In order that I may illustrate the supplementary dietary relationship among certain of our naturally occurring foodstuffs, it is essential first that I offer convincing evidence that we are now in a position to devise experimental procedure which is adequate to reveal such relationships. Secondly, it must be shown that we possess sufficiently complete knowledge of the essential factors which operate in making a diet for a growing animal to enable us to interpret correctly the results of feeding experiments. This can be done perhaps best by briefly reviewing the work of the past decade which has been most significant in advancing our knowledge of the chemical complexes which are indispensable to the growing mammal. Previous to 1909 a number of attempts had been made to nourish animals on diets composed of purified protein, carbohydrates, fats and inorganic salts, usually with a small admixture of cellulose. Such attempts were attended