When McCollum and Davis first observed the necessity in the diet for growth of something which could be furnished by certain fats but not by others or not by non-lipin components of the diet, they were impressed by the experiments reported by Osborne and Mendel (2). The latter appeared to have demonstrated that a pregnant female could cause the development of normal young, and thereafter secrete milk containing everything necessary for their normal nutrition, while herself confined to a diet which lacked, so far as could be seen, not only the amino-acid lysin, but likewise any source of the dietary factor now designated as fat-soluble A. The experimental diet consisted of gliadin, protein-free milk, starch and lard. Osborne and Mendel's experiment in reproduction on this food mixture seemed to prove that the synthesis of vitamins by the mammary gland was possible. It appeared, therefore, that they could be produced by the mother for the preservation of the species, but not by the young for their own preservation. This assumption was at that time believed to be supported by satisfactory experimental data (3). Later studies by McCollum and Simmonds have clearly demonstrated that there can be no adequate nutrition of the young while nursing a mother whose diet is deficient in fat-soluble A, or any other substance which the young rat requires for its nutrition in after-weaning periods.

It has been shown by Osborne and Mendel (4) that body fat of beef cattle contains a small amount of fat-soluble A. It seems certain that the body fats of any animal fed for a time a diet rich in this substance will serve as a reserve supply of this factor, which the mother can draw upon and secrete into the milk. In other experiments of McCollum and Simmonds definite evidence was secured which proved that fat-soluble A is not present in the milk unless it is furnished in the food of the lac-tating animal. The presence of some fat-soluble A in the tissues of the mother makes it especially difficult to obtain milk entirely free from this substance.

Through similar experiments with diets which contain fat-soluble A but not water-soluble B, evidence was also secured that for a time the mother is able to secure this dietary factor from her reserve supply, but none of the growth curves indicated that this substance is present in adequate amounts in the milk when the diet of the mother is lacking therein. It seems certain that neither of these dietary essentials is present in abundance in milk unless the diet of the lactating animal serves as the source.