202 DARWINISM AND HUMAN LIFE 



is given by Prof. H. E. Crampton ^ for the Saturnid 

 moth, Philosamia cynihia. A large number of pupae 

 were collected from a small area, and kept till they 

 hatched. But only 16"6 per cent, of the total 

 number collected gave perfect moths. Many of 

 the pupae were dead within the cocoon (" pupal 

 elimination "), 129 out of 310 died in the period 

 between the formation of the imago and its 

 emergence (" pupal imaginal elimination "). Mr. 

 Crampton compared 134 male pupae that survived 

 pupation with 130 that died immediately after 

 pupation, as regards length, width, and depth of 

 the bust of each, and as regards length and breadth 

 of antennae. Those that survived were longer, 

 narrower in the bust, and had longer, stouter 

 antennae. Similarly, he compared 176 surviving 

 female pupae with 180 that died, and selection, in 

 type was found to be certain for all dimension^, 

 and to be in the same direction as in the male 

 pupae. The survivors were also less variable. 

 We need not discuss the pupal-imaginal elimina- 

 tion, where the results were somewhat different. 

 It is interesting to notice, as Mr. Crampton points 

 out, that the selected characters are not such as 

 seem to be directly or indirectly " useful " to 

 their possessors, yet they are demonstrated to have 

 the high utiUty of determining survival — which 

 is indeed, for the evolutionist, the final criterion 

 of utihty.* 



We cannot do more than allude to the careful 

 statistical methods by which Prof. Karl Pearson 

 and others have proved that there is selective 



1 " Biometrika ' (1904), vol. iii. pp. 113-30. 



2 See " The Evidence of Natural Selection." by E. S. Russell, 

 in Rivista di Scienza (1909), vol. iii. 



