370 MB. H. W. BATES — CATALOGTJE OF EETCIKID^. 



Taking the number of brandies to tbe fore-wing subcostal ner- 

 vure as tlie groundwork, it is found tbat all tlie Old-World 

 genera have four, and the great majority of the New- World genera 

 only tliree. The few New-World genera in which there are foui\ 

 grouped with those of the Old World, form our first subfamily , 

 which I believe to be a perfectly natural one — that is, to comprise 

 forms having a closer blood-relationship to each other than to 

 the members of the other subfamilies. The only exceptions to 

 the number of subcostal branches in the group are presented by 

 two or three small species of Ilesosemia *, in which the wing- 

 neuration is very aberrant ; these, however, offer no practical dif- 

 ficulty in the employment of the character. 



This first subdivision disposed of, we come to a group of species 

 of peculiar fades, in which the number of subcostal branches 

 varies from two to four, and even varies according to sex. These 

 are the Eurygoninae, which there is no difficulty in distinguishing 

 from the other groups, by the mode in which the lower radial (dis- 

 coidal) nervure is connected with the subcostal, especially in the 

 lower wings ; it is so perfectly connected that it seems to be a 

 continuation or a lower branch of the subcostal. A trace of this 

 important peculiarity is seen in the first subfamily ; but in the 

 remaining groups the nervure in question is completely detached, 

 or united at a right angle to the subcostal by an imperfect nervule. 

 This character leads us to place the peculiar genus Metlionella in 

 the Eurygoninae group, which forms our second subfamily. 



In all the rest of the family the subcostal nervure possesses 

 only three branches, if we may set aside one or two species as 

 anomalous exceptions ; and the lower radial is disconnected from 

 the subcostal, as just now explained. This division, which I call 

 the subfamily Erycininse, comprehends the great majority of the 

 New- World genera; and although it is further divisible, the 

 groups thus formed are not of the same rank as the three already 

 defined. To subdivide it, I have taken, in the first place, the 

 colour of the antennse, separating those genera in which these 

 organs are destitute of rings or spots of paler hue, from the re- 

 mainder, in which they are more or less distinctly ringed. These 

 two subdivisions seem to be tolerably natural ; for the first com- 

 prehends every species in which the fore-wing subcostal nervure 

 emits its second branch after the end of the ceU, and the second 

 excludes every such species, with one evidently aberrant ex- 

 ception, viz. Selicojpis Cwpido. 



* M. tenera and M. idotea. 



