Rubner has formulated the law that the expenditure of energy during starvation is diminished in the same proportion as the weight of the body, and it is notable that during the first few days of acute starvation the general metabolism undergoes little or no diminution. The body appears to use up the same amount of material, and makes inroads into its own tissues in lieu of other food supply. In a case under the observation of Voit and Pettenkofer, 2,362 calories were expended whilst being fed, and 2,320 calories during starvation. In other cases similar results were obtained, any increase, apparent or real, being due to the work of digestion and the necessary chemical changes in the food, which would not account for more than an average of 10 per cent. From 30 to 32 calories of energy per kilo of the body-weight are expended by fasting persons daily. During complete repose in this state it is possible to determine accurately the minimal metabolism for the individual.

For purposes of reference, it may be useful at this point to set forth in detail the maintenance diet for a man of 70 kilograms under varying conditions. Von Noorden gives the following figures: -

When these figures are decreased by less than 20 per cent., then the individual is in a condition of mild undernutrition, and when they are reduced by more than 60 per cent. a state of severe undernutrition exists.

During the first few days of deprivation from all food there is still a supply of glycogen in the muscles and the liver, but when this is expended the fasting man is compelled to live on protein and fat, and Tigerstedt has shown that on the fifth day 14.5 per cent. of the energy was derived from protein and 85.5 per cent. from fat. For this reason the respiratory quotient, i.e., the proportion of the oxygen intake to the oxygen output in the shape of carbon dioxide, falls from 1.0, which indicates the combustion of carbohydrates, to a lower figure, .707 indicating fat, and .809 protein. The more rapidly the value falls to this amount the smaller has been the supply of glycogen in the tissues. When it falls below the lowest figure it indicates that the carbon and oxygen, instead of being exhaled entirely by the lungs in the form of carbon dioxide, are excreted by the kidneys as acetone, diacetic, and β oxy-butyric acids. Lehmann and Zuntz have also succeeded in demonstrating that in perfect repose the decomposition of protein permits of the formation of a little glycogen in the liver and muscles. Hence the respiratory quotient is increased during muscular work.

As during fasting all the excretions continue in some degree, the body must depend for its pabulum on what it already contains, apart from any water which is consumed. The decomposition of protein or, as it is called, the katabolism of the protein during fasting indicates the extent to which the tissues have been used for providing energy for the body. Quite 90 per cent. of the nitrogen of the muscles and the glands is in the form of protein, the other 10 per cent. being extractives, and 99 per cent. of the total nitrogen of the blood is in the form of protein. It may therefore be concluded that the nitrogenous metabolism is practically identical with the actual protein metabolism, as the ratio of protein to extractives does not alter during fasting. Therefore, 90 per cent, of the total nitrogen excreted is derived from the breakdown of protein in the blood and tissues. This amount can easily be estimated from the excretion of nitrogen in the urine by multiplying the grams of nitrogen excreted by 6.25 - because there is in the average 1 gram of nitrogen in 6.25 grams of protein - after adding to it .2 for the daily nitrogen-content of the faeces. In every case the excretion of protein nitrogen means that the tissues of the body are being broken down, whereas when protein nitrogen is retained the bodily tissues are being supplemented. The amount of flesh disintegrated is calculated by multiplying the nitrogen excreted by 29.4, i.e., 4.7 x 6.25, as flesh contains 3.4 per cent, nitrogen, while muscle protein contains 16 per cent. nitrogen.

The amount of nitrogen excreted during the first three or four days of fasting is always greater than at any subsequent period, because it is dependent upon the protein intake and the breakdown during the days preceding the fast. This supplies a variable quantity of what is termed labile protein - which corresponds to the "circulating proteid" of Voit, and the excretion of nitrogen will be greater or lesser according to this amount. Voit declares that 8 per cent. of the stable protein or "tissue proteid" and 70 per cent. of the labile protein are decomposed during fasting, and if this be so the latter is all consumed during the first three days. In any case, it is certain that at the end of the third day only 3 per cent. of labile protein can be obtained from the blood, and hence from the fourth or fifth day onwards only stable or tissue protein can possibly be available.

In carnivorous animals like the dog, the fall of the protein katabolism is particularly well marked, although in man there may be an increase on the third or fourth day, owing to the glycogen in the tissues being all utilised. Thus in Breithaupt 10 grams of nitrogen were excreted in the urine on the first day, 9.9 grams on the second day, 13.3 grams on the third day, 12.8 grams on the fourth day, and only on the sixth day was the figure 99 grams again reached, demonstrating the obvious fact that he had been extremely well fed before his fast. After the first two or three days an average loss of 10-13 grams of nitrogen takes place for the next week or ten days, although it is relatively greater in the thin than in the stout, and in the small than in the large men. Fasting women also excrete from 20 to 30 per cent. less nitrogen than fasting men. Cetti, who weighed 57 kilos, excreted daily during the first ten days 10-11 grams nitrogen, equal to 60-70 grams of protein. A young man weighing 60 kilos lost during the same time 11 grams nitrogen daily, equal to 70 grams of protein. In neither case was there much adipose tissue. In fatty subjects the nitrogen excreted was found to be much less, e.g., Succi, who weighed 63 kilos and lost 11 kilos during a thirty days' fast, on the tenth day excreted 6.7 grams nitrogen, 49 per cent. of the output on the first day; on the twentieth day excreted 4.3 grams of nitrogen, 32 per cent. of the output on the first day; on the thirtieth day excreted 3.2 grams nitrogen, 30 per cent. of the output on the first day.

Estimating the average for men of 70 kilos as 13.7 grams nitrogen, or a loss of 90 grams protein, equal to 1.2 grams per kilo of body-weight, this would only account for something less than 400 calories of energy per day. It is manifest that energy must be obtained from some other source than the protein, which is only able to supply 15 per cent. of the metabolic requirements of man during starvation. The remaining portion is satisfied by the stored-up fat, and the greater the quantity of this, the greater proportion of energy will it supply. When the fat stored up as adipose tissue has been almost totally expended - and the remnant resists oxidation in a marked degree - then greater inroads are made upon the tissue protein. Hence a premonition of approaching death is supplied by a rise in the excretion of nitrogen, although at the very close of life this amount again sinks. The addition of carbohydrates and fat alone to the diet will always reduce the excretion of nitrogen, thus displaying their protein-sparing qualities.

In herbivorous animals, at the beginning of a fast there is an active and decided increase of the nitrogen excreted in the urine, because, being compelled to use up their own tissues, they become, for the time being, carnivorous. In the case of Voit and Pettenkofer already mentioned, 1,999.5 calories were evolved from fat and only 320 from protein. The longer the fast is continued the smaller is the protein decomposition - the valuable protein being saved at the expense of the less valuable fat. During the progress of a fast the share of protein in the total metabolism amounts to from 7 to 17 per cent., calculated in calories. This is, of course, derived from the tissues, and as the proportion is much less than the normal, it probably indicates that tissue protein is not so easily decomposed as food protein, of all food-stuffs the most freely disintegrated, although it is doubtless also a mark of the value placed by the body on its own tissue.

The protein consumed is derived chiefly from muscle, although the liver, the pancreas, and other glandular organs contribute a share, as they are diminished in size. Voit has shown, however, that this is not necessarily due to any structural alteration, because they are prone to regain their size rapidly when water is administered. When actual loss of tissue substance has been sustained sufficient to modify the structure of the organ, reconstruction is, even in favourable circumstances, a slow process.

The exhalation of carbon dioxide from the lungs is in direct proportion to the body-weight and to the work done, and in inverse proportion to the surrounding temperature. A man weighing 71 kilos excreted on the first day 201.3 grams carbon by the respiration and 5.8 grams carbon by the urine, equal to a consumption of 78 grams of protein and 215 grams of fat, or about 370 grams of flesh. During the first working day 75 grams of protein and 380 grams of fat, equal to nearly 478 grams of flesh, were lost. During the first day 760 grams oxygen were inspired and 889 grams carbon dioxide were exhaled, while during the work day 1,072 grams oxygen were inspired and 1,777 grams carbon dioxide were exhaled.

In another case weighing 70 kilos, on the second day 8 grams nitrogen and 3.7 grams carbon were found in the urine and 180.9 grams carbon were exhaled by the lungs, equal to 50 grams of protein (235 grams flesh) and 208 grams of fat. In these calculations it should be remembered that for every gram of nitrogen in muscle protein there are 3.28 grams of carbon, and that any excess of carbon in the excretions signifies fat destruction, fat containing 76.52 per cent, of carbon.

It should be noted that during fasting work has no influence in increasing protein metabolism, but enormously increases the combustion of fat. This, of course, has a direct effect on the excretion of carbon dioxide, which is reduced to about 20 grams per hour during sleep, as compared with from 30 to 40 or more grams per hour, according to the severity of the occupation, during the daytime. Sitting up in bed or even a moderate degree of restlessness or nervous activity suffices to increase the metabolism, and therefore the output of carbon dioxide.