COMPARATIVE TOXICITY OP COTTONSEED PRODUCTS 
By W. A. Withers and F. E. Carruth 
Chemical Division , North Carolina Agricultural Experiment Station 
INTRODUCTION 
Notwithstanding the isolation by us of a distinctly toxic substance 
(gossypol) from cottonseed (15) ,* six papers have recently appeared 
offering to explain cottonseed-meal poisoning or “ injury ” on the theory 
of dietary deficiencies. The chief support for this view is found (a) by 
Rommel and Vedder (ij) who fed pigs on polished rice and tankage 
developed symptoms similar to those fed at the same time on com 
meal and cottonseed meal, and (b) by Richardson and Green (jo, 11 , 12 ) 
and by Osborne and Mendel (7, 8 ), who show that, while white rats 
will ultimately fail on diets containing cottonseed meal as the sole 
source of protein, minerals, and vitamines, the animals will grow nor¬ 
mally on cottonseed-meal or cottonseed-flour diets if supplemented by 
milk powder, protein-free milk, butter, etc. 
Wells and Ewing ( 14 ) have also adopted the dietary-deficiency view 
as being the most plausible explanation of cottonseed-meal injury. 
They base their conclusion that cottonseed meal is an incomplete food 
on experiments with very young pigs confined in metabolism cages, the 
cottonseed ration being supplemented with starch, sugar, and small quan¬ 
tities of milk. In some cases they admit that the injury may have been 
due to the presence of a toxic substance. 
The fallacy in concluding, from the ultimate failure of rats on diets in 
which cottonseed meal is the sole source of protein, minerals, and vita- 
mines, that dietary deficiencies are the cause of cottonseed-meal injury 
in swine, for example, is evident if we base a somewhat similar argu¬ 
ment as to other seeds on the results of feeding them to rats as the sole 
source of these nutritive factors. For example, rats similarly restricted 
to any seed will show failure, and in most cases much sooner than on 
cottonseed meal, but we know that in the diet of swine these seed diets 
do not cause the sudden death which is typical of cottonseed-meal injury. 
Another distinct difference between the failure of rats on restricted 
cottonseed-meal diets and the failure of swine on cottonseed meal under 
farm conditions is that with rats death usually follows a period of low 
food intake and consequent emaciation, while with swine it is very well 
known that death often follows a period of excellent growth and with the 
animals in splendid nutrition. None of these investigators has fed 
1 Reference is made by number (italic) to “Literature cited,” p. 451-452. 
Journal of Agricultural Research, 
Washington, D. C. 
pc 
(425) 
Vol. XIV, No. 10 
Sept. 2,1918 
Key No. N. C.-13 
