FEEDING WORKING ANIMALS 393 



walking on a smooth level without load, and when per- 

 forming both light and heavy work. First of all, it appears 

 from his observations that 31.6 per cent, or about one- 

 third, of the total food energy can be converted into useful 

 work. This is much less than the coefficient of useful 

 work found by Wolff, whose conclusions Zuntz regards as 

 erroneous. But even if Zuntz's figures are none too low, 

 it is evident that the animal machine uses fuel with much 

 greater economy than a steam engine where the coefficient 

 of usefulness might not be over 10 per cent. The figures 

 he reached show further that the total expenditure of 

 energy by a horse weighing 1,000 pounds in walking 1 

 mile equaled 453 foot-tons, which would be furnished by 

 .164 poimd of digestible food. As 44.4 per cent of this, 

 or 201 foot-tons, was due to the effort of walking over and 

 above the needs for maintenance, the extra digestible 

 food needed per mile of walking was .07216 pound. 



Zimtz also found that when a horse increases the 

 external mechanical labor performed, such increase costs 

 .001155 pound digestible dry matter for each foot-ton of 

 work. On this basis the 264 foot-tons of energy which is 

 needed for pulling 1 mile a load with a draft of 100 pounds 

 would be furnished by .3049 pound of food matter. The 

 total food expenditure, therefore, for walking and a 

 draft of 100 pounds over a smooth, level road for 1 mile, 

 would be .377 pound digestible nutrients, and for 20 

 miles 7.54 pounds. If we add to this the 6.4 pounds needed 

 for mere maintenance, we have 13.94 pounds digestible 

 matter as the proper ration for a horse doing the work 

 stated for a distance of 20 miles. These figures are cer- 

 tainly not inconsistent with the standard reached by 

 other methods for a horse doing average work. Such a 

 calculation is at least useful in showing the direct rela- 



