MANUAL OF CATTLE-FEEDING. 



205 



over three days. The following results are averages for 

 twenty-four hours : 



Number of Experiment 



IL 



III. 



IV. 



Meat eaten. 

 Grms. 



Best, 



1,500 



f "Work. 

 ( Best. 



Wattr 

 diiuik. 

 Grms, 



ITrme 



excreted. 



Grms. 



258 



186 



873 



518 



123 



145 



527 



186 



125 



143 



182 



1,060 



657 



1,880 



140 



1,081 



412 



1,164 



G3 



1,040 



TJrca 



excr» ted. 



Grmfc. 



14.3 



16.6 



11 9 



13.3 



10.9 



109,8 



117.3 



109 9 



114.1 



110.6 



The 1,500 grammes of meat eaten in Experiments III, 

 and IV. contained 1,138 grammes of w^ater, w^hich is to he 

 added to the amount drunk. 



These results show, at most, only a very slight increase 

 of the protein consumption on the working days. On the 

 days of fasting the increase, as measured by the excretion of 

 urea, was 11.5 per cent., and with the meat i^ation only 

 4.8 per cent. ; and even this small increase appears to be 

 due, not to the work but to the greater amount of water 

 drunk and excreted in the urine, since, as we have already 

 learned (p. 135), an increased excretion of water in the 

 urine causes an increase in the protein consumption. Voit 

 f(3und that an increase of 100 grammes of water in the 

 urine caused, in the dog, a plus of 1.7 grammes of urea. 

 If this relation is applicable to the above experiments, it 



