71 



cow in twelve weeks for the consumption of an average of 6 Ib. 

 of cake and meal per day for the same period, is, in the absence 

 of a great difference in the milk yield, scarcely a profitable 

 increase. 



EXPERIMENT H. 



The results described in the preceding pages were of so 

 interesting a character, and so different from what is alleged 

 to be the experience of practical men, that it was decided to 

 repeat the experiment during the summer of 1906. Accord- 

 ingly, after a series of preliminary trials had been carried 

 out in the usual manner, a set of cows was selected for the pur- 

 pose, eight out of the ten being entirely fresh. The two 

 cows which were in the previous year's experiment, Nos. 96 

 and 98, were not placed in the same lot as in 1905. The 

 cows which had all calved before Christmas, were again 

 divided into two lots according to the quantity and quality of 

 their milk. At the end of the preliminary trials, the two 

 lots were practically on a level in this respect, as will be seen 

 from the following figures : 



Average yield of 



milk per cow Per cent, of Average 



per day. butter-fat. live-weight. 



Lot I. ... 30-2 pints ... 3'48 ... 1,086 Ib. 



Lot II. ... 30-2 pints ... 3-46 ... 1,122 Ib. 



Owing to unavoidable causes, the experiment was not 

 commenced for nearly a fortnight after the preliminary trials 

 came to an end. It was continued for a period of ten weeks, 

 viz., from June 18 to August 27, 1905. 



Rations. The experiment, as regards the feeding, was 

 conducted on exactly the same lines as in the previous season. 

 The period of the experiment was a fortnight shorter this 

 time, otherwise there was no difference as compared with 1905. 

 The ten weeks were divided into two equal periods of five 

 weeks each, during the first of which lot I. received per cow 

 per day, 2 Ib. Bombay cotton cake, 2 Ib. maize meal, and 



