1912.] Cross-breeding of Two Races of Acidalia virgularia. 51 



It would be natural to suppose that such a great difference in size between 

 two broods would be due to heredity or to some great difference in the 

 conditions in the larval stage, but I can find nothing of the sort to account 

 for it. 



The parents of Brood 40 measured £ 18-0, ? 203, whilst those of Brood 34 

 measured <$ 17"9, ? 19*6. 



The parents of 40 were both members of 33, which averaged S ' s 17"7, 

 ? 's 19 7 ; the $ parent of 34 belonged to 30, which averaged c?'sl7"0, 

 ? 's 18"7, whilst the ? parent of 34 belonged to 28, which averaged 

 d's 17'5, ? 's 18*2. It is true that the ancestors of Brood 34 were dis- 

 tinctly smaller than those of Brood 40, but they did not differ from the 

 normal nearly so widely as does Brood 34 itself ; and, taking into account all 

 broods containing over 10 moths, I find that the average spread of the 

 brood shows no correlation with that of its parent nor with those of the 

 broods of which the parents were members. I conclude, therefore, that 

 breadth of wing is not hereditary. 



It is, of course, impossible to prove that these variations in size are not due 

 to differences of conditions, though most of the conditions have been kept 

 very uniform. The moths were reared throughout in similar boxes and the 

 larvae never fed on anything but dandelions. The main difference of 

 conditions between the broods was due to the different temperatures of 

 different periods of the year. This has a marked effect on the length of 

 time required for development. Eggs laid from April to August only take 

 about two months to reach the imago stage, whilst eggs laid from September 

 to November take about four months if the larvae continue feeding, or from 

 seven to eight months if they definitely hibernate and cease feeding. 



In spite of these great differences, moths of the broods which have 

 hibernated are, on the average, exactly the same size as those which have 

 completed their development in two months. 



If the differences were due to different temperatures at the moment 

 when the moths emerged, those broods which emerged from July to October 

 must' have experienced warmer conditions for emergence than those which 

 emerged from December to June, yet on the average there is no difference 

 between the two groups. I am therefore entirely at a loss to explain the 

 considerable variations in size which occur. 



Summary. 



It will be seen that I agree with Messrs. Prout and Bacot's conclusion 

 that Acidalia virgularia and its variety canteneraria are not two Mendelian 

 forms of the species, though I believe I have shown that one of the 



E 2 



