206 PRINCIPLES OF ANIMAL NUTRITION. 
kgm. per day) done by turning an ergostat. As a result of the in- 
creased supply of proteids in the food the proteid metabolism in- 
creased promptly, reaching its maximum upon the fifth day, when. 
it very slightly exceeded the supply. From that time, however, 
it decreased gradually during the remaining thirteen days of the 
experiment, so that a gain of proteids by the body resulted, which 
was still in active progress when the experiment was discontinued. 
Counting from the time when the proteid metabolism reached its 
maximum the average gain of nitrogen per day was 
First five aySiciccciasg vaca seas ce swe 1.28 grams 
Last five days.............-..000 eee 2.05 “ 
Average of all..................00...... L475 “ 
The author pana ‘liad 22 per cent. of the proteids added to 
the food was stored up in the body. In a previous similar experi- 
ment without work it was found that only 16 per cent. was thus 
stored. - 
Two respiration experiments with the Zuntz apparatus were 
made during the work. The difference between their results and 
those of similar experiments during rest was used as the basis for 
computing the actual amount of energy metabolized in the body for 
the performance of work. This was found to be equal to 0.0100875 
Cal. per kgm. external work, which is equivalent to 171.5 Cals. for 
the whole daily work of 17,000 kgm. Assuming the original ration 
to have been a maintenance ration, Bornstein computes that the 
portion of the added proteids which was actually metabolized was 
insufficient to supply the energy necessary for the work done and 
that some of the fat of the body was drawn upon. The loss in 
live weight was found to agree with this assumption. 
The above investigations seem to show, not only that work may 
be done without increasing the proteid metabolism but that it may 
actually result in diminishing it, a fact which appears in harmony 
with the common observation that the tendency of exercise is to 
build up the muscular tissue. 
Summary.—While the results which have been cited are not in 
all respects conclusive, and while further investigation is required 
to fully elucidate the relations of muscular exertion to proteid metab- 
olism, the following general conclusions seem to be justified by the 
evidence now available: 
