The argument based on this evidence is that while a high proteid animal feels at once, or almost immediately, the deleterious effect of a reduction in the amount of proteid food, an omnivorous animal may be more tardy in manifesting the injurious action, which, however, is sure to follow sooner or later from any material reduction of proteid below the customary standards. In other words, man as a moderate proteid consumer can endure for a time even large reductions in the amount of proteid food, but eventually there will be manifested some of the disastrous results obtained with dogs. Here, we have a somewhat serious indictment, one that merits careful consideration. To be sure, it may be objected that between dog and man there is a wide gulf, and that there is no justification for assuming that these two types of animal life have anything in common. Still, the experience of many years has taught the physiologist that much light can be thrown upon the processes of higher types of life by a study of what occurs in lower forms, and on the subject of nutrition any one of experience would hesitate to cast out of court the evidence gathered from observation of what occurs among the higher animals. It will be the part of wisdom, therefore, to scrutinize somewhat carefully the character of this evidence obtained from a study of the behavior of dogs toward a low proteid diet.

The first series of experiments was made in 1891 by the late Immanuel Munk of Berlin, privat docent of physiology at the University, followed by further experiments in 1893.1 Four dogs in all were studied, The diet made use of was" fleischmchl" (dried meat ground to a powder), fat (suet), and rice boiled together with water. We may refer briefly to the details of one experiment. The dog weighed 10.4 kilograms, and at first was given a daily diet composed of 85 grams of rice, 29 grams of fat, and 30 grams of the flesh meal. This ration contained 30.3 grams of proteid, 31 grams of fat, and 66 grams of carbohydrate, with a total fuel value of 663 calories, or 63 calories per kilogram of body-weight. On this diet, there was at the outset a slight loss of body-weight, after which both body equilibrium and nitrogen equi-brium were practically maintained. After this preliminary period of three weeks, the day's diet was altered by replacing 15 grams of the proteid by 15 grams of rice, bo that the daily ration consisted of 15.3 grams of proteid (with 2.42 grams of nitrogen), 31 grams of fat, and 81 grams of carbohydrate, with essentially the same fuel value per kilo of body-weight as before. Later, the fuel value of the food was further increased by raising the amount of rice to 125 grams per day, the day's ration then consisting of 15.5 grams of proteid, 37 grams of fat, and 96 grams of carbohydrate, with a total fuel value of 780 physiological heat units, or 78 calories per kilo. On this diet, nitrogen equilibrium was maintained and the animal gained somewhat in body-weight. By the seventh week, however, Munk reports that the animal began to show signs of change; there was loss of appetite, absorption of the daily food was impaired, both proteid and fat failing in large degree to be utilized, while nitrogen equilibrium could no longer be maintained. This condition continued during the next week, aggravated by vomiting and accompanied by loss of strength and vigor. At the beginning of the tenth week of this low proteid ration, the animal was in a very poor condition, with complete loss of appetite, little inclination to take food, etc. On feeding a liberal diet of fresh meat, as much as 250 grams per day, with some fat (50 grams a day), the animal speedily recovered its appetite, and in a short time was in normal condition, absorption of food and utilization of the same being as complete as at the beginning of the experiment.

1 Ueber die Folgen einer aunichenden, aber eiweisnermen Nahrung. Ein Bating zur Lehre torn Eiweiubedarf. Virchow's Archiv fur pathologiiche Aoatomie and Phvtiologie, Band 132, p. 01.

It is not necessary to give further details bearing on the three additional experiments. It will suffice to quote the general conclusions which Munk drew from the various results obtained, viz., that a low proteid intake in the case of dogs causes a loss of appetite, weakness, vomiting, etc., while body-weight and nitrogen equilibrium are difficult or impossible to maintain. More specifically, Munk's observations led him to state that for dogs of ten kilograms body-weight a daily intake of 0.255 gram of nitrogen per kilo of body-weight is not sufficient to maintain the normal condition of the body, even when the fuel value of the day's food amounts to more than 100 calories per kilo. In order to have the animal continue in nitrogen and body equilibrium, the daily food must contain at least 0.31 gram of nitrogen per kilogram of body-weight, with sufficient non-nitrogenous food to yield over 100 calories per kilo.

Let us now pass to the experiments made by Rosenheim,1 which were carried on at about the same date as Munk's. In the first experiment, the dog weighed 11.8 kilograms, and was fed daily a low proteid ration having a fuel value of 1447 calories and containing 2.825 grams of nitrogen. This ration was reduced in a short time to a still lower plane, viz., to 1066 calories and 2.526 grams of nitrogen daily. The food as then given was composed of 170 grams of rice, 50 grams of fat, and 25 grams of chopped meat, on which the dog gained weight and preserved nitrogen equilibrium. For six weeks, or thereabouts, the animal maintained its normal condition, after which it began to show symptoms of a general disturbance, with lack of appetite and weakness accompanied by a condition of icterus. Addition of meat extract to the diet to improve the flavor was without any appreciable effect. During the next two weeks, the condition of the animal steadily grew worse, although the body-weight remained practically stationary and nitrogen equilibrium was maintained. A week later, the animal died in a condition of exhaustion, without having manifested any symptoms of disturbed metabolism. There was found a marked catarrhal condition of the mucous membrane of the gastro-intestinal tract, with a fatty degeneration or metamorphosis of the glandular apparatus, but nothing sufficiently specific to account for the peculiar manner of death.

1 Theodor Rosenheim: Ueber den Gesundheitsschadigenden Einflnss eiweiat-armer Nahrung. DuBois-Reymond's Archir fur Physiologie, 1891, p. 841. Also, Weiterer Unterouchungen liber die Schadlichkeit eiweissarmer Nahrung. Pflüger's Archiv f. d. gesammte Physiologic, Band 64, p. 61,1893.