June, 1941] Animal Breeding and Nutrition 7 



not reproduced the desired traits with uniformly satisfactory results. 



With the more recently accepted practice, particularly by dairy- 

 men, of using proven sires, a progressive step has been made toward 

 a definite determination of the breeding potential. This has, how- 

 ever, serious drawbacks, as only a limited number of males can be 

 tested, too much time is lost before the value (which may be nega- 

 tive as well as positive) becomes known, and too many offspring of 

 indifferent value may result in the process. 



This was recognized to be beyond the means of the average breed- 

 er, and as a consequence a new method of selection was devised. 

 This was based on the principle that the individual's somatic traits 

 constitute a partial and imperfect index to its germ plasm and that 

 a better insight into that germ plasm is gained by considering the 

 traits shown by as many close relatives as possible. Naturally the 

 qualities of the proposed mates are considered, but only as members 

 of their families. 



The potency of the germ plasm for a given trait or for a group of 

 traits (i.e., for a type of function) is proportional to the numerical 

 representation of genes present for that trait. When a closely re- 

 lated group of individuals is uniform in type or function it is because 

 the genes which have determined this type uniformity are numer- 

 ically predominant. 



It is thus also clear that improvement of type or function, as well 

 as the establishment of new types that will reproduce their own traits 

 with a relatively greater degree of certainty, depends on a process of 

 intensification of a definite type of germinal determiners and that 

 this can be attained only by inbreeding. 



The older literature on heredity has dilated persistently on the 

 harmful effects of inbreeding without supplying any compensating 

 information on its possible benefits. The problem of avoiding unde- 

 sirable consequences appears to depend on evaluating the germinal 

 constitution of animals intended for parentage, followed by rigid 

 culling of all those offspring in which the undesirable traits appear. 



Many physical and physiological characteristics have been studied 

 by following this method of selection and of inbreeding, and utility 

 traits from different sources have been combined into a new breed 

 thereby. 



One of the most important contributions to this end has been ob- 

 tained in the increased milk yield from a flock of multi-nippled ewes 

 developed by the late Alexander Graham Bell. The flock was don- 

 ated to this station for further research. It represented thirty years 

 of careful experimental breeding with the practical object of in- 

 creasing the fecundity and the milk yield of ewes, and its contribu- 

 tion has spared us much in time and expense. 



The scientific significance of this type of research was recognized 

 as of sufficient importance by the committee which organized the 

 recent International Congress of Genetics, held at Ithaca, New York, 

 to select it as representative of progress in applied genetics with 

 domestic livestock. An exhibit to illustrate the effects of breeding 

 on sheep, wool, and woolen products attracted particular attention 



