Jan. 14, 19x8 Gossypol, the Toxic Substance in Cottonseed 89 



siderable amou nts (about i per cent) of a substance which we have called 

 "D-gossypol." 



"D-gossypol" is very slightly soluble in ether. For rabbits ether 

 extraction does not render the meal nontoxic. But where we find in the 

 meal after 6 hours' extraction with ether considerable amounts of a 

 substance giving color reactions for gossypol or "D-gossypol," it would 

 seem that the substance is bound in some way. To explain this we have 

 assumed that it may be combined with the protein or some other con- 

 stituent. There is some evidence of this in the properties of these sub- 

 stances. We have mentioned that gossypol combines with anilin and 

 with acetic acid to form less-soluble compounds. Possibly similar com- 

 bination may take place with free amino and free carboxyl groups in the 

 protein molecule. Marchlewski (<?) mentioned the fact that gossypol 

 behaves like tannin toward basic dyes. Tannin also precipitates pro- 

 teins as insoluble compounds. A similar combination of gossypol with 

 protein may occur in the cooking of cottonseed. 



Confirmatory evidence that these substances may be the cause of 

 cottonseed-meal poisoning is given in the previous publications of this 

 Station. Thus, when cottonseed meal is treated with an alcoholic 

 alkali (jp), the meal is rendered nontoxic to rabbits. When the meal 

 is fed with iron salts to pigs {18) and rabbits {16) , the toxic effect is greatly 

 diminished. Pigs fed on cottonseed meal and corn meal (1:3) with 

 ferrous sulphate (copperas) did not die in 180 days, whereas without 

 ferrous sulphate all the animals died. Rabbits were fed 106 days with 

 ferric ammonium citrate without harmful results. We have explained 

 this by assuming that the alkali treatment promoted oxidation of the 

 gossypol and by assuming that the iron salt formed an insoluble pre- 

 cipitate with the gossypol, or possibly assisted the organism to oxidize it. 



The experiments referred to in this article support our previous view 

 that gossypol is toxic and that it is the only toxic substance in the raw 

 kernels.^ Extensive experiments with various meals with rats, rabbits, 

 fowls, and swine show that there still remains, even in thoroughly cooked 

 meals, an injurious factor. Such thoroughly cooked meals are harmful 

 to rabbits and swine, but seem to have little effect on rats and fowls when 

 fed on adequate diets. 



PRELIMINARY EXPERIMENTS WITH PIGS 



In order further to test the correctness of our view that cottonseed-meal 

 injury is due to a toxic substance rather than to dietary deficiencies, we 

 have conducted a few preliminary experiments with small pigs. 



It seemed desirable, in view of the extreme position taken by Rommel 

 and Vedder (14) to ascertain (i) whether gossypol is toxic to pigs; (2) 



' The results of other experiments showing that cooking exerts a profound influence on the toxicity will 

 be published at an early date. 



