I7(t MdiiitciHdice Rations, (iiul Starch I'Jtjinrah nix 



14,500 to 18,350 cals. per 1000 lb. live weijjht per day, and is praetically 

 identical with that jjiven in the appendix to Wolff's Farm Foods. 

 We must, therefore, assume either that Kellner copied the figures 

 straight from Wolff's table, or that he made the mistake of reckoning 

 the amount of hay recjuired from tlie starch equivalent for production 

 instead of from the total metabolisable energy of the food. In the 

 latter case, the mistake, the existence of which Murrav rocf)gnises, 

 is the exact reverse of that which Murray assumes, and Murray's 

 estimate of the energy requirements of oxen at rest is much too high. 

 Independently of Kellner's later experiments, a long series of 

 digestibility experiments was conducted by Armsby' in America on 

 cattle. These experiments, which were of long duration, and in which 

 the animals were weighed very frequently, support in their findings 

 the results given by the more exact calorimetry experiments of Kellner. 

 Kellner's experiments are open to the objection that the results given 

 are based on an assumption in most cases of the thing Kellner was 

 endeavouring to find, i.e. an exact knowledge of the maintenance 

 re(|niremeiits of an ox at rest. Where the animals used for the estima- 

 tion of the average had ])nt on weight (and five out of the seven animals 

 used for the average had increased in weight during the ex])eriment), 

 Kellner added a certain percentage of energy to the total stored; 

 since he had found in previous experiments that of 100 parts available 

 energy only 43 parts in the case of meadow hay, and 37-fi parts in the 

 case of oat straw, were available for fat production. But these figures 

 were arrived at by assuming the maintenance requirements of an ox 

 at rest to be 21,300 cals. per 1000 kilos, live weight. In other words 

 Kellner was guilty here of the vicious practice of arguing in a circle. 

 The difference is not large, and does not affect Kellner's estimate to 

 a large extent, but it is essential to realize here that Kellner's estimate 

 of the maintenance requirements of an ox at rest is not an exact one. 

 As a result of Armsby's investigations into the maintoTiance require- 

 ments of cattle, the following estimate was arrived at. and represents 

 the average of 12 different experiments. The food fed was coarse 

 timothy hay, and the energy of the digestible food was estimated 

 by means of the Berthelot bomb calorimeter. As a result of these 

 determinations Armsby came to the conclusion that the energy require- 

 ments of a 500 kg. ox are satisfied with a total digestible energy of 

 12,771 cals. at 51° F., or 11,980 cals. per 1000 lbs. live weight. 



' .'irrasby. liiilhfin 42, IVtiii. Statf Ccill. Kx])t. Stat. ISnS. 



