• 
Vol. 6, 1920 GENETICS: R. PEARL 225 
A CONTRIBUTION OF GENETICS TO THE PRACTICAL BREED- 
ING OF DAIRY CATTLE 
By Raymond Peari^ 
Department of Biometry and Vital Statistics, Johns Hopkins University 
Communicated February 10, 1920 
When the writer began his work at the Maine Agricultural Experiment 
Station some 13 years ago the only genetic concept which played 
any part in or influenced the actual operations of the animal breeder was 
that of phaenotypic selection. The slogan and the practice, so far as 
there was any intelligent practice, was to "breed from the best to get the 
best." This was perhaps theoretically a sound enough doctrine, but there 
was lacking one essential to its practical operation vvith the anticipated 
degree of success. This lack, as the rapidly advancing knowledge in the 
field of genetics soon clearly showed, was of any precise criterion as to 
how the best genotypes were to be selected, when the breeder operated on 
the basis only of a knowledge of somatic performance or characters alone. 
As a matter of fact it was easy to show, as the writer did show, in the case 
of egg production in fowls, ^ and milk production in cattle, ^ that the phaeno- 
typic performance was an exceedingly unreliable guide to real breeding 
worth. 
The real value of the measure of an animal as a breeder — as a factor in 
the improvement of a herd or flock — is the kind of progeny which it pro- 
duces. Confining the discussion to performance in respect of milk or 
butter-fat production, a cow may have herself a perfectly enormous record 
of performance but she is worthless as a breeder if her progeny are uni- 
formly poor producers, which can in fact be easily shown to be often the 
case. Again a bull may come from ancestry of the highest performing 
ability in respect of milk and fat production, and consequently he may as 
a calf command a price in five figures, but such ancestry has repeatedly 
been shown to be no guarantee in fact that his progeny will be great per- 
formers. 
The paramount importance of having a bull of such genotypic consti- 
tution as to insure relatively high dairy production in his offspring, if 
one is to attain success in the breeding of dairy cattle, is obvious. Nearly 
10 years ago the writer began the intensive study of the practical prob- 
lem of scientifically measuring the breeding worth or value of bulls of the 
dairy breeds, to the end that the farmer and breeder might operate more 
intelligently and successfully. It appeared evident that the most important 
thing which a breeder of dairy cattle desires to know is whether the animals 
which he breeds are transmitting productive qualities to their offspring. 
This information is particularly desired for the herd bull since his offspring 
are much more numerous than those of any single cow and he constitutes 
one-half the heritage given to them. It would appear beyond question that 
