010 THE PAXORPOID COMPLEX, iii'., 



panorpa are the distal forkings of R 2 and K ;i . Elimination of 

 all the branches of R 4 , and all but the most basal of R 3 , would 

 give us the condition found in Belmontia. R 4 only forks once 

 in Archipanorpa; the loss of this fork would give the unforked 

 condition of this vein seen in Belmontia. The latter genus, how- 

 ever, has the more archaic Ri 5) with three branches, as against 

 the two present in Archipanorpa; and this difference shows that 

 reduction is not proceeding on the same lines in these two forms. 



As regards the 1 media, M 1 and M 2 are both forked in Archi- 

 panorpa, simple in Belmontia; while M 4 is simple in the former, 

 but forked in the latter. So that, both as regards Rs and as 

 regards M, Belmontia tends to keep the more posteriorly placed 

 forks, Archipanorpa the more anteriorly placed ones. 



Archipanorpa is more specialised than Belmontia in that it 

 already possesses a partial fusion between M 4 and Cu la , on 

 the same lines as that found in the well-known aberrant hind- 

 wing of the genus Sthenopis of the family Jlepialidae (15, fig. 

 337), and therefore indicating the method by which the present 

 completely fused condition ot these two veins has been brought 

 about in the Lepidoptera ( Section vi . ) . Further, M has already 

 attained the transverse position in this fossil, — a fact which 

 makes it much easier to understand the condition of this vein 

 in almost all Lepidoptera to-day. This evolution of M in the 

 Lepidoptera and in the fossil Archipanorpa is an excellent ex- 

 ample of that parallel development which is so frequently notice- 

 able in widely separated types which have sprung from the same 

 stem, and have therefore inherited the same tendencies. 



In both Belmontia and Archipanorpa, it is Cu lb .vhich con- 

 tinues the line of the main stem of Cuj distad, while Cu la 

 arches up away 'from this line, towards M 4< The condition is 

 of interest, as showing the strong tendency there must have been, 

 from the very first, as soon as stenogenesis set in, for at least 

 a partial fusion of these two veins to take place; or, failing 

 that, for Cu la to be eliminated entirely, leaving the strong, 

 straight Cuj that is so typical of the Mecoptera, Paratrichoptera 

 and Diptera. 



It does not seem possible to compare very closely the vein Se 

 in Archipanorpa and Belmontia, seeing that we do not know 

 how far back towards the base the dichotomy of this vein took 

 place in the former. All we can say for certain is that Archi- 

 panorpa exhibits the more archaic condition, and that the sub- 



