8o mains: AGRICUIvTURAL E^XPE^RIMi^NT STATION. I909. 



ments with beans. Jennings t reaches the same result in one 

 of the most extensive and thorough studies of the effects of 

 selection ever made. In summarizing his conclusions at the 

 end of the paper (loc. cit. p. 522) he makes the following state- 

 ment : "Certainly, therefore, until some one can show that 

 selection is effective within pure lines, it is only a statement of 

 fact to say that all the experimental evidence we have is against 

 this." Finally, to cite a single further instance, the case recently 

 published by Arenander * showing that true mutants may occur 

 among dairy cattle in respect to fat content of the milk makes 

 it inferentially probable that selection in regard to this character 

 can only bring about improvement by isolation of superior pure 

 strains not by really increasing fat content within a strain. 



All together much evidence is accumulating from widely 

 different sources to show that simple selection of superior indi- 

 viduals as breeders will not insure definite or continued 

 improvement in a strain. Some improvement may possibly 

 follow this method of breeding at the very start but the limits 

 both in time and amount are very quickly reached. In support 

 of this view of the possibilities of selective breeding the results 

 of the present paper and of the nine-year selection experiment 

 carried on at the Station furnish definite and positive confirma- 

 tory evidence. The experience of the Station shows that in 

 order to establish a strain of hens in which high egg production 

 shall be a fixed characteristic it is necessary to do something 

 more than simply breed from high producers. 



In relation to the general subject of the inheritance of 

 fecundity the results of the present work are of considerable 

 interest. This arises in particular from the fact that in statis- 

 tics of egg production in poultry we are dealing with ,data 

 measuring the most fundamental factor in fecundity and fertil- 

 ity, namely, ovulation. Practically all of the work which has 

 hitherto been done regarding variation and inheritance of 

 fecundity and fertility has been upon mammals. In all vivi- 

 parous, as compared with oviparous, animals, the study of these 

 subjects is greatly complicated by the fact that the magnitude 

 of the apparent or recorded productiveness is influenced by 



* Arenander, E. O. Eine Mutation bei der Fjellrasse (Kullarasse) 

 Jahrb. f. wiss. u. prakt. Tierzucht. Bd. 3, pp. LXXXVII— LL. 1908. 

 t For foot note to Jennings see preceding page. 



