500 GENETICS IN RELATION TO AGRICULTURE 
tion which has been made use of in the establishment of modern breeds 
of domestic animals has come from amphimixis, the result of polyphyletic 
origin. But modern breeds have become highly standardized and they 
are closely guarded by rigid rules of pedigree registration. In conse- 
quence improvement within them has been effected purely by methods 
of selection without hybridization between breeds. In poultry, however, 
and here again we turn to Pearl for data, new breeds are still being 
created, and almost entirely by methods of hybridization. Thus for 
practically every variety of larger domestic fowl there is a corresponding 
bantam variety. Pearl sent queries to bantam breeders in all parts of the 
world for the purpose of obtaining information upon the method of 
creating new varieties of bantams. Here it would seem was an ideal test 
for the utility of Darwinian selection in the establishment of breeds, for 
it would appear to be a very direct mode of attack in breeding bantams 
corresponding to a given larger variety of fowl simply to select for smaller 
size within the larger breed. Pearl’s enquiries brought out the fact, 
however, that in no case were bantam breeds created in this fashion, but 
always by crossing the larger breed with some bantam variety and then 
selecting within the hybrid progeny of subsequent generations. In view 
of the demonstration of the relatively simple Mendelian heredity of body 
weight in Seabright * Hamburg crosses which has been demonstrated 
by Punnett and Bailey, we can readily understand how this method 
should be quickly and uniformly successful. The creation of bantam 
varieties of fowls, therefore, strikingly bears out this general thesis, that 
so far as the results of selection go in actual practice, the isolation 
interpretation is sufficient to account for all facts. The creative view 
of selection is an uncertain doctrine to accept for guidance. 
Selection and Breeding Methods.—I'inally it becomes necessary to 
again point out that a change in interpretation does not necessarily call 
for a change in breeding methods. In fact it can be shown that in general 
those breeding methods have been most successful which are most closely 
in accord with the isolation view of heredity. This fact is particularly 
patent when the earlier histories of breeds is taken into account, and the 
vast amount of inbreeding which has been employed by the best of these 
old pioneer breeders is given a true valuation. For inbreeding determines 
the fixing of a given genotypic constitution, because in such a system of 
breeding family excellence is the basis of judgment, just as in the suc- 
cessful selection for high winter egg production progeny test was the 
basis of selection for breeding stock. That most famous of all early 
breeders, Robert Bakewell, when he set about establishing his herds 
made excursions all over Britain for the purpose of selecting and purchas- 
ing the best of all sorts for his foundation stock. However, after he had 
brought this stock together he used only the progeny of these animals in 
Digitized by Microsoft® 
