222 THE CHEMISTRY OF CATTLE FEEDING 



oats about sixty, but 2 Ibs. of the former cannot be substituted 

 for i Ib. of the latter in the rations of working horses. 



The system of starch equivalents is based upon the 

 relative fattening powers of the several nutrients. It is, 

 therefore, applicable, strictly, only to rations for fattening. 

 When applied as it has been to rations for work, milk 

 production, and maintenance, the system becomes purely 

 empirical, and certain precautions must be observed in 

 using it. 



Apologists have urged that no considerable error is 

 involved if the starch values of the foods are used in com- 

 puting the rations for work. This, as it happens, is true; 

 but it is true only because such rations need not, and as a 

 rule do not, contain a large amount of protein. According 

 to Kellner, the relative fattening powers of protein and starch 

 are as 0*94 to i. As a source of dynamic energy, they stand 

 in the relation of 4*93 to 3*76 (p. 123), i.e. 1*3 to i. If, 

 therefore, a large amount of protein were involved, the error 

 would be considerable. 



The converse of the above statement, viz. that no con- 

 siderable error is involved if the dynamic values of the foods 

 are used in computing rations for fattening, is necessarily true 

 in the same degree. This close correspondence, of course, 

 arises from the fact that, except as regards protein, the fatten- 

 ing powers ascribed to the nutrients bear practically the same 

 relation as the dynamic values of the same. In fact, if the 

 dynamic values of the foods are divided by that of starch 

 (3*76), the figures obtained are approximately the starch 

 values, except in the case of foods rich in protein. The 

 dynamic values of the foods may, therefore, be used instead 

 of the starch equivalents for comparison of nutritive values, 

 and for most other purposes ; and they are probably more 

 reliable. 



The advisability of applying the system of starch equiva- 

 lents to the case of rations for milk production is questionable. 

 Such rations necessarily contain a large amount of protein, of 

 which the larger part ultimately appears, as such, in the milk. 

 The protein is not oxidised, like that in rations for work, nor 



