160 The Starch Eqnira/eiif Thcori/ 



The author's estimate^ of the maintonaucc ro(juir('monts of slioop of 

 100 lb. live weight is from 7 J to 8i lb. per week of "total di<^estible 

 nutrients" corresponding to about 5 or 5J lb. of Kellner's starch equiva- 

 lent. Even so, there is reason to l)elieve that sheep retain in their 

 bodies a larger percentage of the food in excess of the rnaiiileiuince 

 ration tliau o.xen do. 



The whole question stands in urgent need of re-e.xaniination on 

 a fundamental basis. The estimate tliat the fattening increase comprises 

 "67 per cent, of dry matter, chiefly fat, and 33 per cent, of water" 

 is probably true only under conditions similar to those of the Rotham- 

 sted experiments in which the animals- appear to have been not fully 

 grown. Comparison of the figures' relating to the fat ox and the half- 

 fat ox shows that of a fattening increase of 200 lb. something like 

 190 lb. was fat and only 10 lb., i.e. about 5 per cent., was water. It 

 may ultimately prove that the true fattening increase apart from 

 growth — consists almost entirely of pure fat. 



It is to be noticed also in this connection that in the attempts to 

 estimate the maintenance rations of oxen of different size by means 

 of graphs or formulae it is assumcil that ail the animals are in store 

 condition, i.e. that the difference in live weight is due in all cases to 

 greater or less giowth aiul not to fattening. Fat animals retpiire more 

 food to maintain them without gain or loss of weight than those in 

 lean condition. It is not easy to see how a single graph (■(iiilii tn^ 

 adajjterl to meet tiiis difficulty, but the formula given above might 

 be modified in some such manner as that shown below: 



logiV= I logM- ] •03152a;, 



where a; is a number that varies inversely as the fatness of the animal. 

 If any means can l)e found to express the degree of fatness in numerical 

 terms it would probably not be very dilKcult to determine ,/. All that 

 can be said at present is that when the animals are in store condition — 

 whatever that may be tiie value of x is 1, ami that when they are 

 fatter it is less than 1. 



It may also be found eventually that herein lies the ex])lanati(in 

 of the gradually diminishing returns in the shape of increase for food 

 consumed as the animal becomes fatter. To say that it is due to the 

 fact that a fat aiiitnal requires more food for maintenance is not merely 



' Cliemintry of Cattle Feeding and J^nirijing (Lonfimans). pp. i:!(l ami L'(I4. 

 ' Cf. Scientific Feeding of Animalx, pp. 254 ami 2.').'). 



' Rotlmmstcd Metrwirs, Vol. rii. pp. 520 ami 55S. Cf. Chrmi.ili-ii of faltir Feeding. 

 pp. 189 and li)0. 



