H. G. MAY. 



Return selections from the low lines produced on the average 

 slightly higher numbers than low selections in the same genera- 

 tions. Individual cases, however, do not seem to show any 

 definite response. The few return selections made from the 



200 



180 

 160 

 140 

 120 



too 



80 

 60 

 40 

 



Feb. 6 8 10 12 14 16 18 20 22 24 26 28 Mr. 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 Apr4 6 



220, , 



200 



180 



160 



140 



120 



100 



80 



60 



40 



20 







A 



ttpr.6 8 10 12 14 16 18 20 22 24 26 28 30 Ilay4 68 1O 12 14 16 18 20 22 24 26 28 30 Je.3 5 7 



FIG. 8. Daily averages of high and low lines in Ba from February 6 to June 

 7, 1916. The high lines are represented by continuous lines and the low ones 

 by broken lines. The two upper lines show the daily facet averages while the lower 

 ones represent the actual numbers of flies from which those averages were obtained. 



high lines were even less definite. Only one gave offspring that 

 could be distinguished from those of the generation in which 

 they hatched (u4f4-i9). 



The offspring from the crosses between low and high lines are 

 intermediate and show a very slightly wider range than the low 

 lines, but not nearly so wide a range as the high lines. There 

 is no significant difference between the offspring from high males 

 and low females and those from low males and high females. 

 The apparent difference in the extremes is accounted for by the 

 fact that the standard of all the lines changed during the period 

 of hatching. The results of these matings are given in actual 

 daily counts in Tables III. and IV., and are also shown in Figs. 

 5 and 7. A comparison of Tables III. and IV. with Fig. 8 shows 



