He thinks it is difficult to understand why the men, especially the athletes, should have returned afterwards practically to their former diet, considering the improved physique they obtained, and can only explain this on the assumption that they were obsessed by an irresistible craving for more protein food. He controverts the statement that large quantities of protein food are liable to damage the kidneys, and points out that athletes who habitually consume excess of protein during training have no difficulty in excreting all the waste products without harm, nor are they liable to be hampered in their activities or grace of moment by gouty or rheumatic infirmities. When large quantities of pure water are administered to patients there is no obstacle to its excretion by the kidney. It is indeed more reasonable to suppose that a more concentrated urine of high specific gravity would impose greater strain upon the kidneys. The lungs have no difficulty in augmenting their excreting capacity, and the kidneys are not likely to be in a different category.

He proceeds to give details of other investigations, from a consideration of which one must infer that a reduction of the protein content is directly hostile to the best interests of the body. These observations cover a wide field, are comprehensive in character, and quite emphatic in their deductions. Experiments upon carnivorous, omnivorous, and herbivorous animals are cited which demonstrate the danger to life and health attending the diminution of the protein ration. Munk, Rosenheim, and Jagerroos have all concluded that a lessening of the protein in the diet of the dog is invariably attended by rapid exhaustion, and if unduly persisted in, death supervenes quite suddenly. This they attributed to a degradation of the epithelial cells of the intestinal canal inducing a loss of the power of absorption, as well as to a diminution in the secretion of the digestive fluids. These bad results are ascribed by Chittenden to the monotony of the diet employed, and he has with the assistance of his coadjutors at Yale conducted a series of experiments, the results of which controvert the conclusions arrived at by the investigators just mentioned. The dogs remained in good health, maintained their strength, increased their weight, and continued in nitrogenous equilibrium on less than two-thirds of the protein which caused the death of the animals under Munk's observations. He is at present engaged in further researches on the effects of low-protein diet on dogs, and we await with interest the publication of his results.

The flesh of hogs fed on either an excessively low or an excessively high protein diet is characterised by "softness." Reduction of the protein, as, e.g., by administering Indian corn, produces pork of a very inferior quality, and an alimentary canal of abnormal friability when dressed. These animals must be indeed very susceptible to any alteration in their diet, for a slight increase in the protein ration results in their return to a normal growth and a healthy condition.

Another well-known experiment made at the Experiment Station of the University of Minnesota by Professor T. L. Haecker is detailed, in which two groups of ten cows each were fed, the one on the usual diet and the other on a low-protein ration, for three years. At the end of two years the latter group had lost weight, but their physical tone was good. In the third winter, however, they began to fall off in flesh and their coats became exceedingly harsh, a recognised indication of malnutrition, so that it became necessary to increase the protein. This experiment should be contrasted with that recently reported from the University of Wisconsin on the protein requirements of dairy cows. The authors state that, on an average, rations of medium protein-content proved more economical and more effective both absolutely and relatively for dairy cows than rations of a high protein-content, such as approach the German standard rations.

Benedict rightly infers from these experiments that "men may subsist for six months or even longer on a low-protein diet without serious disturbance of the alimentary tract or of nutrition in general." He points out, however, that from an examination of the faecal evacuations of the soldier squad there was considerable evidence of wide variability in the quantity of the faecal nitrogen, as much as a quarter of the amount ingested being excreted instead of about one-ninth, which he himself had previously determined was the normal amount.

General Considerations

Other objections that might be urged against a low-protein diet relate to a comparison of the general dietetic habits of successful nations and races of people as compared with the illiberal allowances of inferior and unsuccessful races, and it is notable in this connection that Italians from the northern provinces, as well as the Japanese, by adopting a high-protein ration on their advent in America, markedly increase their working capacity. As I mentioned in the last chapter, it has been frequently asserted that the Japanese subsist on a low-protein vegetarian regime, and the performances of their jinricksha men are truly cited as wonderful feats of strength. But Professor Oshima, who made a very careful study of this subject, clearly established the fact that the diet of the jinricksha men contained a much larger quantity of protein than was proposed in the Voit diet. 'Much more recently Colonel Melville conducted an experi-/ment on twenty infantry soldiers, to determine, on a measured quantity of mixed food and a definite amount of work, the effect on the health and well-being of the men. For six consecutive days they marched 13 miles each day, then had one day's rest and marched 13 miles on five further consecutive days. During the first week each man consumed on an average 190 grams of protein, 510 grams of carbohydrate, and 58 grams of fat each day, and during the remaining part of the experiment 145 grams of protein, 450 grams of carbohydrate, and 110 grams of fat daily, making a daily average for the whole period of 168 grams of protein, 480 grams carbohydrate, and 84 grams of fat. This amounted in all to 3,481 calories, or, with 10 per cent. deducted for loss in the excretions, an available daily food supply of 3,140 calories. Briefly, the result of the experiment showed a gain of weight for the first three days, which could only have been attributable to a retention of water, of which a liberal allowance was drunk, a maintenance of the weight for the next two days, and then a steady fall of weight for the rest of the time. The average fall of weight per man was 1.22 kilos, or a calculated daily loss of 60 grams of fat and 170 grams of flesh, the effect of which was visible in the sunken eyes and hollow cheeks of the men. Quite evidently the men had been subsisting on an amount far short of their nutritive requirement, which was estimated by Melville at 190 grams of protein, 480 grams of carbohydrate, and 150 grams of fat per man per day, an amount well over 4,000 calories.