Bulletin 476
UNIVERSITY OF FLORIDA
AGRICULTURAL EXPERIMENT STATIONS
Willard M. Fifield, Director
GAINESVILLE, FLORIDA JOM
JUN 3 195
Toxic Factor in Citrus
Seed Meal
EXTRACTION, CHICK FEEDING TRIALS AND
CHEMICAL CHARACTERISTICS
J. CLYDE DRIGGERS, GEORGE K. DAVIS
AND N. R. MEHRHOF
TECHNICAL BULLETIN
Single copies free to Florida residents upon request to
AGRICULTURAL EXPERIMENT STATION
GAINESVILLE, FLORIDA
21951
May 1951
BOARD OF CONTROL
Frank M. Harris, Chairman, St. Peters-
burg
N. B. Jordan, Quincy
Hollis Rinehart, Miami
Eli H. Fink, Jacksonville
George J. White, Sr., Mount Dora
W. F. Powers, Secretary, Tallahassee
EXECUTIVE STAFF
J. Hillis Miller, Ph.D., President'
J. Wayne Reitz. Ph.D., Provost for Agr.s
Willard M. Fifield, M.S., Director
J. R. Beckenbach, Ph.D., Asso. Director
L. O. Gratz, Ph.D., Asst. Dir., Research
Geo. F. Baughman, M.S., Business Mgr.'
Powers L Ba'-tley. B.S., Admin. Mgr.'
Claranelle Alderman, Accountant"
MAIN STATION, GAINESVILLE
AGRICULTURAL ECONOMICS
H. G. Hamilton, Ph.D., Agr. Econo-
mist' a
R. E. L. Greene, Ph.D., Agr. Economist
Zach Savage, M.S.A., Associate
A. H. Spurlock. M.S.A., Associate
D. E. Alleger, M.S., Associate
D. L. Brooke, M.S.A., Associate
M. R. Godwin, Ph.D., Associate
H. W. Little, M.S., Assistant'
Tallmadge Bergen, B.S., Assistant
D. C. Kimmel, Ph.D., Assistant
A. L. Larson, Ph.D., Agr. Economist
Orlando, Florida (Cooperative USDA)
G. Norman Rose, B.S., Asso. Agr.
Economist
J. C. Townsend, Jr., B.S.A., Agr.
Statistician"
J. B. O-:vens, B.S.A., Agr. Statistician
AGRICULTURAL ENGINEERING
Frazier Rogers, M.S.A., Agr. Engineer' "
J. M. Johnson, B.S.A.E., Asso. Agr. Eng.'
J. M. Myers, B.S., Asso. Agr. Engineer
R. E. Choate, B.S.A.E., Asst. Agr. Engr.'
A. M. Pettis, B.S.A.E., Asst. Agr. Eng.2
AGRONOMY
Fred. H. Hull, Ph.D., Agronomist'
G. B. Killinger, Ph.D., Agronomist'
H. C. H-rris, Pl D. Arr nomist
R. W. Bledsoe, Ph.D., Agronomist
W. A. Carver, Ph.D., Associate
Darrel D. Morey, Ph.D., Associate
Fred A. Clark, B.S., Assistant
Myron C. Grennell, B.S.A.E., Assistant
E. S. Horner, Ph.D., Assistant
A. T. Wallace, Ph.D., Assistant
D. E. McCloud, Ph.D.. Assistant
ANIMAL HUSB. AND NUTRITION
'T. T C'1nh,. Ph.r. An. Hush.' I
R. S. Glasscock, Ph.D., An. Husb.8
G. 4. Davis, Ph.D., Animal Nutritionist'
R. L. Sltirley, PhP.., Biochemist'
J. E. Parp. M.S.. AFst. An. Husb.'
S. John Folks, B.S.A., Asst. An. Husb.4
Katherine Boney. B.S., Asst. Chem.
James T. Wing, M.S.Aksst. Dairy Hush.
A. M Pearson, Ph.D., Asst. An. Husb.a
John D. Feaster, Ph.D., An. Nutritionist
H. D. Wallace, Ph.D., Asst. An. Husb.
DAIRY SCIENCE
E. L. Fouts, Ph.D., Dairy Tech.' '
R. B. Becker, Ph.D., Dairy Husb.'
S. P. Marshall, Ph.D., Asso. Dairy
Husb.
W. A. Krienke, 1M.S., Asso. in Dairy Mfs.'
P. T. Dix Arnold, M.S.A., Asst. Dairy
Husb.'
Leon Mull, Ph.D., Asst. Dairy Tech.
H. Wilkowske, Ph.D., Asst. Dairy Tech.
EDITORIAL
J. Francis Cooper, M.S.A., Editors
Clyde Beale, A.B.J., Associate Editor3
L. Odell Griffith, B.A.J., Asst. Editors
J. N. Joiner, B.S.A., Assistant Editor' 4
ENTOMOLOGY
A. N. Tissot, Ph.D., Entomologist'
L. C. Kuitert, Ph.D., Associate
H. E. Bratley, M.S.A., Assistant
F. A. Robinson, M.S., Asst. Apiculturist
HOME ECONOMICS
Ouida D. Abbott, Ph.D., Home Econ.1
R. B. French, Ph.D., Biochemist
HORTICULTURE
G. H. Blackmon, M.S.A., Horticulturist'
F. S. Jamison, Ph.D., Horticulturist'
Albert P. Lorz, Ph.D.. Horticulturist
H. M. Reed, B.S., Chem., Veg. Processing
R. K. Showalter, M.S., Asso. Hort.
R. A. Dennison. Ph.D., Asso. Hort.
R. H. Sharpe, M.S., Asso. Horticulturist
F. S. Lagasse, Ph.D., Asso. Hort.'
R. D. Dickey, M.S.A., Asst. Hort.
L. H. Halsey, M.S.A., Asst. Hort.
C. D. Hall, Ph.D., Asst. Horticulturist
S. E. McFaddlen, Ph.D., Asst. Hort.
Austin Griffiths, Jr., B.S., Asst. Hort.
LIBRARY
Ida Keeling Cresap, Librarian
PLANT PATHOLOGY
W. B. Tisdale, Ph.D., Plant Patholo-
gist'
Phares Decker, Ph.D.. Plant Pathologist
Erdman West, M.S., Mycologist and
Botanist
Robert W. Earhart, Ph.D., Plant Path.2
Howard N. Miller, Ph.D., Asso. Plant
Path.
Lillian E. Arnold, M.S., Asst. Botanist
C. W. Anderson, Ph.D., Asst. Plant Path.
POULTRY HUSBANDRY
N. R. Mehrhof, M.Agr., Poultry Husb.1 '
J. C. Driggers, Ph.D., Asso. Poultry
Husb.
SOILS
F. B. Smith, Ph.D., Microbiologist' V
Gaylord M. Volk, Ph.D., Chemist
J. R. Henderson, M.S.A., Soil Technolo-
gists
J. R. Neller, Ph.D., Soils Chemist
Nathan Gammon, Jr., Ph.D., Soils
Chemist
R. A. Carrigan, Ph.D., Biochemists
Ralph G. Leighty, B.S., Asso. Soil
SurvPvnr'
G. D. Thornton, Ph.D., Asso.'
Microbiologist'
Charles F. Eno, Ph.D., Asst. Soils
Microbiologist
H. W. Winsor. B.S.A., Assistant Chemist
R. E. Caldwell, M.S.A., Asst. Chemist'
V. V. carlisle. B.S.. Asst. Soil Surveyor
James H. Walker, M.S.A., Asst. Soil
Surveyor
S. N. Edson, M. S., Asst. Microbiologist
William K. Rcbertson, Ph.D., Asst.
bhemist
0. E. Cruz, B.S.A., Asst. Soil Surveyor
W. G. Blue, Ph.D., Asst. Biochemist
VETERINARY SCIENCE
D. A. Sanders, D.V.M., Veterinarian'
M. W. Emmel, D.V.M., Veterinarian':'
C. F. Simpson, D.V.M., Asso.
Veterinarian
L. E. Swanson, D.V.M.. Parasitologist
Glenn Van Ness, D.V.M., Asso. Poultry
Pathologist
G. E. Batte, D.V.M., Asso. Parasitologist
BRANCH STATIONS
NORTH FLORIDA STATION, QUINCY
J. D. Warner, M.S., Vice-Director in
Charge
R. R. Kincaid, Ph.D., Plant Pathologist
L. G. Thompson, Ph.D., Soils Chemist
W. C. Rhoads, M.S., Entomologist
W. H. Chapman, M.S., Asso. Agronomist
Frank S. Baker, Jr., B.S., Asst. An.
Husb.
Mobile Unit, Monticello
R. W. Wallace, B.S., Associate
Agronomist
Mobile Unit, Marianna
R. W. Lipscomb, M.S., Associate
Agronomist
Mobile Unit, Pensacola
R. L. Smith, M.S., Associate Agronomist
Mobile Unit, Chipley
J. B. White, B.S.A., Associate
Agronomist
CITRUS STATION, LAKE ALFRED
A. F. Camp, Ph.D., Vice-Director in
Charge
W. L. Thompson, B.S., Entomologist
J. T. Griffiths, Ph.D., Asso.
Entomologist
R. F. Suit, Ph.D., Plant Pathologist
E. P. Ducharme, Ph.D., Asso. Plant
Path.4
R. K. Voorhees, Ph.D., Asso.
Horticulturist
C. R. Stearns, Jr., B.S.A., Asso. Chemist
J. W. Sites, M.S.A., Horticulturist
H. 0. Sterling, B.S., Asst. Horticulturist
H. J. Reitz, Ph.D., Asso. Horticulturist
Francine Fisher, M.S., Asst. Plant Path.
I. W. Wander, Ph.D., Soils Chemist
J. W. Kesterson, M.S., Asso. Chemist
R N. Hendrickson, B.S., Asst. Chemist
J. C. Bowers, M.S., Asst. Chemist
D. S. Prosser, Jr., B.S., Asst.
Horticulturist
R. W. Olsen, B.S., Biochemist
F. W. Wenzel, Jr., Ph.D., Supervisory
Chem.
Alvin H. Rouse, M.S., Asso. Chemist
H. W. Ford, Ph.D., Asst. Horticulturist
L. W. Faville, Ph.D., Asst. Chemist
L. C. Knorr, Ph.D., Asso. Histologist'
R. M. Pratt, B.S., Asso. Ent.-Pathologist
W. A. Simanton, Ph.D., Entomologist
E. J. Desyck, Ph.D., Asso. Horticulturist
C. D. Leonard, Ph.D., Asso. Horticul-
turist
EVERGLADES STATION,
BELLE GLADE
R. V. Allison, Ph.D., Vice-Director in
Charge
Thomas Bregger, Ph.D., Sugar
Physiologist
J. W. Randolph, M.S., Agricultural Egr.
W. T. Forsee, Jr., Ph.D., Chemist
R. W. Kidder, M.S., Asso. Animal Husb.
T. C. Erwin, Assistant Chemist
C. C. Seale, Asso. Agronomist
N. C. Hayslip, B.S.A., Asso. Entomolo-
gist
E. A. Wolf, M.S., Asst. Horticulturist
W. H. Thames, M.S., Asst. Entomologist
W. N. Stoner, Ph.D., Asst. Plant Path.
W. A. Hills, M.S., Asso. Horticulturist
W. G. Genung, B.S.A., Asst. Entomologist
D. W. Smith, B.S., Asst. Chemist
Frank V. Stevenson, M.S., Asso. Plant
Pathologist
Raymond H. Webster, Ph.D., Asst.
Agronomist
Robert J. Allen, M.S., Asst. Agronomist
SUB-TROPICAL STATION,
HOMESTEAD
Geo. D. Ruehle, Ph.D., Vice-Dir. in
Charge
D. 0. Wolfenbarger, Ph.D., Entomologist
Francis B. Lincoln, Ph.D., Horticulturist
Milton Cobin, B.S., Asso. Horticulturist
Robert A. Conover, Ph.D., Plant Path.
John L. Malcolm, Ph.D., Asso. Soils
Chemist
R. W. Harkness, Ph.D., Asst. Chemist
W. CENT. FLA. STATION,
BROOKSVILLE
William Jackson. B.S.A., Animal
Husbandman in Charge2
RANGE CATTLE STATION, ONA
W. G. Kirk, Ph.D., Vice-Director in
Charge
E. M. Hodges, Ph.D., Agronomist
D. W. Jones, M.S., Asst. Soil
Technologist
CENTRAL FLORIDA STATION,
SANFORD
R. W. Ruprecht, Ph.D., Vice-Dir. in
Charge
J. W. Wilson, Sc.D., Entomologist
P. J. Westgate, Ph.D., Asso. Hort.
Ben. F. Whitner, Jr., B.S.A., Asst. Hort.
Geo. Swank, Jr., Ph.D., Asst. Plant Path.
W. FLA. STATION, JAY
C. E. Hutton, Ph.D., Vice-Director in
Charge
H. W. Lundy, B.S.A., Associate
Agronomist
SUWVANEE VALLEY STA., LIVE OAK
G. E. Ritchey, M.S., Agronomist in
Charge
GULF COAST STA., BRADENTON
E. L. Spencer, Ph.D., Soils Chemist in
Charge
E. G. Kelsheimer, Ph.D., Entomologist
David G. Kelbert, Asso. Horticulturist
Robert O. Magie, Ph.D., Gladioli Hort.
J. M. Walter, Ph.D., Plant Pathologist
Donald S. Burgis, M.S.A., Asst. Hort.
L. M. Geraldson, Ph.D., Asst. Hort.
FIELD LABORATORIES
Watermelon, Grape, Pasture-Leesburg
G. K. Parris, Ph.D., Plant Path. in
Charge
C. C. Helms, Jr., B.S., Asst. Agronomist
Strawberry-Plant City
A. N. Brooks, Ph.D., Plant Pathologist
Vegetables-Hastings
A. H. Eddins, Ph.D., Plant Path. in
Charge
E. N. McCubbin, Ph.D., Horticulturist
Pecans-MIonticello
A. M. Phillips, B.S., Asso. Entomologist2
John R. Large, M.S., Asso. Plant Path.
Frost Forecasting-Lakeland
Warren O. Johnson, B.S., Meterologist2
SHead of Department
2 In cooperation with U. S.
s Cooperative, other divisions, U. of F.
SOn leave.
Contents
INTRODUCTION AND STATEMENT OF PROBLEM 5
REVIEW OF LITERATURE ........------- 6
PREPARATION OF CITRUS SEED MEAL--....- ------.-------- 7
EXPERIMENTAL PROCEDURE AND RESULTS ------ 8
Statement of General Procedure ...---------- 8
Experiment 1.-Orange and Grapefruit Seed Meal 10
Experiment 2.-Ether-Extracted Seed Meal 14
Experiment 3.-Hydrolyzed, Water-Extracted and Alcohol-
Extracted Citrus Seed Meal----.------- 15
Experiment 4.-Alcohol-Extracted Citrus Seed Meal --------- 18
Experiment 5.-Acetone-Extracted Citrus Seed Meal ------------ 20
Experiment 6.-Ether, Acetone and Alcohol-Extracted Citrus
Seed Meal ... ....-------------------------------- 26
Experiment 7.-Ether, Acetone and Alcohol-Extracted Citrus
Seed Meal -.----- 29
CHEMICAL COMPOSITION OF FEEDS ------ 32
IDENTIFICATION OF THE TOXIC PRINCIPLE ---- ---- 33
DISCUSSION OF RESULTS -------- 34
SUMMARY OF CONCLUSIONS 35
ACKNOWLEDGMENTS -- ----------- .....------------- 33
BIBLIOGRAPHY---------------- 35
Toxic Factor in Citrus Seed Meal
Extraction, Chick Feeding Trials & Chemical Characteristics
J. CLYDE DRIGGERS', GEORGE K. DAVIS AND N. R. MEHRHOF
Introduction and Statement of Problem
Citrus seed meal, a by-product of the citrus canning and, sub-
sequently, the citrus seed oil industries, is potentially an impor-
tant-feed ingredient for growing chicks because the availability of
this high protein product parallels an acknowledged need for
vegetable protein in chick feeding.
As now produced, commercial citrus seed meal compares fa-
vorably in chemical composition with acceptable sources of vege-
table protein such as soybean oil meal, peanut meal and cotton-
seed meal. Average analysis indicates a protein content in excess
of 33 percent, fat content higher than 5 percent, fiber less than 8
percent and a mineral content of approximately 6 percent.
Each year Florida produces approximately 50,000,000 boxes
of oranges and 15,000,000 boxes of seeded grapefruit, of which
more than one-half are processed for concentrates, juice and sec-
tions (1)2. Approximately one pound of seed can be secured
from each box of fruit so processed; but, only about one-half of
this quantity of seed is processed for oil. This should give a
working potential of 16,000,000 pounds of seed annually. From
each pound of citrus seed approximately one-third pound of hulls
is removed and one-third pound of oil is extracted, leaving one-
third pound of seed cake, or, when ground, citrus seed meal.
Using the above figures as a basis, approximately 2,500 tons of
33 percent protein supplement of vegetable origin would be avail-
able annually from the citrus industry of the state.
The number of chicks raised plus the number of broilers pro-
duced annually in Florida is approximately 10,000,000 (1). Each
of these birds consumes 10 pounds of feed to 12 weeks of age, and
of this amount probably 90 percent is purchased as commercial
feed. Most commercial chick mashes contain more than 25 per-
1 Also presented to the Graduate Council of the University of Florida in
June 1949 as a Dissertation in partial fulfillment of the requirements for
the degree of Doctor of Philosophy.
2 Italic figures in parenthesis refer to Literature Cited in the back of
this bulletin.
Florida Agricultural Experiment Station
cent vegetable protein supplements. If as much as one-fifth of
this amount could be replaced by citrus seed meal this would be
of great value to the citrus and poultry industries, provided that
the citrus seed meal could be produced as economically as the
other vegetable proteins and also that it was equally as efficient
for chick feeding.
With the foregoing situation recognized, this study was initi-
ated to investigate the possibilities of incorporating citrus seed
meal as a feed ingredient in the diet of growing chicks.
The proposed method of attack was: (1) to determine by
feeding trials whether or not citrus seed meal was toxic to grow-
ing chicks; (2) in the event of toxicity to destroy the harmful
factor by removal by dissolution, acid or alkaline hydrolysis, or
by wet and dry heat treatment; (3) to concentrate, purify and
endeavor to identify chemically the toxic principle if present and
(4) to determine by further feeding trials the relative value of
citrus seed meal, if not harmful, or detoxified citrus seed meal as
compared with soybean meal.
Review of Literature
There are few reports of experimental work with citrus by-
products as a poultry feed, and no reports have been located of
experiments with citrus seed meal, per se, for chicks.
Neal, as reported by Mehrhof and Rusoff (11), showed un-
satisfactory growth and high mortality in chicks fed a high per-
centage of dried grapefruit cannery refuse. Almquist 3 noted
that orange peel and pulp meal in practical chick rearing rations
had no value and seemed to pass through the bird like an inert
ingredient. Sherwood, according to Mehrhof and Rusoff (11),
reported that citrus meal was not palatable to laying birds and
that its use resulted in low egg production. Mehrhof and Rusoff
(11) demonstrated that chicks fed citrus meal (ground citrus
pulp) grew more slowly than the controls during the first four
weeks, and the growth rate was depressed more at the higher
levels. Twenty percent of citrus meal in the chick diet resulted
in the death of almost all the birds. Pectin and naringin, at
levels found in the citrus meal, gave decreased growth rates but
apparently did not explain adequately the heavy mortality caused
by the citrus meal.
However, Neal, Becker and Arnold (12, 13) found dried
citrus pulp and Becker, Marshall and Arnold in unreported work
3 Personal communication to the author.
Toxic Factor in Citrus Seed Meal
found dried citrus seed meal to be palatable for dairy cattle. Also,
Becker, Arnold and Davis (2) noted that citrus cannery by-prod-
ucts have furnished a variety of feeds used largely by dairy cattle.
Preparation of Citrus Seed Meal
Whole citrus seed are a by-product of the citrus juice, concen-
trate and section canning operations. After the separation of the
sections or extraction of the juice from the fruit, the pulp and
seed are carried through a hydrated lime solution with a pH of
approximately 12.6, after which the seed are separated from the
pulp. They then are passed through a steam-heated drum (see
Fig. 1) for drying. Temperature of the drum is approximately
250 F. and the seed are subjected to this temperature for ap-
proximately one hour. After removal from the dryer, they are
cooled and aerated until the moisture content approximates 5
percent.
Since the seed are processed primarily for the citrus seed oils
which are used in manufacturing processes, considerable effort is
made to remove as much of the oil as possible from the seed. The
hulls are removed by first cracking the seed between closely ad-
justed steel plates (see Fig. 2) and are then carried down a series
of steel vibrators which separate the hull from the kernel. As
Fig. 1.-A battery of Louisville steam dryers used in dehydrating
citrus seed to less than 5 percent moisture.
Florida Agricultural Experiment Station
the seed are conveyed beyond the big vibrators, they are passed
through a tunnel where a current of air removes the hulls, leaving
the seed kernels on the conveyor.
The kernels are then conveyed to a screw press where the oil
is extracted by means of mechanical pressure (see Fig. 3). Citrus
seed meal is the ground residue remaining after the oil which can
be extracted by this method is removed.
Experimental Procedure and Results
Statement of General Procedure
Unsexed, day-old Single Comb White Leghorn chicks from the
University of Florida breeding flock were used as experimental
birds, except in one experiment. In each experiment an effort was
Fig. 2.-Mechanical huller used for removing the hulls from whole
citrus seed. This machine is also used for hulling tung nuts, some of
which were being processed at the time this photograph was taken. The
only operation necessary to prepare for hulling citrus seed is changing of
plates, the finer ones being used for citrus seed.
Toxic Factor in Citrus Seed Meal
Fig. 3.-Screw press used in expelling the oil
from the citrus seed kernels.
made to segregate the chicks into as nearly uniform groups as
possible, taking into consideration breeding, weight and physical
condition. At the beginning of each experiment the chicks were
kept in 25-chick metallic starting batteries which were heated by
a 40-watt electric light bulb (see Fig. 4). At the beginning of the
third week the covers to the batteries were elevated by wood-lath
frames to allow more headspace for the chick (see Fig. 5). At
the beginning of the fifth week the chicks were moved to un-
heated intermediate batteries and kept there until the experiment
was terminated (see Fig. 6). Throughout the experimental period
the chicks were held in a room approximately 12' X 20' equipped
with a 12" exhaust fan operating constantly to provide ventilation
and artificially lighted by means of two 40-watt bulbs. Feed and
water were available to the chicks at all times. Weekly records
Florida Agricultural Experiment Station
of chick weight and feed consumption were taken and observa-
tions of abnormalities as well as post mortem examinations were
recorded as deaths occurred.
EXPERIMENT 1.-Orange and Grapefruit Seed Meal
Experiment 1 was designed to determine whether or not citrus
seed meal was toxic and also to determine if there were any differ-
ence in the effect of grapefruit and orange seed meal on starting
chicks.
Fifty White Leghorn chicks from the University of Florida
breeding flocks were used for this experiment and were divided
into five lots of 10 chicks each.
Lot 1 was used as a positive standard group and had access
to the University of Florida chick starting diet which is referred
to as the "positive control" feed.
Lot 2 received a diet of 90 grams positive control feed plus 10
grams orange seed meal. Lot 3 was fed a diet of 80 grams posi-
tive control feed plus 20 grams orange seed meal. Lot 4 received
90 grams positive control feed plus 10 grams grapefruit seed
meal. Lot 5 received 80 grams of positive control feed plus 20
grams of grapefruit seed meal. Results of this experiment are
shown in Table 2.
Fig. 4.-Experimental room, showing batteries for starting chicks.
TABLE 1.-FOR iULA FOR THIE POSITIVE CONTROL FEED.
Yellow corn meal.. ----------
41% soybean meal .. ------
Standard wheat middlings --
Alfalfa leaf meal .-- ---------
Finely ground oats ------- -
Wheat bran ----- ----
Dried whey -------------
55% meat scrap .. --------
Dried distillers' solubles ..-------
Steamed bone meal ----
Oyster shell flour -------
Salt --------- -----
Manganese sulfate --- -
Delsterol (2000 AOAC units/gram)
Pounds
S36
30
7
6
5
3
3
3
3
2
1
1
5.7 grams
9.0 grams
Fig. 5.-Experimental room showing batteries with elevated covers for
holding chicks four weeks. The multideck battery shown on the left was
not used in this experiment.
Florida Agricultural Experiment Station
Fig. 6.-Intermediate batteries used in keeping chicks from five weeks
to termination of experiment.
The higher the percentage of orange or grapefruit seed meal
in the diet the more droopy and generally more unthrifty the
chicks appeared. The droppings were increasingly dark at the
higher levels of citrus seed meal. On post mortem examination
the livers were generally light and occasionally enlarged, the gall
bladders were enlarged and distended with a dark brownish-black
fluid, and generally there was a congestion of the intestinal tract.
TABLE 2.-AVERAGE WEIGHT, FEED CONSUMPTION AND MORTALITY
OF CHICKS FED ORANGE AND GRAPEFRUIT SEED MEAL.
Lot Feed Chicks Av. Weight Feed Con- Mortality
Started sumption
Begin. 6 Wks. per Chick
NUMBER GRAMS GRAMS GRAMS PERCENT
1 Positive control -- --. 10 42.4 429.4 1094.1 0
2 10% orange seed meal --.. 10 42.9 264.2 697.2 40
3 20% orange seed meal -- 10 41.0 161.5 947.6 80
4 10% grapefruit seed meal 10 43.7 235.5 589.2 20
5 20% grapefruit seed meal 10 42.0 235.3 709.0 70
Toxic Factor in Citrus Seed Meal
S 11I
Fig. 7.-Glass Soxhlet extractor used in solvent
extraction of one kilogram of citrus seed meal.
The mortality in each of the lots was experienced prior to the end
of the fourth week. Consequently, the average weight at 6 weeks
was based on 10, 6, 2, 8 and 3 chicks, respectively, for Lots 1
through 5.
Even though 10 chicks only were used in each lot, the results
obtained demonstrate quite conclusively that orange and grape-
fruit seed meal contain a harmful substance and that this factor
is present in approximately the same quantity in each of the
products.
With the establishment of the presence of a harmful factor in
citrus seed meal, a series of experiments was then designed to
remove it.
Florida Agricultural Experiment Station
EXPERIMENT 2.-Ether-Extracted Citrus Seed Meal
Experiment 2 was conducted with the object of determining
whether or not ether extraction would remove the harmful factor
from citrus seed meal. One kilogram of commercial citrus seed
meal dried at 1000 C. for 72 hours was placed in a linen bag and
extracted with diethyl ether in a six-liter capacity Soxhlet ex-
tractor for 72 hours (see Fig. 7). The excess ether was allowed
to evaporate from both the extract and the extracted meal,
which was dried for another 72 hours at 1000 C. and reweighed.
The yield of ether extract from this procedure was 97 grams, or
9.7 percent of the dehydrated citrus seed meal.
For this experiment 20 unsexed day-old Single Comb White
Leghorn chicks were divided into two lots of 10 chicks each. They
were kept in batteries and managed in the usual manner. The
positive control diet was the same as given in Table 1. Lot 1
received a diet composed of 75 grams of the positive control feed
plus 25 grams of ether-extracted meal. Lot 2 received 98 grams
of the positive control feed plus 2 grams of dehydrated ether ex-
tract from citrus seed meal. Results are shown in Table 3.
All chicks fed the ether-extracted citrus seed meal were dead
before they were three weeks old, hence the experiment was ter-
minated at the end of three weeks. As a matter of fact, eight of
the chicks died during the second week of the experiment and the
last of the remaining two chicks died 19 days after the experi-
ment began.
TABLE 3.-AVERAGE WEIGHT, FEED CONSUMPTION AND MORTALITY OF CHICKS
FED ETHER-EXTRACTED CITRUS SEED MEAL AND DEHYDRATED
CITRUS SEED MEAL ETHER EXTRACT.
Lot Feed Chicks Av. Weight Feed Con-
Started -sumption Mortality
Begin. 3 Wks.per Chick
NUMBER GRAMS GRAMS GRAMS PERCENT
1 25% ether-extracted
citrus seed meal ..---- 10 37.0 173.7 100
2 2% dehydrated ether
extract .----------- -- 10 36.6 142.8 262.2 0
Post mortem examination of the dead chicks revealed the
same manifestations as in Experiment 1, namely, enlarged gall
bladders, light-colored livers and congestion of intestinal tract.
Inasmuch as this experiment was designed as a pilot trial to
serve as the first step in a series of experiments, no positive or
Toxic Factor in Citrus Seed Meal
negative groups were used; however, the results were quite con-
clusive in demonstrating that the primary harmful factor had
not been removed from the ether-extracted meal. Average weight
of the chicks fed the dehydrated ether extract from the citrus
seed meal was slightly less at the termination of this experiment
than was that of the positive control group of Experiment 1 at
the end of three weeks; nevertheless, they appeared thrifty and
quite normal in every respect.
Further experiments were planned to fractionate the defatted
citrus seed meal by dissolution, using the several fractions for
feeding trials.
EXPERIMENT 3.-Hydrolyzed, Water-Extracted and Alcohol-
Extracted Citrus Seed Meal
This experiment was conducted with New Hampshire chicks
of a fast growing strain from the flock of a Florida breeder-
hatcheryman. There were seven lots of chicks, 10 chicks to
each lot except number 7, which had only two chicks. All chicks
were kept in 25-chick batteries and managed in the usual manner.
Lot 1 chicks received a diet consisting of 80 grams of the posi-
tive control feed plus 20 grams of untreated commercial citrus
seed meal.
The chicks in Lot 2 were fed a diet of 80 grams positive con-
trol feed plus 20 grams of commercial citrus seed meal which
had been hydrolyzed with hydrochloric acid. For the hydrolysis,
1,800 milliliters of 5 percent HCL were added to each kilogram
of citrus seed meal and autoclaved at 15 pounds pressure for 60
minutes in a steam pressure cooker. This mass was then de-
hydrated at 1000 C. for 96 hours. When the hydrolyzed citrus
seed meal was ready to be incorporated with the positive control
diet it was quite dark in appearance.
Water-extracted citrus seed meal in the amount of 17.75
grams was added to 82.25 grams of positive control feed and
given to Lot 3. This meal was prepared by extraction in a
Soxhlet extractor for 96 hours and then oven-dried at 100' C.
for 72 hours. The yield of dehydrated extract was 12.5 percent.
Because of considerable expansion of the citrus seed meal on
wetting, the linen cloth container became slightly distended dur-
ing the extraction, which made for possible inefficiency of extrac-
tion, hence, the longer extraction period.
Lot 4 chicks received a diet of 97.75 grams of positive control
feed plus 2.25 grams of dehydrated water extract.
Florida Agricultural Experiment Station
Lot 5 diet was composed of 83.6 grams positive control feed
plus 16.4 grams of alcohol-extracted citrus seed meal. The ex-
traction was made in the same extractor as the previous ether
and water extractions. Ninety-five percent alcohol was used and
after the citrus seed meal was dried for 96 hours at 1000 C. it
was extracted for 72 hours. The yield of alcohol extract after
dehydration was 18.2 percent.
The dehydrated alcohol extract was incorporated in the feed
of Lot 6 at the rate of 3.6 grams per 96.4 grams positive control
feed.
Limonin, the bitter principle of citrus seed, was considered as
a possible toxic factor in citrus seed meal. Because of the small
quantity of this compound available at this time only two chicks
were started in Lot 7. The diet of this lot was composed of 99.9
grams of positive control feed plus 0.1 grams of limonin4.
Results of this experiment are shown in Table 4.
On post mortem examination dead chicks from Lots 1, 2, 3
and 6 had enlarged gall bladders and showed signs of intestinal
congestion. In addition, the chicks from Lot 2 showed light-
colored kidneys.
Although it was apparent from these results that untreated
citrus seed meal is unsatisfactory as a chick feed, it appears that
the strain of New Hampshire chicks secured for this experiment
were more resistant to the ill effects of this product than were
the Single Comb White Leghorns which had been used previously.
Comparing the mortality of Lots 1 and 2, it was obvious that
the hydrolyzed citrus seed meal was more toxic than the com-
mercial, untreated product. Whether this was due to an excess
quantity of hydrochloric acid in the hydrolysis or the deleterious
change brought about by the process itself was not determined.
Regardless of the reason therefore, it was believed sufficient to
state that acid hydrolysis did not render the citrus seed meal
acceptable as a chick feed.
From results obtained in Lots 3 and 4, it appeared that the
harmful factor was not removed by water extraction. The
water-extracted meal gave nearly as poor results as the untreated
meal, whereas excellent growth was obtained from feeding the
dehydrated citrus seed meal water extract.
According to results obtained in Lot 5, the ethyl alcohol ex-
4 Secured through the courtesy of the USDA Western Regional Research
Laboratory, Albany, California.
TABLE 4.-AVERAGE EIGHTH, FEED CONSUMPTION AND MORTALITY OF CHICKS FED HYDROLYZED, \~ATER-EXTRACTED AND ALCOHOL-
EXTRACTED CITRUS SEED MEAL.
Chicks Av. Weight
Started Begin. 6 Wks.
Feed Consump-
tion per Chick Mortality
cx.
NUMBER GRAMS
20% citrus seed meal ... ...---- ...---. 10 40.1
20% hydrolyzed citrus seed meal ------.----. .. 10 39.0
Water-extracted citrus seed meal (20% equiv.)._..- 10 41.7
Dehydrated water extract from citrus seed meal (20% equiv.) 10 37.9
Ethyl alcohol-extracted citrus seed meal (20% equiv.) ...--- 10 40.6
Dehydrated alcohol extract from citrus seed meal (20% equiv.) 10 42.3
0.1% Limonin .............---------------------- ---2 37.0
Lot
Feed
GRAMS
246.9
177.0
292.0
478.7
421.7
337.9
Sacrificed
GRAMS
735.93
684.31
783.06
1425.20
1124.20
977.43
at I week
PERCENT
30
90
30
0
0
10
Florida Agricultural Experiment Station
traction removed practically all of the harmful factor present
in the citrus seed meal. When the dehydrated alcohol extract
was added to the positive control diet in Lot 6 there was a marked
diminution in chick growth.
The two chicks in Lot 7 were sacrificed at one week of age, due
to the unavailability of additional limonin. No abnormal condi-
tions were noted in these chicks.
EXPERIMENT 4.-Alcohol-Extracted Citrus Seed Meal
Based on the favorable results obtained from feeding alcohol-
extracted citrus seed meal in the previous experiment, Experi-
ment 4 was designed to corroborate these findings.
For this experiment 80-day-old White Leghorn chicks were
used and divided into eight lots of 10 chicks each. They were all
managed as in previous experiments.
Lot 1 received the positive control diet while Lot 2 was
designated the negative control group and received 80 grams
positive control feed plus 20 grams commercial citrus seed meal.
Lot 3 received 83.75 grams positive control feed plus 16.25
grams of alcohol-extracted citrus seed meal which was prepared
in the same way as for Experiment 3.
Because of occasional manifestations resembling those of a
deficiency of certain vitamins of the B-complex or amino acids, it
was decided to supplement the alcohol-extracted citrus seed meal
with these. Table 5 gives the supplement and units added to 100
grams of alcohol-extracted citrus seed meal. The chicks fed
16.25 grams of this supplemented meal incorporated with 83.75
grams of positive control feed were designated as Lot 4.
TABLE 5.-SUPPLEMENTAL INGREDIENTS USED WITH 100 CHAMS
ALCOHOL-EXTRACTED CITRUS SEED MEAL.
Ingredients Units
dl-Methionine -----. ... 0.9 grams
Vitamin A (Viadex) .............. 441.0 International Units
Vitamin D (Delsterol).......... 40.0 AOAC Units
Thiamine hydrochloride ..----- 0.2 mgms.
Riboflavin ------- ------.-...... ..... 0.4 mgms.
Calcium Pantothenate .---............ 1.1 mgms.
Nicotinic acid ------- -------........... 1.8 mgms.
Pyridoxine hydrochloride ----.....-- .. 0.4 mgms.
Biotin -- -------...... .......... .._.... 3.1 mcgms.
Choline chloride -............... .......153.8 mgms.
Alpha tocopherol ..- --.. ......... 0.3 mgms.
p-Amino benzoic acid... .......--- --- --....---.-- .. .... 30.8 mgms.
Inositol -- ---------................ 2.2 mgms.
Toxic Factor in Citrus Seed Meal
Fig. 8.-Barnstead Soxhlet extractor used in
solvent extraction of four kilograms of citrus seed
meal.
During the process of concentrating the alcohol extract from
the citrus seed meal a crystalline formation was observed ad-
hering to the evaporating dish. Because of the profuseness of
these crystals it was decided that they should be incorporated
in the positive control diet to indicate whether or not this mate-
rial was a toxic factor. Two grams of the crystalline unknown,
approximately one-half the weight of the dried alcohol extract,
was added to 98 grams of the positive control feed for Lot 5.
Whole citrus seed which had been air-dried and from which
the hulls had been removed were used as a basis for Lot 6. The
dried citrus kernels from which the hulls had been removed by
hand were ground in a moderately fine hand-operated food chop-
per. Twenty grams of the ground citrus seed without hulls were
Florida Agricultural Experiment Station
added to 80 grams of positive control feed composing the feed for
the lot.
For Lot 7 the dried citrus kernels which had been ground
were extracted with alcohol in the same manner as previously
described. One part of this extracted product was added to nine
parts of positive control feed.
Lot 8 received 10 grams of the dehydrated alcohol extract
from ground citrus seed plus 90 grams of positive control feed.
Results of this experiment are summarized in Table 6. Lots 2
and 6 demonstrate that citrus seed meal is a toxic ingredient,
whether prepared commercially after the oil has been expelled
or whether ground and fed without oil extraction. Post mortem
examination of the chicks in each of these lots revealed the same
typical citrus seed meal manifestations. Also, growth in Lots
3 and 4 was only slightly less and Lot 7 slightly more than Lot 1.
This demonstrated that the toxic factor can be partially if not
completely removed by 95 percent ethyl alcohol extraction alone
and that vitamins and amino acids when added in amounts tested
apparently did not improve the growth rate. A comparison of
the chicks in each lot can be observed in Fig. 9.
There was a significant retardation of growth in Lot 5, indi-
cating that the alcohol-soluble crystals, even though impure, gave
deleterious results. The alcohol extract from the ground citrus
seed gave results so poor that they were comparable to those
from Lot 2, the negative control group.
In an effort to gain further substantiation of the indication
that alcohol-soluble crystals were the cause of the depressed
growth, chicks in Lots 1 and 5 were continued on the experiment
through 10 weeks of age, but the alcohol-soluble crystals were
removed from the diet of Lot 5 at the end of six weeks. This
resulted in a positive control feed being consumed by both lots
after six weeks of age. The bi-weekly weights of Lots 1 and 5
are presented in Fig. 10.
The chicks in Lot 5 gained 392.3 grams from the end of
sixth to end of tenth weeks, 13.7 grams more than the average
gain of 378.6 grams in Lot 1, demonstrating that when crystals
were removed from feed the chicks gained weight at approximate-
ly a normal rate after having been significantly retarded.
EXPERIMENT 5.-Acetone-Extracted Citrus Seed Meal
Recognizing the possibility that the impure crystals fed in
Lot 5 of Experiment 4 contained the harmful factor in citrus seed
Toxic Factor in Citrus Seed Meal
Fig. 9.-Photographic comparison of chicks in Experiment 5 at six
weeks of age. Lot 1 is the positive control group; Lot 2 received 20%
citrus seed meal; Lot 3 alcohol-extracted citrus seed meal; Lot 4 same as
Lot 3 plus vitamin supplement; Lot 5 alcohol-soluble citrus seed crystals;
Lot 7 alcohol-extracted ground citrus seed; Lot 8 alcohol extract from
ground citrus seed.
.OT
LOT 4
.OT I
LOT 1
LOT 5
)I 'B*
TABLE 6.-AVERAGE V EIGHT, FEED CONSUMPTION AND MORTALITY OF CHICKS FED ALCOHOL-EXTRACTED COMMERCIAL AND GROUND
CITRUS MEAL AND ALCOHOL-SOLUBLE CITRUS CRYSTALS.
Feed
NUMBER GRAMS
Positive control .....-------.. ---------------- 10 37
20% citrus seed meal--..- .... ------ 10 38
16/4% alcohol-extracted commercial citrus seed meal ._----- 10 38
161/4% alcohol-extracted commercial citrus seed meal plus
Vitamin and Amino Acid Supplements .-------..... .------ 10 37
2% alcohol-soluble citrus seed crystals... ---.._.....-------- 10 37
20% ground citrus seed with hulls removed --- 10 37
10% alcohol-extracted ground citrus seed ----------- 10 38
10% alcohol-extract from ground citrus seed.. ---- 10 38
Lot
Chicks Av. Weight
Started Begin. 6 Wks.
Feed per
Chick
GRAMS
1060.8
410.2
1073.3
1031.1
707.4
526.2
1151.1
439.3
GRAMS
394
128
364
348
227
415
146
Mortality
PERCENT
20
90
20
10
10
100
10
90
Toxic Factor in Citrus Seed Meal
meal, solubility tests were made with several solvents in an effort
to find one more economical than ethyl alcohol and yet as efficient.
Solvents used were: acetone, furfuryl alcohol, isopropyl alcohol,
xylene, Skellysolve B, methyl alcohol, petroleum ether, glycerol,
absolute ether, water, and methyl cellisolve. Of these solvents
only acetone and furfuryl alcohol appeared satisfactory, inasmuch
as they were the only ones in which the crystals were easily solu-
ble, and because of the availability of acetone it was the solvent
chosen for this experiment.
Five lots of 12 Single Comb White Leghorn chicks were used
for this trial. These chick were of the same general breeding as
the ones used in previous experiments and were managed in a
similar fashion.
Lot 1 received the positive control diet while Lot 2 received
83.75 grams of positive control feed plus 16.25 grams of acetone-
extracted commercial citrus seed meal. This meal was dehydrated
at 96 to 1000 C. for 72 hours, acetone-extracted for 72 hours in a
Soxhlet extractor and dried again at 96 to 1000 C. for 72 hours, as
had been the previous extractions.
The acetone extract was dried at 96 to 1000 C. for 72 hours
and 3.75 grams of this material were added to 96.25 grams of the
positive control feed for Lot 3.
Whole citrus seed with the hulls removed were ground in a
hand food chopper similar to that in Experiment 4 and acetone
extracted as the commercial citrus seed meal for Lot 2. Twelve
and one-half grams of this material was added to 87.5 grams of
positive control feed for Lot 4.
Seven and one-half grams of the dried acetone extract from
the ground citrus kernels was added to 92.5 grams of positive
control feed to compose the formula for Lot 5.
Because of the lack of some feed ingredients, this experiment
was terminated at five weeks and results are presented in Table 7.
Very little mortality was experienced in this experiment, as
can be seen in Table 7. However, there was a marked depression
of growth in Lots 3 and 5 which received the dried acetone extract
from citrus seed meal, indicating that a harmful factor had been
extracted by acetone. The failure of the chicks in Lots 2 and 4 to
gain weight comparable to those in Lot 1, on the other hand, was
strong evidence that the acetone extraction did not completely
remove the toxic element from the citrus seed meal.
Florida Agricultural Experiment Station
W-EKS
r ams
750
700
650
6oo
550
500
450
4oo .
300 o-
200
150 -
100
L I ee
Legend
* Lot 1 Positive Control
Lot 5 Positive Control plus
Alcohol Soluble Crystals first 6
weeks. Yo crystals 7 10 weeks.
I
I
1
/
/
I
I Receive(I Positive
/ Control. Diet Only.
I
/
I
I
/ ,eceived Positive
a Control Diet Only.
Fig. 10.-Bi-weekly weights of chicks fed positive control diet and
alcohol-soluble crystals.
TABLE 7.-AVERAGE WEIGHT, FEED CONSUMPTION AND MORTALITY OF CHICKS FED ACETONE-EXTIACTEDI CITRUS SEED MEAL.
Feed
Av. Weight
Chicks
Started Begin. 5 Wks.
NUMBER GRAMS
1 Positive control ------------- 12 35.8
2 16.25% acetone-extracted commercial citrus seed meal...--- 12 35.2
3 3.75% dried acetone extract from commercial citrus seed meal 12 36.0
4 12.50% acetone-extracted ground citrus kernels. ----- 12 36.7
5 7.50% dried acetone extract from ground citrus kernels--- 12 36.9
GRAMS
306.4
264.5
214.5
250.5
152.4
Feed per c
Chick Mortality
GRAMS PERCENT
808.5 0 i
769.5 0
583.4 8.3
793.5 8.3 '
589.7 16.7
Florida Agricultural Experiment Station
EXPERIMENT 6.-Ether, Acetone and Alcohol-Extracted
Citrus Seed Meal
Considerable difficulty in completely removing the harmful
factor from citrus seed meal was experienced in the previous
trials. This difficulty suggested Experiment 6 to determine
whether a combination of solvents would remove all toxicity from
the citrus seed meal.
For this experiment 120 White Leghorn chicks were used and
divided into 12 lots of 10 each. They were managed as in previous
experiments, and due to difficulties in securing sufficient feed in-
gredients, the experiment was terminated at four weeks.
Lot 1 was designated as the positive control and Lot 2 was the
negative control citrus seed meal group.
For the third lot, 16.25 grams of ether and alcohol-extracted
citrus seed meal was added to 83.75 grams of the positive control
feed. The extracted meal was prepared as follows: the citrus
seed meal was dehydrated at 96 to 1000 C. for 72 hours, extracted
with diethyl ether in a Soxhlet extractor for 72 hours, dried
at 96 to 1000 C. for 72 hours, extracted with 95 percent ethyl alco-
hol for 72 hours and then dried again for 72 hours at 96 to 1000 C.
to completely remove the alcohol.
Lot 4 received 3.75 grams of dried ether extract added to
96.25 grams positive control feed.
Sixteen and one-fourth grams ether and acetone-extracted
meal was added to 83.75 grams positive control feed to compose
the diet of Lot 5. In this instance the citrus seed meal was pro-
cessed in the same manner as for Lot 3, except that acetone was
used instead of the alcohol.
A quantity of commercial citrus seed meal was ether-ex-
tracted to remove the fat and then water-extracted in a similar
manner as for the previous lot. Then 650 grams of this material
were added to 3,350 grams of positive control feed for Lot 6.
The chicks in Lot 7 were fed 3.75 grams of dried acetone ex-
tract added to 96.25 grams positive control feed. The acetone
extract used was that collected after the ether extraction and
subsequent drying of the meal.
The dried water extract was collected after defatting the meal
with ether and later dried at 96 to 1000 C. for 72 hours. Three
and three-fourths grams of this material were added to 96.25
grams of positive control feed for Lot 8.
Toxic Factor in Citrus Seed Meal
Ether, acetone and alcohol extracted commercial citrus seed
meal was used for Lot 9. The meal was dehydrated at 96 to
1000 C. for 72 hours, ether extracted for 72 hours, dried for 72
hours, acetone extracted for 72 hours, dried for 72 hours, alcohol
extracted for 72 hours and finally dried for 72 hours at 96 to
1000 C. After the final extraction, the dried meal had an agree-
able taste with no bitterness. The alternate drying and extrac-
tion method was used to insure completeness of extraction by the
different solvents. This procedure also allowed a relatively large
quantity to be extracted at one time and it was possible to keep
the still in operation at all times extracting meal for the other
lots. Sixteen and one-fourth grams of this meal were added to
83.75 grams of positive control feed for Lot 9 diet.
For Lot 10 dried alcohol extract in the amount of 3.75 grams,
obtained after the meal had been ether and acetone-extracted,
was added to 96.25 grams of the positive control feed.
Three and three-fourths grams dried alcohol extract obtained
from the meal after it had been defatted with ether was added to
96.25 grams of positive control feed for Lot 11.
Lot 12 received a diet of 16.25 grams ether-extracted meal
plus 83.75 grams positive control feed.
Results of this experiment are summarized in Table 8. The
difference in weight between Lots 1 and 9 was insignificant, indi-
cating that the combination of extractions using ether, acetone
and alcohol completely removed the harmful factor from citrus
seed meal.
Lot 8 gains were fairly good; however, they were 10 percent
less than the control lot, indicating again that little if any harm-
ful factor is contained in a dried water extract. Because the fat
content of Lot 4 was increased by approximately 3.75 percent, it
is possible that the lower growth in Lot 4, which received the
dried ether extract from the citrus seed meal, was due to the high
fat content in the diet (5).
Average chick weight at four weeks was 30.6 grams less in
Lot 3, receiving the ether-alcohol extracted citrus seed meal, and
49.6 grams less in Lot 5, which received the ether-acetone ex-
tracted citrus seed meal, than in the positive control lot. It is
possible that the citrus seed meal was incompletely extracted in
both instances.
Growth rate of chicks in Lot 11 was better than for chicks
previously fed alcohol extract from citrus seed meal. Presumably
TABLE 8.-AVERAGE WEIGHT, FEED CONSUMPTION AND MORTALITY OF CHICKS FED ETHER, ACETONE
AND ALCOHOL EXTRACTED CITRUS SEED MEAL.
Lot
1 Positive control --...---..--....----- ---------------
2 20% citrus seed meal --.--.---- ---------
3 16.25% ether and alcohol-extracted citrus seed meal -------
4 3.75% dried ether extract from citrus seed meal --.--
5 16.25% ether and acetone-extracted citrus seed meal -- -
6 16.25% ether and water-extracted citrus seed meal ---
7 3.75% dried acetone extract from citrus seed meal after ether
extraction ....---------
8 3.75% dried water extract from citrus seed meal after ether
extraction ------.. ----------.
9 16.25% ether, acetone and alcohol-extracted citrus seed meal
10 3.75% dried alcohol extract from citrus seed meal after ether
and acetone extraction .--.-- -------- -----
11 3.75% dried alcohol extract from citrus seed meal after ether
extraction ------...------------
12 16.25% ether extracted citrus seed meal
Chicks Av. Weight
Started Begin. 4 Wks.
NUMBER GRAMS GRAMS
10 35.1 246.0
10 37.8 93.2
10 38.4 215.4
10 36.9 178.1
10 36.5 196.4
10 36.0 103.8
10 36.8
10 36.3
10 36.0
10 37.8
81.8
222.5
242.2
195.3
176.7
139.2
Feed per
Chick
GRAMS
517.5
226.2
468.5
440.4
482.6
226.1
175.5
525.7
545.7
448.4
407.2
331.0
Mortality
PERCENT
10
40
0
20
20 ;
20
50
50
0
10 1
0
0
20
N
Toxic Factor in Citrus Seed Meal
there was less of the toxic factor present in the extract or it was
less toxic than in previous trials. Results from Lot 10, which re-
ceived the dried alcohol extract after ether and acetone extrac-
tion, indicate that some but not all of the growth-retarding sub-
stance had been previously removed by acetone extraction.
As in previous trials, chicks receiving 20 percent untreated
citrus seed meal, meal extracted with ether and water, dried ace-
tone extract from citrus seed meal, and ether-extracted citrus
seed meal in Lots 2, 6, 7 and 12, respectively, had the poorest
growth and highest mortality. Apparently neither ether nor
water removed the toxic substance from the seed meal, but at
least part of the factor was removed by acetone extraction. On
post mortem examination the chicks in these lots showed the
usual toxicity lesions resulting from feeding citrus seed meal.
A comparison of the chicks in each lot can be made in Fig. 11.
EXPERIMENT 7.-Ether, Acetone and Alcohol-Extracted
Citrus Seed Meal
This experiment was designed to (1) corroborate the findings
of the previous experiment which indicated the apparent detoxifi-
cation of citrus seed meal by ether, acetone and alcohol extrac-
tion; (2) study the relative efficiency of the detoxified seed meal
as a vegetable protein in comparison with soybean meal; and (3)
study the effect of the acetone-soluble crystals when purified and
incorporated in the chick diet.
Sixty Single Comb White Leghorn chicks were used in this ex-
periment and were divided into 5 lots of 12 chicks each. The
chicks were managed as in previous experiments.
Lot 1 was the positive control group and received the positive
control diet, while Lot 2 was designated the negative control
group and received 20 grams of untreated citrus seed meal added
to the 80 grams of positive control feed.
In preparing the citrus seed meal for Lot 3 a Barnstead
Soxhlet extractor (see Fig. 8) was used so that larger quantities
(four kilograms) could be extracted at one time. The extraction
time, drying time and procedure were the same as had been used
in the previous experiment using ether, acetone and alcohol as
solvents. It was noted that the finished product was free from
disagreeable flavors of bitterness. In preparing the feed for Lot 3
a basal feed was prepared of the same ingredients and in the same
Lot 3, Ether and alcohol extracted citrus
seed meal
Lot 5, ztner and acetone extracted citrus
seed meal
Lot 4, Dried ether extract from citrus
seed meal
Lot 7, Dried acetone extract after ether
extraction
LCt ., riner ana water extractea citrus
seed meal
Lot 8, Dried water extract after ether
extraction
Lot 11, Dried alcohol extract after ether
extraction
Lot 10, Dried alcohol extract after ether
and acetone extraction
Lot 9, Ether, acetone and alcohol ex-
tracted citrus seed meal
Fig. 11.-Photographic comparisons of
chicks in Experiment 6 at four weeks of
age. Lot 1 chicks are used as positive con-
trcl standards and are stationed at the left
in each photograph.
Lot 12, Ether extracted citrus seed meal
Toxic Factor in Citrus Seed Meal
quantities as the positive control feed except that soybean meal
was omitted.
For Lot 3, 70 grams basal feed, 10 grams soybean meal and
24.85 grams extracted citrus seed meal constituted the diet. The
citrus seed meal was calculated to replace the protein in 20 grams
of soybean meal which was 20 percent of the positive control feed.
The crystals, which had been observed previously and found
to be acetone-soluble but water-insoluble, were purified by repeat-
edly dissolving in hot acetone and crystallizing by the addition of
water. Three and three-fourths grams of this crystalline material
were added to 96.25 grams of positive control feed for Lot 4.
Lot 5 received 3.75 grams of alcohol extract from citrus seed
meal after ether and acetone extraction added to 96.25 grams of
positive control. The alcohol was removed by distillation under
reduced pressure.
Results of this experiment are shown in Table 9.
TABLE 9.-BoDY WEIGHT, FEED CONSUMPTION AND MORTALITY OF CHICKS FED
ETHER, ACETONE AND ALCOHOL-EXTRACTED CITRUS SEED MEAL.
Lot Feed Chicks Av. Weight Feed Per Mortality
Started Begin. 4 Wks. Chick
NUMBER GRAMS GRAMS GRAMS PERCENT
1 Positive control -12 37.4 247.1 570.5 0
2 20% untreated citrus seed
meal-negative control- 12 -37.6 71.3 227.6 75
3 Ether, acetone, alcohol-ex-
tracted citrus seed meal
(replacing 20% sybean
meal) 12 40.0 240.9 562.7 0
4 3.75% acetone-soluble
crystals ------ 12 40.2 191.5 100
5 3.75% alcohol-soluble ex-
tract after ether and ace-
tone extraction .. 12 42.3 147.2 355.0 33
Results of this experiment are in agreement with those of the
previous trials as regards the feed value of citrus seed meal when
completely detoxified. The alcohol-soluble extract after ether and
acetone extraction demonstrated a growth-retarding activity and
some mortality.
Differences in body weight, feed consumption and mortality in
Lots 1 and 3 are not significant. Chicks in Lot 3 appeared as
thrifty as those in Lot 1 in every respect with the exception that
three individuals in Lot 3 exhibited a slight leg bone malforma-
tion indicating an improper mineral balance. Analysis of the feeds
Florida Agricultural Experiment Station
for ash showed considerably higher quantities in Lot 3 than in
Lot 1, indicating an improper mineral balance.
As in previous experiments, all chicks were autopsied and
typical citrus seed meal manifestations were observed in all chicks
from Lots 2, 4 and 5. Other outstanding pathological symptoms
noted were mottled livers and occasionally ascites.
A comparison of the internal organs of chicks from Lots 2 and
4 in this experiment can be made from the photographs in Fig. 12.
Chemical Composition of Feeds
Because the results of these experiments appeared to be af-
fected primarily by complete extraction of the toxic factors in
citrus seed meal rather than by changes in the feed formulas, only
routine analyses were made on those ingredients which were
varied in the formulas and on the formulas after complete mixing.
The average composition of the ingredients and formulas is
given in Table 10.
Fig. 12.-Photographic comparison of postmortems performed on
chicks from Lot 2 (left) and Lot 4 (right). The chick from Lot 2 received
20% untreated citrus seed meal, while the one from Lot 4 received the
positive control diet to which had been added 3.75% acetone-soluble,
water-insoluble crystals.
LIVER.
N
6ALL % ILIVLk
4 %
bLADDEk i I
.. VK m
SMALL
INTESTINE.
GIZZARDL
.1 ,' BLADDEUL
'SMALL
INTESTINE.
Toxic Factor in Citrus Seed Meal
TABLE 10.-AVERAGE COMPOSITION OF FEEDS USED IN EXPERIMENTAL
CHICK RATIONS.
Feeding Stuff Dry Protein Fat Fiber N-Free Mineral
Matter Extract Matter
Percent
Soybean meal (41%) 86.3 44.7 0.5 5.4 29.0 6.7
Citrus seed meal (commercial). 85.0 33.9 5.7 7.5 31.9 6.0
Citrus seed meal detoxifiedd)_ 87.7 43.0 0.1 9.6 26.9 8.1
Positive control diet 85.0 22.0 0.9 5.5 45.9 10.7
20% detoxified citrus seed meal
diet- 86.8 23.8 0.8 4.9 44.1 13.2
Identification of the Toxic Principle
Bernays (3) reported that limonin obtained from citrus seed
has a bitter taste, is insoluble in water and ether, and soluble in
alcohol. Nolte and Von Loesecke (14) found that the bitter frac-
tion of grapefruit seed is limonin. Emerson (4) obtained limonin
from orange seed by grinding, defatting with petroleum ether and
extracting with acetone. He subsequently extracted more limonin
with iso-butanol saturated with water.
In view of the procedure used in the complete detoxification of
the citrus seed meal, i. e., dehydration, ether-acetone-alcohol ex-
traction, and drying, and recognizing that the meal which was not
toxic was not bitter, it was suspected that the toxic principle
might be limonin.
The melting point of the purified crystals was observed to be
between 261 and 265 C., which is in close agreement with the
melting point of 262 to 264 C. observed by Higby (8) for a com-
pound which he called isolimonin but which he crystallized from
alcohol. Because of the difference in melting point obtained by
Higby from presumably the same compound but crystallized by
different methods, the toxic crystals recovered from an acetone
solution by an addition of water were dissolved in hot ethanol and
allowed to recrystallize on cooling. The melting point of the crys-
tals so prepared was 275 to 278 C., which is in close agreement
with the 277 to 280 C. melting point determined on a supply of
limonin received from the USDA Western Regional Research
Laboratory, Albany, California. Melting points of 2800 C, 292-
294 C, 299-300 C and 3040 C have been reported by Koller and
Czerny (10), Seshadri and Veeraraghaviah (17), Schechter and
Haller (15, 16) and Feist and Schulte (6).
Chemical analysis of the acetone-soluble compound showed the
composition to be: Carbon, 66.37 percent; hydrogen, 6.43 percent;
Florida Agricultural Experiment Station
oxygen, 26.20 percent. This composition suggests a compound
with the empirical formula CM H. Os, which is in close agreement
with that reported by Feist and Schulte (6). Other formulae
which have been proposed for limonin are: C2. H,- 07 by Koller and
Czerny (9) and CG H~o O by Seshadri (18).
Geissman and Tulagin (7) presented some observations on the
structure of limonin, but the determination of the structural
formula of our toxic principle was not undertaken.
Lacking further information, it is impossible to state that our
substance is identical with that of Emerson which he has desig-
nated as limonin, but similarity of melting point and composition
makes this possible.
Discussion of Results
Complete detoxification of citrus seed meal was necessary to
obtain favorable results in feeding chicks. This was effected in
these experiments by complete solvent extraction. The non-agi-
tating type of extraction equipment available made it difficult to
extract completely the large samples required. Finely ground
particles of seed appeared to block the small openings in the linen
cloth bag and when the solvent was being siphoned off into the
distilling flask the sample could not be freed of the residual sol-
vent containing the toxic principle. The efficiency was enhanced
somewhat by elevating the cloth container in the condensing
chamber, which allowed considerable quantities to drip out before
fresh solvent was condensed over the sample. However, the
greater the elevation the less sample extracted, hence there was a
definite limit in this direction. Hand agitation of the sample ap-
proximately every four hours aided considerably, but reducing
the size of the sample extracted to where it was not more than
three inches in depth, it appeared, was the most efficacious method
of completing the extraction. It was found that refluxing the
sample in the flask of the solvent desired gave excellent results
very rapidly but was expensive, due to the excessive solvent loss.
Because the ether, acetone and alcohol solutions were very
clear during the process of extracting the fat and toxic principle
from citrus seed meal, it was difficult to determine whether ex-
traction was complete without dismantling the extraction ap-
paratus. In every instance where the meal gave good results,
there was no bitter or otherwise offensive taste. Consequently,
the taste test was the best criterion of complete extraction found.
Toxic Factor in Citrus Seed Meal
The small amount of feed which could be extracted at one time
limited the number of chicks which could be used in one experi-
ment. However, the number of repetitions demonstrated that, as
now prepared, commercial citrus seed meal is toxic to chicks but
that toxicity can be removed by ether, acetone and 95 percent
ethyl alcohol extraction. It was further demonstrated that de-
toxified citrus seed meal could be included in the chick diet up to
20 percent of the total, and that acetone-soluble crystals were the
primary cause of chick mortality. Results in Experiments 7 and 8
indicate that a residual deleterious factor after ether and acetone
extraction is removed by subsequent alcohol extraction.
The difficulty experienced in extracting the meal raises the
question as to whether efficient acetone or alcohol extraction alone
after ether extraction might not give a satisfactory product for
growing chicks. This can be answered only by the extraction with
improved equipment. Separate extractions with these solvents
did remove all toxicity, whereas neither ether nor water appeared
to remove much, if any, of the toxic principle.
Summary and Conclusions
Experiments including 20 percent commercially prepared cit-
rus seed meal in the diet of White Leghorns demonstrated that,
as now prepared, this by-product of the citrus canning and citrus
seed oil industries is unsatisfactory for chick growth. A high
mortality resulted during the first three weeks in each of these
studies. Enlarged gall bladder and sometimes mottled liver,
ascites and congestion of the intestinal tract were the manifesta-
tions observed.
The primary deleterious factor was found to be a white crys-
talline compound soluble in acetone and 95 percent ethyl alcohol
but insoluble in water and diethyl ether. Chemical analysis sug-
gests that this compound is limonin, the bitter principle of the
seed. In certain experiments there were indications that a second
factor soluble in alcohol after separate extraction with ether and
acetone was a contributing cause of the unsatisfactory results.
Dehydrated citrus seed meal, when extracted successively
with diethyl ether, acetone and 95 percent ethyl alcohol and added
to a good chick mash at a level of nearly 20 percent, gave satis-
factory results. Also, when used to replace slightly above 20 per-
cent soybean meal in the University of Florida chick starting diet,
completely detoxified citrus seed meal proved quite satisfactory as
a source of vegetable protein for chicks to five and six weeks old.
Florida Agricultural Experiment Station
ACKNOWLEDGMENTS
The authors wish to express their sincere appreciation to Mr. J. E.
Hutchman, Lakeland, Florida, who provided the citrus seed meal, and Dr.
Ralph Miller, Plymouth, who supplied the citrus seed.
Dr. S. G. Gilbert, Dr. F. S. Lagasse, Dr. E. G. Rietz, Mr. G. H. Blackmon
and Miss Katherine Boney, University of Florida, rendered very valuable
technical assistance. Mr. H. S. Massey, Dade City, and Mr. A. H. Miller,
Brooker, provided photographs of their industries. Student assistants at the
University Poultry Laboratory, Verbon Pugh, Lester Kalch, Eugene Fortner,
John Scussel, James Burkhalter and Rollin McNutt, were enthusiastic and
faithful in caring for the chicks. Mrs. John Coleman recorded much of the
data contained in this manuscript. To all of these and still many others
who were so helpful, the authors are truly grateful.
Literature Cited
1. Agricultural statistics. USDA. Pp. 171-295; 434-474. 1948.
2. Becker, R. B., P. T. Dix Arnold and George K. Davis. Citrus by-prod-
ucts as feeds for cattle. Fla. Agr. Exp. Sta. Press Bull. 644. 1948.
3. Bernays, Fr. Repert. Fur Die Pharm., 71:306-15. Also, Zweiter Reihe,
Bd XXI. 1840.
4. Emerson, Oliver H. Isolation of limonin from Valencia orange seeds.
Jour. Am. Chem. Soc., 70:545-9. 1948.
5. Ewing, W. Ray. Poultry nutrition. W. Ray Ewing, Pasadena, Cali-
fornia. p. 390. 1947.
6 Feist, K., and L. Schulte. Citrolimonin. Ber. Der. Deut. Chem. Ges.,
69B:1322-3. C. A. 30:5586. 1936.
7. Geissman, T. A., and Vsevold Tulagin. Some observations on the
structure of limonin. J. Org. Chem., 11:760-70. 1946.
8. Higby, Ralph H. Bitter constituents of Navel and Valencia oranges,
Jour. Am. Chem. Soc., 60:3013-18. 1938.
9. Koller, G., and H. Czerny. Limonin. the bitter principle of the orange
kernel. Monatsh. Fur Chemie und Ver. Teile Anderer Wissenschaft,
67:248-68. C. A. 30:3831. 1936.
10. Koller, G., and H. Czerny. Limonin, the bitter principle of the orange
kernel. II. Monatsh. Fur Chemie und Ver. Teile Anderer Wissen-
schaft, 70:26-9. C. A. 31:4334. 1937.
11. Mehrhof, N. R., and L. L. Rusoff. Utilization of citrus meal for poultry
Proc. of the Seventh World's Poultry Congress and Exposition. Pp.
209-12. 1939.
12. Neal, W. M., R. B. Becker and P. T. Dix Arnold. The feed value and
nutritive properties of citrus by-products. Fla. Agr. Exp. Sta. Bull.
275. 1935.
13. Neal, W. M., R. B. Becker and P. T. Dix Arnold. Dried grapefruit
refuse-a valuable feed. Fla. Agr. Exp. Sta. Press Bull. 466. 1934.
14. Nolte, A. J., and H. W. von Loesecke. Grapefruit seed oil and its prop-
erties. Ind. and Eng. Chem., 32:1244-6. 1940.
15. Schechter, Milton S., and H. L. Haller. Identity of obaculactone,
evodin and dictamnolactone with limonin. Jour. Am. Chem. Soc.,
62:1307-9. 1940.
16. Schechter, Milton S., and H. L. Haller. The insecticidal principle in
the fruit of the Amur cork tree. Jour. Org. Chem., 8:194-7. C.A.
37:3530. 1943.
17. Seshadri, T. R., and J. Veeraraghaviah. Chemical investigation of In-
dian fruits. I. Bitter principles of pamparapanas (Indian shaddock).
Proc. Indian Acad. Sci., 11A:505-11. C. A. 34:8090. 1940.
18. Seshadri, T. R. Chemical investigation of Indian fruits. IV. Bitter
principle of a variety of Citrus limetta. Proc. Indian Acad. Sci.,
18A:201-3. C. A. 39:1438. 1943.
|