THE FEEDING OF MILCH CATTLE 331 



milk fat requires special notice here. The older 

 investigations on this subject led to very varying 

 results ; at one time there was found to be an in- 

 crease in the milk fat and at another a decrease, 

 whilst in many cases no effect at all was noticed. 

 Ten series of experiments carried out on the same 

 lines in different parts and with cattle of various 

 breeds there being almost 200 cows under observa- 

 tion have given a trustworthy reply to this im- 

 portant question. The plan of these experiments 

 was to feed a constant basal ration in each series, 

 and then add to this at one time a rice meal rich in 

 fat, and at another time rye meal and enough potato 

 starch to make the starch equivalent of each food 

 mixture the same. In one ration there was i kg. 

 digestible fat per 1000 kg. live weight, and in the 

 other an equivalent quantity of digestible carbo- 

 hydrates and very little fat. The average results 

 of the ten experiments showed that with the food 

 rich in fat there were 0-31 kg. =27% less milk and 

 13 g.==37% less fat obtained than with the food 

 poor in fat differences which are so small that 

 they hardly come into consideration. 



In separate cases the individual peculiarities of 

 the cows were very striking, as has been seen in 

 an experiment previously quoted (p. 310). There 

 were also considerable differences in the several 

 groups, and if the results are arranged according 

 to the changes in the milk yield and several groups 



