The investigation of Palmer and Kennedy, referred to above, appears to add the last word to this discussion, and to leave no further room for doubt that the presence of liberal amounts of fat-soluble A in certain yellow pigmented vegetable tissues, and its absence from other varieties of the same species not possessing yellow color, is fortuitous. These investigators succeeded in demonstrating that approximately normal growth and reproduction can be secured in the albino rat on rations free from carotinoid pigments but rich in fat-soluble A.

They first made a critical study of the presence of carotinoid pigments in the albino rat and then instituted growth experiments with this species, using colorless ewe's milk fat as the sole source of fat-soluble A. In other experiments they used, instead of ewe milk fat, colorless egg yolk as a source of this factor. The liver, blood, spleen, suprarenals, adipose tissue, ovary and the fat of milk recovered from the stomachs of very young rats were examined with every precaution for thoroughness to determine the presence of carotinoid pigments. The rat milk was absolutely colorless, as was also the fat separated from the adipose tissue. The fat of the blood, suprarenals, spleen and ovaries, and also the liver tissue of new-born rats was entirely devoid of color. The liver of adult rats fed a diet containing an abundance of plant carotinoids, yielded an oil which was yellowish, but the pigment did not give any of the tests characteristic of carotin. This substance cannot, therefore, be regarded as in any way related to those plant pigments with which fat-soluble A may be concerned.

The milk of the ewe is almost colorless, but gives a very faint test for plant carotinoids. Nine per cent of the diet consisted of this fat, and it was estimated that the ration contained thereof but 0.127 parts in a million. The diet employed consisted of purified casein, a suitable salt mixture, agar, ewe milk fat, and dextrin carrying the alcoholic extract of ether extracted wheat embryo or germ to serve as a source of water-soluble B. The latter proved an inadequate source of the factor B and was afterwards supplemented by 2 per cent of baker's yeast.

On this diet the rats grew fairly well during two and a half to three months, and three females each produced litters of eleven young. Two of these mothers were able to nurse their litters, which were reduced to six in each case. The young grew in an essentially normal manner. On a similar diet in which 15 per cent of colorless egg yolks served as the sole source of fat-soluble A, six rats grew well, two increasing in weight from about 70 grams to about 230 grams. Two females produced each a litter of young which were normally nourished.

Palmer and Kennedy also pointed out certain numerical relations between the carotin content and fat-soluble A efficiency of various foods. These varied within very wide limits. Thus, 15 per cent of their egg yolks containing no carotin, were at least equivalent as sources of fat-soluble A to 85 per cent of yellow maize, although the latter amount furnished 1,400 parts per million of carotin. Five per cent of ewe milk fat, they estimated to be equivalent as a source of fat-soluble A to 1 per cent of dry spinach. The carotin content of one million parts of rations containing these amounts of materials would amount to 0.073 and 160 parts per million, respectively. The second of these two diets containing about the same amounts of fat-soluble A would contain about twenty-two hundred times as much pigment as the former.