BY R. J. TILLYARD. G23 



between the most archaic types of the Cyclorrhapha, such as 

 the Syrphiduc, and the Myiodaria. The former have the original 

 arculus, or cubito-median Y-vein, strongly formed near the base 

 of the wing; while the secondary arculus, formed from the base 

 of M 3+4 and the cross-vein m-cu, is only just beginning to move 

 basad, and is little altered from its normal position as seen in 

 the Tabanidae. The line of evolution of the forewings of the 

 Myiodaria, then, parallels that of the forewings of the higher 

 Heteroneurous Lepidoptera, in that it consists chiefly of the 

 movement basad, and finally the suppression, of the cubito- 

 median Y-vein, and a great expansion of the more distal por- 

 tions of the wing. We may profitably contrast the position of 

 the cross-vein r-m in the wings of the Syrphidae and the Taeh- 

 inidae, and note also the immense increase in the area occupied 

 by the median cell (mc) in the latter. 



It does not seem necessary here to say much upon the internal 

 Phylogeny of the Diptera. It is quite clear, from the evidence 

 of the wing-venation, that the primary dichotomy must have 

 been into Nemooerr and Brachycera, and not into Orthorrhapha 

 and Cyclorrhapha. The Nemocera are properly defined as that 

 line of evolution in which reduction of Rs has been brought 

 about by loss of the fork of R 4+5 , and in which little or no 

 tendency to a reduction in the distal joints of the antennae is 

 manifest. The Brachycera, on the other hand, are that line of 

 evolution in which the reduction of Rs has been brought about 

 reduction is evident in the antennal distalia, leading to the three- 

 by the loss of the fork of R 2+ g, and in which also a progressive 

 jointed type, with arista, as its highest expression. Nature only 

 makes such dichotomies as these clear-cut, through the dying-out 

 of intermediate forms. Thus we should neither be surprised nor 

 annoyed, (though it may be very awkward for the framing of 

 good systematic definitions), when we still find, in this Order, 

 a few types which have preserved all four branches of Rs in- 

 tact (e.g. Psychoda, Pericoma, Protoplasa), or a few more which, 

 while obviously Brachycera as regards the condition of Rs, have 

 still fairly numerous distalia in their antennae (e.g. Rhyphus, 

 Xyhphagus, Arthroceras) . The Phylogenist will welcome the 

 presence of these forms, just as heartily as some Systematists 

 will anathematise them; and it should be remembered that the 

 attitude of mind, that puts systematic convenience before natural 



