Dec. 17,1917 
Influence of Age of Cow on Milk and Milk Fat 
651 
In calculating the data in Table II, which may be referred to as a 
frequency table, it was found in certain cases that a cow showed the 
same average test in two lactation periods, which usually followed in 
succession. Those cases in which the average percentage of fat in two 
periods did not differ more than 0.05 per cent were assigned to this class. 
In calculating the frequency tables for these cases the assumption was 
made that the period showing the highest average test meant that no 
other period showed a higher test, and the frequency calculation was 
made in favor of the first period which showed the high test. 
There are several outstanding features of the tables and figures so far 
cited, of which mention may be made. * In the first place one is impressed 
by the fact that the differences in the average fat percentage from lac¬ 
tation period to lactation period among the different breeds are not 
large. Indeed, it is necessary to plot the fat percentages on a rather 
large scale in order to make apparent the differences which occurred. 
Inasmuch as the figures do not show all the data given in the tables, it is 
necessary to turn to them to see that the differences in the average per¬ 
centage of fat from lactation period to lactation period accumulate to 
rather large figures as the number of lactation periods of a cow's life 
becomes greater. 
A second striking feature of the data is the marked similarity between 
the fat percentage and distribution curves. This fact indicates that the 
distribution curves and the data from which they are drawn are in 
reality of much greater significance than the curves of percentage of 
fat. They show clearly that a high percentage of fat in any one period 
among a large number of cows is due in most cases to the fact that a 
greater number of cows in that group showed a higher test in that period. 
This similarity between the curves of fat percentage and distribution of 
the highest test is especially striking for the 30 Jersey cows shown in 
figure 1, and for the Holstein cows shown in figure 2. 
A third feature of the data is the striking breed characteristic with 
respect to the frequency of distribution of the highest test in the differ¬ 
ent lactation periods. In the case of the Jersey cows the data in Table 
II show that, when only the first two lactation periods are involved, the 
tendency is for the second period to give a slightly higher average test than 
the first period, but that this tendency is greater among a few cows than 
among a large number. In fact, the data show that with the 74 cows 
having only two lactation periods the distribution was practically equal 
between the two periods. For cows of the Jersey breed in the herd the 
tendency for the second period to show the highest average test dimin¬ 
ished greatly as the number of lactation periods increased, and in many 
cases passed to the third period. In general the data show that Jersey 
cows may be expected to show their highest average test in either the 
first, second, or third periods, but the chances favor the second and third 
periods over the first to a rather marked degree. From the third period 
