J. W. H. Harrison 49 



his poultry ; since his mortality rate was negligible, in all probability 

 we have to look to both of the suggested causes for a correct explanation 

 in this instance. 



(4) The progeny of the treated batches w^as greatly superior in 

 several respects to that of the untreated controls. 



(5) The offspring from the cross between a treated male and an 

 untreated female was superior to that of the reverse mating to much the 

 same degree as the latter was better than the controls. 



(6) The superiority of the broods originating with treated parents 

 was exhibited in : 



(a) The smaller percentage of embryos perishing in the shell, the 

 exact figures being 201% and 2*9% against 7'8%. 



(6) The quicker development of the embryo. 



(c) The speedier rate of feeding up in the larva. 



(d) The lower rate of larva mortality, % , and 1 8 "/„ against 46 °/^ . 



(e) The greater mean weight of the male pupa. 

 (/) The greater mean weight of the female pupa. 



(^f) The quicker development of the imago and its earlier 

 appearance. 



(h) Its richer pigmentation. 



(7) In sex ratio and in the percentage of fertile eggs no differences 

 were perceptible. 



Except, therefore, in the percentage of fertile eggs and in characters 

 only capable of manifestation in the lepidoptera, my data in the main 

 tend to point in the same direction as those of Pearl. 



To explain the results I think we have to look to the cumulative 

 effects of selection of two different types acting at different stages : 

 (I) the first of these acts on the parental zygotes when the alcohol 

 weeds out the weaker insects and preserves the stronger to perpetuate 

 their race ; (II) the second works directly on the germ cells of these 

 selected, stronger, and sturdier survivors. 



The first point needs little or no elaboration as it is almost self- 

 explanatory ; still perhaps a few words are necessary. As only insects 

 superior in size and vigour survived the action of the alcohol to act 

 as parents it is a fair inference that their gametes would be of a 

 superior type. Of necessity, then, zygotes into whose composition 

 these gametes enter would manifest greater strength and vitality. 

 Thus if the qualities in which the broods from alcoholised parents excel 

 depend in part at least on parental excellence then such broods would 



Journ. of Gen. ix 4 



