82 H. E. EWING. 



The results obtained under these more definite conditions are 

 shown in Fig. 8 and Table VIII. By comparing these results 

 with those obtained for the first 20 generations we note, first of 

 all, much less fluctuation in fraternal means. We note that the 

 extreme variants do not vary nearly as much from the mean of 

 the line as they did in the first 20 generations. 



We find that the heavy line in Fig. 8 joining the various fra- 

 ternal means shows no general trend upwards, so we conclude 

 that selection has produced no perceptible inherited change in 

 body length. The last mean obtained, 1.611 mm., is just 6 

 thousandths of a millimeter less than the mean of the fraternity 

 started with, 1.617 mm. 



The Effect of the Use of Older Wheat Plants, the Last Four Genera- 



ations of Isolation No. n. 



During the last 4 generations of subline II the aphids were 

 reared on wheat plants several days older than those previously 

 used. During the first and second score of generations the young 

 aphids were placed on wheat shoots to be reared when these 

 shoots were barely out of the soil. In almost every case the first 

 leaf had not been put out. Most of these shoots were less than 

 twenty-four hours old when the aphids were placed on them. 

 In the last four generations all of the wheat plants were several 

 days older. They were from 5 to II days old when the young 

 aphids were added, and by the time these insects reached ma- 

 turity and gave birth to their progeny most of these plants were 

 two weeks old or older. 



The offspring of iigsi were the first to be placed on these older 

 wheat plants. After reaching maturity their lengths were ob- 

 tained and the fraternal mean computed. It was found to be 

 1.343 mm., a big drop from that of the previous fraternity, 

 which was 1.611 mm. The means of the three following genera- 

 tions were also greatly below- this point and the mean of the line. 

 The last mean obtained for these 4 generations and for the 

 whole subline, or isolation u was 1.107 mm., which it is observed 

 is greatly lower than any mean previously obtained and only 

 u hundredths of a millimeter above the lowest variant obtained 

 in all the first 40 generations where younger wheat shoots were 

 used. 



