It is clear that on the one hand the theoretical reasoning from the fact that an excess of nitrogen is so rapidly eliminated from the body, and on the other the result of experiment, show-that about 50 grammes of protein is sufficient to supply the needs of a man of average weight and to keep him in health for a considerable period of time. It is claimed by some, for instance, by Chittenden, that health is improved by such a diet, and the view has been put forward that a larger amount of protein is harmful, and involves the doing of an unnecessary amount of work by the kidney in excreting the extra nitrogen, and may cause disease. If this is so we ought to reduce by more than half the quantities of meat and eggs usually taken. The protein contained in the remainder and in the bread, puddings, and other foods would be ample to supply all our requirements. Such a course would also effect a considerable reduction in the cost of food.

No doubt a great many people eat meat to excess. That is, however, outside the question which is, is 100 grammes of protein per day, being that contained in a moderate English dietary, of benefit or the reverse?

In discussing this question we may first point out that it cannot be said to be proved that any benefit to health results from a low protein diet. Most physicians know some who have tried it and given it up, having thought after some months that they were less energetic than before, and were more liable to contract colds and other infections, and others who have convinced themselves that they are better on it, but have not convinced those who knew them on the old regime and on the new. Chittenden's subjects have many of them returned to an ordinary diet. This may be regarded by some as a backsliding, a moral delinquency, a yielding to the pleasures of the palate, but by others as evidence that a larger protein ration was more satisfactory to them. The influence of suggestion must also be borne in mind in considering the above figures. If a man knows that his nitrogenous metabolism is being investigated he is likely to unconsciously limit his protein food. The fact that most of the nitrogen of a liberal protein diet is rapidly eliminated is certainly of great interest, but cannot in the present state of our knowledge be said to prove that the protein which contained it was redundant. The suggestion that the excretion of this nitrogen is harmful to the healthy kidney is totally without foundation. Carnivorous animals live entirely on protein and fat, and no evidence has been produced to show that their kidneys suffer on this account. It has been said by Chittenden that the universal choice of a liberal protein diet by man cannot be taken as evidence that that food-stuff is beneficial in quantity any more than the very general taste for alcohol proves it to be a desirable food-stuff. This analogy is a false one. Alcohol, though physiologically capable of acting as a food-stuff, is not at all comparable to either protein, carbo-hydrate, or fat. Before it can be classed with them, it must be shown that a man or animal can exist for long periods of time, in health, while drawing a large proportion of his energy needs, say a third, from alcohol. This has been shown for protein innumerable times.

We may now consider some experiments on animals bearing on this question. Rosenheim, in 1891 and in 1893, found that dogs fed upon meat, fat and rice in which the proportion of meat, 2 grammes of protein per kilo with a total food value of 110 calories per kilo, was low for this animal, suffered from digestive disturbances, such as vomiting and poor appetite, which were relieved by giving meat, and from an incomplete absorption of the food, and, after several months, died in an apathetic condition. Bad results had previously been obtained by Munk in dogs fed upon a low protein diet. Hagemann also reported similar observations. Jagerroos believes that the deductions made from these experiments are not valid, and kept two bitches 6 1/2 months upon a low protein diet without any disturbances of digestion. He states that 1.2 grammes of protein per kilogramme is sufficient quantity for these animals; dogs, as we have seen above, require mere food per kilo than man, on account of their greater relative surface, and this must be allowed for in comparing the figures. Jageroos' two bitches, however, succumbed rapidly to an acute infection contracted after the premature delivery of one of them, a fact which may have no bearing on the point at issue, but does not inspire confidence as to their power of resistance to disease. All these experiments on dogs are open to the criticism that the dog is a carnivorous animal and cannot be expected to do well on a diet so far removed from its natural habit. The pig, being like man omnivorous, is free from this objection. Benedict, in his valuable paper on this subject, quotes the following experiments. Shutt found that when the pig is fed upon Indian corn, which contains a low proportion of protein, the pork is of inferior quality, which is improved by the addition of skim milk, or beans. Pork butchers find that the intestines of hogs fed on low protein rations tear more easily when the carcass is dressed. A bad quality of pork is also produced if too much protein be given. Skinner also showed that in pigs a low protein diet of corn-meal caused impairment of the digestive capacity, especially if continued for a long time. The diet resulted in " poor appetites, light bone, deficient development in valuable portions of the carcass and a general state of unthrift, as shown by the hair, skin, and hungry look of the animals." The addition of a small amount of protein to the food resulted in a natural growth, and a healthy condition. In herbivorous animals, Benedict goes on to say, we should expect less marked effects, as they are accustomed to a low protein diet. Haecker, however, has found that cows do not thrive on a diet poor in protein. Ten cows were fed on their natural food and ten others on a diet containing less protein : for two years very little difference could be observed, but in the latter part of the third winter the second group of animals became thin and their coats harsh, and in two months it became necessary to increase the protein. These observations are not directly transferable to man, but are of considerable importance. Especially should it be marked that the effect of a low protein diet on pigs and cows was a very gradual one, but one which, nevertheless, led in time to a general deterioration of health. They do not give us information as to the level of daily protein intake suitable for a man, but they teach definitely, first, that in case of doubt it is better to be above that level than below it, and secondly, that experiments on man to be conclusive and show that men can live and work on a low protein diet as well as on the usual one, should last for several years, since in animals the reduction of the protein in the diet below that to which they are accustomed is harmful, if continued over a sufficiently long period, which in the case of cows is two or three years.