142 SWINE 



protein and in the other too much carbohydrate. Thus 

 in a test of this kind there is simply a balance between 

 two evils, one of which is as great as the other. 



Another American experiment has shown that rape is 

 a better feed for swine than clover. Here again the con- 

 ditions under which the results were determined are of 

 the greatest importance. In this particular case the basal 

 ration, or the feeds that were used supplementary to pas- 

 ture, was made up of middlings and skim milk. Mid- 

 dlings and skim milk are both classified as nitrogenous 

 concentrates and when fed in connection with clover pas- 

 ture, three nitrogenous feeds were used in the ration. 

 When fed in connection with .rape, two nitrogenous and 

 one carbonaceous feed were used, the rape being car- 

 bonaceous in nature. In the former case it is quite evi- 

 dent that an excess of protein was fed and a deficiency 

 of carbohydrate which clearly would not produce as good 

 results as in the latter case where the ration was more 

 nearly balanced. 



Thus very many results from experimental data need 

 proper interpretation. In the main, however, much valua- 

 ble work has been done by the experiment stations of the 

 world, which needs only a little more study and a little 

 more supplementary experimentation to put the entire 

 system of swine feeding upon a much firmer foundation. 



