|
Bulletin 431
UNIVERSITY OF FLORIDA
AGRICULTURAL EXPERIMENT STATION
HAROLD MOWRY, Director
GAINESVILLE, FLORIDA
THE TOXIC PRINCIPLE OF
THE TUNG TREE
M. W. EMMEL
Fig. 1.-Both foliage and fruit of the tung tree
contain toxic principles.
TECHNICAL BULLETIN
Single copies free to Florida residents upon request to
AGRICULTURAL EXPERIMENT STATION
GAINESVILLE, FLORIDA
June, 1947
BOARD OF CONTROL
J. Thos. Gurney, Chairman, Orlando
N. B. Jordan, Quincy
Thos. W. Bryant, Lakeland
M. L. Mershon, Miami
J. Henson Markham, Jacksonville
J. T. Diamond, Secretary, Tallahassee
EXECUTIVE STAFF
John J. Tigert, M.A., LL.D., President of the
University3
H. Harold Hume, D.Sc., Provost for Agricul-
ture
Harold Mowry, M.S.A., Director
L. O. Gratz, Ph.D., Asst. Dir., Research
W. M. Fifield, M.S., Asst. Dir., Admin.
J. Francis Cooper, M.S.A., Editors
Clyde Beale, A.B.J., Associate Editors
Jefferson Thomas, Assistant Editors
Ida Keeling Cresap, Librarian
Ruby Newhall, Administrative Managers
K. H. Graham, LL.D., Business Managers
Claranelle Alderman, Accountants
MAIN STATION, GAINESVILLE
AGRONOMY
W. E. Stokes, M.S., Agronomist1
Fred H. Hull, Ph.D., Agronomist
G. E. Ritchey, M.S., Agronomist3
G. B. Killinger, Ph.D., Agronomist
W. A. Carver, Ph.D., Associate
H. C. Harris, Ph.D., Associate
Fred A. Clark, B.S., Assistant
ANIMAL INDUSTRY
A. L. Shealy, D.V.M, An. Industrialist'1
R. B. Becker, Ph.D., Dairy Husbandmans
E. L. Fouts, Ph.D., Dairy Technologists
D. A. Sanders, D.V.M., Veterinarian
M. W. Emmel, D.V.M., Veterinarians
L. E. Swanson, D.V.M., Parasitologist
N. R. Mehrhof, M.Agr., Poultry Husb.3
G. K. Davis, Ph.D., Animal Nutritionist
R. S. Glasscock, Ph.D., An. Husbandman
P. T. Dix Arnold, M.S.A., Asst. Dairy Husb.8
C. L. Comar, Ph.D., Asso. Biochemist
L. E. Mull, M.S., Asst. in Dairy Tech.
Katherine Boney, B.S., Asst. Chem.
J. C. Driggers, B.S.A., Asst. Poultry Husb.
Glenn Van Ness, D.V.M., Asso. Poultry
Pathologist
John S. Folks, B.S.A., Asst. An. Husb.
ECONOMICS, AGRICULTURAL
C. V. Noble, Ph.D., Agri. Economist1 '
Zach Savage, M.S.A., Associates
A. H. Spurlock, M.S.A., Associate
I. E. Allegar, .M.S., Associate
D. L. Brooke, M.S.A., Associate
Orlando, Florida (Cooperative USDA)
G. Norman Rose, B.S., Asso. Agr. Economist
J. C. Townsend, Jr., B.S.A., Agr. Statistician'
J. B. Owens, B.S.A., Agr. Statisticians
ECONOMICS, HOME
Ouida D. Abbott, Ph.D., Home Econ.'
R. B. French, Ph.D., Biochemist
ENTOMOLOGY
A. N. Tissot, Ph.D., Entomologist and Act-
ing Head of Dept.
H. E. Bratley, M.S.A., Assistant
HORTICULTURE
G. H. Blackmon, M.S.A., Horticulturist'
A. L. Stahl, Ph.D., Asso. Horticulturist
F. S. Jamison, Ph.D., Truck Hort.
Byron E. Janes, Ph.D., Asso. Hort.
R. A. Dennison, Ph.D., Asso. Hort.
R. K. Showalter, M.S., Asso. Hort.
R. J. Wilmot, M.S.A., Asst. Hort.
R. D. Dickey, M.S.A., Asst. Hort.
Victor F. Nettles, M.S.A., Asst. Hort.
F. S. Lagasse, Ph.D., Asso. Hort.2
PLANT PATHOLOGY
W. B. Tisdale, Ph.D., Plant Pathologist1
Phares Decker, Ph.D., Asso. Plant Path.
Erdman West, M.S., Mycologist and Botanist
Lillian E. Arnold, M.S., Asst. Botanist
SOILS
F. B. Smith, Ph.D., Chemist'1
Gaylord M. Volk, Ph.D., Chemist
J. R. Henderson, M.S.A., Soil Technologist
J. R. Neller, Ph.D., Soils Chemist
Nathan Gammon, Jr., Ph.D., Soils Chemist
C. E. Bell, Ph.D., Associate Chemist
L. H. Rogers, Ph.D., Associate Biochemist
R. A. Carrigan, B.S., Asso. Biochemist
H. W. Winsor, B.S.A., Assistant Chemist
Geo. D. Thornton, M.S., Asst. Microbiologists
R. E. Caldwell, M.S.A., Asst. Soil Surveyor
Wade McCall, B.S., Asst. Chemist
J. B. Cromartie, B.S.A., Asst. Soil Surveyor
1 Head of Department.
2 In cooperation with U. S. D. A.
3 Cooperative, other divisions, U. of F.
SIn Military Service.
6 On leave.
BRANCH STATIONS
NORTH FLORIDA STATION, QUINCY
J. D. Warner, M.S., Vice-Director in Charge
R. R. Kincaid, Ph.D., Plant Pathologist
W. H. Chapman, M.S., Asso. Agron.
R. C. Bond, M.S.A., Asso. Agronomist
L. G. Thompson, Ph.D., Soils Chemist
Frank D. Baker, Jr., B.S., Asst. An. Hush.
Mobile Unit, Monticello
R. W. Wallace, B.S., Associate Agronomist
Mobile Unit, Marianna
R. W. Lipscomb, M.S., Associate Agronomist
Mobile Unit, Wewahitchka
J. B. White, B.S.A., Associate Agronomist
CITRUS STATION, LAKE ALFRED
A. F. Camp, Ph.D., Vice-Director in Charge
V. C. Jamison, Ph.D., Soils Chemist
W. L. Thompson, B.S., Entomologist
J. T. Griffiths, Ph.D., Entomologist
R. F. Suit, Ph.D., Plant Pathologist
E. P. Ducharme, M.S., Plant Pathologist5
J. E. Benedict, B.S., Horticulturist
B. R. Fudge, Ph.D., Associate Chemist
C. R. Stearns, Jr., B.S.A., Asso. Chemist
James K. Colehour, M.S., Research Chemist
T. W. Young, Ph.D., Asso. Horticulturist
J. W. Sites, M.S.A., Asso. Horticulturist
H. 0. Sterling, B.S., Asst. Horticulturist
J. A. Grange'- B.S.A., Asst. Horticulturist
H. J. Reitz, M.S., Asso. Plant Path.
Francine Fisher, M.S., Asso. P1. Path.
EVERGLADES STA., BELLE GLADE
R. V. Allison, Ph.D., Vice-Director in Charge
J. W. Wilson, Sc.D., Entomologist
F. D. Stevens, B.S., Sugarcane Agron.
Thomas Bregger, Ph.D., Sugarcane
Physiologist
B. S. Clayton, B.S.C.E., Drainage Eng.2
W. D. Wylie, Ph.D., Entomologist
W. T. Forsee, Jr., Ph.D., Asso. Chemist
R. W. Kidder, M.S., Asst. An. Husb.
T. C. Erwin, Assistant Chemist
R. A. Bair, Ph.D., Asst. Agronomist
C. C. Scale, Asst. Agronomist
L. O. Payne, B.S.A., Asst. Agronomist
Russel Desrosie's, M.S., Asst. Plant Path.
N. C. Hayslip, B.S.A., Asst. Hort.
SUB-TROPICAL STA., HOMESTEAD
Geo. D. Ruehle, Ph.D., Vice-Director in
Charge
H. I. Borders, M.S., Asso. Plant Path.5
D. 0O Wolfenbarger, Ph.D., Asso. Ento.
R. W. Harkness, Ph.D., Asst. Chemist
W. CENT. FLA. STA., BROOKSVILLE
Clement D. Gorden, Ph.D., Poultry Geneticist
in Charge2
RANGE CATTLE STATION, ONA
W. G. Kirk, Ph.D.. Vice-Director in Charge
E. M. Hodges, Ph.D., Associate Agronomist
D. W. Jones, B.S.A., Asst. An. Hush.
E. R. Felton, B.S.A., Asst. An. Hush.
CENTRAL FLORIDA STATION, SANFORD
R. W. Ruprecht, Ph.D., Chemist in Charge
A. Alfred Foster, Ph.D., Asso. Hort.
J. C. Russell, M.S., Asst. Entomologist
Ben F. Whitner, Jr., B.S., Asst. Hort.
WEST FLORIDA STATION, MILTON
H. W. Lundy, B.S.A., Asst. Agronomist
FIELD STATIONS
Leesburg
G. K. Parris, Ph.D., Plant Path. in Charge
Plant City
A. N. Brooks, Ph.D., Plant Pathologist
Hastings
A. H. Eddins, Ph.D., Plant Pathologist
E. N. McCubbin, Ph.D., Truck Horticulturist
Monticello
S. O. Hill, B.S., Asst. Entomologist' 4
A. M. Phillips, B.S., Asst. Entomologist'
Bradenton
J. R. Beckenbach, Ph.D., Horticulturist in
Charge
E. G. Kelsheimer, Ph.D., Entomologist
David G. Kelbert, Asso. Horticulturist
E. L. Spencer, Ph.D., Soils Chemist
Robert O. Magie, Ph.D., Hort., Glad. Inv.
Donald S. Burgis, M.S.A., Asst. Hort.
Lakeland
Warren O. Johnson, Meteorolbgist'
1 Head of Department.
SIn cooperation with U. S.
3 Cooperative, other divisions, U. of F.
SIn Military Service.
SOn leave.
CONTENTS
PAGE
REVIEW OF LITERATURE.. -------..................... ------- -................... 5
EXPERIMENTAL DATA-----------.......... ---- .......................---- 7
Plan of Procedure-----....------............--...-.------- 7
Isolation of a Toxic Principle. .... ----------..................--- 8
Tests Applied to the Toxic Principle.............................. .... ........ 8
Relative Toxicity of Foliage of Tung Species.....----.............-- ... ........ 10
Saponin Content of Commercial Tung Meal versus Tung Kernels.... 11
Removal or Destruction of Saponin Content ...................--- .....--...... 13
SRelative Toxicity of Hydrolyzed Aged and Freshly-Milled
Commercial Tung Meal.. ------------..................--............... 16
Solvent-Extracted Tung Meal Feeding Experiments. ..........-................ 17
Detoxification of Commercial Tung Meal and Tung Kernel Flakes.... 19
Isolation of the Second Toxic Substance from Commercial Tung
Meal and Tung Kernel Flakes................................. ................ 24
Growth Curves and Lesions of Chicks Fed Commercial Tung Meal.... 27
DISCUSSION .--...---- -.......-. .........---.---- .... 30
SUMMARY ......----............--------------------- --- 33
LITERATURE CITED ....------............. ------- ----- ------------------ 34
THE TOXIC PRINCIPLE OF THE
TUNG TREE
M. W. EMMEL
The tung tree, (Aleurites spp.), a native of China, was in-
troduced into the United States in 1905 by the office of Foreign
Plant Introduction of the United States Department of Agricul-
ture (2)1 in an attempt to establish a domestic supply of tung
oil which is expressed from the fruit of the trees. The acreage
devoted to the tung industry in this country has increased grad-
ually until in 1946 commercial plantings consisted of 200,000
acres located chiefly in the Gulf Coast States. In the United
States in 1944, 60,800,000 pounds of fruit were harvested, from
which approximately 9,728,000 pounds of oil were extracted.
The war gave a great impetus to the tung industry in this
country. Imports from China, the primary pre-war source of
tung oil, were reduced drastically with the result that the in-
dustry in the United States has become established on a more
firm and scientific basis. The oil expressed from the seeds has
exceptional drying qualities and is used chiefly in paints, varn-
ishes and waterproofing materials. It is superior to other oils
for these purposes.
The dry tung seed contains 20 to 30 percent of oil. It is pro-
cessed commercially by hulling, grinding the seeds into a meal
and extracting the oil by pressure. The residue after extraction
of the oil, "pressed cake," contains about 6 percent of ammonia
(8) and has been used in some instances as a fertilizer or fer-
tilizer filler; it is comparable in this respect to castor pomace.
Investigations by numerous workers have indicated that com-
mercial pressed cake contains a toxic substance or substances,
heretofore undetermined, which render it unsuitable as a feed
for livestock.
It was the purpose of this investigation to determine the na-
ture of the substances responsible for the toxicity of commercial
pressed cake and possible methods of detoxification. The term
commercial tung meal will be used hereinafter for ground com-
mercial pressed cake.
REVIEW OF LITERATURE
A number of references are found in the literature pertaining
to the toxicity of commercial tung meal [Lewkowitsch (11),
1 Italic figures in parentheses refer to Literature Cited.
Florida Agricultural Experiment Station
Newell (12) and Jamieson (10)]. Godden (9) fed commercial
tung meal to rats, poultry, cattle and pigs and found it to be un-
palatable; also, it contained some irritating substances which
had a harmful effect upon the mucous membrane of the intes-
tinal tract of animals to which it was fed. Sanders, Emmel and
Henley2 observed that commercial tung meal fed as 20 percent
of the ration to young pigs was unpalatable and induced severe
inflammation of the mucous membrane of the gastro-intestinal
tract. Thus, it is generally agreed among workers who have
fed commercial tung meal to animals that it is unpalatable and
toxic.
Erickson and Brown (7) found that oil-free commercial tung
meal heated to 230" F. for 2 hours was non-toxic for rats.
Commercial tung meal contains approximately 5 percent of
residual oil. It has been thought by some that oil-free tung
meal perhaps would be non-toxic. Feeding experiments in which
solvent-extracted tung meal was used have been reported.
Jones3 found that solvent-extracted tung meal was unpalatable
to rats but observed no symptoms of toxicity. Rusoff, Mehrhof
and McKinney (13) have reported that solvent-extracted tung
meal constituting 5 to 20 percent of the ration of young chicks
was unpalatable but did not induce noticeable symptoms of tox-
icity. Both of these investigations were concerned with com-
mercial tung meal which was extracted with an oil solvent to
remove the residual oil. It should be stated that the meal used
by Rusoff et al was heated after solvent extraction for 21/2
hours under steam pressure gradually increased to 90 pounds.
Emmel, Sanders and Swanson (4, 14) have reported that the
foliage of the tung tree which had been discarded in a 20-acre
pasture was responsible for the death of 14 of a herd of 44 pure-
bred cattle. Experimental feeding indicated that 1:1/4 pounds of
the fresh green leaves was sufficient to kill a 300-pound steer. A
profuse watery diarrhea which was sometimes hemorrhagic,
stasis of the rumen, inappetence and depression were the most
prominent symptoms observed. An intense inflammation of the
mucous membrane of the intestinal tract occurred in the ani-
mals autopsied.
* D. A. Sanders, M. W. Emmel and W. W. Henley-Unpublished data
Fla. Agr. Expt. Sta., 1933.
a D. B. Jones, U. S. D. A., Bur. Agr., Chem. and Eng.-Unpublished
data (cited by Rusoff, et al.)
The Toxic Principle of the Tung Tree
More recently Emmel (5, 6) has reported that the toxic prin-
ciple of tung foliage (Aleurites fordi Hemsl.) is a saponin
and that the viscous tenacious sap of the tree is rich in this
substance.
EXPERIMENTAL DATA
PLAN OF PROCEDURE
Since the author (5, 6) found the toxic principle in the foliage
of A. fordi was a saponin, it seemed desirable to conduct ex-
periments to determine if commercial tung meal and the tung
kernel contained the same toxic substance. It also seemed ad-
visable to determine the toxicity of the foliage of other species
of tung.
As these studies progressed and it was determined that
saponin was contained in commercial tung meal, experiments
were conducted to compare the saponin content of commercial
tung meal with that of tung kernels. Experiments also were
conducted in which attempts were made to destroy the saponin
content of commercial tung meal, the resultant meal constituting
20 percent of the basal ration and being fed to chicks 4 to 18
days of age for a period of 2 weeks.
These later experiments led to a comparative study of the tox-
icity of aged and freshly-milled commercial tung meal in which
the saponin had been destroyed. As the claim has been made
that solvent-extracted tung meal is not toxic, feeding tests were
conducted on young chicks with meal extracted with petroleum
ether and commercial acetone, naphtha and hexane.
During the progress of these experiments it became evident
that tung meal in which the saponin was destroyed was still
toxic, as indicated by chick feeding tests. Attempts were made
to detoxify tung meal completely by destruction of the saponin
content and extraction of the remaining toxic substance. The
extraction of the latter was verified by both baby chick feeding
tests on the extracted meal and similar tests on the extract
itself.
A method was perfected for isolating the toxic substance in
tung meal after the destruction of saponin and purifying it to
the extent that approximately 50 percent of the 4 to 6-day old
chicks to which it was fed in an amount equivalent to 20 percent
of tung meal in the ration died during the 2-week feeding period.
Growth curves were determined for young chicks fed oil-free,
hydrolyzed (to destroy saponin) and detoxified tung meal as well
Florida Agricultural Experiment Station
as the isolated second toxic substance for a period of 5 weeks.
The tung, kernel flakes and commercial tung meal used in these
experiments were from Aleurites fordi Hemsl.
ISOLATION OF A TOXIC PRINCIPLE
Fresh Green Leaves.-The fresh green leaves of A. fordi were
ground in a meat grinder, placed in a large flask, covered with
distilled water and allowed to stand overnight. The liquid then
was strained through 6 thicknesses of cheesecloth, saturated
with ammonium sulfate and filtered through filter paper. The
filter paper was washed with a small amount of distilled water.
After the liquid was dialyzed overnight in parchment paper in
running water, it was concentrated to 15 to 20 ml. in an evapo-
rating dish over a water bath. The liquid was divided into 2
parts and poured into 100 ml. centrifuge tubes; sufficient abso-
lute alcohol was added to fill the tubes. After centrifuging, the
precipitate was collected from the bottom of the tubes, washed
with ether and allowed to dry.
The fresh green leaves of Aleurites montana (Lour.) Wils,
A. moluccana Willd. (A. triloba Forst.), and A. trisperma
Blanco were treated in the same manner.
Commercial Tung Meal.-Two samples of tung meal approxi-
mately 1 year old and 1 sample 2 years old were used. These
samples had been stored under good conditions. The tung meal
was soaked overnight in distilled water. The fluid was strained
through 8 thicknesses of cheesecloth and saturated with am-
monium sulfate. The subsequent procedure was the same as
that used with fresh green leaves.
Solvent-Extracted Tung Kernel Flakes.-Fresh tung kernels
were sliced with a plane and extracted with acetone in a Soxhlet
extractor until free of oil. The toxic principle then was extracted
from the flakes by the same procedure used in the case of com-
mercial tung meal.
TESTS APPLIED TO THE TOXIC PRINCIPLE
Preliminary tests indicated that the toxic principle of the tung
tree is a saponin. The following tests were applied to the sub-
stance isolated under the above procedures (1).
Froehde's Test.-Five milligrams of sodium molybdate are
dissolved in 1 ml. of concentrated sulfuric acid by gently heating
in a test tube. After cooling, 1 ml. of the solution to be tested
The Toxic Principle of the Tung Tree
is added slowly so the separation of the 2 liquids is distinct. The
formation of a blue violet or green ring at the juncture of the 2
liquids is considered a positive test for saponin.
Sulfuric Acid Test.-The procedure is similar to that used in
Froehde's test except that concentrated sulfuric acid is used as
the testing agent. The formation of a ring yellowish in color
changing to red, purplish red or reddish violet is considered a
positive test for saponin.
Hydrolysis.-By boiling in a 5 percent solution of hydrochloric
acid, a saponin can be hydrolyzed into 2 molecules, 1 of which
is a sugar. The precipitate is dissolved in water and divided into
2 portions. One portion is tested for the presence of sugar with
Fehling's solution. An equal amount of 5 percent hydrochloric
acid is added to the second portion which then is boiled until
major foaming ceases. The hydrolysate is tested for sugar with
Fehling's solution. A negative sugar test for the first portion'
and a positive test for the second portion indicates that sugar
formed a portion of the original molecule.
Hemolysis Test.-One ml. of defibrinated sheep's blood is
added to 99 ml. of physiological salt solution. The dried sub-
stance to be tested is dissolved in a small quantity of physiologi-
cal salt solution; One ml. of this latter solution is mixed with
an equal volume of diluted sheep's blood. Laking of the blood
is considered a positive test for saponin.
Cholesteride Test.-Saponins unite with cholesterol to form
cholesterides. The dried substance to be tested is dissolved in
a small quantity of physiological salt solution. It is added to
an equal volume of a solution of cholesterol in ether, incubated
in a water bath at 36" C. for several hours and the hemolysis
test applied as described above. A negative hemolysis test indi-
cates that the saponin has united with cholesterol to form a
cholesteride which does not hemolyze blood.
Results.-In all instances extracts from the fresh green leaves,
commercial tung meal and oil-free tung kernel flakes gave a
positive test for saponin. These positive tests, together with
the fact that the substance foamed markedly when agitated with
water, was non-dialyzable or at least incompletely so, was water-
soluble and was precipitated by saturation with ammonium sul-
fate indicated that the leaves and meal contained saponin. In
some instances the tests conducted on preparations from com-
mercial tung meal were not as clear as on those from the green
Florida Agricultural Experiment Station
leaves or tung kernel flakes, due to the brown color which per-
sisted in the former preparations.
It also is characteristic of a poisonous saponin to induce gas-
tro-enteritis in animals to which it is fed. Most available reports
of feeding experiments with green tung leaves and unaltered
commercial tung meal in which autopsies have been performed
on the animals fed have indicated that gastro-enteritis was the
predominant lesion. Super-purgation as a result of gastro-enteri-
tis is a common symptom of tung poisoning in some species of
animals, particularly cattle. Two mature rabbits fed 100 mg.
of the extract prepared from oil-free tung kernels developed an
extremely profuse watery diarrhea within 24 hours and con-
tinued for 4 days, during which time symptoms of abdominal dis-
tress were observed.
RELATIVE TOXICITY OF FOLIAGE OF TUNG SPECIES
The species A. fordi predominates in commercial plantings
throughout the South. However, 3 other species, A. montana,
A. moluccana and A. trisperma, are found in experimental
plots in this area. The following experiments were conducted
to determine the relative toxicity of these species.
Ground fresh leaves of each of the 4 species were given in
capsule in addition to other feed to a group of 4 Barred Plym-
outh Rock hens of similar age and breeding in dosages of 300
grams administered to each bird at the rate of 100 grams on 3
consecutive days. Preliminary experiments had indicated that
the minimum lethal dose of A. fordi foliage force-fed in this
manner is approximately 250 to 300 grams. The tissues of all
birds were studied microscopically and this formed the basis for
judging the relative toxicity of the species.
All of the birds fed foliage of A. ford died within 2 days
after completion of exposure. The birds appeared indisposed,
listless and depressed. Autopsy revealed the following lesions:
necrotic proventriculitis, necrotic enteritis, muddy discolora-
tion of the body fat, congestion and very small irregular foci of
necrosis on the surface of the liver, pale and slightly enlarged
kidneys, and flaccid heart and gizzard muscle; 2 birds showed
the presence of a fibrinous exudate on the serosa of the duode-
num. When the tissues were examined microscopically the kid-
neys, intestines and liver showed the most pronounced lesions.
Extensive necrosis and cloudy swelling occurred in the tubular
epithelium of the kidneys. Many tubules were completely de-
The Toxic Principle of the Tung Tree
nuded of epithelium. Necrosis of the tips of the villi, cloudy
swelling of the epithelium, occasional edema of the musculature
and foci of proliferation of histiocytes at the base of the villi
with infiltration of histiocytes into adjacent tissues occurred in
the duodenum. Hydropic degeneration occurred in the hepatic
cells; numerous foci of the necrosis and occasional congestion
and small foci of necrosis, chiefly perivascular, were observed in
the lungs. The spleen showed marked congestion. Necrosis of
the surface epithelium and numerous foci of infiltration of histio-
cytes in the submucosa occurred in the proventriculus.
Birds which had been fed the foliage of the other 3 species
did not develop symptoms of illness. They were killed for patho-
logical study 4 days after the completion of exposure. Two birds
fed A. montana showed marked congestion of the liver and in 1
of these congestion of the lungs, discoloration of the body fat
and a fibrinous exudate on the serosa of the duodenum occurred.
The microscopic lesions were essentially the same as occurred in
A. fordi poisoning except they were much less intense. Although
the hepatic cells showed well advanced hydropic degeneration,
foci of necrosis were infrequent in the liver and almost totally
absent in the lungs. While some necrosis occurred in the tubu-
lar epithelium of the kidneys, cloudy swelling predominated.
The birds fed foliage of A. moluccana and A. trisperma did
not show gross lesions on postmortem examination. The micro-
scopic lesions, although similar in character to those observed in
the instance of A. montana, were less intense. These lesions
were less intense in the birds fed A. trisperma than in those
fed A. moluccana.
SAPONIN CONTENT OF COMMERCIAL TUNG MEAL
VERSUS TUNG KERNELS
The observation was made in the preceding experiments that
extracts of oil-free tung kernels appeared more abundant in
saponin content than extracts prepared from commercial tung
meal. This immediately raised the question as to whether or not
the commercial method for the extraction of oil from tung nuts
reduced the original saponin content.
In the present commercial method of tung oil extraction the
nuts are passed through a machine which removes the hull and
a portion of the hard shell immediately surrounding each kernel.
After grinding, the meal is heated to approximately 180* F. for
30 minutes and then subjected to high pressure in a continuous
Florida Agricultural Experiment Station
process which removes all but about 5 percent of the oil. The oil
which is expressed from the meal in this process is very cloudy
and is passed through a filter machine while heated to clarify it.
The filter residue which contains about 40 percent of oil is
passed again through the extraction process. All of the filter
residue eventually passes into the commercial tung meal, as
there is no accumulation of this residue in the oil mill.
Two volumes of ether were added to 650 ml. of a fresh sample
of unfiltered oil. After being agitated it was filtered through
filter paper. The residue remaining on the paper was washed 3
times with fresh ether and air dried. One hundred grams of
residue, an almost impalpable powder, were recovered. An extract
was prepared from this residue by the method previously de-
scribed. Tests on this extract indicated the residue was rich in
saponin.
According to McKinley,4 approximately 26 gms. of filter resi-
due is derived from the processing of 220 gms. of commercial
tung meal, which in turn contains approximately 99 gms. of
kernel residue; the remainder consists of shell and moisture.
These amounts of fresh filter residue, commercial tung meal
and ground kernel were saved during the milling of the same
lot of nuts and were weighed after the materials had been ex-
tracted with acetone in a Soxhlet extractor to render them oil-
free. They were allowed to stand overnight in sufficient dis-
tilled water to cover the mass.
Each was filtered through 8 thicknesses of gauze, allowed to
stand for several hours, decanted and evaporated over a water
bath until about 50 ml. remained. Two volumes of absolute alco-
hol were added to each sample. After centrifuging, the precipi-
tate was collected from the bottom of the tubes, washed with
ether and allowed to dry.
As previously stated, a saponin can be hydrolyzed into 2 mole-
cules of which 1 is a sugar. Each of the 3 samples was taken
up in an equal volume of water. Each gave a negative test for
sugar when tested with Benedict's solution. A measured portion
of each of the 3 samples was hydrolyzed by boiling with an equal
volume of a 5 percent aqueous solution of hydrochloric acid for
30 minutes. A quantitative test for sugar was made with Bene-
dict's solution. The extract from the 26 gms. of filter residue
contained 0.118 gms. of sugar, that from the 99 gms. of oil-
free tung kernels contained 0.0982 gms. of sugar, while that from
4 Personal communication.
The Toxic Principle of the Tung Tree
the 220 gms. of commercial tung meal contained 0.0148 gms. of
sugar. The test was repeated on a second group of samples se-
cured at another time. The.extract from 26 gms. of filter residue
contained 0.121 gms. of sugar, that from the 99 gms. of oil-free
tung kernels contained 0.0871 gms. of sugar, while that from
the 220 gms. of commercial tung meal contained 0.0114 gms. of
sugar.
Assuming that saponin extraction was accomplished equally
well in all samples, these quantitative tests indicate that the
commercial tung meal contained approximately 1/6 as much sa-
ponin as the tung kernels. As the sugar content of the filter
residue extract was higher than that from the oil-free tung
kernels, it appears that some accumulation of saponin content
occurs in the residue.
As this residue is remilled and does not accumulate in the oil
mill, it must pass into the commercial tung meal. The low sapon-
in content of this meal indicates that a portion of the original
saponin content of the tung kernel is destroyed in the milling
process.
REMOVAL OR DESTRUCTION OF SAPONIN CONTENT
There are 3 properties of a saponin which might prove of value
in the 'development of a detoxification process for commercial
tung meal; (1) water solubility, (2) destruction by heat and
(3) hydrolysis by acids. The, following feeding experiments were
conducted with young chicks to evaluate these properties in de-
toxification processes.
The basal ration was as follows:
Yellow corn meal 50.0 pounds
Bran 20.0 pounds
Ground wheat 14.0 pounds
Ground oats 10.0 pounds
Alfalfa leaf meal 2.5 pounds
Meat scrap 20.0 pounds
Ground shell 2.5 pounds
Salt 1.0 pounds
Dried whey 5.0 pounds
Commercial tung meal treated in various ways was added
to the basal ration at the rate of 20 percent by weight. Fresh
batches were prepared for each experiment. The chicks were
fed for a period of 2 weeks, weights being recorded at the start
of each experiment and every week until conclusion. A sum-
mary of the results of these experiments is given in Table 1.
TABLE 1.-SUMMARY OF FEEDING EXPERIMENTS IN WHICH ATTEMPTS WERE MADE TO REMOVE OR
DESTROY THE SAPONIN CONTENT OF TUNG MEAL.
Average
Number Age Average Weight Average Gain Feed
Exp. of Start- Commercial Tung Meal Added1 I Con- Mortality
Chicks ed (20 Percent) First Second sump-
Initial Final Week Week Total tion
Days grams grams I grams grams grams grams
19 no tung meal 49.15 124.94 38.37 37.42 75.79
I 19 5 raw meal 50.47 105.55 29.89 25.19 55.08 ....... 1
20 heated 4 hrs. 20 lbs. pressure. 47.25 124.10 36.10 40.75 76.85
19 water-extracted 47.35 114.52 34.70 32.47 67.17 ...... ....
15 no tung meal 44.80 126.00 30.40 50.80 81.20 208.84
II 15 4 raw meal 45.80 102.75 16.20 40.75 56.95 199.76 3
15 heated 4 hrs. 20 lbs. pressure. 42.33 99.06 21.33 35.40 56.73 186.14
15 water-extracted 43.60 106.07 19.00 43.60 62.60 186.14
18 no tung meal 51.00 127.22 29.50 46.72 76.22 233.05
III 15 5 raw meal 50.13 111.08 20.79 40.16 60.95 210.00 3
16 heated 4 hrs. 20 lbs. pressure. 53.81 114.00 28.00 32.19 60.19 245.31
16 hydrolyzedz 60 min. 15 lbs. pres. 53.31 136.31 35.62 47.38 83.00 266.25 1
11 no tung meal 102.27 208.63 44.54 61.82 106.36 464.81 ..
IV 11 18 hydrolyzed2 15 min. 15 lbs. pres. 106.25 171.75 27.75 37.75 65.50 472.08 ....
11 hydrolyzeds 30 min. 15 lbs. pres. 104.90 182.77 31.73 46.14 77.87 489.20 ....
11 hydrolyzed2 60 min. 15 lbs. pres. 104.60 194.20 38.20 51.40 89.60 494.00
18 no tung meal 53.44 123.44 29.89 40.11 70.00 279.72 .
V 19 6 hydrolyzed& 15 min. 15 lbs. pres. 54.36 107.58 22.64 30.58 53.22 260.22
19 hydrolyzeds 30 min. 15 lbs. pres. 56.47 114.00 25.26 32.27 57.53 270.57
19 hydrolyzed2 60 min. 15 lbs. pres. 56.19 114.36 24.02 34.15 58.17 261.10
1 Tung meal at least 1 year old and stored under good conditions.
2 Hydrolyzed meal was dampened with 600 ml. 5 percent aqueous solution of hydrochloric acid per liter of meal, allowed to stand 30
minutes and heated in autoclave at the interval and pressure specified.
The Toxic Principle of the Tung Tree
Commercial Tung Meal.-Forty-nine chicks were fed commer-
cial tung meal in 3 experiments. They gained averages of 55.08,
56.95 and 60.95 gms., as compared with gains of 75.79, 81.20,
and 76.22 gms. in the control groups. Thus the gains of chicks
fed tung meal were 20.71, 24.25 and 15.27 gms. less than the
respective control groups.
Water-Extracted Tung Meal.-Commercial tung meal was
placed in a sack made of 6 thicknesses of cheesecloth and
suspended in running water for 24 hours. The meal then was
drained, spread in a thin layer and air dried. Extracts of the
water-extracted meal failed to give a positive test for saponin
by the technique previously described.
A total of 34 chicks were fed water-extracted tung meal in
2 experiments. They gained an average of 67.17 and 62.60 gms.
as compared with gains of 75.79 and 81.20 gms. in the control
groups, their gains being 8.62 and 18.60 gms. less than the re-
spective control groups.
Water extraction probably lowered the nutritional value of the
meal as all water soluble constituents were removed. Also, there
was loss of nutritional value, as many particles of finely ground
meal passed through the cheesecloth.
Heated Tung Meal.-A preliminary experiment was conducted
in which commercial tung meal was heated in an autoclave un-
der 15 pounds pressure for 2 hours. The meal then was spread
in a thin layer and air dried. Meal heated in this manner be-
came darker in color, which tended to obscure the test of ex-
tracts for saponin when made as previously described. A feeding
trial involving 16 chicks resulted in an average gain of 28.26
gms. less than the control group.
To determine the effects of additional heat 3 groups of chicks
were fed tung meal heated for 4 hours at 20 pounds pressure.
In Experiment 1, 20 chicks during the 2-week feeding period
made an average gain of 76.85 gms. compared with 75.79 gms.
by the control group. In 2 other experiments, however, the
chicks fed the heated meal gained an average of 56.73 and 60.19
gms., compared with 81.20 and 76.22 gms. by the control group,
an average gain of 24.47 and 16.03 gms. less than the respective
control groups. Comparisons between these latter chicks and
those receiving unheated tung meal indicated that the heated
meal still retained considerable of its original toxicity.
Florida Agricultural Experiment Station
Hydrolyzed Tung Meal.-Commercial tung meal was dampened
with 600 ml.. of 5 percent aqueous solution of hydrochloric acid
per liter of meal, allowed to stand for 30 minutes and heated in
an autoclave for 60 minutes at 15 pounds pressure. This process
destroys saponin by splitting off the sugar molecule. In Experi-
ment III tung meal treated in this manner was fed to 16 chicks.
They gained an average of 83.00 gms. during the 2-week feed-
ing period, compared with 76.22 gms. in the control group. This
represented an average gain of 6.78 gms. mdre than the controls.
As a result, Experiments IV and.V were conducted with hydro-
lyzed meal which had been heated at 15 pounds pressure for
varying periods-15, 30 and 60 minutes-for purposes of com-
parison. Even though extracts from these prepared meals gave
negative saponin tests, none of the chicks in either experiment
made gains comparable to the control groups. The average gains
of 11 chicks fed hydrolyzed meal heated for 15, 30 and 60 min-
utes were 65.50, 77.87 and 89.60 gms., respectively, compared
with 106.36 gms. in the control group. In another experiment 19
chicks fed hydrolyzed meal heated for 15, 30 and 60 minutes
gained 53.22, 57.53, and 58.17 gms., respectively, compared with
70.00 gms. for the controls. In both experiments better gains
were made as the heating period was increased. Groups fed meal
heated the longest period (60 minutes) made average gains 16.76
and 11.83 gms. less than their respective control groups.
These 2 experiments led to the assumption that commercial
tung meal in addition to containing a saponin contained another
toxic substance which apparently was partially destroyed as the
heating period was progressively increased.
RELATIVE TOXICITY OF HYDROLYZED AGED AND
FRESHLY-MILLED COMMERCIAL TUNG MEAL
During some of the preliminary experiments samples of hy-
drolyzed freshly milled tung meal were used in a few feeding
tests in which aged tung meal (at least 12 months old) similarly
treated also was used. In all of these tests the fresh tung meal
appeared to be more toxic than the old meal. As the preceding
experiments indicated that commercial tung meal contained a
toxic substance after the destruction of its saponin content, ex-
periments were conducted to determine the relative toxicity of
this substance in hydrolyzed aged and freshly milled meal.
STung meal which had been hydrolyzed (30 minutes at 15
pounds pressure as previously described) to destroy its saponin
The Toxic Principle of the Tung Tree 17
content was fed to young chicks for a period of 2 weeks, the
meal constituting 20 percent of the ration. Samples 1 to 3 were
at least 12 months old; samples 4 to 6 were freshly milled in
Experiments I to III and 5 months old in Experiments IV and V.
A summary of these feeding experiments is given in Table 2.
The chicks fed the hydrolyzed freshly milled tung meals gained
an average of 37.14, 36.22 and 42.24 gms., compared with 49.85,
46.07 and 50.25 gms., respectively, in the groups fed hydrolyzed
aged meal. The latter groups gained 12.71, 9.85 and 8.01 gms.
more than the groups fed hydrolyzed freshly milled meal.
The chicks fed hydrolyzed freshly milled tung meal which had
been stored for 5 months made average gains of 49.86 and 49.65
gms., compared with 50.73 and 50.23 gms., respectively, in the
groups fed hydrolyzed aged meal. The latter group gained only
0:87 and 0.68 gms. more than the groups fed 5-month old meal.
These experiments clearly indicate that hydrolyzed freshly
milled tung meal is more toxic than 12-month old tung meal
similarly treated. As the meal used in these experiments was
hydrolyzed to destroy the saponin content, one would assume
that the freshly milled product contained toxic substances other
than saponin in greater amount than the 12-month old meal.
Experiments IV and V indicate that much of this toxicity of hy-
drolyzed freshly milled meal was lost during 5 months of storage
under good conditions.
SOLVENT-EXTRACTED TUNG MEAL FEEDING EXPERIMENTS
The tung oil industry applies the general term "solvent-ex-
tracted" to meals from which the oil has been extracted by a
solvent. As Jones5 and Rusoff et al (13) observed no indica-
tions of toxicity in solvent-extracted meal, although it did ap-
pear to be unpalatable, 5 feeding experiments as previously de-
scribed were conducted in which the oil was extracted from com-
mercial tung meal in a Soxhlet extractor with petroleum ether,
and commercial acetone, naptha and hexane. The same lot of tung
meal was used in all instances. Four groups of chicks were in-
cluded in each experiment and were fed the following: (a) basal
ration with no tung meal added; (b) basal ration containing 20
percent solvent-extracted tung meal air-dried for 1 day; (c) the
same as (b) but air-dried for 7 days; (d) basal ration containing
20 percent solvent-extracted tung meal air-dried for 1 day, fol-
5 D. B. Jones, U. S. D. A. Bur. Agr. Chem. and Eng. Unpublished data.
(Cited by Rusoff et al.)
TABLE 2.-SUMMARY OF FEEDING EXPERIMENTS INDICATING THE RELATIVE TOXICITY OF AGED AND FRESH
HYDROLYZED COMMERCIAL TUNG MEAL.
I | Average
Number Age Commercial Tung Meal Added' I Average Weight Average Gain Feed
Exp. of Start- First Second Con-
_____ Chicks ed (20 Percent) Initial Final Week Week Total sumption
Days grams grams grams grams grams grams
12 no tung meal 53.83 118.00 26.17 38.00 64.17 281.25
I 13 5 sample No. 3 49.76 99.61 21.08 28.77 49.85 215.30
13 sample No. 4 49.21 86.35 12.86 24.28 37.14 196.64
14 no tung meal 51.50 107.07 21.92 33.65 55.57 222.42
II 13 6 sample No. 1 53.46 99.53 16.30 29.77 46.07 238.46
__13 sample No. 5 54.61 90.83 13.40 22.82 36.22 173.15
16 no tung meal 56.77 117.44 32.11 28.56 60.67 213.88
III 17 7 sample No. 2 56.12 106.37 25.88 24.37 50.25 220.87
16 sample No. 6 57.87 100.11 19.81 22.43 42.24 198.46
15 no tung meal 50.81 107.00 22.48 33.71 56.19 I 251.72
IV 1 16 6 sample No. 2 51.43 102.16 20.15 30.58 50.73 I 230.34
__16 __ sample No. 6 51.96 101.82 20.02 29.84 49.86 231.19
17 no tung meal 52.31 108.65 24.06 32.28 56.34 236.49
V 16 6 sample No. 1 53.04 103.27 22.16 28.07 50.23 234.64
16 sample No. 5 52.82 102.47 21.59 28.06 49.65 240.17
SSamples No. 1-3 at least 1 year old; No. 4-6 freshly-milled in Exp. I-III, and 5 months old in Exp. IV and V.
0;
X.1
0e
The Toxic Principle of the Tung Tree
lowed by extraction with 95 percent ethyl alcohol in a Soxhlet
extractor and then hydrolyzed by wetting with 600 ml. of
aqueous 5 percent hydrochloric acid per liter of meal and auto-
claving at 15 pounds pressure for 30 minutes.
Results are set forth in Table 3. None of the groups fed sol-
vent-extracted meals made as good gains as the control groups
which received no tung meal. The groups fed meal extracted with
naphtha and hexane made considerably less comparative gains
than those which received meal extracted with ether and ace-
tone, indicating a considerable residual toxic effect of naphtha
and hexane.
The 13 chicks fed ether-alcohol-extracted hydrolyzed meal
gained an average of 69.77 gms. during the 2-week feeding pe-
riod, compared with 64.17 gms. in the 12 control chicks. Thirteen
chicks fed acetone-alcohol-extracted hydrolyzed meal gained an
average of 95.70 gms., compared with 89.64 gms. for the 13 con-
trol chicks. In another experiment 12 chicks fed acetone-alcohol-
extracted hydrolyzed meal gained an average of 55.66 gms.,
compared with 54.78 gms. in the control group. Alcohol extrac-
tion followed by hydrolysis of the meal extracted with naphtha
and hexane diminished the residual toxic effects of naphtha and
hexane but chicks fed these meals gained only 45.36 and 39.37
gms., compared with 55.41 and 83.27 gms., respectively, in the
control groups.
DETOXIFICATION OF COMMERCIAL TUNG MEAL
AND TUNG KERNEL FLAKES
It has been indicated in previous experiments that hydrolysis
with 5 percent aqueous hydrochloric acid destroys the saponin
content of commercial tung meal but that another toxic sub-
stance still remains as shown by the retarded growth of chicks in
feeding experiments. Preliminary experiments indicated that this
second toxic substance could be removed from the meal by ex-
traction for 6 to 8 hours with 95 percent ethyl alcohol in a Sox-
hlet extractor.
Feeding experiments were conducted in which different lots
of tung meal were used for each experiment and in which the
groups of chicks were fed as follows: (a) basal ration with no
tung meal added; (b) basal ration which included 20 percent
alcohol-extracted meal; (c). basal ration which included 20 per-
cent alcohol-extracted hydrolyzed meal; (d) basal ration which
included 20 percent ether-alcohol-extracted hydrolyzed meal. In
TABLE 3.-SUMMARY OF EXPERIMENTS IN WHICH SOLVENT-EXTRACTED COMMERCIAL
TUNG MEAL WAS FED.
Number Age
Exp. of Start-
Chicks ed
S Days
12
I 13 6
13
S13
13
II 13
13
S 13
12
III 12
12
S 12
12
IV 12
11
11
12
V 12
12
12
8
4
7
7
Commercial Tung Meal Added Average Weight
(20 Percent).
no tung meal
ether-extracted (dried 1 day)
ether-extracted (dried 7 days)
ether-alcohol-extracted-hydrolyzedi
no tung meal
acetone-extracted (dried 1 day)
acetone-extracted (dried 7 days)
acetone-alcohol-ext'd-hydrolyzedi
no tung meal
acetone-extracted (dried 1 day)
acetone-extracted (dried 7 days)
acetone-alcohol-ext'd-hydrolyzedi
no tung meal
naphtha-extracted (dried 1 day)
naphtha-extracted (dried 7 days)
naphtha-alcohol-ext'd-hydrolyzedi
no tung meal
hexane-extracted (dried 1 day)
hexane-extracted (dried 7 days)
hexane-alcohol-ext'd-hydrolyzedi
Initial
grams
53.83
51.23
52.00
50.07
71.72
70.90
71.00
70.00
46.12
45.50
46.00
45.40
61.75
59.58
61.75
61.09
63.09
62.66
62.91
62.66
Final
grams
118.00
92.14
92.40
119.84
161.36
112.00
113.20
165.70
100.90
78.67
78.92
101.06
117.16
70.72
74.25
106.45
146.36
77.93
80.29
102.03
Average
Average Gain Feed
First Second Con-
Week Week Total sumption
grams grams grams grams
26.17 38.00 64.17 281.25
19.06 21.85 40.91 248.12
20.05 20.35 40.40 252.66
29.77 40.00 69.77 266.30
53.83 35.81 89.64 288.84
19.82 21.28 41.10 250.18
20.19 22.01 42.20 262.94
43.81 51.89 95.70 330.18
24.62 30.16 54.78 204.62
15.06 18.11 33.17 182.45
16.01 16.91 32.92 196.63
25.41 30.25 55.66 210.74
22.66 32.75 55.41 235.25
6.14 5.00 11.14 271.81
5.48 7.22 12.50 261.50
18.45 26.91 45.36 252.54
42.02 41.25 83.27 300.72
7.00 8.27 15.27 259.66
17.38 279.54
28.50 10.87 39.37 274.66
1 Hydrolyzed as in preceding experiments at 15 pounds pressure for 30 minutes.
The Toxic Principle of the Tung Tree 21
(b) the second toxic substance was rerhoved by alcohol extrac-
tion; any toxic effects would be attributable to saponin. In (c)
the second toxic substance was removed by alcohol extraction
and the saponin content was destroyed by hydrolysis. In (d) the
meal was prepared as in (c) after first extracting with ether
to render it oil-free.
Results of the 2 experiments in which this procedure was
followed are shown in Table 4. The groups which received ether-
alcohol-extracted hydrolyzed meal made average gains of 69.77
and 54.39 gms., compared with 64.17 and 55.57 gms., respective-
ly, in the control groups. The 2 groups fed meal which was al-
cohol-extracted and hydrolyzed but not ether-extracted original-
ly made average gains of 6.65 and 0.78 gms. less than those fed
meal similarly treated but also rendered oil-free by ether extrac-
tion, indicating that the residual' oil content either has some
toxic effects or interferes slightly with the detoxification process
used.
The 2 groups which received alcohol-extracted meal and in
which the second toxic substance only was removed made aver-
age gains of 45.83 and 47.92 gms., compared with 64.17 and
55.57 gms., respectively, in the control groups.
Two additional experiments are reported in Table 4 in which
the groups of chicks were fed as follows: (a) basal ration to
which no tung meal was added; (b) basal ration which included
ether-alcohol-extracted hydrolyzed meal; (c) basal ration to
which the alcohol extract from (b) was hydrolyzed and added
and allowed to air dry for 72 hours. The alcohol extract was
hydrolyzed by boiling with an equal amount of 5 percent aqueous
solution of hydrochloric acid for 30 minutes. The groups which
received ether-alcohol-extracted hydrolyzed meal made average
gains of 79.00 and 54.72 gms., compared with 61.51 and 59.46
gms., respectively, in the control groups. Chicks fed the hy-
drolyzed alcohol extract from the above meal gained an aver-
age of 29.67 and 44.47 gms., compared with 61 and 59.46 gms.,
respectively, in the control groups. These latter experiments
further attest.the toxicity of this second substance.
As the above detoxification experiments were concerned only
with commercial tung meal it appeared advisable to attempt to
detoxify tung kernel flakes in the same manner. Table 5 sets,
forth the results of 3 experiments in which tung kernel flakes
were detoxified as in the experiments just described. Tung ker-
nel flakes were added to the basal ration in an amount approxi-
TABLE 4.-SUMMARY OF FEEDING EXPERIMENTS IN WHICH EXTRACTED COMMERCIAL
TUNG MEAL WAS USED
NubeI ]Average
Number Age Commercial Tung Meal Added Average Weight | Average Gain Feged
Exp. of Start- (20 Percent) First Second Con-
Chicks ed Initial Final Week Week Total sumption
| Days grams grams grams grams grams grams
| 12 no tung meal 53.83 118.00 26.17 38.00 64.17 281.25
I | 13 6 alcohol-extracted 51.23 97.06 19.46 26.37 45.83 201.43
S13 alchohol-extracted-hydrolyzedi 52.00 115.12 26.02 37.10 63.12 285.58
S13 ( ether-alcohol-extracted-hydrolyzedx 50.07 119.84 I 29.77 40.00 69.77 266.30
S14 | no tung meal 51.50 107.07 21.92 33.67 55.57 222.42
II 13 6 alcohol extracted 51.38 99.30 17.92 30.00 47.92 221.69
13 alchohol-extracted-hydrolyzed, 53.46 107.07 22.38 31.23 53.61 229.15
13 | ether-alcohol-extracted-hydrolyzedi 54.61 109.00 24.08 30.31 54.39 223.53
17 no tung meal 48.15 109.66 21.85 39.66 61.51 220.63
III 16 4 ether-alcohol-extracted-hydrolyzedi 48.75 127.75 34.66 44.34 79.00 275.08
I 16 alcohol ext. from above-hydrolyzed2 48.33 78.50 14.08 15.59 29.67 152.08
I 16 no tung meal 52.84 112.30 21.57 37.89 59.46 218.52
IV 16 6 J ether-alchohol-extracted-hydrolyzedl 54.30 109.02 21.03 33.67 54.72 265.11
S16 ] alcohol ext. from above-hydrolyzed2 53.63 98.10 15.82 28.65 44.47 217.80
'Hydrolyzed as in preceding experiments at 15 pounds pressure for 30 minutes.
2 Hydrolyzed by boiling with equal volume 5 percent aqueous solution hydrochloric acid for 30 minutes.
TABLE 5.-SUMMARY OF FEEDING EXPERIMENTS INVOLVING EXTRACTED TUNG KERNEL FLAKES AND SHELL.
I Average
Number Age Material Added Average Weight Average Gain Feed
Exp. of Start- (Equivalent to 20 Percent Cor- First Second Con-
Chicks ed mercial Tung Meal) Initial Final Week Week Total sumption
I Days grams grams grams grams grams grams
15 no tung flakes 54.38 148.11 48.00 45.23 93.23 295.45
I 15 6 ether-alcohol-ext'd flakes-hydl'yzedi 55.62 148.12 38.88 53.62 92.50 244.62
15 alcohol ext. from above-hydrolyzed 55.98 114.87 22.87 36.02 58.89 221.93
14 no tung flakes 42.15 96.50 20.69 33.66 54.35 216.40
II 14 4 ether-alcohol-ext'd flakes-hydl'yzed 46.33 97.55 18.11 33.11 51.22 230.00
14 alcohol ext. from above-hydrolyzed 46.11 86.88 13.11 27.66 40.77 197.22
14 no tung flakes 56.40 156.25 44.47 55.38 99.85 268.75
III 14 6 ether-alcohol-ext'd flakes-hydl'yzed 55.75 156.50 46.00 54.75 100.75 275.46
14 alcohol ext. from above-hydrolyzed 55.12 98.25 15.13 28.00 43.13 173.75
14 no tung shell 56.40 156.25 44.47 55.38 99.85 268.75
IV 13 6 alcohol-extd shell-hydrolyzed 56.00 156.80 44.50 56.30 100.80 270.50
13 alcohol ext. from above-hydrolyzed 55.62 156.87 42.13 59.12 101.25 257.50
14 no tung shell 43.26 98.45 24.32 30.87 55.19 227.46
V 14 4 alcohol-ext'd shell-hydrolyzed 43.00 99.60 25.81 30.19 56.00 260.17
14 alcohol ext. from above-hydrolyzed 42.16 97.91 23.49 32.26 55.75 217.50
1 Hydrolyzed by boiling with equal volume 5 percent aqueous solution of hydrochloric acid for 30 minutes.
Florida Agricultural Experiment Station
mately equivalent to 20 percent commercial tung meal (on the
basis that commercial tung meal contains 45 percent kernel, 45
percent shell, 5 percent moisture and 5 percent residual oil).
Hydrolysis of the flakes was accomplished by boiling in an equal
volume of 5 percent aqueous solution of hydrochloric acid, as
hydrolysis by autoclaving proved unsatisfactory due to loss of
material.
The 3 groups, consisting of a total of 43 chicks, which were
fed ether-alcohol-extracted hydrolyzed tung kernel flakes made
average gains of 92.50, 51.22 and 100.75 gms., compared with
93.23, 54.35 and 99.85 gms., respectively, in the control groups.
Thus, 2 of the groups gained 0.73 and 3.13 gms. less, while the
other gained 0.90 gms. more than the respective control groups.
The 3 groups which received the alcohol extracts in the pre-
ceding 3 experiments gained an average of 58.89, 40.77 and 43.13
gms., compared with 93.23, 54.35 and 99.85 gms., respectively,
in the control groups.
These experiments indicate the toxicity of the second toxic
substance and that it can be extracted from tung kernel flakes
by means of 95 percent ethyl alcohol.
Two additional experiments were conducted (Table 5) to de-
termine the possibility of shell (the hard coating immediately
around the kernel and which constitutes approximately 45 per-
cent of commercial tung meal) containing a portion of the second
toxic substance. The 2 groups of chicks fed alcohol-extracted
hydrolyzed shell in an amount equivalent to 20 percent of tung
meal in the ration and those fed the hydrolyzed alcohol extract
of shell from the above made average gains slightly better than
the control groups which received no tung shell, indicating that
tung shell is not a factor as far as the second toxic principle is
concerned and that tung shell in the amounts fed was not detri-
mental.
ISOLATION OF THE SECOND TOXIC SUBSTANCE FROM COM-
MERCIAL TUNG MEAL AND TUNG KERNEL FLAKES
In all preceding experiments the second toxic substance was
contained in the alcohol extract of tung meal and tung kernel
flakes. Attempts were made to obtain the toxic substance in
purer form. Feeding experiments were conducted on various
fractions of the alcohol extract and by the elimination of non-
toxic fractions the following procedure was developed. Com-
The Toxic Principle of the Tung Tree
mercial tung meal or tung kernel flakes were extracted with
ether to render them oil-free. The oil-free material then was
extracted with 95 percent ethyl alcohol in a Soxhlet extractor
(approximately 6 to 8 hours). The alcohol extract, while hot,
was filtered through filter paper and stored in the refrigerator
until the liquid was clear. If the liquid did not become clear with-
in 48 hours it was evaporated over a water bath to one-half its
volume and again placed in the refrigerator. When clear it was
filtered through filter paper, placed in an evaporating dish over
a water bath and evaporated until the liquid was rather syrupy
in consistency in the case of the extract from tung meal and un-
til the liquid began to show milky streaks in the case of tung
kernel flakes. Three volumes of acetone were added and the liquid
allowed to boil while stirring for about 5 minutes. The liquid
was filtered through filter paper while hot and the filtrate was
stored in the refrigerator until it became clear. Usually in the
case of extracts from tung meal no difficulty was experienced
in this respect. However, in the case of tung kernel flakes it
often was necessary to filter repeatedly in the refrigerator and
re-store before the acetone solution finally cleared. The ace-
tone solution finally was filtered through filter paper and re-
mained clear.
This fraction, herein called "acetone-soluble fraction", pre-
pared from commercial tung meal was reddish chocolate brown
in color; that prepared from tung kernel flakes was pale canary
yellow in color. In feeding this fraction, the acetone solution
from an amount of tung meal and kernel flakes equivalent to 20
percent of tung meal in the ration was mixed with the mash,
spread in a layer and allowed to dry for 3 or 4 days. A sum-
mary of feeding experiments in which the "acetone-soluble frac-
tion" was used is tabulated in Table 6. The ether-alcohol-ex-
tracted hydrolyzed meal and flakes were prepared as described
previously.
In 2 experiments in which the "acetone-soluble fraction"
was extracted from tung meal the 26 chicks fed this fraction
gained during the 2-week feeding period an average of 8.80 and
9.60 gms., as compared with 42.71 and 54.31 gms., respectively,
in the control groups; 7 chicks fed the "acetone-0oluble fraction"
in the first experiment died while 6 chicks similarly fed in the
second experiment died. The 2 groups fed ether-alcohol-ex-
tracted hydrolyzed meal gained an average of 44.63 and 52.91
gms., which represents a gain of 1.92 gms. more than the con-
TABLE 6.-SUMMARY OF FEEDING EXPERIMENTS TO DETERMINE TOXIC FRACTION OTHER THAN SAPONIN
IN TUNG MEAL AND FLAKES
Number Age
Exp. of Start-
Chicks ed
Days
12
I 12 4
12
Material Added
(Equivalent to 20 Percent Com-
mercial Tung Meal)
no tung meal
acetone-soluble fraction
ether-alcohol-ext'd-hydrolyzedi
Average
Average Weight Average Gain Feed
__ Con- Mortality
| First Second sump-
Initial Final Week Week Total tion
grams
45.14
46.58
45.91
grams grams grams grams
87.85 18.07 24.64 42.71
55.38 1.78 7.02 8.80
90.54 17.68 26.95 44.63
grams
159.50
127.62
159.76
14 no tung meal 52.88 1 107.19 18.69 35.62 54.31 254.05
II 14 6 acetone-soluble fraction 54.90 64.50 1.97 7.63 9.60 187.52 6
14 ether-alcohol-ext'd-hydrolyzedi 53.27 106.18 23.27 29.64 52.91 228.63
14 no tung flakes 48.20 88.22 17.65 22.37 40.42 192.33
III 14 4 acetone soluble fraction 48.66 60.87 2.15 10.06 12.21 170.33 7
14 ether-alcohol-ext'd-hydrolyzedz 48.66 88.09 19.42 20.01 39.43 184.55
10 no tung flakes 45.77 98.92 22.00 31.15 53.15 237.40
IV 10 4 acetone soluble fraction 45.55 58.64 3.29 9.80 12.09 184.06 5
10 ether-alcohol-ext'd-hydrolyzeds 45.88 98.11 20.89 31.34 52.23 229.66
1 Hydrolyzed as in preceding experiments at 15 pounds pressure for 30 minutes.
2 Hydrolyzed by boiling with equal volume 5 percent aqueous solution hydrochloric acid for 30 minutes.
The Toxic Principle of the Tung Tree
trol group in the first experiment and 1.40 gms. less than the
control group in the second experiment.
In 2 additional experiments in which the "acetone-soluble frac-
tion" was extracted from tung kernel flakes the 24 chicks fed
this fraction gained during the 2-week feeding period an average
of 12.21 and 12.09 gms., as compared with 40.02 and 53.15 gms.,
respectively, in the control groups; 7 chicks in the first experi-
ment and 5 in the second fed the "acetone-soluble fraction" died.
The 2 groups fed ether-alcohol-extracted hydrolyzed tung flakes
gained an average of 39.43 and 52.23 gms., which represents a
gain of 0.59 and 0.92 gms., respectively, less than the control
groups.
The "acetone-soluble fraction" was fed to 3 groups of chick-
ens from 4 to 6 weeks of age. In each instance none of the
chicks died but the feed which contained this fraction extracted
from an amount of tung meal equivalent to 20 percent of the
ration was not readily consumed after the first 4 or 5 days and
the chicks did not make gains comparable to control groups.
GROWTH CURVES AND LESIONS OF CHICKS FED
COMMERCIAL TUNG MEAL
Oil-free Tung Meal.-Commercial tung meal (stored 30
days) was extracted with petroleum ether in a Soxhlet extractor
to render it oil-free. The oil-free meal was added to the basal
ration so as to constitute 10 percent of the total ration and fed
to young chicks for a period of 5 weeks. The control group re-
ceived the basal ration only. The 10 chicks in the control group
weighed an average of 126.8 gms. at the start of the experiment
and gained an average of 418.2 gms. over the 5-week period.
Those fed the tung meal ration weighed an average of 132.9
gms. at the start of the experiment and gained an average of
170.3 gms.; 6 of the 10 chicks fed the tung meal ration died dur-
ing the 5-week period. A second group of chicks was fed the
basal ration consisting of 5 percent of oil-free tung meal for a
period of 5 weeks. The 10 control chicks which received the
basal ration only'weighed an average of 131.0 and 538.3 gms.,
respectively, at the beginning and conclusion of the experiment,
an average gain of 407.3 gms. The 10 chicks fed the tung meal
ration weighed an average of 129.7 and 358.5 gms., respectively,
at the beginning and conclusion of the experiment, an average
gain of 228.8 gms.; 2 of the 10 chicks died during the experiment.
The growth curves are illustrated in Fig. 2.
Florida Agricultural Experiment Station
545.0 558.5
5Z0
480
1oo
100
3Go ,3558.5
5Z0 / ,505.- / /'
280 /. /
160 1
S. / /
2/ 0 / /
/ 3 / ./
sent 2 groups of 10 chicks each which received no tung meal. The 10
chicks represented by the broken line on the left received 10 percent of
oil-free tung meal in the ration. Those on the right received 5 percent
Fig. 2.-Growth curves of New Hampshire chicks showing effect of
feeding oil-free commercial tung meal for 5 weeks. The solid lines repre-
sent 2 groups of 10 chicks each which received no tung meal. The 10
chicks represented by the broken line on the left received 10 percent of
oil-free tung meal in the ration. Those on the right received 5 percent
of oil-free tung meal in the ration. Approximate time of death of chicks
represented by dots under the broken line.
The chicks which died as a result of receiving the tung meal
ration revealed the following lesions: the gallbladder was dis-
tended greatly; the liver was somewhat mottled with irregular
congestion, in 2 chicks the liver showed atrophic degeneration;
the heart was flabby; the duodenum was thickened and the mu-
cosa showed various degrees of catarrhal inflammation with oc-
casional petechia on the mucous membrane; in 4 chicks which
lived longest considerable necrotic enteritis was observed.
"Acetone-Soluble Fraction".-Another feeding experiment was
conducted in which the "acetone-soluble fraction" was added to
the basal ration in an amount equivalent to 20 percent of tung
The Toxic Principle of the Tung Tree
meal in the ration. The 10 control chicks weighed an average of
126.2 and 552.0 gms., respectively, at the beginning and conclu-
sion of the experiment, the average gain being 425.8 gms. The
10 chicks fed the "acetone-soluble fraction" from commercial
tung meal weighed an average of 127.7 and 320.2 gms., respec-
tively, at the beginning and conclusion of the experiment, an
average gain of 192.5 gms. The 10 chicks fed the "acetone-solu-
ble fraction" from an amount of tung kernel flakes equivalent
to 20 percent of commercial tung meal in the ration weighed an
average of 126.8 and 292.4 gms., respectively, at the beginning
,S52.0 552.0
52.0.42.
/ .7,9z'?.4.
I 2 5 4 5 I 2. 3 4 5
Weeks ed
Fig. 3.-Growth curves of New Hampshire chicks showing effect
of feeding "acetone-soluble fraction" extracted from commercial tung
meal and from tung kernel flakes. The solid line in each instance is the
growth curve of 10 control chicks for a 5-week period. The broken line on
the left is the growth curve of 10 chicks which received "acetone-soluble
fraction" extracted from an amount of tung meal equivalent to 20 percent
of the ration. The broken line on the right is the growth curve of 10
chicks which received "acetone-soluble fraction" extracted from an amount
of tung kernel flakes equivalent to 20 percent of tung meal in the
ration.
Florida Agricultural Experiment Station
and conclusion of the experiment, an average gain of 165.6 gms.
The growth curves are illustrated in Figure 3.
At the conclusion of the experiment the chicks fed the "ace-
tone-soluble fraction" looked slightly unthrifty; the breast was
poorly fleshed. The following was observed on postmortem ex-
amination: the gallbladder usually was distended, often mark-
edly; the liver was irregularly congested; the duodenum often
was thickened greatly, but the mucosa showed only slight gross
evidence of enteritis. Even though the duodenum was more
thickened than in the birds receiving the oil-free tung ration
the degree of enteritis was much less than in these latter chicks.
Detoxified and Hydrolyzed Tung Meal.-Commercial tung meal
was extracted with petroleum ether in a Soxhlet extractor to
render it oil-free. The detoxified meal was extracted for 6 to 8
hours in a Soxhlet extractor with 95 percent ethyl alcohol and
hydrolyzed. The hydrolyzed meal was hydrolyzed as has been
described previously to destroy the saponin content; the second
toxic substance still remained. These 2 processed meals were
mixed with the feed so as to constitute 20 percent of the ration
and each fed to a group of 10 chicks over a period of 5 weeks.
Ten control chicks received the basal ration alone.
In the first experiment the chicks fed the detoxified meal
weighed an average of 48 gms. at the start and 364.9 gms. at
the conclusion of the 5-week feeding period, an average gain of
316.9 gms. The control chicks weighed an average of 48 and
326.8 gms., at the beginning and conclusion, respectively, an
average gain of 278.8 gms. The chicks fed hydrolyzed meal
gained an.average of 131.8 gms. during the 5-week period.
In the second experiment freshly milled tung meal was used.
The chicks fed the detoxified meal weighed an average of 54.6
gms. at the start and 377.8 gms. at the conclusion of the 5-week
feeding period, an average gain of 323.2 gms. The control chicks
gained from 54.7 gms. to 386.4 gms., or 331.8 gms. The chicks
fed hydrolyzed meal gained an average of 128.9 gms.
DISCUSSION
These experiments indicate that commercial tung meal con-
tains 2 toxic substances, a saponin and a second toxic substance
which has not been identified. The saponin content was de-
stroyed to best advantage, as judged by feeding the meal to young
chicks, by hydrolysis. Chemically, a saponin is composed of 2
molecules, 1 of which is a sugar; when hydrolyzed, sugar
The Toxic Principle of the Tung Tree
4oo0
384.4
577.8
2bo / /
ZOO / / 52.8
/o / .5.8 / 85
2.8 0 /
o/ /
feeding detoxified commercial tung meal and tung meal in which the
zL^ 'I
zo I---------
saponin was destroyed by hydrolysis for a 5-week period. The solid line
in each instance is the growth curve of 10 control chicks. The broken line
on the left is the growth curve for 10 chicks whose ration contained 20
percent of detoxified tung meal (ether-alcohol-extracted-hydrolyzed) 12
months old. The broken line on the right is the growth curve for 10 chicks
whose ration contained 20 percent of detoxified tung meal which was
freshly milled. The dotted line in each instance represents the growth
curve of 10 chicks whose ration contained 20 percent of hydrolyzed tung
meal from the respective meals used. Hydrolysis destroys the saponin but
not the second toxic substance.
splits off, rendering the saponin innoxious when fed to livestock.
The second toxic substance can be extracted with 95 percent
ethyl alcohol in a Soxhlet extractor over a period of 6 to 8 hours.
It has been shown that hydrolyzed freshly milled tung meal is
more toxic than 12-month-old tung meal similarly treated. Ex-
periments also indicated that a considerable portion of this
increased toxicity of hydrolyzed freshly-milled tung meal was
lost during 5 months of storage. Thus, tung meal under good
storage conditions gradually becomes less toxic within certain
limits, with at least 5 months of storage. This factor has not re-
ceived consideration from investigators working on the toxicity
5J4.1
Florida Agricultural Experiment Station
of this product and may, in fact, be responsible for considerable
variation in the results obtained from feeding meals treated in
different manners in efforts to detoxify them.
In experiments reported herein the feeding of tung meal 12
months of age or older and detoxified by the alcohol-extraction-
hydrolysis process described in this paper resulted in consistently
higher gains when composing 20 percent of the ration than in
control chicks which received no tung meal. In the experiments
in which freshly-milled tung meal was detoxified by the same
process considerable variation occurred in this respect in indi-
vidual experiments, although average results indicated detoxifi-
cation to the extent that chicks fed 20 percent detoxified tung
meal made gains comparable with chicks which received no tung
meal.
These experiments indicate that tung meal rendered oil-free by
extraction with petroleum ether, commercial acetone, naphtha
and hexane is toxic. The toxic saponin contained in tung meal
is not soluble in any of these oil solvents. Even though the sec-
ond toxic substance was soluble in acetone and finally designated
as the "acetone-soluble fraction", the extraction of commercial
tung meal with acetone only partially removed this second toxic
substance. Apparently it was completely soluble in acetone
after it had been originally extracted with 95 percent ethyl
alcohol. Thus, neither the saponin nor the second toxic sub-
stance was removed from tung meal by extraction with pe-
troleum ether, naphtha or hexane. The second toxic substance
was partially removed by extraction with acetone.
These experiments also indicate that heat tends to reduce
the toxic properties of these 2 toxic substances in tung meal.
The results of chick feeding experiments showed that meal sub-
jected to heat (15 pounds pressure in an autoclave) for 4 hours
was less toxic than meal similarly treated for 2 hours. Experi-
ments were not conducted to determine if the meal could be
detoxified completely by heating it for long periods. The recent
report of Davis, Mehrhof and McKinney (3) confirms the previ-
ous report of this author that the subjection of tung meal to
heat does not fully detoxify the meal.
Rusoff, Mehrhof and McKinney (13) concluded from their ex-
periments that the solvent-extracted tung meal they fed to young
chicks showed no definite indications of toxicity. The experi-
ments reported herein indicate that solvent-extracted tung meal
was toxic. The meal fed by Rusoff et al, in addition to being
The Toxic Principle of the Tung Tree
solvent-extracted, was heated for 21/2 hours under steam pressure
gradually increased to 90 pounds. While this probably was done
to exhaust the meal of naphtha, it has been shown in the experi-
ments reported herein that heat partially destroys some of the
toxic substances contained in the meal. These authors did not
take into account the effects of heat upon the toxic substances
contained in the meal. Their conclusion that solvent-extracted
tung meal showed no indications of toxicity probably was based
upon the fact that none of the chicks included in their experi-
ments died as a result of being fed this meal.
SUMMARY
Saponin, a toxic substance, was isolated from the foliage of
Aleurites fordi Hemsl., A. montana (Lour.) Wils., A. moluccana
Willd. (A. triloba Forst.) and A. trisperma Blanco. It was more
abundant in A. fordi than in the other 3 species. Saponin was
isolated also from kernels of A. fordi as well as from commer-
cial tung meal (ground pressed cake from the seeds of A. fordi).
The kernel of the fruit of A. fordi and commercial tung meal
from the same species contain a toxic substance in addition to
saponin.
This second toxic substance was more abundant in 3 samples
of freshly milled tung meal than in 3 samples of tung meal at
least 12 months of age, as determined by 2-week feeding tests
to chicks 5 to 7 days old. Freshly milled tung meal decreased in
toxicity during a 5-month storage period until the presence of
this second substance in 5-month old samples was comparable
to that of older samples of tung meal.
Two-week feeding tests with chicks 4 to 8 days old indicated
that solvent-extracted (ether, acetone, naphtha and hexane)
tung meal was toxic.
It was found that the saponin content of tung kernel flakes
and commercial tung meal could be destroyed to best advantage
in meal intended for feeding tests by hydrolysis (600 ml. 5 per-
cent aqueous hydrochloric acid was added to 1 liter of meal and
heated in autoclave at 15 pounds pressure for 30 minutes after
being allowed to stand 30 minutes).
It was found that the second toxic substance could be removed
from tung meal by 6 to 8 hour extraction in a Soxhlet extractor
with 95 percent ethyl alcohol.
Florida Agricultural Experiment Station
In 7 trials a total of 92 chicks fed a ration consisting of 20
percent of detoxified (ether-alcohol extracted followed by hy-
drolysis) commercial tung meal gained an average of 52.58 grams
during a 2-week feeding period, as compared with an average
gain of 52.77 grams in chicks which received no tung meal. A
total of 43 chicks in 3 trials gained an average of 81.49 grams
during a 2-week feeding period in which an amount of detoxi-
fied tung kernel flakes equivalent to 20 percent of tung meal
was included in the ration, while an equal number of chicks which
received no tung kernel flakes gained an average of 82.47 grams
during a similar period. The trials were started when the chicks
were 4 to 8 days of age.
A method of isolation of the second toxic substance in an "ace-
tone-soluble fraction" is described. In this form it caused the
death of 50 percent of the chicks to which it was fed in an
amount equivalent to 20 percent of commercial tung meal during
a 2-week period; the chicks were 4 to 6 days of age when placed
on feeding test. The feeding of the "acetone-soluble fraction"
to chicks when 4 to 6 weeks of age at the same rate did not
cause death but resulted in retarded growth.
Two trials in which an amount of tung shell equivalent to 20
percent tung meal was added to the ration and fed for a 2-week
period to chicks 4 to 6 days indicated that the shell did not con-
tain the second toxic substance.
ACKNOWLEDGEMENTS
The author expresses his thanks to the following for assistance ren-
dered during the progress of the experiments reported herein: J. Wershaw,
Beverly Hills Plantation, Alachua, Florida, for supplying tung meal and
flakes; 0. K. Moore, Florida Agricultural Experiment Station (now re-
signed), for establishing the basal feed formula; Dr. Geo. D. Ruehle, vice-
director and plant pathologist, Sub-tropical Experiment Station, Home-
stead, Florida, for supplying leaves of A. trisperma; Albertus Miller,
Brooker, Florida, for supplying samples of freshly milled commercial tung
meal; and Dr. F. S. Lagasse, senior pomologist, U. S. Field Laboratory
for Tung Investigations, BPISAE, USDA, Gainesville, Florida, for supply-
ing tung flakes.
LITERATURE CITED
1. AUTENRIETH, W. Detection of poison (Trans. by Warren). 6th ed. P.
Blakiston's Sons & Co., Philadelphia. 1928.
2. BLACKMON, G. H. The tung-oil industry. Bot. Rev. 11: 1-40. 1943.
3. DAVIS, GEORGE K., N. R. MEHRHOF and R. S. MCKINNEY. Effect of
tung meal in rations for growing chicks. Poultry Sci. 25: 74-79.
1946.
Florida Agricultural Experiment Station
In 7 trials a total of 92 chicks fed a ration consisting of 20
percent of detoxified (ether-alcohol extracted followed by hy-
drolysis) commercial tung meal gained an average of 52.58 grams
during a 2-week feeding period, as compared with an average
gain of 52.77 grams in chicks which received no tung meal. A
total of 43 chicks in 3 trials gained an average of 81.49 grams
during a 2-week feeding period in which an amount of detoxi-
fied tung kernel flakes equivalent to 20 percent of tung meal
was included in the ration, while an equal number of chicks which
received no tung kernel flakes gained an average of 82.47 grams
during a similar period. The trials were started when the chicks
were 4 to 8 days of age.
A method of isolation of the second toxic substance in an "ace-
tone-soluble fraction" is described. In this form it caused the
death of 50 percent of the chicks to which it was fed in an
amount equivalent to 20 percent of commercial tung meal during
a 2-week period; the chicks were 4 to 6 days of age when placed
on feeding test. The feeding of the "acetone-soluble fraction"
to chicks when 4 to 6 weeks of age at the same rate did not
cause death but resulted in retarded growth.
Two trials in which an amount of tung shell equivalent to 20
percent tung meal was added to the ration and fed for a 2-week
period to chicks 4 to 6 days indicated that the shell did not con-
tain the second toxic substance.
ACKNOWLEDGEMENTS
The author expresses his thanks to the following for assistance ren-
dered during the progress of the experiments reported herein: J. Wershaw,
Beverly Hills Plantation, Alachua, Florida, for supplying tung meal and
flakes; 0. K. Moore, Florida Agricultural Experiment Station (now re-
signed), for establishing the basal feed formula; Dr. Geo. D. Ruehle, vice-
director and plant pathologist, Sub-tropical Experiment Station, Home-
stead, Florida, for supplying leaves of A. trisperma; Albertus Miller,
Brooker, Florida, for supplying samples of freshly milled commercial tung
meal; and Dr. F. S. Lagasse, senior pomologist, U. S. Field Laboratory
for Tung Investigations, BPISAE, USDA, Gainesville, Florida, for supply-
ing tung flakes.
LITERATURE CITED
1. AUTENRIETH, W. Detection of poison (Trans. by Warren). 6th ed. P.
Blakiston's Sons & Co., Philadelphia. 1928.
2. BLACKMON, G. H. The tung-oil industry. Bot. Rev. 11: 1-40. 1943.
3. DAVIS, GEORGE K., N. R. MEHRHOF and R. S. MCKINNEY. Effect of
tung meal in rations for growing chicks. Poultry Sci. 25: 74-79.
1946.
The Toxic Principle of the Tung Tree 35
4. EMMEL, M. W., D. A. SANDERS and L. E. SWANSON. The toxicity of
foliage of Aleurites fordi Hemsl. for cattle. Jour A. V. M. A. 101:
136-137. 1942.
5. EMMEL, M. W. The toxic principle of Aleurites fordi Hemsl. Jour.
A. V. M. A. 103: 162. 1943.
6. EMMEL, M. W. The toxic principle of the tung tree. Proc. Amer.
Tung Oil Asso., pp. 38-42. 1945.
7. ERICKSON, J. L. E., and J. H. BROWN, JR. A study of the toxic prop-
erties of tung nuts. Jour. Pharm. and Exp. Ther. 74: 114. 1942.
8. FLEMING, P. B., H. O. HUNTER, F. KERR and W. E. HARKNESS. Tung
Oil. Fla. Dept. of Agri. Bul. 11 (new series). 1942.
9. GODDEN, W. The feeding value of tung seed meal. Bul. Imp. Inst.
30: 3. 1933.
10. JAMIESON, G. S. Vegetable fats and oils. Chemical Catalog Co., New
York. 1932.
11. LEWKOWITSCH, I. Chemical technology and analysis of oil fats and
waxes. 4th ed. Macmillan & Co., London. 1909.
12. NEWELL, WILMON. Preliminary report on experiments with tung oil
tree in Florida. Fla. Agri. Exp. Sta. Bul. 171. 1924.
13. RUSOFF, L. L., N. R. MEHRHOF and R. S. McKINNEY. Chick feeding
experiments with solvent-extracted tung oil meal. Poultry Sci. 21:
451-454. 1942.
ments with solvent-extracted tung oil meal. Poultry Sci. 21: 451-
454. 1942.
14. SANDERS, D. A., M. W. EMMEL and L. E. SWANSON. Tung Tree
(Aleurites fordi Hemsl.) foliage poisoning of cattle. Fla. Agri. Exp.
Sta. Bul. 376. 1942.
|
