Secondly, the elimination of nitrogen during the earlier days of fasting is governed in large measure by the character and extent of the diet on the days just preceding the fast. This is well illustrated by some experiments conducted by C. Voit on a dog. In the first series of experiments, the dog received daily 2500 grams of meat prior to fasting; in the second series, 1500 grams of meat were fed daily before the fast; while in the third series, a mixed diet relatively poor in proteid was given. The following figuresl show the amounts of proteid used up by the dog (calculated from the nitrogen excreted) each day of the fasting period, under the different conditions:

We see very clearly in these experiments the effects of the large quantities of proteid fed on the destruction of proteid in the early days of fasting. When the body is rich in proteid from food previously taken, the metabolism of nitrogenous matter is very large at first, as in the first series of experiments. Indeed, in this series, even on the fifth day of fasting, the amount of proteid metabolized was larger than on the second day of the third series. We have here a forcible illustration of the physiological axiom that ex cess of proteid, matter in the tissues, or in the blood, stimulates proteid metabolism; and it affords convincing proof of the contention that in the first days of fasting the output of nitrogen, or the amount of proteid used up, will depend in large measure upon the proteid condition of the body at the time of the fast. Equally noticeable is the fact that there comes a time - the Bixth day in the above experiment - when the nitrogen output reaches a common level, irrespective of the previous proteid condition of the body. Further, it is easy to see that the greater loss of nitrogen, i. e.t the large breaking down of proteid during the first few days of fasting, in those cases where proteid food has been freely taken (suggests) the existence in the tissues of two forms of proteid.. We may term them, following the nomenclature of Voit, as circulating and morphotic, or tissue, proteid; or, we may designate them as labile and stable forms of proteid. In other words, following the usually accepted view, this circulating or labile proteid represents reserve or surplus material which is easily decomposed and hence rapidly gotten rid of, while the stable proteid is more slowly oxidized, and its metabolism may be taken as representing more nearly the real necessities of the body. However this may be, it is plainly manifest that the nitrogen output, meaning the metabolism of proteid matter, during hunger or fasting is modified by a variety of circumstances, notably the previous nutritive condition of the body as regards both fat and proteid. It is hardly necessary to add that the amount of muscular work performed is another factor of importance in this connection. Fat in the body represents inert material stored up mainly for nutritive purposes; hence, in hunger it is used largely, and serves to protect more important tissues. Thus, experiments have shown that in long periods of fasting, adipose tissue may be consumed to the extent of 97 per cent of the total amount present, while the heart and nervous tissue will not lose over 8 per cent of their tissue substance. The influence of tissue fat upon the consumption of proteid during hunger can thus be fully appreciated.

1 Expreaied in this form from Voit'f figures by A. Magnus-Levy. Loc cit, p. 811.

The output of carbon during fasting may be illustrated by the following experiment1 made upon a young man, the nitrogen data being included for comparison, and likewise the intake of food, in terms of nitrogen and carbon, preceding the fast and for two days following the fast The fasting was of five days' duration.

On the non-fasting days, the intake consisted of an ordinary food mixture of proteids, fats, and carbohydrates, with a small addition of alcohol. The point to be emphasized hero, however, is that the carbon-content was more than sufficient to meet the needs of the body. Thus, it will be observed that on all three of the days when food was taken, the income of carbon was far in excess of the output. In other words, on the day preceding the beginning of the fast the body stored up 135 grams of carbon, and on the day following the fast the body retained 169 grams of carbon to help make good the loss. Similarly, the amount of proteid food taken in on the day prior to the fast was considerably in excess of the needs of the body, 5.1 grams of nitrogen equivalent to 31.8 grams of proteid being stored for future use. Plainly, the man was not in either carbon or nitrogen balance prior to the fast, but was taking far more food than the needs of the body called for. This fact may be emphasized by noting that the total fuel value of the daily food, plus the fuel value of the alcohol, amounted on an average to about 4200 large calories, while the fuel value of the material metabolized on the feeding days averaged only 2500 calories. Looking at the figures showing the output of carbon, as well as of nitrogen, during the fasting days, it is to be seen that in the early days of fasting, the metabolism of the body tends to remain at a fairly constant level, especially when figured per kilogram of body-weight.

1 Taken from Johansson, Landergren, Sondén, and Tiegerotedt: Beiträge zur Eenntni8s des Stoffwechsels beim hungernden Menschen. Skandinavisches Archly fur Physiologic Band 7, p. 29.

2The carbon output represents the total carbon of the expired air, urine, and excrement.