34.8 MAINE AGRICUIvTURAIv EXPEKIMENT STATION. 1C)12. 



It will be seen from this table that these families had on the 

 average fewer than two adult daughters each, too small a num- 

 ber with which to work. This makes clear again the difficulty 

 with which one has always to contend in practice in work with 

 fecundity, namely that of getting even reasonably large families 

 of normal adult daughters. One hatches a large number of 

 chicks in order to supply thieves, crows, rats, hawks, etc., and 

 finally get a small number of adult females available for the 

 study of fecundity. Fecundity in fowls is not, as has been 

 pointed out before, in all respects an ideal character for the 

 investigation of the laws of inheritance. 



Siimmary of results of all pure Barred Rock matings 



The data presented in detail in this section of the paper, which 

 deals with the matings of Barred Plymouth Rock males and 

 females inter se, would appear to demonstrate the following 

 points. 



1. That there is a definite and clean-cut segregation (in the 

 Mendelian sense) of high fecundity and low fecundity, the char- 

 acter 'fecundity' being here measured by winter tgg production. 

 The mode of inheritance is such as to indicate that winter &gg 

 production depends upon two separately inherited physiological 

 factors. The presence of both of these factors (L^ and Lz) is 

 essential to a high fecundity record. The second factor L2, 

 without which high fecundity never appears is inherited in a 

 sex-correlated manner, such that it is never borne in the same 

 gamete that carries the female sex-factor F. 



2. That the things segregated are perfectly definite and dis- 

 tinct. This is shown by the mean or average production records 

 of the birds falling into the several fecundity classes. The birds 

 bearing the factors for high fecundity have mean winter produc- 

 tion records ranging from two to five or six times as great as 

 the mean production records of birds lacking these high fecun- 

 dity factors. Such difl^erences as these do not depend upon re- 

 fined statistical analysis for their detection and appreciation. 



While by no m^ans all the possible gametic combinations in 

 respect to fecundity within the Barred Rock breed have yet been 

 made, still the range covered by the data given above is fairly 

 wide. All classes of females except the zero producers (class 



