CHAPTER XIV 



CLASSIFICATION 



The Classification of recent forms adopted in this chapter 

 differs from all hitherto published systems chiefly in attempting 

 a more comprehensive scheme for the Zygoptera. The classi- 

 fication of the Anisoptera has engaged the attention of many 

 authors, prominent amongst whom have been de Selys, Ris, 

 Calvert, Needham, R. Martin, Williamson, Karsch and others. 

 Consequently the points still at issue are only minor ones. We 

 still need a thorough study of the Gomphinae, the present division 

 into two tribes being only tentative. The position of the Petaliini 

 as a tribe within the Aeschninae is open to question on the ground 

 that the differences between them and the rest of the subfamily 

 are perhaps great enough to warrant their elevation to subfamily 

 rank. The true relationships of the Gomphinae with the Aeschninae 

 on the one hand, and the lower Libellulidae on the other, still 

 need careful study. It may be necessary to elevate all the 

 subfamilies of the Anisoptera, except the Libellulinae and Cordu- 

 liinae, to the rank of families, the differences between the two 

 latter being of a lower order than those between the others. The 

 division here made of the immense subfamily Libellulinae into 

 tribes follows in the main Ris's excellent modern classification, 

 with the exception that Old and New World parallel groups are 

 reduced to the rank of series within a single tribe. 



In the Zygoptera, the two chief alterations proposed are the 

 separation of the family Lestidae on entirely new characters, and 

 the elevation of the " legions " of de Selys to the rank of subfamilies. 

 Only in the case of the Platycneminae is this step open to doubt, 

 since these forms grade very closely into the Protoneurinae. The 

 tendency of asthenogenesis has been all along towards the 



