The small intestine of the rabbit was carefully separated from the mesentery and from the pancreatic gland, and the upper portion cut open and quickly washed free from any contained matter or adherent secretions, by repeated immersion in 0.5 per cent. salt solution warmed at 40° C. This was repeated until the tissue was quite free from all impurities, after which it was cut into small pieces and immersed for a moment in a 0.5 per cent. solution of sodium chloride containing 1.25 per cent. of peptone. The tissue was then carefully collected on coarse muslin, allowed to drain, and then quickly transferred to the flask containing the warm blood and peptone. This mixture was kept at 40° C. for two hours, a slow current of air being bubbled through the fluid during the entire period. At the expiration of this time the fluid was separated from the pieces of tissue by nitration through muslin, and then saturated with ammonium sulphate after the usual method for the separation of albumoses, etc. On now testing a portion of the clear filtrate for peptone by the biuret test, not a trace of a reaction could be obtained. The entire amount of proteid matter present was precipitated by the ammonium salt, thus showing that the peptone originally added had been completely transformed into something precipitable by saturation of the fluid with ammonium sulphate. That this transformation of the peptone was accomplished mainly through the action of the intestine, was shown by a parallel experiment, in which all of the above conditions were duplicated, omitting only the pieces of intestine. Here, however, on testing the filtrate from the ammonium sulphate-precipitate, a strong biuret reaction was obtained, thus proving the presence of at least some unaltered peptone. This experiment is almost a counterpart of one reported by Neumeister, and like his, testifies to the probability that the peptones formed in the alimentary tract, as a result of proteolysis, undergo retrogression through the agency of the epithelial cells of the intestinal walls during their absorption. I have tried similar experiments with deuteroproteose, notably with deuterocaseose, and have obtained corresponding results. The same method may be employed as that already outlined, although of course the deuterocaseose is in great part precipitated by saturation with ammonium sulphate. Still, this form of deuteroproteose, β deuterocaseose, as I have elsewhere noted, is very slowly precipitated by the ammonium salt. Consequently, it is an easy matter to demonstrate that this proteose, on treatment with the intestinal mucosa in the presence of blood at the body-temperature, is transformed into something completely and readily precipitable by ammonium sulphate; the filtrate from the latter failing to show any biuret reaction, although the corresponding control experiment without the intestine gives a bright violet color with cupric sulphate and potassium hydroxide.