66 ENTOMOLOGICAL NEWS [Mar., '19 



is strongly similar to the Embiids and Dermaptera) the re- 

 semblances are fairly evenly divided between the Panplecoptera 

 and the Pandictyoptera. Grylloblatta has many features in 

 common with the Embiids, Dermaptera, Isoptera, Zoraptera 

 and Mantids, and fewer in common with the Blattids and 

 Plecoptera, so that the choice which we make as to what 

 forms more nearly represent the ancestors of the Grylloblattids 

 depends upon what structures we consider the most important 

 for a phylogenetic study. From my own studies of a rather 

 wide range of anatomical structures, I would be more inclined 

 to regard the neck and prothoracic sclerites as the most de- 

 pendable features, and taking the evidence as a whole, I have 

 become convinced that Grylloblattids arose from a Plecopteroid 

 stock* (Panplecoptera) rather than from a Blattoid stock 

 (Panisoptera), although it is quite evident that the Grylloblat- 

 tids branched off from the Plecopteroid stock very near the 

 point at which the Blattoid lines of development likewise 

 branched off from the same Plecopteroid stock, as indicated in 

 the diagram (page 43). 



While much attention has been paid to Grylloblatta as the 

 most primitive representative of the Orthopteroid group, it 

 must be borne in mind that all of the evidence of relationship 

 must be considered from every available source, and in this 

 connection it would be a very grave error to slight the evi- 

 dences of relationship presented by that other very primitive 

 Orthopteroid insect Timema calif ornica Scud., (a small wing- 

 less Phasmid), since Timema has preserved certain features 

 which even Grylloblatta has lost. Timema, like Grylloblatta, is 

 wingless in both sexes, and anyone who will compare Timema 

 and Grylloblatta with the wingless females of the Embiids or 

 Dermaptera will certainly concede that the general appearance 

 of the body is very much more similar in these insects than is 

 the case when one compares the Grylloblattids with the average 

 wingless Blattid or Mantid (or even with the Isoptera, for 

 that matter). Again, while the tarsi of Grylloblatta are five- 

 jointed and the legs are quite like those of the Blattids, the 



*See footnote, page 48. 



