No. 556] SHORTER ARTICLES AND DISCUSSION 



2?3 



compare only offspring from similarly fed mothers. All data 

 considered below are from offspring of opium-fed mothers. 



To decide whether or not we have "pure lines" in Aphis it 

 is necessary to breed two generations of offspring; it is better 



Opiu 



to breed more. It is necessary to breed two lines through these 

 generations ; it is better to breed three or more. The results so 

 far obtained are inadequate since they continue only one line 

 through two generations of offspring. The data obtained are as 

 follows: 



A wingless mother, whose antennal index was not obtained, 

 was fed on opium poppy and produced 18 offspring. The dis- 

 tribution of the antennal ratios of these offspring is as given 

 in Fig. l. The mode is at 1.50-1.59, the average is 1.53. From 

 this fraternity three individuals were now selected as mothers 

 of the next. We may call them 5a, 56 and 5c (Figs. 4, 5, 6). 5a 

 has the highest antennal ratio, 1.46. The mode of the progeny is at 

 1.55 (1.50-1.59) and the average ratio is 1.49. 5c has the next 

 lowest ratio, 1.32. Her progeny also have the mode at 1.50-1.59. 

 The third mother (56) has a ratio of 1.22. The progeny is 

 few in numbers and has two modes of which the major is 

 1-50-1.59 like the two others; and the average is 1.47. 



The foregoing series of facts may be tabulated as in Table 



: 



