The publications in this collection do
not reflect current scientific knowledge
or recommendations. These texts
represent the historic publishing
record of the Institute for Food and
Agricultural Sciences and should be
used only to trace the historic work of
the Institute and its staff. Current IFAS
research may be found on the
Electronic Data Information Source
site maintained by the Florida
Cooperative Extension Service.
Copyright 2005, Board of Trustees, University
Bulletin 255 November, 1932
UNIVERSITY OF FLORIDA
AGRICULTURAL EXPERIMENT STATION
Wilmon Newell, Director
SOYBEANS FOR SILAGE
R. B. BECKER
W. M. NEAL
C. R. DAWSON
P. T. Dix ARNOLD
Fig. 1.-These soybeans were utilized efficiently as silage.
Bulletins will be sent free upon application to the
AGRICULTURAL EXPERIMENT STATION
EXECUTIVE STAFF BOARD OF CONTROL
John J. Tigert, M.A., LL.D., President of the P. K. Yonge, Chairman, Pensacola
University A. H. Blanding, Bartow
Wilmon Newell, D.Sc., Director Raymer F. Maguire, Orlando
H. Harold Hume, M.S., Asst. Dir., Research Frank J. Wideman, West Palm Beach
J. Francis Cooper, M.S.A., Editor Geo. H. Baldwin, Jacksonville
R. M. Fulghum, B.S.A., Assistant Editor J. T. Diamond, Secretary, Tallahassee
Ida Keeling Cresap, Librarian
Ruby Newhall, Administrative Manager
K. H. Graham, Business Manager BRANCH STATIONS
Rachel McQuarrie, Accountant
NORTH FLORIDA STATION, QUINCY
L. O. Gratz, Ph.D., Plant Pathologist in Charge
MAIN STATION, GAINESVILLE R. R. Kincaid, M.S., Asst. Plant Pathologist
W. A. Carver, Ph.D., Associate Agronomist
AGRONOMY R. M. Crown, B.S.A., Assistant Agronomist
W. E. Stokes, M.S., Agronomist** Jesse Reeves, Farm Superintendent
W. A. Leukel, Ph.D., AgronomistU
G. E. Ritchey, M.S.A., Associate* CITRUS STATION, LAKE ALFRED
Fred H. Hull, M.S., Associate John H. Jefferies, Superintendent
J. D. Warner, M.S., Associate Geo. D. Ruehle, Ph.D., Associate Plant Pathol-
John P. Camp, M.S., Assistant ogist
W. A. Kuntz, A.M., Associate Plant Pathologist
ANIMAL HUSBANDRY B. R. Fudge, Ph.D., Associate Chemist
A. L. Shealy, D.V.M., Animal Husbandman** W. L. Thompson, B.S., Assistant Entomologist
R. B. Becker, Ph.D., Specialist in Dairy Hus-
bandry EVERGLADES STATION, BELLE GLADE
W. M. Neal, Ph.D., Associate in Animal Nutri-
tion R. V. Allison, Ph.D., Soils Specialist in Charge
E. F. Thomas, D.V.M., Assistant Veterinarian R. N. Lobdell, M.S., Entomologist
W. W. Henley, B.S.A., Assistant Animal Hus- F. D. Stevens, B.S., Sugarcane Agronomist
bandman G. R. Townsend, Ph.D., Asst. Plant Pathologist
P. T. Dix Arnold, B.S.A., Assistant in Dairy In- B. A. Bourne, M.S., Sugarcane Physiologist
vestigations J. R. Neller, Ph.D., Biochemist
A. Daane, Ph.D., Agronomist
CHEMISTRY AND SOILS R. W. Kidder, B.S., Asst. Animal Husbandman
R. W. Ruprecht, Ph.D., Chemist** Ross E. Robertson, B.S., Assistant Chemist
R. M. Barnette, Ph.D., Chemist
C. E. Bell, Ph.D., Assistant SUB-TROPICAL STATION, HOMESTEAD
J. M. Coleman. B.S., Assistant H. S. Wolfe, Ph.D., Horticulturist in Charge
H. W. Jones, M.S., Assistant W. M. Fifield, M.S., Assistant Horticulturist
H. W. Jones, MS., AssistaStacy O. Hawkins, M.A., Assistant Plant
ECONOMICS, AGRICULTURAL Pathologist
C. V. Noble, Ph.D., Agricultural Economist**
Bruce McKinley, A.B., B.S.A., Associate
M. A. Brooker, Ph.D., Associate
Zach Savage, M.S.A., Assistant FIELD STATIONS
Ouida Davis Abbott, Ph.D., Specialist** Leesbur
L. W. Gaddum, Ph.D., Biochemist M. N. Walker, Ph.D., Plant Pathologist in
C. F. Ahmann, Ph.D., Physiologist Charge
"W. B. Shippy, Ph.D., Associate Plant Pathol-
J. R. Watson, AM, Entomologist** K. W. Loucks, M.S., Asst. Plant Pathologist
. Grossman, M.A., Entomologist JW. Wilson, Ph.D., Associate Entomologist
A. N. Tissot, Ph.D., Associate C. C. Goff, M.S., Assistant Entomologist
H. E. Bratley, M.S.A, Assistant Plant City
P. W. Calhoun, Assistant, Cotton Insects Bros, Pt Pt t
A N. Brooks, Ph.D., Plant Pathologist
HORTICULTURE R. E. Nolen, M.S.A., Asst. Plant Pathologist
A. F. Camp, Ph.D., Horticulturist** Cocoa
Harold Mowry, B.S.A., Horticulturist A. S. Rhoads, Ph.D., Plant Pathologist
M. R. Ensign, M.S., Associate
A. L. Stahl, Ph.D., Associate Hastings
G. If. Blackmon, M.S.A., Pecan Culturist Hastings
C. B. Van Cleef, M.S.A., Greenhouse Foreman A. H. Eddins, Ph.D., Asso. Plant Pathologist
PLANT PATHOLOGY West Palm Beach
W. B. Tisdale, Ph.D., Plant Pathologist** D. A. Sanders, D.V.M., Veterinarian
George F. Weber, Ph.D., Plant Pathologist Monticello
R. K. Voorhees, M.S., Assistant
Erdman West, M.S., Mycologist Fred W., Walker, Assistant Entomologist
"*In cooperation with U.S.D.A. Bradenton
"**Head of Department. David G. Kelbert, Asst. Plant Pathologist
SOY BEANS FOR SILAGE
By R. B. BECKER, W. M. NEAL, C. R. DAWSON* AND P. T. Dix ARNOLD
Review of literature ............................................ 4
Plan of investigation ......................................... 5
Presentation of results ........................................ 7
Growing the soybeans ................................... 7
Capacity of the silo ..................................... 7
Changes in composition during the ensiling process......... 9
Losses in natural curing into hay during the summer........ 11
Comparative feeding trials ................................ 14
Soybean silage versus No. 1 alfalfa hay............. 14
Palatability of the feeds .......................... 17
Effect of soybean silage on flavor and odor of milk.. 18-
Proximate and mineral contents of feeds used........ 19
Consumption of mineral supplements................ 21
Discussion of results ........................................... 22
Summary and conclusions ....................................... 23
Acknowledgements .................... ........................ 23
Literature cited ................................................. 24
Legume roughages are considered the basis of ideal dairy ra-
tions, since they provide more home-grown protein and mineral
matter than do grasses grown under the same conditions. Where
legumes yield satisfactorily, there is need to study methods of
harvest in order to utilize these crops with the least loss of nu-
trients and at a minimum cost.
In regions having summer rains, difficulty often is experienced
in curing forages into hay. Rains, dews, and even heavy fogs
cause losses of nutrients and palatability of the product, and
prolong the period required for curing the hay. Such prolonga-
tion increases the losses that occur from shattering, from enzy-
matic action in the wilted plants, and from molds and mildews
in storage due to improper and incomplete curing of the crop.
These losses even mount to the point where cattle refuse to eat
appreciable quantities of the finished "hay." As the season ad-
vances in a humid, semi-tropical climate, forage crops are more
subject to insect damage and plant diseases, making harvest
imperative regardless of the weather, in order to reduce loss of
the crop. All of these factors are involved in hay-making during
the summer rainy season in Florida (12)1.
The labor of ensiling a crop is less than that of making hay
under practices in vogue in the Southeastern states. A ton of
dry matter requires less storage space in a silo than as bulk or
baled hay in a modern barn. Some losses of dry matter and of
nutrients' occur in the ensiling process, as well as in the current
practices of making hay.
In order to obtain reliable information as to the feasibility of
ensiling legume forages under Florida conditions, a study of sev-
*Formerly Assistant in Dairy Investigations, Fla. Agr. Exp. Sta.
1Figures in parentheses (italic) refer to "Literature Cited" in the back of
4 Florida Agricultural Experiment Station
eral factors involved in this problem was undertaken with soy-
beans. Earlier observations having a direct bearing on utiliza-
tion of soybeans as silage, have been summarized from the litera-
REVIEW OF LITERATURE
Many stations have investigated mixed silages of soybeans
with corn and other crops, but few have dealt with pure soybean
silage. Studies dealing with these mixed silages have been re-
viewed by Slate and Brown (18), who noted an increased yield
of protein per acre from the corn-soybean combination. Hills (9)
ensiled 564 pounds of green soybeans. The silage kept well and
had a pleasant aromatic odor. Cows ate it readily. Analyses of
freshly-cut soybeans and of the silage after six months in the
silo, showed small losses in total dry matter, crude protein, ni-
trogen-free extract and ether extract from the ensiling process,
based on single sample comparisons.
Soybeans cut into two-inch lengths when the beans were full
size and firm but the leaves and stems still green, were reported
by Woll and Humphrey (20) tohave made a rank distasteful silage
less palatable than corn silage. Three out of 21 cows refused to
eat it, an average refusal of 10 percent having been observed.
These investigators claimed that it imparted an odor to milk,
butter, and cheese.
Eckles (3) stated that "soybeans containing between 30 and 40
percent of dry matter when put into the silo kept well, showing a
loss of feeding value no more than corn, and made a silage palat-
able for livestock. . Silages require about 40 percent as much
storage space per ton as does the same material as hay."
Hayden and Perkins (7) found practically no difference in the
value of good quality soybean hay and soybean silage fed in equal
amounts on the dry matter basis.
Hopkins (11) determined the digestibility of silage made from
fairly mature soybeans, and obtained similar results from each
of four 2-year-old Shorthorn steers, the average digestibility of
the nutrients being as follows:
Coetficients of digestibility
Dry matter. .............................. 49.8
A sh ...................................... 28.0
Crude protein ............................. 55.3
Ether extract .......................... 48.9
Nitrogen-free extract ..... .............. 61.2
Crude fiber ............................. 42.9
Bulletin 255, Soybeans for Silage 5
Emery and Kilgore (5) obtained less uniform results, due, per-
haps, to variety or to stage of maturity of the soybeans, to indi-
viduality of the experimental animals, or to other factors.
Alfalfa has been used successfully as silage over a period of
years by a dairyman near Oklahoma City, soybean silage by a
Jersey breeder in North Carolina, and cowpea silage by a Florida
dairyman near Daytona Beach*.
From the literature reviewed, one may note the incomplete-
ness of information on pure soybean silage, although the increas-
ing use of this crop on dairy farms justifies wider knowledge of
Several questions arise relative to soybean silage. Will soy-
beans make good silage? How many tons will a silo hold? Are
the losses in the ensiling process of significant magnitude? Can
soybean silage replace a good grade of legume hay in dairy ra-
tions ? What effect, if any, will soybean silage have on the quality
of market milk? The answers to these questions have a direct
bearing upon the use of soybeans as pure or mixed silage in
feeding dairy cows.
PLAN OF INVESTIGATION
Otootan and Biloxi soybeans yielded more green forage than
did other varieties grown under the supervision of W. E. Stokes
and C. R. Enlowt on small plots at this station. Since Otootan
appeared to rank slightly higher in average yield, this variety
was grown during the first year of the investigation. Their
decumbent habit of growth made machine harvest so inefficient
that in the second and third years they were replaced by the more
erect growing Biloxi variety.
The soybeans were weighed when harvested, and the area was
measured as a basis for calculating yields. The main part of the
crop was cut in 11/2-inch lengths and ensiled in a monolithic con-
crete silo 10 feet in diameter by 20 feet 10 inches in height, while
loads selected in connection with the green soybean samples were
cured into hay. To study changes that occur during the ensiling
process, five 10-kilogram samples of freshly-cut soybeans were
placed in moistened muslin bags and distributed at various depths
in the silo. Each bag was covered with a thin layer of soybeans
over which moistened burlap was placed to guard against damage
"*Information supplied from these sources.
tUnpublished data based on work at the Florida Agricultural Experiment
6 Florida Agricultural Experiment Station
of the sample bag when removing the silage. Corresponding
samples of fresh soybeans were dried at once for chemical anal-
ysis. The contents of these bags were weighed and sampled
upon removal from the silo during the course of the feeding trials.
All silage was weighed when removed, and the depth measured
at regular intervals to determine the capacity of the silo for this
In each of three winters, a group of eight or 10 Jersey cows
was selected, and divided into two similar lots according to age,
weight, yield and richness of milk, stage of lactation and gesta-
tion. They were used in feeding trials, comparing soybean silage
with No. 1 federal graded alfalfa hay according to the standard
double reversal method (6). Three 30-day periods were used,
with 10-day preliminary and 20-day experimental feeding periods.
The basal ration consisted of corn silage, and mixed grain in
the proportion of four parts of corn feed meal, three parts wheat
bran, two parts velvet bean feed meal (ground in the pods) and
one part of cottonseed meal (41 percent protein). In addition,
either soybean silage or No. 1 federal graded alfalfa hay, pur-
chased on the Kansas City market, was fed these lots. During
the first feeding trial, 31/ pounds of soybean silage were allowed
per 100 pounds of live weight, but in other trials this was reduced
to three pounds. Corn silage was allowed at the rate of three
pounds, and alfalfa hay at one pound per hundredweight of cow.
Allowances of grain were calculated at the beginning of each 30-
day period. according to the Morrison standard (8). The cows
were weighed at 8:00 A.M. on three consecutive days at the be-
ginning and end of each 30-day period. Individual records of
feed consumption, milk yields, and butterfat tests were kept.
The cows were kept in dry lot where they had access to common
salt, feeding bonemeal and water. They were stabled only while
being fed and milked. Since the second feeding trial occurred
during the tick eradication campaign in Alachua County, the
influence of dipping upon milk yield was determined (1).
After the second and third feeding trials were over, the same
groups of cows were fed the soybean hay mentioned previously,
records being kept of the amounts consumed and refused. Rela-
tive.losses in composition of hay and silage were calculated. Con-
sumption of salt and bonemeal per cow per month, was deter-
Influence of the feeds upon flavor and odor of milk was studied
during the second feeding trial.
Bulletin 255, Soybeans for Silage 7
PRESENTATION OF RESULTS
GROWING THE SOYBEANS
For three successive years, the soybeans used in this investi-
gation were grown on one field. The soil was a Gainesville sandy
loam (of residual phosphatic origin). It was unfertilized, except
by turning under a heavy crop of Austrian winter peas before
planting the first crop. The first seeding of soybeans was almost
totally destroyed by cutworms, making it necessary to turn the
soil again, distribute poisoned bran mash, and replant the crop.
Soybeans were drilled in rows, and cultivated to insure a weed-
free stand. The dates and rates of planting, yields of freshly-cut
soybeans and of dry matter per acre, are presented in Table I.
The average acre yield of 5.92 tons of green soybeans contained
337 pounds of crude protein, 33.8 pounds of calcium and 11.9
pounds of phosphorus. Applying digestibility coefficients (8),
the yield of total digestible nutrients was calculated to have been
1,709 pounds per acre.
TABLE I.-ACRE YIELDS OF GREEN SOYBEANS FOR 1929-30-31.
Seed Date Date Stage harvested
Year Variety per of of of
acre planting harvest cutting Fresh Dry
pounds tons tons
1929 Otootan.. 16.8 May 12*... August 22.. Early bloom 4.33 1.18
1930 Biloxi... 37.2 March 19.. July 31.... Earlybloom 8.87 1.90
1931 Biloxi... 48.4 April 6.... Sept. 1.... Small pods. 4.57 1.16
*Date of replanting; original seeding of April 29 destroyed by cutworms.
The dates of harvest in 1929 and 1930 were determined by
Attacks of army worms and other leaf-eating insects. Dry weather
in the summer of 1931 retarded the plant growth. Insect at-
tacks forced an early harvest. Harvest at this early stage of
maturity (early bloom) is not in the interest of maximum yield
of nutrients per acre, according to Hilton (10) and Uhland (19).
The character of growth attained by the soybeans at harvest is
shown in Fig. 1.
CAPACITY OF THE SILO
The silo was opened 80 days after being filled, and there was
an average of 1.62 feet of surface spoilage. The amount of sur-
face spoilage could be reduced by methods under investigation
8 Florida Agricultural Experiment Station
at the Illinois station .(13) in which roofing paper is weighted
down on the top surface so as to exclude air from the silage. An
average depth of 14 feet of settled sound silage was available,
which was fed out over a 123-day period.
TABLE II.-RECOVERY OF GREEN SOYBEANS AS SILAGE IN A SMALL
Interval ensiled r Fresh Percentage re-
Year Variety so--reen weight of Spoiled cover of original
before during nisoyb silage silage dry matter as
feeding feeding enle recovered silage
days days pounds pounds pounds percent
1929 Otootan 79 118 52,122 43,298 2,760 78.94
1930 Biloxi.. 96 122 61,910 54,696 4,205 69.85
1931 Biloxi.. 64 128 61,910 47,155 5,762 64.27
Average. ........ -80 123 58,647 48,383 4,242 71.02
The soybeans were weighed as cut into the silo, and all silage
weighed as removed (Table II), an average of 48,383 pounds of
sound silage being produced from 58,647 pounds of soybeans en-
siled. From the analyses, it was calculated that 71.02 percent of
the dry matter in the fresh soybeans was removed in the sound
silage, full allowance having been made for the surface spoilage.
This efficiency appears lower than under good commercial condi-
tions, and probably is due to the small height of the experimental
silo (20 ft. 10 in.) and the consequent large proportion of surface
The density of the silage at different depths in the silo has
been computed from the weights of silage removed day by day,
and from periodic measurements of depths. These data are com-
pared with similar data for corn silage in Table III, and are pre-
sented graphically in Fig. 2. The surface six feet of settled soy-
bean silage weighed practically the same as corn silage (4). Be-
low this depth, the soybeans compressed to a greater density.
Fifteen feet of settled soybean silage weighed 17 percent more
than an equal depth of corn silage, Density in the bottom four
feet of the silage was increased by the presence of gravitational
Bulletin 255, Soybeans for Silage 9
TABLE III.-CAPACITY OF SILOS FOR CORN AND SOYBEAN SILAGE.
Weight of silage Cumulative weight of
Depth of per cubic foot silage in 10-foot silo
Corn* Soybean Corn* Soybean
feet pounds pounds tons tons
1 32.0 28.2 1.26 1.10
2 32.7 30.4 2.54 2.30
3 33.4 32.7 3.85 3.58
4 34.1 35.1 5.19 4.46
5 34.S 37.3 6.55 6.43
6 35.4 39.1 7.94 7.96
7 36.0 41.5 9.37 9.59
8 36.6 42.8 10.80 11.27
9 37.4 44.2 12.26 13.01
10 38.0 46.2 13.74 14.82
11 38.4 47.5 15.25 16.69
12 38.8 49.4 x 16.77 18.63 x
13 39.2 52.4 x 18.32 20.68x
14 39.6 55.2 x 19.90 22.85 x
15 40.0 57.99 x 21.44. 25.09 x
16 40.2 ........
17 40.4 ..
18 40.6 ...... 26.22..
19 40.8 ........
21 41.2 ...... 31.00 ..
24 41.8 ...... 35.90 ..
25 42.0 ..
26 42.2 ... ..
27 42.4 ...... 40.92..
28 42.6 ......
29 42.8 ... ..
30 43.0 ...... 45.94 .
*Data of Eckles, Reed and Fitch.(4)
xlnfluenced by presence of gravitational moisture in bottom of silo.
CHANGES IN COMPOSITION DURING THE ENSILING PROCESS
Changes in the composition of soybeans incidental to the en-
siling process were measured by means of two series of parallel
samples, one taken of the fresh soybeans at the silo, and the
other from corresponding 10-kilogram samples placed in the silo,
as previously described. The latter were recovered, weighed,
sampled and analysed during the course of the feeding trials.
Comparative average analyses of these series, together with a
10 Florida Agricultural Experiment Station
S20 ---- -Soybean silage
0 --- Gravitational
S10 15 20 Legend 30
Depth of silage feet
"Fig. 2 Comarison of weight per cubic foot of soybean and corn silages
at various depths.
summary of comparisons of soybeans with soybean silage noted
in the literature, are presented in Table IV.sture present
The percentage losses of nutrients in the ensiling process,
00 5 10 15 20 25 30
Depbased on actual amounts of dry matter taken from the samplefeet
bags, are showmparin n Tablweight per cubic foot of soybean an were computed orn silagesthe two
at various depths.
summparallel series of sacomparisonles mentioned above. As previously noted
in the literature, are presented in Table IV.was gravitational
The percentage losses of nutrients in the silo. Losses occurred in
based on actual amounts of dry matter, involving principallytaken from the nitrogen-free extractmple
bags, are shown in Table V. These were computed from the two
parallel series of samples mentioned above. As previously noted
with ensiled corn (4, 15, 16 and 17), there was gravitational
movement of soluble nutrients in the silo. Losses occurred in
total dry matter, involving principally the nitrogen-free extract
and crude protein (nitrogen x 6.25). Crude fiber was affected
little. Slight migration of mineral matter was evident. An in-
crease of ether extract was noted, this phenomenon being due to
the formation of ether-soluble products during the ensiling proc-
Bulletin 255, Soybeans for Silage 11
TABLE IV.-COMPOSITION OF GREEN SOYBEANS AND OF SOYBEAN SILAGE, AS
REPORTED BY THE FLORIDA, MISSOURI, VERMONT AND WISCONSIN STATIONS.
On dry matter basis
moisture Crude Crude Ether Ash
protein N-F E fiber extract
percent percent percent percent percent percent
fresh soybeans........... 74.61 12.12 40.00 36.38 2.41 9.07
soybean silage............. 76.94 8.77 36.35 41.10 3.34 10.44
fresh soybeans............ 75.00 17.60 40.80 25.60 4.40 11.60
soybean silage............. 76.00 18.33 34.17 26.67 8.33 12.50
fresh soybeans ............ 69.80 15.89 37.09 28.48 6.62 11.92
soybean silage............. 62.10 16.36 44.06 26.39 4.49 8.71
fresh soybeans............ 75.36 20.29 41.56 22.53 3.46 12.16
soybean silage............ 77.16 17.61 34.44 26.09 2.86 19.00
fresh soybeans ............ 73.89 14.67 41.82 26.46 4.86 12.18
soybean silage............. 73.92 15.68 33.70 30.44 8.05 12.12
"*The Florida values are average analyses of 15 pairs of samples of Biloxi
and Otootan soybeans cut in the early bloom and small pod stages of maturity.
All others are of earlier varieties harvested when more mature.
LOSSES IN NATURAL CURING OF SOYBEANS INTO HAY DURING
The average rainfall during the summer rainy season at
Gainesville, Florida (12), is in excess of si inches per month.
This precipitation occurs irregularly in showers, the frequency
and irregularity of which almost preclude natural curing of hay
during these months. Rainfall during the autumn averages less
than two inches monthly; however, its irregularity is said to per-
mit satisfactory hay-curing in perhaps two out of three years.
Soybeans seldom can be held over'for fall haying because of dam-
age incurred by army worms and other leaf-eating insects. In
fact, the date of harvesting silage in two out of three years,
was set to avoid excessive insect damage.
During 1930 and 1931, five loads of green soybeans (corre-
sponding to samples taken of fresh soybeans) were cured into
"hay." These loads were representative of the quality of green
material ensiled, and of the growth of forage over the field of
soybeans. They were cured on frames, as shown in Fig. 3, and
TABLE V.-CHANGES IN COMPOSITION DURING THE ENSILING PROCESS, BASED ON ACTUAL AMOUNTS OF DRY MATTER IN FRESH
SOYBEANS AND IN BAG OF SILAGE._
Level in Net weight of Dry Crude Ether Crude Nitrogen-
the silo fresh material Moisture matter protein extract fiber free extract Ash Calcium Magnesium Phosphorus
kilograms percent percent percent percent percent percent percent percent percent percent
"Top........ 10.00 72.11 27.89 3.15 0.62 10.49 11.22 2.41 0.3088 0.1017 0.1119
Second...... 10.00 76.53 23.47 2.77 0.53 8.24 9.41 2.52 0.2607 0.0825 0.0900
Third....... 10.00 74.37 25.63 3.13 0.64 9.44 10.41 2.01 0.3044 0.0941 0.1078
Fourth...... 10.00 75.10 24.90 3.45 0.59 8.80 9.93 2.14 0.2819 0.0895 0.0992 .
Bottom..... 10.00 76.47 23.53 2.71 0.65 8.67 9.21 2.29 0.2729 0.0928 0.0916
Average..... 10.00 74.92 25.08 3.04 0.61 9.13 10.03 2.27 0.2857 .0.0921 0.1001
Top........ 9.252 72.64 27.36 2.46 0.78 11.56 9.91 2.65 0.3296 0.1051 0.1042
Second..... 9.321 74.81 25.19 2.13 0.86 9.66 9.12 3.41 0.2880 0.0958 0.0892
Third....... 9.929 77.33 22.67. 2.07 0.73 9.31 8.42 2.14 0.3084 0.0903 0.0990 t
Fourth...... 10.263 79.03 20.97 1.86 0.65 9.03 7.67 1.77 0.2560 0.0849 0.0848 T
Bottom.... 11.143 80.81 19.19 1.63 0.66 7.92 6.89 2.09 0.2369 0.0819 0.0839
Average.... 9.982 76.92 23.08 2.03 0.74 9.50 8.40 2.41 0.2838 0.0916 0.0922
Percentage recovery of constituents of freshly-cut soybeans in soybean silage
Weighted average.. ...... 102.75 91.05 66.07 i20.26 103.09 82.85 104.54 98.44 98.67 91.56
____ ____ Ivrg
Bulletin 255, Soybeans for Silage 13
stored under a shed until the close of the corresponding feeding
trials with soybean silage. The same groups of cows were offered
the soybean hay.
Fig. 3.-Soybean hay being cured on racks.
The soybean hay was quite unpalatable in 1930, the average
consumption being 21.7 percent. Thunder showers fell during
two afternoons of the harvest. The hay was moldy and dusty,
leached and discolored badly. Slightly better success was at-
tained in 1931, when only one cow refused to eat the hay. The
14 Florida Agricultural Experiment Station
maximum consumption of 23.04 percent, however, indicates its
lack of quality as hay. In no case did any cow refuse as much
soybean silage as was consumed of the soybean hay. The humidi-
ty and rainfall during the summer rainy season interfered with
prompt and complete curing of the soybeans into hay.
Changes in composition of soybeans during the curing process
are presented in Table VI. These analyses do not represent feed
TABLE VI.-LOSSES OF NUTRIENTS IN NATURAL CURING OF SOYBEANS INTO
On dry matter basis
moisture Crude Nitrogen- Crude Ether
protein extract fiber extract Ash
percent percent percent percent percent percent
1930 Fresh soybeans....... 78.59 12.06 36.24 40.45 2.49 8.76
Soybean hay.......... 7.76 9.00 32.88 48.17 1.42 8.53
1931 Fresh soybeans....... 73.39 13.28 42.94 32.61 2.40 8.78
Soybean hay......... 9.53 11.83 35.06 42.93 1.78 8.40
Percentage of dry matter in
1931 crop recovered in
Total dry matter...... 82.85 73.73 67.79 109.10 61.44 78.59
value when the cows refuse to eat such unpalatable material. The
lack of quality is due to leaching, shattering, and production of
undesirable odors and flavors by molds.
It is regretted that the climatic conditions prevented produc-
tion of a high quality hay in order to permit a comparison of
the expected losses with products of equal commercial quality.
The silage was of high quality, but the hay was so damaged as to
be designated as sample grade (14). There is need to measure
the efficiency of methods of harvest in other iregions where dam-
age of the hay from climatic conditions is less serious.
COMPARATIVE FEEDING TRIALS
SOYBEAN SILAGE VS. NO. 1 ALFALFA HAY
Soybean silage was compared with No. 1 federal graded alfalfa
hay by the standard double reversal method. The alfalfa was
purchased on the Kansas City market. Eight Jersey cows were
used during the first and second feeding trials, and 10 during the
third trial. The second feeding trial took place during the course
Bulletin 255, Soybeans for Silage 15
of the tick eradication campaign in Alachua County. It was neces-
sary under state and federal regulations to dip all cows at the pre-
scribed times in a standard solution containing 0.18 to 0.20 percent
arsenious oxide (As203). This involved a total of five dip days,
all of which were during the 20-day experimental periods. A total
decrease in milk yield of 3.32 percent due to dipping during the
winter months was observed (1). No allowance has been made
for this decrease, as both groups of cows were subjected to the
same conditions, and the dip days were well distributed.
The record of one cow was discarded from the second trial be-
cause of the influence of advancing gestation. One record was
dropped from the third trial because the cow went off-feed during
a 20-day experimental period. Twenty-four complete cow records
were obtained upon which to base the calculations comparing the
feeding value of soybean silage with alfalfa hay. These records
were computed by averaging the feed consumption and milk pro-
duction for the first and third 20-day experimental periods, for
comparison with those of the second experimental period. The
summaries of these tabulations is shown in Table VII.
TABLE VII.-SUMMARY OF FEED CONSUMPTION, MILK AND BUTTERFAT
PRODUCTION, AND CHANGES IN BODY WEIGHT.
Feed consumption Production Two-
Number ----thirds of
Year ecow Soybean Alfalfa Corn Mixed Milk Butter changes
silage hay silage grain fat weights
pounds pounds pounds pounds pounds pounds pounds
1929-30... 8 ........ 1,260.0 4,402.5 1,109.5 2,207.7 150.59 43.00
1929-30... 8 2,632.5........ 4,427.0 1,170.0 2,090.6 142.87 31.33
1930-31... 7 ........ 1,099.5 3,483.0 650.0 1,554.5 100.60 14.67
1930-31... 7 2,939.5........ 3,371.0 780.0 1,456.5 97.06 25.00
1931-32... 9 ........ 1,419.0 4,766.0 1,050.0 2,341.6 132.41 -42.33
1931-32... 9 3,510.5....... 4,515.5 1,130.0 2,211.5 134.73 5.65
Weighted.. .. ........ 62.38 210.03 45.77 100.0 6.28 0.25
Average*.. .. 160.26 ........ 214.65 53.43 100.0 6.50 1.10
*This average is weighted on the basis of the number of individual cow-
In order to compute direct comparative value of soybean silage
and No. 1 federal graded alfalfa hay, it was necessary to calculate
slight differences in feeds consumed, and changes in body weights
in terms of some common unit. Morrison (8) computed the total
16 Florida Agricultural E.p,' rintct Station
digestible nutrient intake during the time required for 100-pound
increases in body weights of steers (Minn. Bul. 155 and 193).
From this; the total digestible nutrients per pound of gain, in
excess of maintenance, have been computed.* Since feeds can be
calculated readily upon the basis of total digestible nutrients,
this indirect calculation has been carried out as illustrated below.
The amounts of feeds consumed for maintenance, milk produc-
tion and gains in body weight, per 100 pounds of milk produced,
FEEDS CONSUMED, AND WEIGHTED AVERAGE RETURNS PER
100 POUNDS OF MILK
Soybean Alfalfa Corn Mixed Milk changes in
silage hay silage grain yield body weights
pounds pounds pounds pounds pounds pounds
160.26 .... 214.63 53.43 100 1.10
..... 62.38 210.03 45.77 100 0.25
Difference ... 160.26 -62.38 4.60 7.66 ... 0.85
In other words, by difference, 62.38 pounds of alfalfa hay may
be assumed to have been the equivalent of 160.26 pounds of soy-
bean silage, 4.60 pounds of corn silage, 7.66 pounds of mixed grain,
less 0.85 pounds increase in body weight. As calculated,** 2.93
pounds of the quality of soybean silage used, were equivalent to
one pound of No. 1 alfalfa hay.
"*An 800-900-lb. steer gaining at the rate of 1.53 lbs. daily required 708 lbs.
of T. D. N. per 100 Ibs. gain, including maintenance during this period of
65.36 days. To maintain a 1,000-lb. steer one day requires 7.925 lbs. T. D. N.
On this basis, an 850-lb. steer required .850 x 7.925 x 65.36, or a total of
440.28 lbs. T. D. N. for maintenance while making 100 lbs. gain.
-- = 2.6772 lbs. T. D. N. per pound of grain, in excess of
100 requirements for maintenance during the same
"**Differences in feeds consumed, and in changes in body weight, per 100
lbs. of milk produced, were computed as follows:
4.6 lbs. of corn silage at 0.177 is equivalent to 0.8142 lbs. T. D. N.
7.66 lbs. of mixed grain at 0.7125 is equivalent to 5.4578 lbs. T. D. N.
Total ..... ............ .................... 6.2720 lbs. T. D. N.
0.85 lbs. gain, in excess of maintenance, at 2.6772......2.269 lbs. T. D. N.
Then, 160.26 lbs. of soybean silage plus 6.272 lbs. of T. D. N. were equivalent
to 2.269 lbs. of T. D. N. (0.85 lbs. gain) plus 62.38 lbs. of alfalfa hay. Or
160.26 Ibs.' of soybean silage is the equivalent of 62.38 lbs. of alfalfa hay
plus 4.003 (6.272-2.269) Ibs. T. D. N. Alfalfa hay containing 51.6 lbs. of
T. D. N. per cwt., 4.003 lbs. of T. D. N. is contained in 7.76 Ibs. of alfalfa hay.
Then, -- = --, or one pound of alfalfa hay may be assumed to be
62.38 7.76 1.00
the equivalent of 2.93 pounds of the quality of soybean silage herein repre-
Bulletin 255, Soybeans for Silage 17
PALATABILITY OF THE FEEDS
In the first feeding trial, the cows were offered 31/2 pounds of
soybean silage per hundred pounds live weight, in addition to
the basal ration of three pounds of corn silage per hundredweight,
and sufficient grain to meet the requirements calculated accord-
ing to Morrison standard (8). Alfalfa hay, when fed in place of
soybean silage, was offered at the rate of one pound per hundred
pounds live weight. The proportion of soybean silage during the
second and third feeding trials was reduced to three pounds per
hundred pounds live weight. With this reduction, there was a
drop in the amounts of soybean silage refused from 38.28 percent
during the first year, to 18.26 and 24.58 percent, respectively, in
the second and third years. The cows made no efforts to separate
soybean leaves from stems, as shown by the character of the
The refusal of alfalfa hay was 17.12, 10.76 and 14.85 percent;
and of corn silage, 0.39, 3.20 and 2.54 percent, respectively, dur-
ing the three feeding trials. Grain was refused by only six cows
during very brief periods, and amounted to less than one-fourth
of 1 percent of the total grain offered. The refused roughages
were fed to a group of young heifers that consumed these feeds
with practically no waste.
The above statements are based upon data presented in Tables
VIII and IX. In Table VIII are shown the proportions of all feeds
refused, calculated from 25 individual cow-records covering both
TABLE VIII.-PERCENTAGE OF FEEDS REFUSED BY COWS DURING THE
Number Proportion of feeds refused
records Soybean Alfalfa Corn Mixed
silage hay silage grain
percent percent percent percent
1929-30................. 8 38.28 17.12 0.39 0.56
1930-31.................. 8 18.26 10.76 3.20 none
1931-32 ................. 9 24.58 14.85 2.54 .002
Mean*.................. 25 27.09 12.77 2.05 .03
Range.................. .. 2.40-68.51 1.02-32.96 0-7.19 (a)
Standard deviation........ .. 15.94 8.45 2.11 ....
"*Average of individual percentages of feeds refused by 25 cows.
(a) Grain was refused by only six cows.
18 Florida Agricultural Experiment Station
the preliminary and experimental periods of each feeding trial.
This is a total of 2,240 cow-feed-days, one cow having been dis-
continued for cause at the end of the 80th day. Little tendency
toward an increased liking for soybean silage in succeeding years
was noted from a study of the records, as presented in Table IX.
TABLE IX.-PALATABILITY OF SOYBEAN SILAGE WHEN OFFERED TO DAIRY COWS
FOR THE FIRST TIME, COMPARED WITH TOTAL REFUSAL DURING THREE FEED;
Number Daily offering of silage Daily offering of silage
Year of -- rus --- --
cows offered refused offered refused
pounds pounds percent pounds pounds I percent
First preliminary 10 days First 20- day experimental
1929.............. 8 27.75 12.54 45.18 27.74 12.07 43.51
1930.............. 6 25.00 3.65 14.60 25.00 4.40 17.60
1931.............. 2 24.50 4.20 17.14 24.50, 8.10 33.06
Weighted average.. 16 ................ 31.02 ........ ........ 33.06
All preliminary periods All experimental periods
1929.............. 8 27.58 10.36 37.55 27.58 10.65 38.64
1930.............. 8 25.75 4.16 16.15 23.50 4.55 19.34
1931.............. 9 18.23 4.24 23.25 18.23 5.03 27.57
Weighted average.. 25 ........ ......... 26.15 ............... 29.15
EFFECT OF SOYBEAN SILAGE ON FLAVOR AND ODOR OF MILK
A brief study was made in 1930 to observe whether flavor and
odor of milk were affected when cows ate soybean silage. Three
groups of cows received mixed concentrates and corn silage. In
addition, one of the groups was fed soybean silage, a second group
alfalfa hay, and the third group had access to grass pasture once
daily. Individual. samples of milk were bottled warm and un-
strained from cows in these three groups. A sample of mixed
herd milk was secured also. Fifty-eight samples of milk were
used in this study.
Odor and flavor of each sample of milk were passed upon sepa-
rately by 12 judges, largely instructors and technical workers on
the staff, in an effort to detect which milk was from cows eating
soybean silage. A slight "soybean" flavor in a judge's opinion
Bulletin 255, Soybeans for Silage 19
was weighted one-half point, and an "adriitted" flavor at one
point. Not all judges felt qualified to rate odor of milk, hence
fewer opinions were obtained on this point.
The total number of points (presence of "soybean" flavor or
odor, in the opinion of each judge) was tabulated as shown in
Table X. The opinions were distributed nearly equally among
samples of milk from the four sources, as to which was produced
by cows receiving soybean silage. In other words, no distinguish-
able odor or flavor was imparted to the milk by the cows consum-
ing soybean silage. No tests were made on the physical or chem-
ical constants of the milk fat.
TABLE X.-OPINIONS AS TO THE PRESENCE OF ODOR OR FLAVOR OF SOYBEAN
SILAGE IN MILK.
Samples from Mixed
individual cows samples
Soybean Alfalfa General General
silage hay herd herd
Number of opinions................... 152 54 116
Number of demerits. ................... 43.0 14.5 31.0
Percentage of demerits. ................. 28.3 26.9 26.7
Number of opinions..................... 298 116 226 22
Number of demerits.................... 81.5 39.5 54.0 5.5
Percentage of demerits.................. 27.3 34.0 23.9 25.0
PROXIMATE AND MINERAL CONTENTS OF FEEDS USED
Since relatively few analyses of soybean silage are available in
the literature, and since the mineral contents of feeds for livestock
are of economic importance in Florida, chemical analyses of the
feeds employed in the three feeding trials are presented. The
soybeans were grown on a Gainesville sandy loam (of residual
phosphatic origin), the subsoil containing some clay. Corn for
silage was grown the first and third years on Portsmouth fine
sand, and on a Fellowship sandy loam during the second year.
The No. 1 federal grade alfalfa hay was grown in the Kansas
City trade territory, and purchased there on the open market.
Unfortunately, samples were taken of the mixed concentrates
rather than of the individual components.
The growing season of 1931 was one of low rainfall, which re-
TABLE XI.-PROXIMATE AND MINERAL ANALYSES OF FEEDS USED IN FEEDING TRIALS.
On moist basis, as fed On dry matter basis
Number -- -- -
Feed Year of Crude Nitrogen- Crude Ether Calcium Magnesium Phosphorus
samples Moisture protein free extract fiber extract Ash (Ca.) (Mg.) (P.)
percent percent percent percent percent percent percent percent percent
1929 5 72.86 3.00 10.88 9.98 0.64 2.65 1.267 0.344 0.380
Fresh soybeans..... 1930 5 78.59 2.57 7.77 8.69 .54 1.84 1.159 .398 .451
1931 5 73.28 3.55 11.45 8.74 .64 2.33 1.007 .363 .376
1929 7 71.42 2.39 10.74 11.85 .61 3.00 1.264 .367 .374 "
Soybean silage..... 1930 8 79.92 1.78 7.05 8.01 .84 2.09 1.280 .377 .417
1931 8 74.46 2.56 9.44 10.22 .78 2.55 1.160 .365 .448
1929 3 69.34 2.33 19.18 7.23 .75 1.17 .253 .181 .290
Corn silage........ 1930 3 65.11 2.34 21.61 8.90 .84 1.21 .271 .171 .250
1931 3 72.01 2.28 17.37 6.36 .65 1.32 .322 .185 .255
1929 3 9.99 13.35 44.06 34.05 2.35 6.19 1.169 .146 .181
Alfalfa hay, No. 1.. 1930 3 7.90 15.62 43.54 29.84 2.23 8.76 1.389 .169 .256
1931 3 7.17 16.97 36.83 30.52 1.56 6.97 1.096 .178 .289
1929 1 11.84 17.39 63.41 8.64 6.04 4.52 .220 .331 .619
Mixed concentrates. 1930 3 12.63 16.08 56.60 8.16 2.15 4.38 .147 .390 .697
1931 3 10.27 15.97 59.04 7.25 3.86 3.60 .159 .273 .497
Soybean hay....... 1930 1 7.76 9.00 32.88 48.17 1.42 8.53 .877 .402 .369
1931 1 9.53 10.70 31.72 38.84 1.61 7.60 1.229 .402 .432
Bulletin 255, Soybeans for Silage 21
suited in decreased yields of soybeans and corn. The mineral con-
tent of these crops was lower also. A summary of these analyses
is presented in Table XI.
CONSUMPTION OF MINERAL SUPPLEMENTS
Small quantities of common salt and of finely ground feeding
bonemeal were accessible in the exercise lot. All of the cows were
observed to take these supplements during the course of the feed-
ing trials. The supply was replenished when necessary, and any
excess was weighed out at the close of the feeding trial. Records
of salt and bonemeal consumption per cow per month are pre-
sented in Table XII. These compare closely with similar records
obtained at the Oklahoma station (2) with cows receiving alfalfa
hay, a 16 percent protein concentrate mixture, and cane or kafir
TABLE XII.-AVERAGE CONSUMPTION OF BONEMEAL AND COMMON SALT BY
cow per month Average Average
Cow -- body daily
Year records* Feeding weight milk
Salt bone- yield
pounds pounds pounds pounds
1929-1930.................... 8. 2.25 0.71 918 14.0
1930-1931 .................. 7 8/9 2.33 .66 836 10.6
1931-1932 .................... 10 3.73 1.89x 867 15.2
Oklahoma (2) dairy cows....... 24 2.31 .86 .... ....
*Cow records cover entire 90-day periods for each feeding trial.
xMarked lack of rainfall occurred during the preceding pasture season,
which may have affected mineral contents of feeds during the summer, as
well as those during the feeding trials.
Rainfall was below normal during the growing season of 1931,
and was reflected in a lower calcium and phosphorus content. of
forages grown during that season, as previously noted in Table
XI. The effect of this was noticed in increased appetites of the
cows for both salt and bonemeal. (Steers on grazing experiments
were noted also to take more bonemeal during that summer.)
22 Florida Agricultural Experiment Station
DISCUSSION OF RESULTS
A good quality of silage was made from soybeans harvested
during the rainy season, even when the dry matter content was
as low as 25 percent. Yields of protein and mineral matter per
acre were larger than with corn grown under similar environ-
ment, except soil, but the yields of total digestible nutrients
were not as large.
Soybean silage was usually less palatable to the cows than corn
silage, although considerable variation was noted in the appetites
of different cows for all feeds used in the experimental rations.
It was not possible to secure a complete consumption of the feeds
when offered at the rate of 3.5 pounds of soybean silage and 3
pounds of corn silage per hundred pounds live weight, as when
the offering of soybean silage was reduced one-half pound from
this rate. This observation lends further support to the general
"Rule of Thumb" in feeding roughages. This rule calls for-
2 pounds of dry hay per 100 pounds live weight, or
3 pounds of silage and 1 of hay per hundred pounds live weight,
5 pounds of mangel beets and 1 of hay per hundred pounds live
Apparently, the total bulk of the offering of roughages beyond
the.e rates is a limiting factor in the ability of the average cow
to consume thlem. Exceptions noted were few.
In comparative feeding value, 2.93 pouIds of soybean silage
were the equivalent of one pound of No. 1 alfalfa hay. No char-
acteristic odor or flavor typical of this silage could be detected in
the milk of cows fed this product, even though the milk was
handled in such a manner as to detect any such influence. How-
ever, no opportunity was allowed for the milk to absorb odors or
flavors in the barn after being milked.
Little success was attained in natural curing of soybeans into
hay during the summer rainy season in two successive years. At-
tacks of leaf-eating insects determined the time of harvest, ir-
respective of stage of maturity, maximum yield of nutrients, or
optimum (desirable) weather for curing the soybeans into hay.
These climatic conditions impeded the ensiling process but little.
Under conditions of heavy rainfall, it appears entirely practical
and economical to ensile soybeans rather than attempt to cure
them into hay. Losses in the ensiling process due to enzymatic.
bacterial and chemical actions are closely comparable to those
observed experimentally by others with the corn plant, sunflowers
Bulletin 255, Soybeans for Silage 23
SUMMARY AND CONCLUSIONS
More protein and mineral matter and less total digestible nu-
trients were produced per acre with soybeans than with corn
used as a silage crop.
A silage of good quality was produced even when the green soy-
beans contained as little as 25 percent dry matter. Putrefactive
odors were apparent in the surface spoilage only.
The principal losses of nutrients in the ensiling process were
of the protein and nitrogen-free extract. The crude fiber was
affected very slightly. Some increase was noted in the amount of
ether extract, due to the ensiling processes.
Soybean silage required less space, ton for ton, than corn silage.
Soybean silage is less palatable than corn silage of equal quali-
ty. The cow's capacity seems to be about three pounds of corn
silage and three pounds 'of soybean silage per hundred pounds
live weight. A greater offering resulted in increased refusals.
Measured in terms of No. 1 federal graded alfalfa hay, 2.93
pounds of soybean silage were equivalent to one pound of alfalfa
No flavor or odor characteristic of soybean silage could be
detected in the milk.
The mineral composition of the soybeans varied from year to
year with the amount of rainfall. Soybean silage contained about
four times as much calcium and one and one-half times as much
phosphorus as the corn silage.
Even with the legumes used in the experimental rations, the
cows took additional calcium and phosphorus in the form of bone-
meal, the amount taken being greater after the dry season of
The success of these investigations is due to cooperation of
several persons. W. E. Stokes and C. R. Enlow gave access to
Experiment Station records, from which it was noted that Otoo-
tan and Biloxi soybeans had outyielded several other varieties lo-
cally. Henry Zeigler grew and ensiled the corn and soybeans.
Dr. O. C. Bryan identified the soil types on which these crops
were grown. K. S. McMullen conducted the first and second
feeding trials, and Dan McCloud the third, in an efficient manner.
Will Dunn assisted with analyses of samples representing the
third feeding trial.
24 Florida Agricultural E.cperiment Station
1. Arnold, P. T. Dix, Neal, W. M. and Becker, R. B.: Influence of arsenical
dipping on yield of milk by dairy cows. Jour. Dairy Sci. 15: 407-412.
2. Becker, R. B. and Gallup, Willis D.: Comparison 6f cane and kafir silage
for milk production. Okla. Agr. Exp. Sta. Bul. 177: 1-8. 1928.
3. Eckles, C. H.: Legumes, sudan grass and cereal crops for silage. Mo.
Agr. Exp. Sta. Bul. 162: 3-25. 1919.
4. Eckles, C. H., Reed, O. E. and Fitch, J. B.: Capacities of silos and weights
of silage. Mo. Agr. Exp. Sta. Bul. 164: 3-24 and Kans. Agr. Exp. Sta.
Bul. 222: 3-22. 1919.
5. Emery, F. E. and Kilgore, B. W.: Digestion experiments with pulled
fodder, crimson clover hay, cowpea-vine hay, corn silage, .soja bean
silage and cottonseed, raw, roasted, hulls and meal. N. Car. Agr. Exp.
Sta. Tech. Bul. 4: 3-53. 1892.
6. Forbes, E. B. and Grindley, H. S.: On the formulation of methods of ex-
perimentation in animal production. Bul. Natl. Research Council,
Vol. 6, Part 2, No. 32: 3-16. 1923.
7. Hayden, C. C. and Perkins, A. E.: Soybean hay and soybean silage. Ohio
Agr. Exp. Sta. Bimo. Bul. 11: 5: 178-179. 1926.
8. Henry, W. A. and Morrison, F. B.: Feeds and Feeding. 18th ed. (illus.)
Henry-Morrison Company, Madison, Wis. Pages 449, 726 and 746. 1923.
9. Hills, J. L.: Ensiling soja beans and vetch. Vt. Agr. Exp. Sta. Rpt.:
10. Hilton, J. H., Wilbur, J. W. and Epple, W. F.: Early, intermediate and
late cut soybean hay for milk and butterfat production. Purdue Univ.
Agr. Exp. Sta. Bul. 346: 2-24. 1931.
11. Hopkins, C. G.: Composition and dig -tibility' of corn ensilage, cowpea
ensilage, soja bean ensilage and coi n fodder. Ill. Agr. Exp. Sta. Bul.
43: 181-208. 1896.
12. Mitchell, A. J. and Ensign, M. R.: The climate of Florida. Fla. Agr.
Exp. Sta. Bul. 200: 89-300. 1928.
13. Nevens, W. B.: Preventing surface losses in the silo. Paper presented at
27th Meeting, Amer. Dairy Sci. Assoc., Lexington, Ky., June 30, 1932.
14. Parker, E. C. and Seeds, K. B.: Handbook of official hay standards. U. S.
Dept. Agr. Bur. Agr. Econ.. Pages 2-48. 1925.
15. Perkins, A. E.: Losses and exchanges of material during the storage of
corn as silage. Ohio Agr. Exp. Sta. Bul. 370: 291-306. 1923.
16. Peterson, W. H., Ha -tigii-. E. G. and Fred, E. B.: A study of the principal
changes which take place in the making of silage. Wis. Agr. Exp.
Sta. Res. Bul. 61: 1-32. 1925.
17. Shaw, R. H., Wright, P. A. and Deysher, E. F.: Nitr.:.gcn and other losses
during the ensiling of corn. U. S. Dept. Agr. Bul. 953: 1-16. 1921.
18. Slate, W. L., Jr., and Brown, B. A.: Corn and soybeans as a combination
crop for silage. Conn., Agr. Exp. Sta. Bul. 133: 353-376. 1925.
19. Uhland, R. E.: Time of harvesting soybeans in relation to soil improve-
ment and protein content of the hay. Mo. Agr. Exp. Sta. Bu. 279: 3-28.
20. Woll, F. W. and Humphrey, G. C.: Soybean silage as a food for dairy cows.
"Wis. Agr. Exp. Sta. 21st Ann. Rpt.: 67-74. 1904.