144 A. FRANKLIN SHULL 



more line would have shown a higher percentage of male-pro- 

 ducers than the New York line. Nor is this all. Half a dozen 

 successive generations may be selected from any point in either 

 line, and compared with the generations of the other line bred at 

 the same time, and in nearly every case the percentage of male 

 producers will be higher in the Baltimore line than in that from 

 New York. 



It is safe to say, therefore, that we have here two pure lines 

 that differ from one another in a fairly constant manner, and the 

 difference is an internal one. This ev dence seems on the face of it 

 to support Punnett's contention that differences in pure lines 

 might account for the results obtained by himself and previous 

 workers. How far it supports his view, and whether the internal 

 differences discovered are due to zygotic constitution, as Punnett 

 suggested, are discussed elsewhere. 



Such distinct pure lines having been found, it is important to 

 known how the proportion of male-producers will behave if 

 individuals of different pure finds are crossed. Will the propor- 

 tion of male-producers in one of these lines behave as a dominant 

 to that in the other, or will it exhibit no relation to Mendelian 

 phenomena? To answer these and other questions, the several 

 following experiments were performed. 



Experiment XXXIII. A female from the eleventh generation 

 of the Baltimore pure line of the preceding experiment, was paired 

 April 12, 1910, with a male from the New York line. Resting 

 eggs were laid by her on April 14 and several days following. 

 These eggs were kept at room temperature, and from one of them, 

 on April 23, there hatched a female. From this female the line 

 in table 35 A, designated ''Cross," was reared. 



Another female of the Baltimore line was paired April 13, 1910, 

 with a male from the New York line. One of the resting eggs 

 from this female, having been kept at room temperature, hatched 

 April 28. From this female offspring the line designated ''Cross" 

 in table 35 B, was bred. Each pure line designated "Cross" is 

 compared with those parts of the parent (New York and Balti- 

 more) pure lines which occurred simultaneously with it. All lines 



