These are the results, collectively, so frequently used to point a moral for man: Beware of the possible danger of reducing the consumption of proteid food below the commonly accepted dietary standards! We must admit, however, that there is a woeful lack of agreement in these results, and it is difficult to prevent a shadow of doubt from creeping over us as we try to depict for ourselves the way in which a low proteid ration exerts its deleterious effect on dogs. I do not believe that radical changes in diet, whether they involve increase or decrease in total quantities, or in specific elements of the diet, can be made suddenly without danger of some disturbance of the gastro-intestinal tract or other parts of the economy, either in dog or man. It is reasonable to believe also that a high proteid feeder, like a dog, with his more limited dietary, will be far more sensitive to great changes than omnivorous man with his wider range of foodstuffs. Moreover, there is just as good ground for believing that in any animal, excess of proteid is as dangerous as a low proteid diet. Too great a disturbance in the nutritive balance, whether it involves excess or reduction in the amount of a given foodstuff, is liable to be attended with serious disturbance in any sensitive organism.

In illustration of these statements, we have some recent results obtained by Watson and Hunterl upon the influence of diet on growth and nutrition. These investigators find that young rats - two and a half months old - when fed upon a diet composed exclusively of horse-flesh, which is chiefly proteid matter with some fat, succumb very quickly, for some reason. Of fourteen young rats fed on this meat diet, six died on the third day. On the morning of this day, as the authors state, "the rats appeared to be in their usual health, but an hour after feeding one of them was lying on its side apparently unconscious. In a few minutes others were affected. They appeared to be paralyzed, they felt cold to the touch, exhibited symptoms of tetany, and speedily became unconscious. Six succumbed within half-an-hour. Of the remainder, some showed similar symptoms, although in less degree, and they recovered when the diet was changed to bread and skim milk." After two days of the so-called normal diet, composed of bread and skim milk, the remaining eight rats were again placed on an exclusive meat diet. They appeared now to have acquired a certain degree of immunity, for although they exhibited symptoms of deranged nutrition, these were gradually recovered from and they gained in weight. At the end of the eighth month, five of the animals were still alive and in apparent good health, but their growth was permanently stunted. With an exclusive diet of ox-flesh, young rats were much more liable to thrive, although their growth was distinctly retarded.

1 Chalmers Watson, M.D., and Andrew Hunter, M.B.: Obterrations on Diet. The Influence of Diet on Growth and Nutrition. Journal of Physiology, Vol. XXXIV, p. 112, 1906.

This difference in the behavior of the animals towards the two forms of proteid food is to be attributed to the fact that ox-flesh contains more fat than horse-flesh, and consequently the diet with this form of meat was less exclusively proteid in character. Further, there were some indications that horseflesh is less digestible than ox-flesh. Another fact, showing the far-reaching effect of a distinctly unphysiological diet, is the marked influence of pure meat food on the progeny. Thus, of 98 rats born of meat-fed parents only 19 were alive at the end of two months, while of 97 young born of bread and milk-fed rats, 82 were alive and in apparent health at the end of the same period.

As illustrating how foods that have, superficially at least, approximately the same chemical composition may react differently in the animal body we have the observations of Watson on rats fed with porridge, made by boiling oatmeal with water and skim milk, as contrasted with a diet of bread and skim milk, the two diets having essentially the same composition. Of fourteen young rats fed exclusively on porridge, all, with the exception of two that were withdrawn, succumbed within five months, while the bread and milk-fed animals thrived as usual. Adult rats, however, can live for prolonged periods and maintain their weight on a porridge diet It is believed that the difference in the behavior of young rats to these two closely allied forms of diet, is due to a difference in the digestibility of the food, the porridge being presumably less readily digested by the young animals than bread. With the more fully developed digestive powers of the adult animals, however, this difference in availibility practically disappears as a potent factor in their nutrition. Finally, mention may be made of the fact that a pure rice diet, notably deficient in proteid, arrests the growth of young rats and leads to a fatal issue within three months, while adult rats placed on such a diet lose weight rapidly and die in about the same time. All of these facts bearing on the nutrition of animals quite remote from man have significance as showing how any wide departure from a physiological diet, for that particular species or type, may lead to very undesirable results, and they warn us not to be too hasty in drawing far-reaching conclusions and sweeping deductions from a few experiments with a given species of animal.

Recurring now to the experiments made with dogs, there is certainly suggested an element of danger in a low proteid diet, which, if the experiments are taken at their face value and the conclusions derived therefrom applied to man, needs careful consideration. Jägerroos plainly was not inclined toward the belief that a low nitrogen intake was the cause of the unfortunate results that attended his experiments. Still, his animals did die from some cause, and thereby his position was weakened. Munk and Rosenheim, on the other hand, from their experiments were apparently convinced that a low proteid intake was inimical to dogs, and it will be remembered Rosenheim concluded that "a daily ration containing even 0.32 gram of nitrogen per kilogram of body-weight, and with a fuel value of 110 calories per kilo, is not sufficient to maintain the animal in a condition of health." If this is really true, there is some ground for the arguments advanced by critical writers regarding the general subject of nitrogen requirements of man. The evidence and the arguments, however, have always seemed to the present writer frail and faulty; but recognizing the hold they have taken on physiologists and the way they are usually applied to man, I have attempted to test the matter experimentally under conditions which would yield trustworthy and conclusive results.