clxxxiv Proposed Division of Neuroptera 



and incapable of feeding : the much-discussed exception in the instance 

 of Raphidia ophiopsis is apparent, and not real ; in this insect there 

 is no cocoon, and the pupa inhabits exactly the same situations as the 

 larva, namely, excavated passages in the decayed bark of trees ; it 

 moves after a fashion, but only in the same way that the pupa of a 

 Hepialus or an JEgeria may be said to move; and in each of the three 

 instances the character, as regards the tribe, is exceptional, and not 

 distinctive. The pupa in the Neuroptera, on the contrary, is uniformly 

 active and voracious, although not uniformly isomorphous ; the squat 

 toad-like pupa of the Libellulina being very unlike the elegant imago. 

 Natural classification is tripod, and metamorphosis, alary structure 

 and cibarian structure are the three legs on which it stands : to neg- 

 lect metamorphosis, therefore, where so decidedly marked as in the 

 present instance, were to undermine the entire edifice, and to endan- 

 ger the stability of the building. The flight, again, in the typical 

 Neuroptera is pre-eminently strong and long-continued, gradually fail- 

 ing in aberrant orders, and perhaps becoming totally extinct among 

 the most aberrant of all, the Thripsina : in the Stegoptera the flight is 

 generally of short duration, heavy, fluttering and laboured, and the more 

 strongly marked the characters of the class, the more feeble are the 

 powers of flight. Again, as to food : the conspicuous and numerically- 

 preponderating Libellulina in the Neuroptera are predaceous and vo- 

 racious in an eminent degree ; whereas the equally well marked and 

 equally numerous Phryganeina, in the Stegoptera, not only are not 

 predaceous, but are not known to take any food : and when we find 

 the Stegoptera exhibiting a predaceous and raptorial structure, as in 

 Raphidiina and Mantispina, such characters are associated with the 

 assumption of other Neuropterous characters, as the nude, hyaline, 

 glittering, and strongly reticulated character of the wings, and the 

 nudity of the pupa. These considerations lead to the belief that the 

 Neuroptera and Stegoptera are distinct but proximate classes. In 

 addition to these observations, it seems desirable to explain that these 

 classes, although bearing old names, are nevertheless totally new. I 

 feel an absolutely insuperable objection to giving new names where 

 old ones can by any ingenuity be made to answer the purpose ; and 

 although I take on myself the responsibility of the new combinations 

 of groups, and of the present application of the names, I trust that en- 

 tomologists will feel no difficulty in the mental connexion of the word 

 Neuroptera with the dragon-flies, heretofore and still the type of the 

 class, nor of the word Stegoptera, or "roof-winged," with the ant-lions 

 and lace-wings, for which Dumeril originally intended it. 



