204 PRINCIPLES OF ANIMAL NUTRITION. 
food consumption, digestion, and metabolism of three bicyclers 
during a six-day contest.. They find that, in spite of an apparently 
liberal diet containing large amounts of protein, all three riders 
lost considerable proteid tissue during the race. The conditions of 
the investigation were not such as to permit of a determination 
of the sufficiency of the food consumed, but the computations by 
Carpenter of the actual amount of work done seem to render it very 
probable that the body fat must have been drawn upon to a con-. 
siderable extent. 
RECAPITULATION.—TLhe investigations above cited seem to show 
beyond a doubt that when work is performed upon food less than 
sufficient to maintain the body and supply the amount of energy 
required for the work the proteid metabolism is somewhat increased. 
Whether the converse of this is true, namely, that when the 
food is sufficient such an increase in the proteid metabolism does 
not occur, is not so clear, for the reason that in most, if not all, of 
the cases we have no adequate data as to the sufficiency of the 
food. It is plain, however, that the question is not so easily inves- 
tigated as might appear at first sight, and that the final solution of 
the relations of work to proteid metabolism can only be reached by 
means of investigations in which the total metabolism both of matter 
and energy is determined. 
GaIN OF PROTEIDS DURING Work.—Caspari and Bornstein have 
recently made further investigations into the possibility of a gain 
of protein as a result of work which was mentioned above in con- 
nection with Pfliger’s experiments. 
Caspari * experimented upon a dog which received an amount 
of food computed to have been fully sufficient for its maintenance 
and to supply energy for the work done. Furthermore, a consider- 
able portion of the non-nitrogenous nutrients of the ration, consist- 
ing largely of carbohydrates, was given shortly before the work was 
done, while in some cases additional sugar or fat was given at that 
time. In the first experiment, work was performed upon three 
successive days. Upon the second of these there was a consider- 
able increase of the urinary nitrogen, but upon the third its amount 
fell below that of the rest period. The average for the three days 
of work was almost exactly equal to the value found for the last 
* Arch, ges, Physiol., 88, 509. ‘4 
