Studies in Milk Secretion. 535 



are found in all well conducted dairies and the records cover 

 nearly, if not quite, the average yearly milking period of the 

 ordinary cow. On the other hand the cows which are entered 

 for the seven day test-s are the pick of the herd ; they are fed 

 to the highest limit for a short period and thus forced to their 

 utmost production for that time. Moreover the records are for 

 too short a time to furnish an accurate estimate of the relative 

 production of cows of different ages. Forty week records must, 

 in th^ nature of the case, give us a surer basis of comparison 

 than those which run for seven days only. 



Variation in Yield of Milk and its Fat Content as 



Lactation Advances. 



A point often discussed concerns the average rate of decrease 

 in yield of milk as the period of lactation advances and also the 

 average increase, if such there be, in the per cent of fat during 

 the same time. As an indication of what this decrease in milk 

 and increase in per cent of fat may be, a careful study was made 

 of the individual records of all the cows in the herd. This study 

 was conducted in the following manner : Beginning with the 

 third week after calving the weekly records of each cow were 

 divided into periods of four weeks each. The average daily 

 yield of milk and the average per cent of fat were then deter- 

 mined for each of these periods. Considering the averages for 

 the first period of four weeks in each lactation as 100, the aver- 

 age for each succeeding period was calculated to the same basis. 

 The first two weeks after calving were thrown out of the calcu- 

 lation because so many abnormal conditions affect the flow of 

 milk at that time. The records were worked out in this manner 

 for each individual cow in each 3'ear, and from these were 

 obtained the averages for seven years which appear below in 

 Table IV. In this table also are given the yearly records of six 

 individual cows representing various degrees of fluctuation in 

 yield of milk and in per cent of fat. An explanation of the 

 figures may be of value. In 1893-94, Dora (II.) had a daily 

 average of 31 pounds of milk during the first four weeks of her 

 lactation. If we represent this average of 31 as 100, then on the 

 same basis her daily average for the next four weeks of 28.25 



