128 



Isabel McCracken. 



diagram is characteristic of the pedigree of all of the fourth genera- 

 tion individuals specified under total in Table IV. 



Fig. 2 shows, as previously stated, that in the first generation 

 from S X S, pure S broods only were obtained, in the second 

 generation both pure and mixed broods were obtained {i. e., S, and 

 S and B), in the third generation pure S broods only. 



Experimerit 4.. With second generation broods (first genera- 

 tion bred from known parents) some crosses were made between 



13486" 134S9 



144 Sd^ 144 S9 32 Sc? 32 S$ 73 Sc? 73 S 9 



Sd" and 9 



Sd" and 9 



S d'and 9 



Sd" and 9 



Sd* and 9 



Sd" and 9 



Sd" and 9 



Sd" and 9, S Bd" and 9 



Sd" and 9, S Bd* and 9 



Sd^ and 9 



Fig. 3. 

 Diagram showing the pedigree of the broods recorded in Table V, c and d, namely, broods obtained 

 from 8Sd^X24S9 and24S9X8Sd'. 



extremes of color type to determine the relation the color types 

 bore to each other in heredity in case each extreme was found to 

 breed true to itself, in other words to test the validity of Mendel's 

 law of dominance under these conditions. 



Eight pure S broods and eight pure B broods were chosen from 

 Table I, a and b, column 6. Each cross was between a half brood 

 of B's and a half brood of S's. Table VIII shows the results of 

 the reciprocal cross matings, and Tables VI and VII the results 

 of control pure matings. 



