FERTILIZER CONTROL STATION. 341 



cent to 25.7 per cent, and the increase in the second case ranged 

 from .08 per cent to 10.6 per cent. 



(2) The changes in production of total milk solids and also of fat 

 are very nearly the same as with the butter. 



An average of 21.9 per cent less solids and \*jA per cent less fat 

 was produced in the second period than in the first, and in the third 

 period the average increase over the second period was 4.8 per cent 

 of solids and 3 per cent of fat. 



The verdict rendered by this experiment is very pronounced, from 

 the fact that with no one of the four cows is the result an exception 

 to the general result. The same was true of the experiment of 

 1885-6, in which three cows were used, so that in carefully con- 

 ducted tests with seven cows, where every precaution against error 

 has been taken, and where all the important facts have been ob- 

 tained that the scales and chemical analysis could furnish, the re- 

 sults are uniformly in favor of substituting cotton-seed meal for part 

 of the corn meal in a moderate grain ration. As will be seen from 

 a subsequent analysis of these results, the amount of digestible ma- 

 terial was practically the same in the three periods, so that we can- 

 not resist the conclusion that there is such a thing as econom}' in 

 combining a ration so as to secure the maximum work from the mini- 

 mum quantity of food. What is the natural and rational outcome 

 of this truth, then, but the formulating of feeding standards? 



Sources of error. In order to be sure that the conclusions in re- 

 gard to the result of the preceding experiment are not reached by 

 fallacious reasoning, it is necessar}* to know that no other causes 

 except the variations in the ration effected a change in the milk pro- 

 duction, in passing from one period to another. Several causes 

 might influence the result, those most likely* to operate being 



(1) A failure of the animals to consume all the ration. 



(2) A gain or loss of weight by the cows, thus either consuming 

 the food for other purposes than milk production, or else producing 

 milk at the expense of the body and not of the food. 



(3) Marked variations in atmospheric conditions, such as temper- 

 ature, &c. 



(1) Consumption of ration. During the first period the food was 

 wholly eaten. For the two remaining periods the amount of refuse 

 material weighed back was small, except in the case of Cow B in the 

 third period. 



