FEEDING EXPERIMENTS WITH MILCH COWS. 



J. M. Bartlett. 



(i) Gluten meal compared with cotton-seed meal for milch 

 cows. 



(2) Silage compared with grain as food for milch cows. 



(3) Ground oats compared with wheat bran for milch cows. 



(4) Nutriotone for the production of milk. 



Feeding experiments with milch cows under ordinary condi- 

 tions are somewhat unsatisfactory. There are many factors, 

 beyond the experimenter's control, which come in to cause 

 greater variations in milk yield than would be produced by any 

 change in the ration one would care to make use of in rational 

 feeding. Conclusive results are, therefore, not to be expected 

 from single, brief feeding trials. It is only from the accumu- 

 lated data of a large number of such trials that results of a relia- 

 ble nature are obtained. Of course a very wide ration like that 

 of timothy hay and corn meal will show its inferiority in a single 

 feeding experiment, but the intelligent dairy farmer of today is 

 already aware of that fact, and such irrational feeding experi- 

 ments are no longer necessary. Considering the question as to 

 whether a ration with a nutritive ratio of 1 to 4 or 1 to 6 is the 

 more profitable, there are very little reliable data on which to 

 base conclusions. Some of the most intelligent dairy farmers 

 claim the best results from the very narrow ration of 1 to 4, while 

 ethers are equally confident that thewider oneof 1 to6 is the bet- 

 ter. It is in studying such problems as this in which results are 

 not decisive thai the single experiment is the most unsatisfactory, 

 and it is only from the results of a large number of trials that 

 definite information is obtained. The individuality of cows is 

 a perplexing element in a feeding experiment. Cows are some- 

 what like people, and a ration that agrees best with one does not 

 always agree with another. 



