THE UTILIZATION OF ENERGY. 541 
gain of nitrogen, so that Wolff estimates that the work might have 
been increased to 350 revolutions. In Period II (23 days) the 
daily work was increased to 450 revolutions and the same behavior 
was observed, while a further increase to 500 revolutions during 
the last ten days checked the gain of nitrogen without causing a 
decrease in live weight. Taking 350 and 500 revolutions respec- 
tively as representing the maximum amount of work that could be 
done on the two rations, the equivalent of the oats added may be 
computed as follows: 
Equivalent 
Revolutions. Work, 
Kem. 
Perigd: UT. cg e ia wtalg anise Cas taed euiding’y yeah eo Ae 500 1,030,687 
OE cs Mataa's tials x hate Sc meer be Sioa Mote de tae haiioee 350 722,678 
DifherenGe: 3.5 iss es ahsg vac e ea ek me ae 150 308,009 
Work of locomotion for 150 revolutions.........)........... 75,000 
Total difference is 2 scien sa des ee ctrsns Hedda wallaes waa ews 383,009 
Equalt0; Jos negeases ds eee vaged es wae ee veld ea eee tea ee 903 Cals. 
The percentage utilization was therefore 903+ 3321 =27.2 per cent. 
The above figures serve to exemplify the general method of 
computation and likewise to illustrate the weak points in Wolff’s 
experiments, viz., the uncertainty in the determination of the work 
of locomotion and the impossibility of demonstrating the equilib- 
rium of food and work without the use of the respiration apparatus 
or calorimeter. 
Out of the whole number of experiments between 1891 and 1894, 
seven admit of a comparison of this sort, viz., four on oats, two on 
straw, and one on beans. Upon making the computations, how- 
ever, the results are found to be so exceedingly variable (the range 
for oats, e.g., being from 16.89 to 63.96 per cent.) as to demonstrate 
that the data of Wolff’s experiments are not sufficiently exact to be 
used in this way, and that the apparently reasonable result just 
computed is purely accidental. 
UTILIZATION OF FIBER-FREE NUTRIENTS.—But although Wolff’s 
results do not enable us to compute the percentage utilization of 
single feeding-stuffs, if we accept provisionally his conclusions re- 
garding the non-availability of the crude fiber they afford data for 
numerous computations of the uthization of the fiber-free nutrients, 
