THE SELECTION PROBLEM IN ANIMAL BREEDING 



499 



progeny of high performance in the first year she was not retained for 

 further breeding purposes. Males for breeding purposes were selected 

 on a like rigid basis; they were from high-producing mothers, the daugh- 

 ters of which were all high producers, and any male was rejected imme- 

 diately if his progeny failed to measure up to high standards. Complete 

 individual pedigrees were kept during this period. For the sake of 

 comparison low and mediocre strains were also selected on a basis equally 

 rigid for their particular characters. 



The success of this type of selection is strikingly evidenced by the 

 data set forth in Table LXV, which gives the means from which Fig. 



TABLE LXV. MEAN WINTER EGG PRODUCTION OF THE MAINE STATION BARRED 

 PLYMOUTH ROCK FLOCKS FROM 1899-1915 (Data of Pearl) 



185 was constructed. From an interpretative standpoint, therefore, the 

 direct contrast is brought out sufficiently well in this case, for when 

 selection was placed on a fairly rigid genotypic basis it was immediately 

 successful. It seems hardly possible to explain the facts of this series of 

 investigations by any other than an appeal to the isolation view of 

 selection, particularly when consideration is directed toward the rapidity 

 with which genotypic selection established high-producing strains in a 

 flock which had failed to respond to a rigid system of mass selection. 



Bantam Fowls. In a previous chapter evidence was presented tending 

 toward the general conclusion that the most potent source of that varia- 



