FUNCTION OF.CAROTINOIDS IN PLANTS, ANIMALS 269 



carotinoids does not require these pigments for its growth or for the 

 reproduction of its kind. 



While the last experiment was in progress Drummond (1919) re- 

 ported the failure of pure crystalline carotin, fed at the rate of 0.003 

 per cent of the ration to improve the condition of albino rats suffering 

 from vitamin A deficiency; while Steenbock, Boutwell and Kent (1919), 

 on the other hand, were calling attention to certain new associations of 

 yellow pigmentation and vitamin A and were suggesting that the two 

 were at least associated in some way. Although the statement was 

 made that vitamin A "is not carotin," this was later retracted by Steen- 

 bock (1919) and the provisional assumption advanced that this vita- 

 min is one of the carotinoid pigments. In support of this assumption 

 Steenbock and his associates have published a series of papers showing 

 that a rather close correlation exists between carotinoid pigmentation 

 and vitamin A content of roots (Steenbock and Gross, 1919; Steen- 

 bock and Sell, 1922), maize (Steenbock and Boutwell, 1919a), leaves 

 (Steenbock and Gross, 1920) , and peas (Steenbock, Sell and Boutwell, 

 1921), as determined by feeding pigmented and colorless varieties of 

 these plant products to albino rats. However, in studying the extract- 

 ability of vitamin A from carrots, alfalfa, and yellow maize by fat 

 solvents, Steenbock and BoutwelPs (1920b) results show that highly 

 colored extracts l do not exhibit the vitamin activity which would be 

 expected if vitamin A is a carotinoid; and yellow maize, even after 

 extraction w r ith hot ether is shown to have lost very little vitamin A, 

 although there must have been a considerable loss of pigment. On the 

 other hand, especially favorable to Steenbock's theory was the finding 

 in this paper that the fat-soluble vitamin follows the carotin in the 

 application of the phase test to the unsaponifiable extracts from alfalfa 

 leaves. When one bears in mind, however, that the solvents employed 

 in the separation of the carotinoids by this method are respectively 

 exceedingly poor and very excellent fat solvents, it is not surprising 

 that the fat-soluble vitamin will follow the substance which goes into 

 the better fat solvent. 



When Steenbock, Sell and Buell (1921) attempted to obtain support 

 for Steenbock's theory among animal products the correlation between 

 pigmentation and vitamin content was so poor when comparing prac- 

 tically colorless codliver oil with butter fats of high and low color that 



1 Drummond and Zilva (1922) have substantiated this. They find that only "very 

 slight" growth in rnts is promotod !>y as much as 2 grams daily of the crudr oil 

 extracted from yellow maize by petroleum ether. This is a high proportion of the 

 diet of young rats. 



