466 PRINCIPLES OF AGRICULTURAL CHEMISTRY 



Work. Working the cow decreases the quantity of milk and 

 fat produced, while the percentages of protein, fat, and ash in- 

 crease. The effect depends upon the amount of work and the 

 feed of the animal. Light work does not appear to decrease the 

 total production of dry matter. 



Palatability of Food. It has been found that palatable food 

 increases the yield of milk. In experiments in which two rations 

 equal in feeding value were fed, the one tasteless, the other of 

 well flavored hay, etc., the second ration always gave better re- 

 sults than the first. 



Other Conditions. The milk secreting organs are closely re- 

 lated to the nervous system. Rough treatment, disturbances, in- 

 sufficient bedding, a cold stall, and other similar conditions de- 

 crease the quantity of milk and fat produced. 



Methods of Investigation. In studying the effect of various 

 conditions upon the production and composition of milk, it is 

 necessary to eliminate the influence of all factors except the one 

 to be studied. Some of these factors, such as the kind and fre- 

 quency of milking, may be eliminated by treating the animals 

 alike. Others, such as individuality, may be compensated for 

 by taking a sufficient number of animals. Two systems of experi- 

 ment 1 are in use the period system, and the group system, which 

 correspond to the methods used for feeding experiments on 

 fattening animals. 



The Period System. Ten or more cows are fed upon a standard 

 ration for a period of two or three weeks. Next the cows are 

 fed on the rations to be tested during three periods or more of 

 the same length. The standard ration is then fed for another 

 period. In each period, the production of milk and fat is deter- 

 mined. From the production of fat and milk during the first and 

 the last periods, we calculate the quantity which should be pro- 

 duced during the intermediate periods, assuming that a regular 

 decrease in production takes place. We compare the calculated 

 production with that actually found to occur, and the difference 

 gives the effect of the ration we are testing. 



1 See Kellner, Ernahriing d. Landw. Nutztiere, p. 500. 



