Among the vegetable foods, therefore, the distribution of fat-soluble A appears to be associated in a remarkable way with yellow pigmentation. The parallelism seemed to Drummond (6) sufficient to warrant an investigation to determine whether this substance might be one of the yellow pigments of plants. He brought rats into a condition in which they were suffering from fat-soluble A deficiency, and then fed them pure and impure preparations of the pigment carotin, which, together with xanthophyll, constitutes the pigments to which are due the yellow color of many vegetable products. The animals did not respond with growth and did not remain free from ophthalmia, as they would have done had a fat containing fat-soluble A been administered. The results indicated, therefore, that carotin is not identical with fat-soluble A.

About the same time Steenbock (7) expressed the belief that the fat-soluble A is one of the yellow plant pigments, although he and Boutwell (8) had earlier stated that this vitamin is not carotin. In several later papers Steenbock has presented data showing the remarkable association of this substance with the yellow pigment of plants. He has shown that when carotin and xanthophyll are separated, fat-soluble A accompanies carotin, whereas the xanthophyll fraction is poor in it.

Drummond and his co-workers have energetically pushed their investigations in the testing of the hypothesis that there is a relation between yellow pigmentation and content of fat-soluble A in food-stuffs. They tested twenty-four different fats and oils which were pigmented to widely differing degrees for their efficiency as sources of this dietary essential, and could discover no relation between vitamin content and color (9).