REPORT OF THE STATE ENTOMOLOGIST IQO/ 221 



principal veins, the same upward hitch of vein Cu^ against media, 

 and many of the cross veins occupy identical positions. Especially 

 striking are the first two cross veins in the first fork of media, 

 one delimiting, the other traversing cell ist M^. The suggestion 

 has been made before by others, and I think it very possible, that 

 some Panorpidlike neuropteroid mutant got its center of gravity 

 hitched forward, its hind wings reduced, and started the dipterous 

 line of evolution. 



Homologies of cross veins. In my study of the venation of 

 the Odonata, I was quite unable to homologize any of their cross 

 veins with those found in other orders qf insects. And I do not 

 iDelieve that those indicated in the Comstock-Needham typical wing 

 are necessarily homologous, even in those orders in which single 

 cross veins occur at the points indicated for them in our type, for, 

 primarily, cross veins are not formed about strong tracheae (they 

 contain either late developing tracheal twigs or none at all), and 

 they show, so far as I can see, none of the earmarks of homology. 

 I conceive that such cross veins, as we may fairly regard as typical 

 by reason of their frequent recurrence, are the survivors of the 

 long elimination process just discussed. They are the cross veins 

 that happened to stand in the positions most favorable for con- 

 necting together longitudinal veins, ordinarily at the points where 

 dichotomous branches came nearest together. If, as seems prob- 

 able, there were originally many cross veins, and if the forks of 

 the principal veins varied somewhat in length and position in the 

 ancestors of different groups, the same particular cross veins might 

 not, probably would not, be preserved in every case. Those most 

 useful would, naturally, survive the elimination process. Yet, with 

 a similar form of wing and the same general primary disposition of 

 branches of tracheae and veins, the process of elimination might 

 leave a few strong cross veins in corresponding positions in very 

 different insects, for it is always to be remembered that all wings 

 have had to meet like conditions : the air is the same for all. The 

 cross veins of the Comstock-Needham type are such merely as 

 recur in like position in a large proportion of winged insects, and 

 whether strictly homologous or not, it is convenient to designate 

 them by the simple method that Professor Comstock devised. It is 

 in this sense that these designations are used in this paper. 



Some general features of the Tipulid zving 

 The primitive ancestral crane fly doubtless possessed more veins 

 in its wing than were necessary or advantageous, and these were 



