227 



classification would, I believe, arise if we reject antennal 

 characters altogether. 



It will be seen that some of the objections so forcibly 

 urged by Mr. Meyrick (Trans. N. Z'd. Inst., 1916, p. 248) 

 against the generic value of modifications of the discocellulars 

 and origin of vein 5 of the hindwings apply also to characters 

 which he recognizes as valid. If applied with impartial 

 logic, they would destroy his own, and, I believe, any other 

 possible classification of the family. It must be admitted 

 that here also intermediate forms occur, though rarely, 

 but they are not such as should create any real difficulty. 

 Vein 5, which is the second median vein, arises normally 

 opposite the termination of the upper primary branch of the 

 median trachea, which becomes obsolete in the adult wing, 

 but its point of termination is often traceable, often situated 

 centrally, but often considerably nearer the radius than the 

 cubitus. This is the structure in Etiphyia, Xanfhorhoe, and 

 most of the genera of the family. The approximation of 

 5 to 6 is often conspicuous, but I do not attach generic 

 im.portanc6 to it, for 5 appears never to rise from above tbe 

 termination of the upper primary branch of the media as it 

 does in the Geometridae fsensu stricto). Usually with this 

 origin of 5 the discocellulars are straight or nearly so, but 

 not always (see for instance Epirrlioe Kociafa, Bkh.). In 

 many genera such as Cidaria and Larentia a striking modifi- 

 cation occurs. In them 5 arises from well helow the termina- 

 tion of the upper primary branch of the media, and there is 

 a strong bend approximating to a right angle at its point 

 of origin. Usually 5 is also strongly approximated to 4 at 

 origin, but not always. In microcyma, for instance, it is 

 from not much below the middle, but the discocellular is 

 strongly bent at the usual point (not straight, as erroneously 

 ;stated .in my former revision). Tliis structural division as 

 thus understood appears clear-cut, aijd I have not so far met 

 with a really doubtful case. Nor do I find that the genera 

 defined by it are less natural than those defined by the areole 

 or antenna! pectination, when considered as a whole. It 

 must, however, be admitted that, as Mr. Meyrick points 

 out, difficulties occur in the New Zealand fauna. Larewtia 

 cineraria is extremely similar to XanthorJinl phrmhea, but 

 here the similarity of grey coloration (doubtless protective) 

 and very simple pattern is one that might well have been 

 independently acquired, and I think we can here trust 

 •structure before appearance. The case of X. adonis, L. 

 heata, and L. henedicta is more difficult. These certainly at 

 lirst sight appear nearly allied, the last two, however, rather 

 more closely than the first, which, except in colour, is very 



