So, on his return from Russia, he took advantage of the fact that an epidemic of infantile cholera had broken out in order to continue his former investigations of that disease. The numerous cases which he thus studied allowed him finally to establish the specific part of the _B. proteus_ as well as the similarity between infantile cholera and Asiatic cholera. This time he succeeded in contaminating, not only young anthropoid apes, but also new-born rabbits, and that not only through sick children's excreta, but by pure cultures of the _proteus_, which eliminated every doubt of the specificity of this microbe. Metchnikoff explained the contamination of children exclusively breast-fed, either by the presence of a carrier personally refractory, among the entourage, or by the transport of dirt, by means of flies, on the objects which infants so readily put into their mouths. He therefore advised preventive means of absolute hygiene and cleanliness, especially where suckling infants are concerned. During the year 1912, he studied the intestinal flora and the influence of divers food diets. He experimented upon the rat, an omnivorous animal whose mode of feeding resembles that of man. The rats were divided into three lots, of which one was kept to a meat diet, another to a vegetarian régime, and the third to a mixture of both. The meat diet was least favourable, and the best results obtained by the mixed food. These observations led Metchnikoff to the study of other problems intimately connected with the same question. He undertook a series of researches in collaboration with his pupils, MM. Berthelot and Wollman, on the conditions which cause the diminution within the organism of the toxic products of intestinal microbes. They found that the quantity of these products was very small in those animals which feed on vegetable or fruit containing much sugar, such as carrots, beetroot, dates, etc. This is explained by the fact that the products of the decomposition of sugar are acids which prevent the development of putrefying microbes. But the sugar, rapidly absorbed by the walls of the small intestine, only reaches the large intestine in a much reduced quantity, for it is only up to a certain point during its journey that the cellulose of vegetables, rich in sugar, protects that substance. The question, therefore, was to find the means of making it reach the large intestine in greater quantities. In the intestine of a normal dog, an innocuous microbe was found, the _Glycobacter peptonicus_, which decomposes starch into sugar.