200 PRINCIPLES OF ANIMAL NUTRITION. 
Kellner estimates that the maximum amount. of work which 
could be performed on the ration containing starch was 700 rev- 
olutions as compared with a maximum of 500 revolutions without 
starch. Even if this estimate of Kellner’s be regarded as high, it is 
evident from the figures given that the addition of the starch enabled 
materially more work to be performed without an increase in the 
proteid metabolism. The results obtained in this and the subse- 
quent series have been made the basis of interesting computations 
regarding the utilization of the potential energy of the food which 
will be considered in Part II. 
Series V.—This series was precisely similar to the preceding one, 
except that the addition of non-nitrogenous matter to the ration was 
made in the form of oil by substituting flaxseed for linseed meal. 
The protein of the ration remained unchanged, while the fat was 
increased by 203 grams. The results were entirely similar to those 
with starch, as the following table shows: 
Nitrogen. Li 
Period. yee Weight, 
Digested, In Urine, Kg. 
Tms. Grms. 
Tee sete ee Without 500 |) {| 148.9 496.5 
Wrsciecs ae addition 500 159.0 149.2 493 .2 
TG): Saescavncss of fat 550 : 147.5 485.8 
TD Dic esteveysasshent L| 550 153.0 479.4 
TOE sinsasvcist done : . 700 148.1 476.0 ° 
With : 
TO resistance ies 700 153.9 469.0 
Wiccan cuss | ee | 650 | 158.9 | 145.6 | 466.4 
TB ise ss rece idvsito 650 145.0 460.8 
While Kellner’s method of investigation may be regarded as 
somewhat imperfect and necessarily giving but approximate results, 
yet it suffices to bring out in a very striking manner the intimate 
relation existing between the supply of non-nitrogenous nutrients 
in the food of a working animal and the effect of the work upon the 
proteid metabolism. In conclusion, it should be noted that in 
all Kellner’s experiments there was a fairly abundant supply of 
protein. Whether the same result would be obtained on a ration 
containing the minimum amount of proteids required by the organ- 
ism is not shown. In no case was the increase in the proteid metab- 
olism, when observed, sufficient to supply energy equivalent to the - 
additional work done. 
