AGRICULTURAL EXPERIMENT STATION. 37 



pressing one and the investigator who is using the respiration ap- 

 paratus and other scientific facilities in a search for this knowledge 

 has before him great possibilities for valuable service. But this 

 lack of knowledge does not constitute a reason why observations of a 

 practical character should not be continued. Feeding experiments 

 may convince us of certain facts which science shall sometime ex- 

 plain. We should demand, however, that the conclusions derived 

 from these experiments shall be fortified by all the accurate data 

 which it is possible to secure, and so in studying the relation of food 

 to growth, it is essential to know not only the amount and kind of 

 nutrients supplied to the animals but also the extent and character 

 of the growth produced. 



The Problem Studied. 



There is much discussion at the present time, of the relative in- 

 fluence and economy of various food combinations. "Standard ra- 

 tions," "narrow ratio" and "wide ratio" are familiar phrases, all of 

 which have to do with a wide spread conviction that the manner in 

 which foods of different classes are combined has much to do with 

 the character of the product and the profit of feeding anima's. 



It is generally taught that a given amount of digestible food should 

 have not less than a certain proportion of protein in order that it 

 may cause a maximum production, and scientific data, practical 

 feeding experiments and even common experience appear to warrant 

 such teaching. It is claimed, still further, that not only the amount 

 but the kind of product is to an extent under the control of the 

 food, and the experiments of Sanborn, Henry, Eoberts and Georgeson 

 with swine, lambs and steers appear to substantiate this claim, cer- 

 tainly so far as it relates to swine. 



In all these experiments the evidence of the effect of the food 

 in modifying the composition of the carcass is simply the appar- 

 ent relative amount of the fat and lean tissues, save in certain in- 

 stances where a chemical analysis was made of a portion of the 

 carcasses, too small to furnish reliable data. With the swine in some 

 cases the differences in the carcasses in their proportions of lean and 

 fat were too unmistakable to allow an erroneous judgment, but with 

 the sheep and steers no past experiments seem to have been so con- 

 ducted as to prove that the rations differed in effect other than to cause 

 more growth or less growth. Moreover, the experiments with 

 ruminants for the purpose of studying the effect of food upon the 

 land of growth were not begun with the young animals and con- 

 tinued until they reached a somewhat mature growth, but covered 

 only such periods of time as would be required to fatten the animals 

 for the market. 



The experiment which is detailed in the following pages had for 

 its object a study of the effect of widely different rations upon the 

 rate of growth and composition of the bodies of steers, and it is 



