The proteins of cow's milk have been the source of much interesting discussion and controversy among pediatricians. Theories as to their digestibility are changing almost yearly. Whey proteins advocated by Rotch have given place to the "whole proteins" of Czerny, Keller and Walls (which bring forward the digestibility of skim-milk); while others have heated, peptonized and lately precipitated the proteins in further attempts to improve our knowledge and treatment. Up to the present the proteins of milk formulas have not received the attention we should expect. Protein requirements have painstakingly been estimated and the formulas prescribed, but no one has seriously questioned the accuracy of such mixtures. Several days during the summer of 1912, formulas made according to Finklestein's method were found on examination to contain proteins varying from 2.5 to 4.5 per cent. This instance of error led us to investigate the composition of home modifications and laboratory milk