able differences in this respect. With this reservation, not- 

 withstanding, we are able to note frequently considerable 

 differences in flight and habits in flight within the limits 

 even of a single family, and it is significant that we are able 

 to find considerable differences in the neuration of the organs 

 of flight, the wings. 



This theory is well supported by looking at the family 

 Geometridce which contains species with a great variety of 

 habits — species that are of robust build, species that are of 

 slender build, species that fly largely at twilight, species 

 that dart about in the bright sunshine, and species that fly 

 late at night. Even among our British geometrid species 

 there is a vast difference in habit between Oiirapteryx samhu- 

 caria, Nyssia zonaria, Acidalia nversata, Fidonia atomnria, and 

 Boarmia rohoraria. There is also a considerable difference 

 in the neuration of these species. Compare the fore-wing of 

 Acidalia aversata with that of Nyssia zonaria. Vein 3 is in 

 quite a different position, being in the former from long 

 before the end of the cell, and in the latter from the angle 

 of the cell. Again, veins 10 and 11 are quite different in the 

 two species. So with Ouraptcryx sanibucaria and Fidonia 

 atojiiaria there is a considerable difference in the neuration. 

 Oiirapteryx has lost vein 10, while Fidonia retains it. The 

 habits of these geometrids are all different, and the mode 

 of flight is equally varied. Contrast the members of 

 the family Sphingida, whose habits are very much the 

 same : they are all pretty nearly constant to a more or 

 less uniform arrangement of the nervures. There are but 

 very few exceptional cases, like the genus Cephonodes* with 

 a very short cell to the hind-wing, and Daphnnsa with Sc'~ (= 

 vein 7) and R^ (= vein 6) on a long stalk; even this latter 

 case is only a small variation and a frequent one in other 

 groups. The remarkable constancy of the neuration has 

 prevented all authors from finding generic characters from 

 this structure. But, all the same, this constancy is of very 

 great value as a family character, and it is scarcely possible 

 to mistake any member of this family as belonging to any 

 other group, although the general build and outline of some 

 of the species have departed very considerably from the 

 typical build and appearance of the family. 



Take, again, the butterfly family P/trzWo'. How varying 

 are the different species in their habits ! Colias edusa and 

 Colias hyale, as well as all other species of the genus, have a 

 strong and exceptionalh' powerful flight. Encldoc cardamines 



* Vide Rothschild and Jordan, " Review of Sphingidce," p. Ixix. 



