70 
Journal of Agricultural Research voi. xiv, No. 3 
The correlation is negative and very significant for all ages. When 
judged by their probable errors, there are no significant differences among 
the correlations for the different groups. 
The milk yield increases from an average of 6,475 pounds for the 2-to- 
3-year-old class to 8,814.5 pounds for the mature class (fig. i). Since 
there is a marked 
negative correlation 
between the percent- 
age of fat and yield 
at all ages, one might 
look for the percent- 
age of fat to decrease 
as the yield of milk 
increases with the age 
of the cows. The 
class which is 5 years 
and over shows a 
slightly smaller per- 
centage of fat, but 
the 4-to-5-year-old 
class has a higher 
percentage than the 
3-to-4-year-old class, though the difference is not significant (fig. 2). 
On the whole, the percentage remains practically the same. This may 
be due to the relatively stable relation between the amount of fat and 
yield throughout the growing period of the individual cow. Holdaway* 
found this to be true for Holstein-Friesians, using the 7-day records. 
Stated in another way. 
Fig. I. — Graphs showing the averages of the milk yield for the different 
ages of cows. 
Is 
(?e/s/?A/S£y 
the percentage of fat 
seems to be fairly con- 
stant throughout the 
life of a given individ- 
ual, but different indi- 
viduals show differing 
percentages of fat. 
The results show that 
at a given age cows 
with a high milk yield 
tend to produce a 
lower percentage of 
fat than do cows with a low milk yield. This is expressed by the nega- 
tive correlations obtained. Table B illustrates in another way that this 
is true. This table is the result of arbitrarily dividing Table II into 
three parts, cows yielding 4,500 to 6,500, those yielding 7,000 to 9,000, 
^-j ^y-^ ^-J- 5 oi/Sff 
Fig. 3. — Graphs showing the averages of percentages of butter fat for 
different breeds of cows. 
' HOIDAWAY, C. W. STATISTICAI, WEIGHTING FOR AGS OP ADVANCED REGISTRY COWS. In AmtT. 
Nat., V. 50, no. 599. p. 676-687, 2 fig. 1916. 
