EDITORIAL. 109 



pendent experiments which embody nothing in common and are 

 incapable of comparison or combination. 



The importance of the* subject of animal feeding merits the verj^ 

 best eifort which the experiment stations are capable of commanding. 

 The conventional experiments have served a very useful purpose and 

 will continue to be needed, but they should be refined to give ti 

 greater degree of accuracy and should be subjected to more critical 

 examination in their planning and their conduct. But beyond this, 

 one of the ultimate objects of work in this field, as in every other, 

 must be to make practice more intelligent and better understood. 

 This calls for the determination of the reason for what is found 

 in experiment and observed in good practice. Without this the the- 

 ory of feeding can not be developed and the more practical experi- 

 ments can not reach their highest degree of reliability or usefulness. 



One of the greatest needs is more men of training who can see the 

 field in its broader aspects and develop a j)oint of attack. Especially, 

 there should be na question of the encouragement and defense of the 

 higher types of work by the men in 'charge of animal husbandly in 

 the agricultural colleges and experiment stations. 



