EDITORIAL. 351 



ties in the way of obtaining- reliable returns through correspondence. 

 The (lata thus collected are to a large extent rough approximations, 

 often given with more or less indiflerence as to exact accuracy. In the 

 case cited above it is stated that considerable difUculty was encoun- 

 tered in getting correspondents to make full returns, and often three 

 letters had to be sent before the necessary data could be secured. In 

 the second place, it is by uo means certain that the practice of these 

 correspondents, even if representative, is the best. It may be the best 

 financially for the individual cases, but even that can hardly be assumed 

 to be true. The individuals are governed very largely by the feeding 

 stuffs at their Command and the market ])rices, rather than by any 

 scientific consideration. , If their practice were sutficiently perfect the 

 necessity for experiments in feeding would be at an end. liut there are 

 good reasons for believing it is not, especially when the collection of 

 data from different parts of the countr}- shows that the daily ration of 

 protein, for instance, ranges all the way from l.o() to 3,20 lbs. per cow. 

 The cow does not demand more protein in Utah than in Illinois. Both 

 amounts may be profitable, but both can not represent the amount best 

 adapted for cows throughout the country — the amount which the aver- 

 age cow requires to give the best result from her food. Surely there 

 can be no safety in such averages. It is a mere accident if the result 

 is not misleading. It is certainly working in the wrong direction. 



In the case of the other station referred to, a representative of the 

 station was sent to each farm where observations were made and 

 remained there 5 to 12 days, weighing the feed given each cow and taking 

 samples for analysis, weighing and testing the milk, and making a 

 record of the breed, weight, and stage of the milking period of each 

 cow. Any portions of the food left uneaten were weighed and allow- 

 ance made for them. The feeding stuffs were all analyzed at the station, 

 and the rations calculated. Suggestions were then made to the farmers 

 for improving their rations, which usually resulted in cheapening them. 

 Several of the farmers have adopted these changes with marked advant- 

 age. Twenty-two such studies were made during two seasons. In 

 conclusion, a tentative lation is suggested which, however, is con- 

 siderably lower in fat and carbohydrates than the average of the 

 rations studied, although it is believed the latter represent methods of 

 feeding above the average for the State. 



It is stated that this is only a beginning of cooperation of the station 

 directly with dairymen in the study of the methods of feeding their 

 cows and the ways by which improvements may be made. There 

 seems reason for hoping that studies of this kind may prove as valu- 

 able an educational feature as the cooperative fertilizer experiments. 



The collection of statistics on the practice of feeding may be sug- 

 gestive and useful in its .way. but the eiTor should not be made of 

 supposing that a standard can be worked out in this manner. Thor- 

 ough systematic feeding experiments by scientific experts alone can 



