020 ORTHOPTERA AND NEUROPTERA. 



resemblance. This is more marked between the two first-named 

 famiHes, however, than between them and the third. Mouth-parts 

 and female genitaha illustrate this. The Acrididae, however, 

 approach more nearly the Phasmidae and Phyllidae in these 

 features. 



I am of the opinion that we should properly' set off the 

 Blattidae and Mantidae, as an order, using Crampton's term 

 Endictyoptera. I must confess that I am not able to decide 

 whether the other families that I have mentioned should be one 

 order or more. Failing knowledge, I am satisfied to leave the 

 families Gryllidae, Locustidae, Acrididae, Phyllidae, and Phas- 

 midae in one order under the old name Orthoptera.. I might here 

 point out that the Orthoptera thus restricted are less akin to the 

 Eudictyoptera than are either the Isoptera or the Plecoptera. 



The totally aberrant wings of the Forficididae are alone 

 enough to warrant their forming the order Deruiaptera, as in 

 Comstock's scheme of 1895. This order, of course, comes close 

 to the Eudictyoptera. 



To save time, I omit considering in detail the interesting 

 smaller groups, such as the Panorpidae, which cannot be Neurop- 

 tera; and shall close by submitting that classification which 

 appears to me to represent the facts of relationship. I have 

 purposely omitted any grouping into superorders or subclasses. 



