MAINTENANCE: RATION AND BATTENINGS 447 



Feeding Experiments, The values of feeding-stuffs for fat 

 production are often compared by means of feeding experiments. 

 Feeding experiments have also been of great value in establishing 

 standards of feeding for various purposes. Feeding experi- 

 ments do not give such exact values as experiments in which the 

 carbon and the nitrogen balance are determined, but they are very 

 useful in their proper field. 



Two systems of feeding experiments are used. In the first 

 system, the same groups of animals are fed upon the different 

 rations in succession. The method is open to the objection that 

 the effect of any ration will depend to a considerable extent upon 

 the preceding rations, and the same feed may give entirely differ- 

 ent results according^to the character of the feeding which precedes 

 it. The feed, of course, affects the condition of the animal, so that 

 the maintenance requirements vary, and the excess of the ration 

 over the maintenance requirements, which is the portion used for 

 the gain in weight, will depend upon the previous feeding. The 

 results of the feeding will thus depend upon the previous feed, 

 as well as on the ration being studied. 



The second system consists in dividing the animals into groups, 

 and feeding to the different groups the rations to be compared. 

 This method gives good results when properly used. Care 

 should be exercised to have the groups exactly equivalent at the 

 beginning of the test. The groups should consist of 10 or more 

 animals. Each animal should be matched in age, form, live 

 weight, etc., with another animal in the other group. The groups 

 should be compared by means of a preliminary feeding period, in 

 which the animals receive the same ration for thirty days or 

 more. If each group makes the same gain on the same feed, 

 the experiment proper may be begun; but if there is only a 

 slight difference in gain, the groups should be rearranged, and 

 another test made. 



The best results are secured when the ration is about three 

 quarters of the standards. If the production is forced to its 

 upper limit, or if more feed is given than the animal can utilize, 

 the differences, due to the different feeds, may be insignificant. 



