

52 PROCEEDINGS ENTOMOLOGICAL SOCIETY 



lower families also. A species of Monopis, for example, can 

 always be determined generically with certainty from the vena- 

 tion alone, though hardly two specimens can be found in which 

 /the vein course is identical. 



Much more difficult problems arise because the potentiality of 



'vein differentiation after all is a limited one, and that therefore 



genera of different groups, converging toward the same general 



scheme of a wing support, may independently attain results which 



are embarrassingly similar. 



This fact has in many cases long delayed the true appreciation of 

 some groups and only patient study and comparison with allied 

 forms have made it possible to place them where they rightly 

 belong. An example will illustrate these difficulties. The Gra- 

 cilaridse, to which the leafmining genus, Phyllonorycter (Litho- 

 colletis), belongs, are distinguished from all other Micros by the 

 fact that the mature larvae have prolegs on only three of the 

 middle abdominal segments instead of on four or more, and by 

 the fact that the first larval stages exhibit a very peculiar, highly 

 specialized modification of the mouth parts, not approached by 

 any other lepidopterous larva. Until recently the group has 

 been regarded as a part of the Tineidse or Plutellidse, because the 

 pterogostic and oral characters apparently did not preclude a 

 derivation from some generalized form within these families. 

 However, it was felt that they were in some way out of place 

 and about a year ago, our fellow member, Dr. Charles Ely, who 

 was making a thorough study of the American species of Gracilaria, 

 called my attention to an undeniable evidence of their separate 

 family rank, which has been before all students of the group in 

 Stainton's careful plates of wing venation published fifty years 

 ago, but which no one had properly interpreted before. Stainton's 

 figures of the wings of Gracilaria and Ornix both have the full 

 number of veins in the hind-wings, eight, apparently only slightly 

 differently placed. But if we study them carefully, we will see that 

 it is not the same veins that have persisted. Ornix has one costal 

 vein more than Gracilaria and Gracilaria has one dorsal vein more 

 than Ornix. 



The large number of common characters both in the. larvae 

 and the adults prove conclusively that these two genera are truly 

 closely related. The only explanation, then, is that both must 

 have been derived from a cpmrnon ancestor having nine or more 

 veins. By denuding various species of Gracilaria we found some 

 in which all of the nine and even ten veins had persisted. Such 

 a group with this number of veins clearly can not be derived from 

 either the Tineidae or the Plutellidae, which have only eight veins, 

 but must have developed independently, from a point lower down, 

 nearer the Micropterygidae. 



