BY R. J. TILLY ARD. 859 



other than the wing-venation; such as, for instance, the number 

 and position of the ocelli, the form of the legs and prothorax, 

 and the shape of the head. Recent studies of the nymphal 

 traeheation in the various families have much improved our 

 knowledge of the venation, but they have not yielded a really 

 satisfactory basis of classification on venational characters only. 



In this connection, I should like to point out that the vein 

 which has been interpreted by Comstock and his school (1, 2, 4, 

 5, 6) as 1A in the Homoptera is certainly not that vein, but Cu 2 . 

 Its precedent trachea comes off from Cu, not far from its origin, 

 in all families of the Homoptera, and in all instars in which the 

 wing-sheaths are developed far enough for examination. This con- 

 dition is exactly parallel with that to be found in all instars 

 of the Cockroaches, in which I have recently satisfied myself 

 that the vena dividens, or concave vein separating the anal area 

 from the rest of the wing, is undoubtedly Cu 2 ; and, as this vein 

 also borders the clavus or anal area in the Homoptera, there is 

 no longer any need for a far-fetched attempt to name it 1A, by 

 trying to explain that its trachea has become fused with Cu. 

 The condition shown in the Homoptera and the Cockroaches is 

 also exactly the same as that seen in the Panorpoid Orders, in 

 which I have already shown (11) that Comstock makes the same 

 mistake in working out the venation of the Trichoptera and 

 Lepidoptera, though he has got the names of the veins right in 

 the Megaloptera, and thereby himself admitted his error in the 

 other Orders. I have also, finally, studied the venation in all 

 examinable stages of the Psocoptera, in which this same con- 

 dition is also to be found. The only conclusion to be come 

 to is that the cubitus in all these groups is normally three- 

 branched, C\x 1 forking distally into Cu la and Cu lb , while Cu 2 is 

 a straight, concave vein, which, in those Orders which have a 

 separate clavus, borders that area anteriorly. I have therefore 

 marked the vena dividens, in all my figures, as Cu 2 ; in this 

 respect, my nomenclature for the Homoptera differs from all 

 the figures of the venation of this Order so far published except 

 Mutt's (7) . 



It follows from this that the veins on the clavus itself are 1A 

 and 2A. Vein 3A is absent in all recent forms, but is clearly 

 to be seen in the Permian genus Permofulgor (9), and in the 

 Liassic genus Archijassus (3) . 



In working out the Ipswich fossil Homoptera, it is at once 



