[36] A guinea-pig requires about 80 to 100 c.c. of fresh milk per diem to protect it from manifest scurvy, if this constitutes its sole antiscorbutic quota. There is, however, a great difference in the amount of milk which individual animals consume, so that if we furnish it _ad libitum_ our results will be irregular, one animal developing scurvy and another remaining well, according to individual appetite. Frequently guinea-pigs take very little milk at first, learning gradually to drink more and more, so that it comes to pass that only after some weeks, if at all, they take sufficient to afford protection. Under these circumstances, if any addition is made to the dietary after the experiment has progressed, the lack of development or cure of the scurvy may mistakenly be attributed to this newly introduced factor, whereas, in point of fact, it is due to an increased consumption of milk. These reports stimulated renewed interest in experimental scurvy, suggesting new aspects, and therefore directing attention to points requiring investigation. There were no data at hand on the consistency of the contents of the bowel in guinea-pigs, so that it was necessary to make appropriate observations in normal and in scorbutic animals. It was soon reported by various workers, Rappleye, Cohen and Mendel, Hess and Unger, and others, that there was no definite relationship between the occurrence of scurvy and impaction of the cæcum. In Fig. 8 we see portrayed the stool output of a guinea-pig during the period in which it was developing scurvy, and during a subsequent period when it was being cured by means of orange juice. It is clear that there was no significant variation in the output during these divergent periods. A similar conclusion was arrived at by those who examined the cæcums of animals postmortem. Cohen and Mendel write: "Summarizing our experience with nearly one hundred scorbutic animals, we conclude that actual impaction of the fæces in the cæcum occurred in about one-quarter of the cases, and visible damage to wall, _i.e._, congestion or hemorrhage, or impaction, or both, was found in perhaps half of the cases. It should be noted that this statement covers all the diets we have tried." We concur in this conclusion. Not infrequently we found the cæcums of markedly scorbutic pigs to contain semi-fluid fæces, the consistency of its contents depending on the character of the diet, quite apart from its adequacy and lack of the antiscorbutic factor. For example, a diet rich in milk, containing 100 c.c. or more, led to the formation of rather solid fæces; if oats were added to the milk diet the fæces in the cæcum were found to be still more solid, and this portion of the gut more often impacted. This condition could be detected by palpation even during life. Just as we encountered scorbutic animals on an oat, hay and water diet, who had semi-fluid fæces in the cæcums, so we met with others which were on a milk diet, and showed no signs of scurvy,