BY R. J. TILLYARD. 291 



from tables or keys based upon venation, it seems to me that 

 much confusion would be avoided (especially for those whose 

 knowledge of the group is not extensive) if two points were 

 carefully borne in mind. Firstly, it is necessary to be able to 

 pick up the median vein at a glance. But, owing to the basal 

 fusion of this vein with R or Rs, how can we recognise it with 

 certainty, except by recourse to the pupal tracheation 1 This 

 can always be done by looking for the radio-median Jurrow 

 (Text-fig. 2, rm), a deep groove which separates the last (most 

 basal) branch of R in the forewing, or Rs in the hindwing, from 

 M. This furrow runs just anteriorly to M, and, in certain lights, 

 it shows up as a white, shining line, even more conspicuous than 

 the veins themselves. The median vein also is always two- 

 branched in Bemerobiidce, the fork being called the median fork 

 (Text-fig. 2, mf). In the Plates, rm is represented as a dotted 

 line. A similar, very distinct furrow separates Cu from A, and 

 is designated the cubito-anal furrow (Text-fig.2, cm«). Occa- 

 sionally, as in Z)/(Sprt«eji;<ert/.r (Plate xii., fig. 1), a third furrow, 

 the medio-cuhital, separates M from Cu, but usually this is 

 absent. 



Secondly, it must be remembered that the cubitus of the 

 hindwing in Hemerobiidce is a highly specialised vein. In most 

 genera, it forms a high ridge, and is much stronger and thicker 

 than any vein near it. Now the sharp forking of this ridge, 

 which can often be seen at a level distad from that of mf, is tiot 

 the primary fork (cuf) dividing Cu into Cuj and Cu.,, but the 

 secondary fork (cuf) dividing Cuj into Cuj^ and Cun,.* This 

 can be seen at once by referring to Drepanepteryx or Drepanacra 

 (Plate xiii.), where c?t/'may be seen very close to the base of the 

 wing, with Cu, as a weak vein running parallel and close up to 

 Cuj. How Cug has been lost, can be clearly seen in Drepano 

 mina (Plate xiv., fig. 18), where the cubito-anal furrow is double, 



* The re-semblance of this secondary fork to the cubital fork in Myrme- 

 leontidw has led me to re-examine the structure of the cubitus in this latter 

 family. As a result, I have discovered, in the Dendroieontiiue and Acan- 

 thaclis'is, the remnant of the true Cu., near the base of the forewing. It 

 follows that the branch hitherto called Cuj is in reality Cujb for the fore- 

 wing, and probably for the hindwing also. 



