.82813, is the percentage of the residual esters
in the fraction and 2i'9.28o— (0.17187X 180.758), or 188.213, their alkali-
consuming power. The latter divided by the former gives a saponifica-
tion number of 227, a figure between that of ethyl laurate (245.771) and
that of ethyl myristate (218.885), ^-s anticipated.
Let X indicate ethyl laurate and y ethyl myristate, then
rc+r = 0.82813
245. 77ix+2i8.885>'= 188.213
3;= 25.842 per cent of ethyl laurate
)/= 56.971 per cent of ethyl myristate.
The percentages 0.25842, 0.56971, and 0.17187 multiplied by the
weight of the fraction (50.69 gm.), gives the weight of the several esters
in the fraction 13.099, 28.879, ^^^ 8.712, which by their respective fac-
tors 0.87717, 0.89060, and 0.90966 may be converted into the weight of
the several fatty acids in the fraction. The total weight of a fatty acid
divided by the weight of fat taken (300) gives the percentage of that
acid in the butter fat. The same method of calculation is followed in
each of the other fractions.
METHOD IN DETAIL
Reagents. — Alcohol : absolute.
Hydrochloric acid : dry, generated by dropping concentrated sulphuric
acid into a mixture of concentrated hydrochloric acid and sodium chlorid
and dried by passing through concentrated sulphuric acid.
Sulphuric acid: concentrated, heated to 225° C.
Ethyl ether: anhydrous, freshly distilled over metalic sodium.
Magnesium chlorid: dry powder, neutral. Dried at 75° to 80° C.
Magnesium chlorid solution: 250 gm. of dry powder to 500 c. c. of
absolute alcohol.
ESTERIFICATION AND PURIFICATION OF ESTERS. — Into a I,000-C. C. flat-
bottom globe flask are brought 150 gm. of filtered fat, together with
400 c. c. of alcohol previously charged with 8 gm. of dry hydrochloric
acid, or 4 c. c. of concentrated sulphuric acid and a number (25) of glass
beads. The flask is connected with a long spiral or other form of reflux
condenser, and the mixture carefully boiled on 50-nicsh Nichrome wire
gauze lor 24 hours.
728 Journal of AgrimiHural Research voi. xii, No. n
After esterification the contents of the flask are cooled, 50 c. c. of
ether and 150 gm. of magnesium chloride added, rotated to hasten
saturation, transferred to a i,ooo-c. c. pear-shaped separatory funnel
and allowed to stand until a clear separation is secured. The underlying
layer is drawn off into the original flask and the esters carefully shaken
out two or three times with 25 to 50 c. c. of ether and 50 c. c. of an alco-
holic solution of magnesium chlorid. Violent shaking causes a slow
separation. After the removal of the final washing, the clear, purified
esters are filtered through a firm, close-textured paper into a 500-c. c.
round-bottom "low" side-tube distillation flask. The filter is extracted
with ether which is run into the original globe flask containing the
alcoholic layer and washings from the esters. More ether is added until
a ready separation is obtained and the solution again transferred to
the separatory funnel and allowed to stand several hours to recover any
occluded esters which are then washed several times vath ether and
magnesium chlorid solution as previously described, and filtered into
the distillation flask containing the first portion.
Fractionation. — A number (50) of glass beads are placed in the
side-neck distillation flask which is connected with a 12-inch Liebig
condenser and heated in a bath of superheated valve oil for distillation
of the esters. The exposed portion of the flask should be covered with
asbestos paper and the condenser filled with cold water at the outset
but no circulation should be permitted during the distillation. The
temperature should be raised slowly. After the ether and alcohol are
expelled, the entire distillate between 85° and ^65° C. is collected and
constitutes from no to 120 gm. with butter fat.
The distillate from two portions representing 300 gm. of butter fat
and a number (50) of glass beads are brought into a 500-c. c. round-
bottom "high" side-tube distillation flask which is connected with a
Liebig condenser and heated as previously described. Particular care
should be exercised in heating the oil bath so as to insure a slow, steady
rise in temperature taking at least 80 minutes from the beginning of
the first fraction to the completion of the last. The required range of
every fraction must be accurately established with the apparatus em-
ployed by analysis (See Table I).
The volatility of the esters and the readiness with which they hydrolyze
necessitates careful treatment of the fractions, which should be collected
in tared flasks, weighed, and the saponification and iodin numbers
determined as soon as possible.
Mar. 18, 1918
Fatty Acids in Butter Fat
729
APPLICATION OF THE METHOD
A sample of dry filtered butter fat churned from sweet cream from
mixed milk of the Experiment Station herd was taken for examination.
An analysis of the fat and of its insoluble acids gave the following results,
which indicate a normal product.
FAT
Saponification number mgm . . 231. 453
Acid number mgm. . 2. 183
Ether number (e) mgm. . 229. 270
lodin number 27. 999
Equivalent in oleic acid per cent. . 31. 145
Total fatty acids (1.00—0.00022594 e) per cent. . 94. 819
Insoluble fatty acids (by alcoholic saponification) per cent. . 87. 500
Soluble fatty acids (by difference) per cent. . 7- 319
Glycerol (0.00054703 e) per cent. . 12. 542
INSOLUBLE ACIDS
Neutralization number mgm. . 221. 890
lodin number 28. 125
Equivalent in oleic acid per cent. . 31. 285
Stearic acid (by crj'stallization) percent. 13. 010
The fat was esterified, with dry hydrochloric acid as a catalyzer, the
esters purified and fractionated as described in the method.
In Table III are given the range, weight, saponification, and iodin
numbers of the several fractions. From these data the percentage and
weight of the different esters in the fractions were calculated (Table lY)
and the weight of the corresponding acids and their percentage computed
on the basis of the original fat (Table Y).
Table III. — Weight and analysis of fractions
No.
Range of
fraction.
Weight of
fraction.
SaponiSca-
tjon No.
lodin
No.
Ethyl
oleate.
I .
"C.
125-180
180-225
225-270
270-300
300-325
325-365
Gm.
5- 0835
4. 4400
6. 8855
15. 2990
50. 6900
84. 8800
Mgm.
417. 209
346. 396
279. Ill
240. 983
219. 280
205- 955
4-753
8. 091
11. 471
12. 846
14-055
16. 691
Per cent.
5.812
9.894
14.027
15- 709
17. 187
2
■3 .
A
6
20. 410
Table IV. — Percentage and weight of esters recovered
No.
Ethyl
butyrate.
Ethyl
caproate.
Ethyl
caprylate.
Ethyl
ca prate.
Ethyl
laurate.
Ethyl Ethyl
myristate. palmitate.
1
Ethyl
oleate.
P.ct.
42.708
Gm.
2. 171
P.ct. Gm.
51.480 2. 617
SS. 128 2. 448
P.ct.
Gm.
P.ct.
Gm..
P.ct.
Gm.
P.ct.
Gm. P. ct.
Gm.
P.cU
5.812
9.894
14. 027
15-709
17.187
20. 410
Gm.
0.29S
34-978
28. 210
I- 553
1.942
57- 763
IS- 763
3-977
2.412
.966
68.528
25. 842
10. 484
13.099
3. 403
56. 971
55. 736
2d. 879
8. 712
6
::::.. 1
47.309
23- 854
20. 247
17.324
••• 1
Total
2. 171
5-065
3-495
16.389
'23-583
76.188
20.247
3°- 139
730
Journal of Agricultural Research voi.xii.No.it
Table V
. — Weigh
and percentage of fatty acids recovered
Acid.
Gm.
Per cent.
Acid. Gm.
Per cent.
Butyric acid (partial re-
covery)
1.647
4, 080
2. 926
5-494
20. 686
0- 549
1.360
•975
I. 831
6.89s
Myristic acid 67. 853
Palmitic acid (partial
recovery) iS. 250
Oleic acid (partial re-
22. 6r8
Caproic acid
6. 083
Caprylic acid
covery)
27.416
9- 139
Laurie acid
1
1
The recovery of ethyl butyrate was incomplete, owing to its high
solubility and volatility; that of ethyl palmitate was due to inability to
continue the distillation, owing to insufficient volume of higher esters and
excessive heat requirements. Ethyl oleate appeared in all fractions,
gradually increasing with the temperature.
The percentage of the different fatty acids in the butter fat are
presented in Table VI. Butyric acid of the soluble acids and palmitic
acid of the insoluble were determined by difference, stearic acid by
crystallization, as previously stated, and oleic acid from the iodin number
of the insoluble acids. No allowance was made for unsaponifiable matter
in the calculation.
The alkali-consuming and glycerol -combining powers of the several
acids are also recorded and confirm the results secured in large measure.
The glycerol requirements of the constituent acids slightly exceed that of
the fat, for the reason that no allowance was made for the free fatty acids
present.
Table VI. — Percentage of fatty acids in butter fat
Fatty acids.
Amount
present.
Sapoaifi-
cation
niuuber.
Glycerol.
Soluble acids:
Butyric acid. ,
Caproic acid. .
Caprylic acid .
Capric acid. . .
Total
Calculated .
Insoluble acids:
Laurie acid . .
Myristic acid.
Palmitic acid.
Stearic acid. ..
Oleic acid . . .
Per cent.
"■o- 153
I. 360
•975
I. 831
Mgm.
20. 084
6- 571
3- 795
5.966
Per cent.
1.099
■359
.208
.326
7-319
36. 416
37- 299
1.992
Total
Determined ,
Soluble and insoluble acids:
Total
Determined or calculated from fat.
6.89s
22. 618
°' 19. 229
11.384
27-374
87. 500
94. 819
19- 319
55- 588
42. 089
22. 461
54- 395
1-057
3.041
2.302
1. 229
2. 976
193. 852
194. 154
10. 605
230. 268
231-453
12. 597
12. 542
a By difference.
Mar. 18, 1918 Fatty Acids in Butter Fat 731
In a later article attention will be called to additional analyses of
butter fat and to the results reported by other workers employing different
methods.
SUMMARY
The direct esterification of butter fat v/ith subsequent fractionation
of the resulting esters has proved an accurate and practical method for
the determination of five of the fatty acids in butter.
LITERATURE CITED
1) Abderhalden, Emil, and Kautzsch, Karl.
1912. VERSUCHE iJBER VERESTERUNG VON MONOAMINOSAUREN MITTELS
JODATHYL. In Ztschr. Physiol. Chem., Bd. 78, Heft 2, p. 115-127.
2) Bertkelot, Marcellin.
1S54. SUR LES COMBINAISONS DE LA GLYCERINE AVEC LES ACIDES ET SUR LA
SYNTHASE DES PRINCTPES IMM^DTATS DES GRAISSES DES ANIMAUX.
In Ann. Chim. et Phys., s. 3, t. 41, p. 216-319, pi. 2.
3) Elsdon, G. D.
1913. ALCOHOLYSis AND THE COMPOSITION OP cocoANUT OIL. In Analyst, V. 38,
no. 442, p. 8-II.
4)
1914. THE COMPOSITION" OP PALM-KERNEL OIL. In Analyst, v. 39, no. 455,
p. 78-80.
5) Fischer, Emil, and SpEiEr, Arthur.
1895. DARSTELLUNG DER ESTER. In Ber. Deut. Chem. Gesell., Jahrg. 28, No. 19,
P- 3252-3253.
6) Grandmougin, E., Havas, E., and Guyot, G.
I913. METHYLIERUNG ALIPH.\TISCHER VERBINDUNGEN MITTELS DIMETHYL
SULFATS. In Chem. Ztg., Jahrg. 37, No. 81, p. 812-813.
7) GussEROW, C. A.
1828. ueber die einwirkung des bleioxyds auf die organischen korper,
WELCHE IM ALLGEMEINEN UNTER DIE KLASSE DER FETTE GESTELLT
WERDEN UND DIE DADURCH ENTSTEHENDEN VERBINDUNGEN. Ill Afch.
Apoth. Ver. Nordl. Teutschland, Bd. 27, Heft 2, p. 153-244.
8) HallER, a.
1906. SUR l'alcoolyse DES CORPS GR.^s. In Compt. Rend. Acad. Sci. [Paris],
t. 143, no. 19. p. 657-661.
0) -
ic)
1907. alcoolyse DE l'huile dE ricin. In Compt. Rend. Acad. Sci. [Paris],
t. 144, no. 9, p. 462-466.
1908. alcoolyse DE l'huile DE LIN. In Compt. Rend. Acad. Sci. [Paris],
t. 146, no. 6, p. 259-262.
i:) and Youssoufian.
1906. ALCOOLYSE DU BEURRE DE coco. In Compt. Rend. Acad. Sci. [Paris]
t. 143, no. 22, p. 803-806.
12') Hauser, O., and Klotz, A.
I913. K.\T.\LYTISCHE REAKTIONSBESCHLEUNIGUNG MITTELS BERYLLIU.MVER-
BINDUNGEN BEI DARSTELLUNG VON ESTERN ORGANISCHER SAUREN.
In Chem. Ztg., Bd. 37, No. 15, p. 146-148.
13) Holland, E. B.
191 1. PURIFICATION OF INSOLUBLE FATTY ACIDS. In Jouf. Indus. and Engin.
Chem., V. 3, no. 3, p. 171-173.
14') Reed, J. C, and Buckley, J. P., Jr.
1916. DETERMINATION OP STEARIC ACID IN butter FAT. /n Jour. Agr. Research,
V. 6, no. 3, p. ior-113, fig. 1-2. Literature cited, p. 113.
y-^2 Journal of Agricultural Research voi. xii, No. n
(15) JVILLARD, Paul.
1895. surl'acidE dioxyst^arique naturel. In Bui. Soc. Chim. Paris, s. 3 ,
t. 13, p. 238-240.
(16) Kailan, Anton.
1914. UBER DIE durch chlorwasserstofp katalysierte esterbeldung in
LosungsmittelgemischEn. In Ztschr. Phys. Chem., Bd. 88, Heft i,
p. 65-102.
(17) Meyer, V. J.
1907. UBER DAS BAUMWOLLSAMENOL. In Chem. Ztg., Bd. 31, No. 64, p.
793-794.
(18) Phelps, I. K., and Hubbard, J. L.
1907. ON THE ESTERiFiCATiON OF SUCCINIC ACID. In Amer. Jour. Sci., s. 4,
V. 23, no. 137, p. 368-374-
(19) and Phelps, M. A.
1907. THE use op ZINC CHLORIDE IN THE ESTERIFICATION OF SUCCINIC ACID.
In Amer. Jour. Sci., s. 4, v. 24, no. 141, p. 104-196.
(20)
1908. ON THE ESTERIFICATION OF MALONic ACID. In Amer. Jour. Sci., s. 4.
V. 26, no. 153, p. 243-252, I fig.
(21)
1908. RESEARCHES ON THE INFLUENCE OF CATALYTIC .\GENTS IN ESTER FORMA-
TION. ON THE ESTERIFICATION OF CYANACETIC ACID. In Amer. Jour.
Sci., s. 4, V. 26, no. 153, p. 264-266.
(22) . Palmer, H. E., and SmilliE, R.
1908. RESEARCHES ON THE INFLUENCE OP CATALYTIC AGENTS IN ESTER FORMA-
TION. THE EFFECT OF CERTAIN SULPHATES ON BENZOIC AND SUC-
CINIC ACIDS. In Amer. Jour. Sci., s. 4, v. 26, no. 153, p. 290-295.
(23) . Phelps, M. A., and Eddy, E. A.
1908. CONCERNING THE PURIFICATION OF ESTERS. In Amer. Jour. Sci., s. 4,
V. 26, no. 153, p. 253-256.
(24)
1908. RESEARCHES ON THE INFLUENCE OF CATALYTIC AGENTS IN ESTER FORMA-
TION. THE ESTERIFICATION OF BENZOIC .A^CID V/ITH CERTAIN CHLO-
RIDES. Li Amer. Jour. Sci., s. 4, v. 26, no. 153, p. 296-300.
(25)
1908. RESEARCHES ON THE INFLUENCE OF CATALYTIC AGENTS IN ESTERFOR-
MATION. HYDROBROMIC ACID AND ZINC BROMIDE IN THE FORMA-
TION OF ETHYL BENZOATE. In Amer. Jour. Sci., s. 4, v. 26, no.
153, p. 281-289.
(26) and Osborne. R. W.
1908. ON THE ESTERIFICATION OF BENZOIC ACID. In Amer. Jour. Sci., s. 4,
V. 25, no. 141, p. 39-48.
(27) Renard, A.
1871. RECHERCHE ET DOSAGE DE l'hUILE D'aRACHIDE DANS l'hUILE
d'olivE. In Compt. Rend. Acad. Sci. [Paris], t. 63, no. 25, p. 1330-
1332 •
(28) ROCHLEDER, F.
1846. UEBER DAS GLYCERIN. In Ann. Chem. u. Pharm., Bd. 59, Heft 2,
p. 260-261.
(29) VarrEntrapp, Franz.
1840. UEBER DIE OELSAURE. In Ann. Chem. u. Pharm., Bd. 35, Heft
2, p. 196-215.
(30) Wolff, H., and Scholze, E.
I914. t)BER KOLOPHONIUMBESTJMMUNG in FIRNISSEN, OLEN UND SEIFEN.
In Chem. Ztg., Jahrg. 38, No. 34, p. 369-370; No. 35, p. 382-383.
Vol. XII MARCH 25, 1918 No. 12
JOURNAL OF
AGRICULTURAL
RESEARCH
CONTENTS AND INDEX
OF VOLUME XII
PUBUSHED BY AUTHORITY OF THE SECRETARY OF AGRICULTCRE.
WITH THE COOPERATION OF THE ASSOCIATION OF AMERICAN
AGRICULTURAL COUEGES AND EXPERIMENT STATIONS
WASHINGTON, D. C.
WASMINOTONt OOVCRNMCNT PmNTUM OFFIOE I IMI
i
EDITORIAL COMMITTEE OF THE
UNITED STATES DEPARTMENT OF AGRICULTURE AND
THE ASSOCIATION OF AMERICAN AGRICULTURAL
COLLEGES AND EXPERIMENT STATIONS
FOR THE DEPARTMENT
KARL F. KELLERMAN, Chairman
Physiologist and Associate Chief, Bureau
of Plant Industry
EDWIN W. ALLEN
Chief, Office of Experiment Stations
CHARLES L. MARLATT
Enfomoiogist and Assistant Chief, Bttreau
of Entomology
FOR THE ASSOCIATION
RAYMOND PEARL*
Biologist, Maine Agricultural Experiment
Station
H. P. ARMSBY
Director, Institute of A nim al 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.
* Dr. Pearl has tmdertaken special work in connection with the war emergency;
therefore, until further notice all correspondence regarding articles from State Experi-
ment Stations should be addressed to H. P, Armsby, Institute of Animal Nutrition,
State College, Pa.
INDEX
Abelmoschus csculentus — Page
host plant of —
, Fusarium vasiiifectutn 529-546
Veriicillium albo-airum 530-546
wilt diseases of 529-546
Acetic acid. See Acid, acetic.
Acetone, toxicity to insect eggs 5S1-5S7
Acid —
acetic, toxicity to insect eggs 581-582
butyric —
percentage in butter fat 726-731
toxicity to insect eggs 581
capric, percentage in butter fat 726-731
caproic, percentage in butter fat 726-731
caprylic, percentage in butter fat 726-731
fatty, determination in butter fat 719-732
formation in corn-stover silage 591-592
lauric, percentage in butter fat 726-731
myristic, percentage in butter fat 726-731
oleic, percentage in butter fat 726-731
palmitic, percentage in butter fat 726-731
phosphate, effect on soil reaction 27-28
phosphoric, in soil, solubility 674-676
stearic, percentage in butter fat 729-731
valeric, toxicity to insect eggs 581
Acidity, soil, definition of 139
Alcohol —
allyl, toxicity to insect eggs 581
amyl, toxicity to insect eggs 581
benzyl, toxicity to insect eggs 581
ethyl, toxicity to insect eggs 581-586
methyl, toxicity to insect eggs 581-586
Alfalfa. See Medicago sativa.
Allyl—
alcohol, toxicity to insect eggs 581
isosulphocyanate, toxicity to insect eggs. . . 581
Ammonia —
effect of nitrates on production of 200-201
formation in Wisconsin soil 486-487
Ammonium —
nitrate —
influence on —
Bacillus radicicola 210-211
growth of Azotobacter 190
nodule formation 219-220
sulphate —
effect on —
hydrogen-ion concentration of soil-film
water 28-30
solubility of calciixm in soil 674
solubility of phosphoric acid in soil. . . . 674
Amygdalus pcrsica, food plant of Cer otitis
capitate 105
Amyl —
acetate, toxicity to insect eggs 581
alcohol, toxicity to insect eggs 581
nitrite, toxicity to insect eggs 581
Ancylostoma —
caninum, in dogs, effect of anlhclmintics
on 399ff
duodenale. in man, effect of anthelmintics on 415
Page
Anilin, toxicity to insect eggs 581
Anthelmintic, efficacy of 397~477
Apple. See Malus sylvestris.
Apple-blister 132
Areca catechu, use as anthelmintic 419
Areca nut. See Areca catechu.
Ascarid. See Ascaris suum; Belascaris cati;
Belascaris marginata; Toxascaris limbata.
Ascaridia perspicillum, in chickens, effect of
anthelmintics on 42sff
Ascaris suum, in hogs, effect of anthelmintics
on 4oiff
A vena sativa, relation to injury by Mayetiola
destructor 519-527
Averrhoa carambola.iood plant of Ceratitis capi-
tala 105
Azotobacter —
influence of nitrates on —
fLxation of nitrogen by 193-201
formation of volutin bodies in 205-208
growth 187-208
production of pigment by 203-205
influence on nitrates in solution by 200-203
Bacillus —
arnylovorus, causal organism of fireblight. . . 130
azotobacter —
influence of —
calcium carbonate on 490-495
limestone on 490-495
magneshim carbonate on 490-495
radicicola —
influence of —
calcium carbonate on 496-497
limestone on 496-457
magnesium carbonate on 496-497
nitrates on —
fixation of atmospheric nitrogen by. 214-217
growth 20S-226
infection by 220-222
production of gum by 217-219
influence on nitrates in solution by. . 21 1-2 14
Bacteria —
cause of tobacco wildfire 449-458
influence of —
calcium carbonate on 469-499
calcium chlorid on 479-480
dibasic magnesium phosphate on 479-499
limestone on 469-499
magnesium carbonate on 477-499
magnesium chlorid on 479-480
monocalcium phosphate on 469-499
nitrification of, effect on solubility of trical-
cium phosphate 671-683
nitrogen-assimilating, influence of nitrates
on 183-230
Bacterium tabacum —
causal organism of tobacco wildfire 454-457
description 4S4-4SS
Bailey, C. H. and Gurjar, A. M. (jxiper):
Respiration of StorcKl Wheat 685-713
Baldwin-spot. See Bitter-pit.
Barley. See Hordeum spp.
(733)
734
Journal of Agricultural Research
Vol. XII
Batchelor, L. D., and Reed, H. S. (paper): Page
Relation of the Variability of Yields of
Fruit Trees to the Accuracy of Field
Trials 245-283
Bechdel, S. I. Corn-Stover Silage 589-600
Beetle, potato. See Lepiinotarsa decemlincala.
Behavior of Sweet Potatoes in the Ground
(paper) 9-^7
Belascaris —
cati, in cats, effect of anthelmintics on. . . 416,417
viarginata, in dogs, effect of anthelmintics
on 399ff
Benzaldehyde, toxicity to insect eggs 581
Benzene, toxicity to insect eggs 581-582
Benzonitrile, toxicity to insect eggs 581
Benzyl alcohol, toxicity to insect eggs 581
Bestill. See Theveiia neriifoUa.
Black myrobalan. See Tcrminalia chebula.
Bitter-pit—
deFcription iio-iii
effect of heavy irrigation on 126
resemblance of other spot diseases of Malus
sylveslris to 130-i.^S
resemblance to rosy-aphis stigmonose. . . . iio-iii
susceptibility, relation of size of Malus syl-
veslris to 126
" Blister," resemblance to cork disease 134
Blister-rust, white-pine—
caused by Cronarlium ribicola 459
dissemination 461-462
Boiling point, index to toxicity of organic
compounds 580-586
Brazilian plum. See Eugenia braziliensis.
Brommethylphenylketone, toxicity to insect
eggs 581
Bromoform, toxicity to insect eggs 581-586
Bromtoluene, ortho, toxicity to insect eggs . . 581
Bromxylene, toxicity to insect eggs 581
Brooks, Charles, and Fisher, D. F. (paper):
Irrigation Experiments on Apple-Spot Dis-
eases 109-138
Buckley, jr., J. P., and Holland, E. B. (pa-
per): Determination of Fatty Acids in But-
ter Fat 719-732
Bunostomum Irigonocephaluvi, in sheep, effect
of anthelmintics on 404, 410, 412
Burd, J. S. (paper): Water Extractions of
Soils as Criteria of their Crop- Producing
Power 297-309
Butter fat, determination of fatty acids in. . 719-731
Butyric acid. See Acid, butyric.
Calcium. —
carbonate —
effect on — •
solubility of calcium in soil 674
solubility of phosphoric acid in soil .... 674
bacteria in Wisconsin soils 469-499
chlorid, influence on bacteria in Wisconsin
soils 479-480
in soil —
effect of crop growth on 311-364
effect of season on 311-364
solubility 674-676
nitrate —
influence on —
Bacillus radicicola 209
fixation of nitrogen by Azotobacter. . 198-199
growth of Azotobacter 1S9, 192
nodule formation. . ; 219-220
Calf— Page
digestion of starch by S74-S78
use of starch food in addition to milk for . . 577-578
Calomel, use as anthelmintic 399
Calophyllutn inophyllum, food plant of Cera-
iitis capHaia loj
Capric acid. See Acid, capric.
Caproicacid. See Acid, caproic.
Caprylic acid. See Acid, caprylic.
Carbohydrate, in Ipomocc batatas, changes in
content 10-15
See also Starch.
Carbon —
bisulphid, toxicity to insect eggs 581-586
dioxid —
accumulated, influence upon respiration
of stored wheat 706-708
effect of gas of on soil solution 384-385
effect on soil reaction 140-145
effect on water-soluble nutrients in soil . 334-339
tetrachlorid, toxicity to insect eggs 581-586
Carpenter, C. W. (paper); Wilt Diseases of
Okra and the Vcrticillium-Wilt Problem . 529-546
Carruth, F. E., and Withers, W. A. (paper):
Gossypol, the Toxic Substance in Cotton-
seed 83-102
Casein, hydrolysis of in presence of starch,
effect of time of digestion on 1-7
Castor oil, use as anthelmintic 399
Cectun worms. See Heterakis papulosa.
Ceratitis capitata —
control 104
infestation of fruits by larvae of 103-108
parasites of 103-108, 285-296
Chenopodiuni —
ambrosioides anthclminticuni, syn. Cheno-
podiuin anthelminticujn .
anthelminticum, use as anthelmintic 429-439
Chenopodiiun, oil of, use as anthelmintic. . 429-439
Cherry, French. See Eugenia unifiora.
Chinese orange. See Citrus japonica.
Chlorbenzene, toxicity to insect eggs 5S1-586
Chloroform —
toxicity to insect eggs 5S1-586
use as anthelmintic 402-403
Chlorpicrin, toxicty to insect eggs 581-586
Chrysophylluin monopyrenum, food plant of
Ceratitis capitata 105
Citral, toxicity to insect eggs 581
Citrus —
japonica, food plant of Ceratitis capitata 105
limonia —
individual tree yield 255, 258
variability of yield 245-283
sinensis —
individual tree yield 252-255, 258
variability of yield 245-283
spp., effect of mulches on production of. . 513-516.
Cojj'ea arabica, food plant of Ceratitis capitata . . 105
Coffee. See Coffca arabica.
Collins, G. N. (paper): New-Place Effect in
Maize 231-243
Conifer, young, effect of mistletoe on 71S-718
Cooperia sp., in sheep, effect of anthelmin-
tics on 410
Copper sulphate, use as anthelmintic 406-409
Corerot, form of cork disease 132
Cork disease —
description 131-134
resemblance to drouthspot 131-134
Jan. 7-Mar. 25, 1918
Index
735
Com. See Zea mays. Page
Com-stover silage —
fermentation 591-599
importance of water in production of 590-591
practicability 589-591
Corn-Stover Silage (paper) 589-600
Cottonseed —
kernels, toxicity 83-102
meal, toxicity 83-102
toxic substance, method of removing from
ether extract 87
toxicity 83-102
toxicity. See also Gossypol.
Cresol —
meta, toxicity to insect eggs 581
ortho, toxicity to insect eggs 581
para, toxicity to insect eggs 581
Cronartiiim ribicola, causal organism of white-
pine blister-rust 459
Crop growi;h —
effect on soil extract 1 1-368
freezing-point method as index of variations
in soil solution due to 369-395
Crop production, relation of water extraction
of soils to 297-309
Cyladinae, subfamiily of Apionidae 604-607
Cylas —
femoralis 607
formicarius 605
formicariiis elcgantulus 605-607
turcipennis 607
Determination of Fatty Acids in Butter Fat
(paper) 719-732
Diachasma —
fullawayi —
cannibalism 288-290
parasite of Ceratilis capital'a 106-10S, 280-296
tryoni —
cannibalism 286-290
life cycle 294
parasite of Ceratilis capitata, 104, 106, 108, 286-296
prolificness, comparison with Opius hu-
milis -95
seasonal abundance 291
Dibasic magnesium phosphate, influence on
bacteria in Wisconsin soils 469-499
Digestion, effect of time of, on hydrolysis of
casein in presence of starch 1-7
Digestion of Starch by the Young Calf
(paper) 575-578
Dwicoreafcatetaj, weevils affecting 611
Dipylidium caninum, effect of anthelmintics
on 400 ff
Douglas fir. See Pseudotsuga laxifolia.
Drouthspot —
description 130-1J1
resemblance to cork 131-134
Dryrot of Malus sylvestris, form of cork 132
Effect of Nitrifying Bacteria on the Solubility
of Tricalcium Phosphate (paper) 671-683
Effect of Season and Crop Growth in Modify-
ing the Soil Extract (paper) 3 ' '-368
Effect of Time of Digestion on the Hydrolysis
of Casein in the Presence of Starch (paper) . . 1-7
Effects of Mistletoe on Young Conifers
(paper) 715-718
Efficacy of Some Anthehnintics (paper) 397-447
Egg, insect, toxicity of organic compounds
to 579-587
Page
Egg, winter cycle of production in domestic
fowl 547-574
Einkom. See Triticum monococcum.
Emetic, tartar, use as anthelmintic 401-402
Emmer. See Triticum dicoccum.
Epsom salt, use as anthelmintic 400-401
Errata and authors' emendations IV
Ether-
ethyl, toxicity to insect eggs 581-586
petroleiun, toxicity to insect eggs 581
toxicity to insect eggs 582-5S6
use as anthelmintic 405
Ethyl—
aceto-acetate, toxicity to insect eggs 581
alcohol, toxicity to insect eggs 581-586
ether, toxicity to insect eggs 581-586
malonate, toxicity to insect eggs 581
mercaptan, toxicity to insect eggs 581
Eugenia spp., food plants of Ceralitis capitala. 105
Eugenol, toxicity to insect eggs 581
Euscepes batalae —
description 608-610
distribution 608-609
Fat, butter. See Butter fat.
Feed, calf, use of starch ration in addition to
milk 577-578
Fern, male, use as anthelmintic 415
Fertilizer —
dissemination of tobacco wildfire by 456
effect on soil reaction 25-26
Ficuslaurifolia, use as anthelmintic 427-428
Fir, Douglas. See Pseudotsuga taxifolia.
Fireblight, resemblance to drouthspot 130
Fisher, D. F.,and Brooks, Charles (paper): Ir-
rigation Experiments on Apple-Spot Dis-
eases 109-138
Fly, fruit, Mediterranean. See Ceratiiis capi-
tala.
Fly, Hessian. See Mayetiola destructor.
Fames spp. , cultural characters of 4off
Foster, A. C.and Wolf, F. A. (paper): Tobacco
Wildfire 449-458
Foster, W. D. and Hall, M. C. (paper):
Efficacy of Some Anthelmintics. . 397-447
Fowl, domestic. See Poultry-.
Fragaria spp. —
degeneration of microspore 652-655
dieciousness 622-628
inflorescence 613-614
pistils, morphology 623
pistils, steriUty 622-628
pollen, development 628, 641-651
pollen, germination 637-641
pollen, sterility 628
stamens, arrangement 614-621
Freezing-point depression, determination of
in soils 369-395
Freezing-Point Method as an Index of Vari-
ations in the Soil Solution Due to Season
and Crop Growth, The (paper) 369-39S
French chern'. See Eugenia uniflora.
Frogeye, comparison with tobacco wildfire . . . 451
Fruit fly, Mediterranean. See Ceratilis capi-
tala.
Fruit-Fly Parasitism in Hawaii during 1916
(paper) 103-108
736
JouYJial of Agricultural Research
Vol. XII
Page
Fruitpit 109
Fulmer, H. L. (paper): Influence of Carbo-
nates of Magnesium and Calcium on Bac-
teria of Certain Wisconsin Soils 463-504
Fumigation, organic compounds, toxicity to
insect eggs 585-586
Fungus, wood-rotting —
cultural characters, identification by 63-64
hymenium, influence of substratum on
character of "6
mycelium, density of 79
pileus formation, influence of light on 77-78
pore formation, influence of gravity on 78
pore formation, influence of substratum on. 7S-79
sporophore production 65-80
Furfural, toxicity to insect eggs 581
Fusarium vasmfectum —
causal organism of wilt disease of Abel-
moschus esculentus 533-537
description 536-537
occurrence in okra- wilt 537-538
parasitism 538-539
Gas, carbon-dioxid, effect on soil solution . . . 384-385
Gasoline—
toxicity to insect eggs 581-586
use as anthelmintic 409-411
Geranyl acetate, toxicity to insect eggs 581
Germination, pollen, of Fragaria spp 637-641
Gibbs, Joshua, cylindrical plow bottom, de-
scription iSi
Gipsy moth. See Portketria dispar.
Gipsy-Moth Larvae as Agents in the Dissemi-
nation of the White- Pine Blister-Rust
(paper) 459-462
Goodale, H. D. (paper): Winter Cycle of Egg
Production in the Rhode Island Red Breed
of the Domestic Fowl 547-574
Gooseberry. See Rihes spp.
Gossypol —
"acetate," toxicity 87
toxicity to pigs 89-100
toxicity to rabbits 8S-89
Gossypol, the Toxic Substance in Cottonseed
(paper) 83-102
Graham, S. A., and Moore, W. (paper): Tox-
icity of Volatile Organic Compounds to
Insect Eggs 579-5S7
Gravatt. G. F., and Posey, G. B. (paper):
Gipsy-Moth Larvae as Agents in the Dis-
semination of the White-Pine Blister-
Rust 459-462
Grain, cause of heating 6S5-6S6
Guava. See Psidium guajava.
Guava, strawberry. See Psidium caitleianutn.
Gurjar, A. M., and Bailey, C. H. (paper):
Respiration of Stored Wheat ; . 685-713
Haemonchus conlorius, in sheep, effect of
anthelmintics on 404
Hall.M.C.and Foster, W.D., (paper): Effi-
cacy of Some Anthelmintics 397-447
Harsch, R. M., and Long, W. H. (paper):
Pure Cultures of Wood-Rotting Fungi on
Artificial Jledia 33-82
Hasselbring, Heinrich (paper): Behavior of
Sweet Potatoes in the Groimd 9-17
Page
Hawaii, pa.xaisit<isol Ceraliiis capitata'm .... 103-108
285-296
Hessian fly. See Mayetiola destructor.
Hclerakis papulosa, in poultry, effect of
anthelmintic on 425, 439
Hills, T. L. (paper): Influence of Nitrates on
Nitrogen-Assimilating Bacteria 183-230
Hoagland, D. R. (paper): The Freezing-
Point Method as an Index of Variations in
the Soil Solution Due to Season and Crop
Growth 369^395
Hoagland, D. R., and Sharp, L. T. (paper):
Relation of Carbon Dioxid to Soil Reac-
tion as Measured by the Hydrogen Elec-
trode 139-148
Hog, toxicity of cottonseed to 89-100
Holbrook, F. F., plow bottom, description. . 1S2
Hollow-apple 132
Holland, E. B., and Buckley, Jr., J. P.
(paper): Determination of Fatty Acids in
Butter Fat 719-732
Hookworm. See Ancylostoma; Bunosiomum
trigonocephalu'in.
Hordeuni spp., relation to injury by Maye-
tiola destructor 519-527
Humus —
formation in mulched basins 507-513
in soil, relation to orange production 513-516
Humus in >Iulched Basins, Relation of
Humus Content to Orange Production,
and Effect of Mulches on Orange Produc-
tion (paper) 505-518
Hybrid—
Fragaria spp., aborted pollen 631-635
Zea 7nays, new- place effect 231-243
Hydrogen electrode —
use in indicating soil reaction 19-31
use in determining relation of carbon
dioxid to soil reaction 139-148
Hydrogen-ion concentration of soils . . 19-30, 139-146
Incubation, effect on freezing-point depres-
sion 383-384
Influence of Carbonates of JMagnesimn and
Calcium on Bacteria of Certain Wisconsin
Soils (paper) 463-504
Influence of Nitrates on Nitrogen-Assimi-
lating Bacteria (paper) 183-230
Insect eggs, toxicity of organic compounds
to 579-587
Interrelations of Fruit-Fly Parasites in Ha-
waii (paper) 285-296
lodobenzene, toxicity to insect eggs 581
Iodoform, use as anthelmintic 405-406
Ipoinoca batatis —
composition, changes in ground 9-1?
weevils aflecting 604-610
Irish potato. See Solanum tuberosum.
Irrigation Experiments on Apple-Spot Dis-
eases (paper) 109-138
Jefferson, Thomas, plow bottom, descrip-
tion 173-174
Jensen, C. A (paper): Humus in Mulched
Basins, Relation of Humus Content to
Orange Production, and Effect of Mulches
on Orange Production 505-518
Jan. 7-Mar. 25, 1918
Index
737
Jonathan-spot — Page
description 127
resemblance of other spot diseases of Malus
sylvesiris to 130-135
Juglans regia —
individual tree yield 256-258
variability of yield 245-283
Kamani. See Calophyllum inophyllutn: Ter-
minalia catappa.
Kelley, \V. P. (paper): Effect of Nitrifying
Bacteria on the Solubility of Tricalcium
Phosphate 671-683
Kerosene, toxicity to insect eggs 581-584
Knox, Samuel A., plow bottom, descrip-
tion 179-181
Lambruschini, R., plov,- bottom, descrip-
tion 174-175
Laurie acid. See Acid, lauric.
Leaching, soil, effect on freezing-point de-
pressions 386-687
Lemon. See Citrus linionia.
Leplinotarsa dccunlineala, toxicity of organic
compounds to eggs of 580-586
Light, relation to sporophore production in
wood-rotting fungi 65-76
Lime-
effect on soil reaction 27-28
requirement, confusion with soil acidity. . 139-140
Limestone, influence on bacteria of Wiscon-
sin soils 469-499
Long, W. H., and Harsch, R. M. (paper):
Pure Cultures of Wood-Rotting Fungi on
Artificial Media si-ft^
Magnesium —
carbonate, influence on bacteria of Wis-
consin soils 474-499
chlorid, influence on bacteria of Wisconsin
soils 479-480
extraction from soil, effect of crop growth
on 311-368
extraction from soil, effect of season on. . 311-368
phosphate, dibasic, influence on bacteria of
Wisconsin soils 479-499
Maize. See Zea viays.
Male-fem, effect on —
A ncylostoma duodenale 415
Belascaris mcrginala 415
Trickuris depressiuscula 415
use as anthelmintic 415
"Malformation" of Malus syhestris 132
Malus syhestris —
bitter-pit of —
description iio-iii
effect of heavy irrigation on 126
resemblance of other spot diseases to. . . 130-135
corerot of, form of cork disease 132
cork disease of, description 131-134
drouthspot of, description 130-131
dryrot of, form of cork disease 132
individual tree yield 257-238
Jonathan-spot of —
description 127
resemblance of other spot diseases to. . . 130-133
relation of time of picking to development
of spots in storage 119, 127, 129
size, relation to bitter-pit susceptibility 126
spot diseases of, effect of irrigation upon . 109, 137
variability of yield 245-283
Page
Mangifera indica, food plant of Ceratitis capi-
tata 105
Mango. See Mangifera indica.
Manure, mulch of, effect on orange produc-
tion 513-S16
Manure-and-lime, mulch of, effect on orange
production 513-516
Mayetiola destructor, relation of grain to in-
jury by 519-527
McColloch, J. W., ana Sahnon, S. C. (paper):
Relation of Kinds and Varieties of Grain to
Hessian-Fly Injury 519-527
McHargue, J. S. (paper): Effect of Time of
Digestion on the Hydrolysis of Casein in
the Presence of Starch 1-7
Mead, plow bottom, description 181
Medicago saliva —
mulch of —
effect on orange production 513-516
formation of humus in 507-513
roots of, influence of nitrates on 222-226
Mediterranean fruit fly. See Ceratitis capi-
tata.
Mercaptan, ethyl, toxicity to insect eggs 581
Meta-cresol, toxicity to insect eggs 581
Methyl alcohol, toxicity to insect eggs 581-586
Milk, use with starch food for young calf. . . 577-578
Mimusops eleiigi, food plant of Ceratitis cap-
itata los
Mistletoe. See Razoumofskya spp.
Moisture-
content in Ipomoea batatas 10-15
importance in dissemination of tobacco
wildfire 450-451,453,456
in stored wheat, relation to rate of respira-
tion 689-696
soil, effect on soil solutions 370-371
See also Water.
Moniezia spp., in sheep, effect of anthelmin-
tics on 406,412
Monocalcium phosphate, effect on —
bacteria of Wisconsin soils 469-499
hydrogen-ion concentration of soil-fihn
water 28-30
Moore, W., and Graham, S. A. (paper): Page
Toxicity of Volatile Organic Compounds to
Insect Eggs 579-587
Moth, gipsy. See Porthelria dispar.
Mulch —
alfalfa, effect on orange production 513-5x6
alfalfa-and-lime, effect on orange produc-
tion 513-316
alfalfa-hay, effect on orange production. . . . 515
bean-straw, effect on orange production. . 514-515
bur-clover-hay, effect on orange produc-
tion 514-516
manure, effect on orange production S13-51S
manure-and-lime, effect on orange produc-
tion 513-516
pine-shavings, effect on orange production. 515
swcet-clover-hay, effect on orange produc-
tion 314-315
Mulched basins, formation of humus in 507-513
Myristic acid. See Acid, myristic.
Myrobalan, black. See Terminalia chcbula.
Neotopism. Sec New-place effect.
I New-Place Effect in Maize (paper) .'31-243
738
Journal of Agricultural Research
Vol. xn
Nicotiana tabacum — Page
host plant of Bacterium lahacuin 449-458
infected seed of, factor in dissemination of
tobacco wildfire 456
use as anthelmintic 428-429
wildfire of 449-462
Nicotine, toxicity to insect eggs 581
Nitrates —
in Wisconsin soils, accumulation of 486-487
in solution —
influence of Azotobacter on 200-203
influence of Bacillus radicicola 011 21 1-2 14
influence on —
alfaKa roots 222-226
Azotobacter 187-208
Bacillus radicicola 208-226
fixation of atmospheric nitrogen by
Bacillus radicicola 2 14-2 1 7
fixation of nitrogen by Azotobacter 193-201
formation of volutin bodies in Azoto-
bacter 205-208
infection by Bacillus radicicola 220-222
nitrogen-assimilating bacteria 1S3-230
nodule formation 219-220, 222-226
production of gum by Bacillus radi-
cicola 217-219
production of pigment by Azotobacter. 203-205
Nitrification —
effect on solubility of tricalcium phosphate
in soil 675-676
gelatin, influence of —
calcium carbonate in Wisconsin soils on 489-449
limestone in Wisconsin soils on 489-490
magnesium carbonate in Wisconsin soils
on 589-490
raonocalcium phosphate in Wisconsin
soils on 489-490
Nitrobenzene, toxicity to insect eggs 5S1
Nitrogen —
accumulation in Wisconsin soils, influence
of calcium carbonate on 487-488
fixation —
in Wisconsin soils 490-491
influence of nitrates by Azotobacter on . 193-201
influence of nitrates by Bacillus radicicola
on 214-217
in soil, effect of crop growth on 311-364
in soil, effect of season on 311-364
relation to plant life 183-230
Nitromethane, toxicity to insect eggs 581
Nitrotoluene, ortho, toxicity to insect eggs . . 581
Nitroxylene, toxicity to insect eggs 581
Nodular worm. See Oesophagostomum.
Nodule formation, influence of nitrates on. . 222-226
Noronhia emarginata, food plant of Ceratitis
capitate 105
Norton, R. P., et al. (paper): Digestion of
Starch by the Young Calf 575-578
Nut, areca. See Areca catechu.
Oats. See Avena sativa.
Ochrosia elliplica, food plant of Ceratitis capi-
tata 105
Oesophagostomum —
columbiantim, in sheep, effect of anthelmin-
tics on 404, 410, 412
dentatum, in hogs, effect of anthelmintics
on 4ooff
Oil— Page
castor, use as anthelmintic 399
chenopodium, use as anthelmintic 429-439
Okra. S^e Abeimoschus esculentus.
Okra-wilt 529-546
Oleic acid. See Acid, oleic.
Oleoresin aspidii, use as anthelmintic 414-416
Opius humilis —
cannibalism 286-290
life cycle 294
parasite of Ceratitis capitata 104, 286-296
prolificness 295
seasonal abundance 291
Orange. See Citrus.
Orange production —
effect of mulch on 513-516
relation of humus content of soil to 513-516
Organic compounds —
boiling point, index to toxicity 580-586
toxicity to insect eggs 579-587
volatility, index to toxicity 580-586
Orobitidae 608-610
Orobitinae, subfamily of Orobitidae 60S-610
Ortho-brom toluene, toxicity to insect eggs. 581-582
Ortho-cresol, toxicity to insect eggs 581
Ortho-nitrotoluene, toxicity to insect eggs. . . 581
Palaeopus dioscoreae, n. sp 611
Palmitic acid. See Acid, palmitic.
Para-cresol, toxicity to insect eggs 581
Parasites, fruit-fly 103-108, 285-296
Peach. See A mygdalus persica.
Pelletierine tannate, use as anthelmintic. . . 417-418
Pemberton, C. E., and WiUard, H. F. (pa-
per)—
Fruit-Fly Parasitism in Hawaii during
1916 103-108
Interrelations of Fruit-Fly Parasites in
Hawaii 285-296
Petrolemn —
benzin, use as anthelmintic 411-413
ether, toxicity to insect eggs 581, 585-586
Phenol, use as anthelmintic 413-414
Phosphate. See Phosphorus.
Phosphoric acid. See Acid, phosphoric.
Phosphorus —
in soil, effect of crop growth on 311-364
in soil, effect of season on 311-364
Pierce, W. Dwight (paper): Weevils Which
Affect Irish Potato, Sweet Potato, and
Yam 601-612
Pig, toxicity of cottonseed to 89-100
Pine. See Pinus.
Pinene, toxicity to insect eggs 581
Pinus —
ponderosa, effect of mistletoe on 715-718
spp., food plant of Porthclria dispar 459-460
Plow bottom —
forms of 150-173
history of development of 173-183
Plowing, motion of soil particles in 162-173
Plmn, Brazilian. See Eugenia braziliensis.
Plummer, J. K. (paper): Studies in Soil
Reaction as Indicated by the Hydrogen
Electrode 19-31
Plymouth Rock, Barred, egg production
in winter 562-570
Poisoning, cottonseed 83-102
Polyporus spp., cultural characters of. . 45-47, 60-61
Jan. 7-Mar. 25, 1918
Index
739
Page
Polyslictus kirsutus, hymenium, influence of
substratum on character of 76
Porlhetria —
dispar —
distribution 459
factor in dissemination of white-pine
blister-rust 459-462
infestation of Pinus spp. by larvae of . . . 459-460
infestation of Ribes spp. by larvae of . . . 459-462
spp., injury to Ribes spp 461-462
Posey, G. B., and Gravatt, G. F. (paper):
Gipsy-Moth Larvae as Agents in Dissemi-
nation of the White- Pine Blister-Rust. . . 459-462
Potash. See Potassium.
Potassium —
chlorid, effect on hydrogen-ion concen-
tration of soils 146
in soil —
effect of crop growth on 311-364
effect of season on 311-364
nitrate —
influence of Bacillus radicicola on 21 1-2 13
influence on —
alfalfa nodules 224-226
alfalfa roots 222-224
fixation of nitrogen by Azotobacter. . 193-196
fixation of nitrogen by Bacillus radici-
cola 214-217
growth of Azotobacter 188, 191
growth of Bacillus radicicola 208
nodule formation 223-226
sulphate, effect on soil reaction 1 . 26-27
Potato beetle. See Lepiinotarsa decemlincala.
Potato-
Irish. See Solanum tuberosum.
sweet. See I pomoea batatas.
Poultry, egg production in winter 547-574
Premnotrypes solani 602
Propyl acetate, toxicity to insect eggs 581
Prunus persica, syn. Amygdalus persica.
Psaliduridae 602
Pseudotsuga iaxifolia —
effect of mistletoe on 716-718
host plant of Razoumofskya spp 716-718
Psidium —
cattleianum, food plant of Ccratitis capitala . . 105
guajaxa, food plant of Ccratitis capitala .... 105
Pimky disease 132
Pure Cultures of Wood-Rotting Fungi on
Artificial Media (paper) 33-82
Purgative, use as anthelmintic 399-401
Pyridin, toxicity to insect eggs 581
Rabbit—
toxicity of cottonseed to 88-89
toxicity of gossypol to ^ 88-89
Rahm, W. L-, plow bottom, description. . . 178-179
Rain, wind-blown, factor in dissemination of
tobacco wildfire 456
Rat, toxicity of cottonseed to «3-io4
Razoumofskya spp., parasite of Psevdotsuga
taxifolia 716-718
Reed. H. S., and Batchelor, L. D. (paper):
Relation of the Variability of Yields of
Fruit Trees to the Accuracy of Field
Trials 245-283
Page
Relation of Carbon Dioxid to Soil Reaction
as Measured by the Hydrogen Electrode
(paper) 139-148
Relation of Kinds and Varieties of Grain to
Hessian-Fly Injury (paper) 519-527
Relation of the Variability of Yields of Fruit
Trees to the Accuracy of Field Trials
(paper) 245-283
Respiration of Stored Wheat (paper) 685-713
Rhigopsidius iucumanus 602
Rhode Island Red, winter cycle of egg pro-
duction in 547-574
Ribes spp. —
food plants of Porlhetria dispar 459-462
Rose-apple. See Eugenia jainbos.
Rosy-aphis stigmonose, resemblance to
bitter-pit no-iii
Rot, identification by cultural characters. . . 63-64
Roimd-worm. See Ascaridia perspicillum.
Rye. SeeSecale cereale.
Salicylic aldehyde, toxicity to insect eggs 581
Salmon, S. C, and McColloch, J. W. (paper):
Relation of Kinds and Varieties of Grain to
Hessian-Fly Injury 519-527
Salt, Epsom, use as anthelmintic 400-401
Sand. See Soil.
Santonin and calomel, use as anthelmintic. 420-422
Santonin, use as anthelmintic 419-420
Season —
effect on soil extract 31 1-368
freezing-point method as index of variations
in soil solution due to 369-395
relation to abundance of Diachasma tryoni. . 291
relation to abundance of Opiiis humilis 291
variation of concentration of soil solution
with. 371-382
Secale cereale, relation to injury by Mayetiota
destructor 519-527
Seed, dissemination of tobacco wildfire by. . . 456
Sharp, L. T., and Hoagland, D. R. (paper):
Relation of Carbon Dioxid to Soil Reaction
as Measured by the Hydrogen Electrode. 139-148
Shaw, R.H., Woodward, T. E., and Norton,
R. P. (paper): Digestion of Starch by the
Young Calf 575-578
Sherman, J. M., and Bechdel, S. I. (paper):
Corn-Stover Silage 589-600
Silage, corn-stover 5S9-600
fermentation 591-599
imix)rtance of water in production of . . . . 590-591
Small, James, plow bottom, description 175-177
Sodium nitrate —
effect on hydrogen-ion concentration of soil
suspensions 26
influence on —
fixation of nitrogen by Azotobacter 197
fixation of nitrogen by Bacillus radi-
cicola 214-217
growth of Azotobacter 188, 191, 207
growth of Bacillus radicicola 209
nodule formation 219-220
Soil-
acidity, definition 139
solubility of calcium in, effect of —
ammonium sulphate on 674
calcium carbonate on 674
tricalcium phosphate on 674
740
Journal of Agricultural Research
Vol. XII
Soil— Continued. Page
cropped and uncropped —
relation between soluble soil nutrients
in 317-363
seasonal variations 371-382
effect of leaching on freezing-point depres-
sions 386-3S7
extract —
effect of crop growth on 311-368
effect of season on 311-368
relation to crop produced 317-363
filni water, reaction 23-24
humus, relation to orange production. . . . 513-516
hydrogen-ion concentration. See Soil
reaction.
potassiiun chlorid on 146
influence of nitrates on —
Azotobacter in 1S7-208
Bacillus radicicola in 208-226
moisture, effect on soil solutions 370-371
nutrients in —
effect of carbon dioxid on 334-339
effect of temperature on 334-339
particles, motion in plowing 162-173
solubility of phosphoric acid in —
effect of ammonium sulphate on 674
effect of calcium carbonate on 674
effect of tricalcium phosphate on 674
reaction —
effect of acid phosphate on 27-28
effect of ammonium sulphate on 2S-30
effect of carbon dioxid on 140-145
effect of lime on 27-28
effect of monocalciinn phosphate on 28-30
effect of potassium chlorid on 146
effect of potassium sulphate on 26-27
solution —
effect of —
carbon-dioxid gas on 384-385
drying on 385-3S6
incubation on 383-384
moisture on 370-371
relation of water extracts to 387-391
water extracts, index of crop production . . 297-309
Wisconsin, bacteria of —
influence of calcium carbonate on. . . . 469-499
influence of calcium chlorid on 479-480
influence of dibasic magnesium phos-
phate on 479-499
influence of limestone on 469-499
influence of magnesivun carbonate on . 474-499
influence of magnesium chlorid on. . . 479-480
influence of monocalcium phosphate
on 469-499
Solanum tuberosum —
origin 601
weevils affecting 601-604
Species —
emended 454-457
new 603-604, 611
Speck, comparison with tobacco wildfire. . 451
Spelt. See TriticuJn. Page
SpigeUa, use as anthelmintic 428
Sporophore, wood-rotting fungus —
influence of inoculum on development 80
position on media 79
production of 65-80
relation of hght to production of 65-76
relation of temperature to production of . . . 67-68
Starch— Page
content of Ipomoea batatas 10-15
digestion by calf S74-S78
hydrolysis of casein in presence of, effect of
time of digestion on 1-7
use as feed in addition to milk for young
calf 577-578
Stearic acid. Sec Acid, stearic.
Stephen, plow bottom, description 177-178
Sterility in the Strawberry (paper) 613-670
Stewart, G. R. (paper): Effect of Season and
Crop Growth in Modifying the Soil Ex-
tract 311-360
Stippen. See Bitter-pit.
Stomach worms. See Haemonchus contortus.
Storage —
apple, relation of time of picking to develop-
ment of spots in 119, 128, 129
wheat 685-714
Strawberry. See Fragaria spp.
Strawberry guava. See Psidium cattleianum.
Studies in Soil Reaction as Indicated by the
Hydrogen Electrode (paper) 19-31
Study of the Plow Bottom and Its Action
upon the Furrow Slice, A (paper) 149-182
Sugar content in Ipomoea batatas 10-15
Sweet-polato. See Ipomoea batatas.
Taenia spp., in dogs, effect of anthelmintics
on 4oofi
Tapeworm. See Dipylidium caninum; Monie-
zia spp.. Taenia spp.
Tartar enjetic, use as anthelmintic 401-402
Temperature —
effect on water-soluble nutrients in soil. . 334-339
influence on respiration of stored wheat. . 703-706
relation to corn-stover silage fermenta-
tion S92~594
relation to sporophore production in wood-
rotting fungi 67-68
Terminalia spp., food plants of Ceraiitis capi-
tata los
Terpineol. toxicity to insect eggs 581
Tetraslichus giffardianus —
cannibalism 288-296
parasite of Ceratitis capitata 106-108, 288-296
Thievelia neriifolia, food plant of Ceratitis capi-
tata 10s
Thiophene, toxicity to insect eggs 581-586
Thiophenol, toxicity to insect eggs 581
Thymol, use as anthelmintic 423-425
Tobacco. See Nicotiana tabacuin.
Tobacco wildfire —
anatomy of plant tissues affected with. . . 453-454
appearance of 450-451
causal organism of 4S4-4S7
comparison with frogeye 451
comparison with speck 451
dissemination 456-4S7
distribution 449-450
economic importance 450
history 449-450
inoculation 4SI-4S3
isolation 45I-4S3
origin 456
symptoms 45'
Tobacco Wildfire (paper) 449-4s8
Toluene, toxicity to insect eggs 581-586
Jan. 7-Mar.23, 191S
Index
741
Page
Toxascaris limbata, in dogs, effect of anthel-
mintics on 427-433
Toxicity, cottonseed 83-102
Toxicity of Volatile Organic Compounds to
Insect Eggs (paper) 579-587
Tree-
coniferous, effect of mistletoe on 715-718
fruit, relation of variability of yield to ac-
curacy of field trials 245-283
walnut, relation of variability of yield to
accuracy of field trials 245-2S3
Tricalcium phosphate —
effect on solubility of —
calcium in soil 674
phosphoric acid in soil 674
solubility of, effect of iiitrifying bacteria
on 671-683
Trichuris depressiuscula, in dog, effect of an-
thelmintics on 399ff
Trimethylene —
bromid, toxicity to insect eggs 581
cyanid, toxicity to insect eggs 581
Triticum spp. —
relation to injury by Mayetiola destructor . 519-527
stored, cause of heating 686-686
stored, respiration of —
consistency of kernel to 694-696
influence of accumulated carbon dioxid
on 706-708
influence of temperature on 703-706
location in kernel 627-688
measurement 688-6S9
oxidizable material in 686-687
period of dampness in relation to. . . . 701-703
rate in oxygen-free atmosphere 708
relation of moisture to 689-694
relation of plumpness of kernel to. . . . 696-697
soundness of kernel in relation to. . . . 697-701
Trypopremnon —
lalithorax —
description of lar\'a 602-603
description of pupa 603
sanfordi, n. sp 603-604
Turpentine, use as anthelmintic 425-427
Valeric acid. See Acid, valeric.
Valleau, W. D. (paper): Sterility in the Straw-
berry 613-670
Van Slyke method for protein analysis, use. . 1-7
Vermifuge. See Anthelmintic.
Verticilliutn albo-atruwi —
causal organism of okra-wilt 533-537
description 537
inoculations 539-544
occurrence in okra-wilt S37-538
parasitism 538-539
Verticillium-wilt 529-546
Walnut. See Juglans regia. Page
Water-
content of Iponwea batatas 10-15
important factor in production of com-
stover silage 590-591
Water — Continued. Page
soil, effect of crop growth on extracts
from 311-368
soil, effect of season on extracts from 311-368
soU-film, reaction of 23-24
See aho Moisture.
Water Extractions of Soils as Criteria of
Their Crop-Producing Power (paper) . . . 297-309
Weather, factor in dissemination of tobacco
wildfire 456
Weevil, occurrence on —
Ipomoea batatas 604-610
Dioscorea batatas 611
Solarium tuberosum 601-604
Weevils Which Affect the Irish Potato, Sweet
Potato, and Yam (paper) 601-612
Weir, J. R. (paper): Effects of Mistletoe on
Young Conifers 715-718
Wheat. See Triticum spp.
Whipworm. See Trichuris depressiuscula.
White, E. A. (paper): A Study of the Plow
Bottom and Its Action upon the Furrow
Slice 149-182
White- pine blister- rust. See BUster-rust,
white-pine.
Wildfire, tobacco. See Tobacco wildfire.
Willard. H. F., and Pemberton, C. E.
(paper)—
Fruit-Fly Parasitism in Hawaii During
1916 103-108
Interrelations of Fruit-Fly Parasites in
Hawaii 285-296
Wilt Diseases of Okra and the Verticillium-
Wilt Problem (paper) 529-546
Wind, factor in dissemination of —
Portheiria dispar 461-462
Tobacco wildfire 456
WinterCycIeof Egg Production in the Rhode
Island Red Breed of Domestic Fowl
(paper) 547-574
Wisconsin soil. See Soil, Wisconsin.
Withers, W. A., and Carruth, F. E. (paper):
Gossypol, the Toxic Substance in Cotton-
seed 83-102
Wolf, F. A., and Foster. A. C. (paper): To-
bacco Wildfire 449-458
Wood-rotting fungus. See Fungus, wood-
rotting.
Woodward, T. E., ct al. (paper): Digestion
of Starch by the Young Calf 575-578
Worm —
cecum. See Helerakis papulosa.
nodular. See Oesophagostomuin.
round. See Ascaridia pcrspicillum.
stomach. See Haemonchus conlortus.
whip. See Trichuris.
Wyandotte, White, egg production in win-
ter 562-570
Xylene, toxicity to insect eggs 581-586
Yam. See Dioscorea batatas.
York-spot, resemblance to cork 132
Zea mays, new-place effect 231-243
o
New York Botanical Garden Library
3 5185 00263 3889
/