32 BULLETIN 1372, U. S. DEPARTMENT OF AGRICULTURE 
the yield of milk and increasing the percentage of butterfat, or he 
may be prepotent in either raising or lowering both the milk yield 
and the percentage of butterfat. 
This ability of the sire seems to depend upon the combination of 
factors governing the yield of milk and percentage of butterfat that 
he has inherited from his parents. If he is homozygous for dominant 
factors that will determine high milk yield and high percentage of 
butterfat, he will be prepotent in impressing these characters on his 
offspring. If he and the dams he is mated with are heterozygous for 
these factors, as most sires and dams are, a variety of combinations 
in the different offspring will follow, and they will be of varying 
degrees of producing ability. 
5. The percentage of butterfat and the milk yield seem to be 
inherited independently in Holstein-Friesian cattle. This is contrary 
to the findings of other investigators. The theory generally accepted 
is that, as the milk yield increases, there will be a decrease in the 
percentage of butterfat. Though this study showed that in the 
majority of cases there was a tendency toward a negative correlation, 
there was a sufficient number showing a tendency toward a positive 
correlation to indicate that the two are independent of each other. 
It is also shown that it is possible for a sire to increase both the milk 
yield and the percentage of butterfat of his daughters. 
6. A great sire of production is one whose daughters have a high 
average yield of milk and butterfat, a high average increase in milk 
and butterfat yield over the yield of their dams, and a high percentage 
of their number better than their dams. All these things must be 
considered. No one of them alone offers sufficient evidence of the 
sire's worth. The production of each sire's daughters must be con- 
sidered in comparison with the production of other sires' daughters. 
Provided a sufficient number of tested daughters are available for 
each sire the ranking system shown in this bulletin seems to indicate 
the comparative merit of the sires in a group. 
7. The production records of the dams of the 10 highest-ranking 
sires average higher than the records of the dams of the 10 lowest- 
ranking sires. In neither of these cases, however, does the rank of 
the sire follow the size of the record of his dam. 
8. Six of the 10 highest-ranking sires and 6 of the 10 lowest-ranking 
sires are classed as outbred. This seems to indicate that the mere 
fact that an individual, his sire, or his dam, is line bred, inbred, or 
outbred is not indicative of the prepotency of that individual for 
high production. 
LITERATURE CITED 
(1) Graves, R. R. A genealogical study of Holstein-Friesian sires. (In un- 
published manuscript.) 
(2) Babcock, E. B., and Clausen, R. E. 1918. Genetics in relation to agri- 
culture. 
(3) Roberts, Elmer. Journal of Agricultural Research, vol. 14, no. 2. 
(4) Wilson, James. Sci. Proc. Royal Dublin Soc, vol. 12, no. 33. 
(5) Pearson, Karl. Biometrika, vol. 7, no. 4. 
WASHINGTON : GOVEENMENT PRINTING OFFICE : 1926 
