PROC. ENT. SOC. WASH., VOL. 23, NO. 2, FEB., 1921 



39 



As was mentioned above, when we trace the ancestry of the 

 Hymenoptera a step further, the line of development of their 

 precursors would quickly merge with the lines of development of 

 the Isoptera (with the Zoraptera) and the Dermaptera (with the 

 Coleoptera), and since at this level of development, the differ- 

 ent groups under discussion become connected together on all 

 sides by mutual bonds of relationship in the form of annectant 

 types linking them together in all directions, it would be more 

 accurate to represent these interrelationships by a solid cone, 

 instead of depicting the lines of descent as distinct at this level, 



PSOCIDA, 



Zoraptera- 

 ISOPTERA 



HYME.'IOPTERA 



Mecoptera- 

 NEUROPTERA 



Mantida- 

 PROTOBLATTIDA 



FIG. 1 Ancestry of the Hymenoptera. 



as is done in the diagram. Furthermore, the similarities 

 between the precursors of the Hymenoptera, Psocida, Neurop- 

 tera, etc., become so marked as we trace them back to their 

 point of origin (where the lines of descent of the Dermaptera 

 and Isoptera merge) that at this level, or a little below it, the 

 differences between the ancestral types were doubtless no greater 

 than the differences between the families of a single order of 

 insects. 



It is a comparatively simple matter to trace the lines of devel- 

 opment of the Coleoptera, Neuroptera, Hymenoptera, Psocida 

 back to ancestors anatomically intermediate between the 

 Dermaptera and the Isoptera (with the Zoraptera), and these 

 ancestral forms in turn quickly merge with the common ances- 

 tors of the Dermaptera and Isoptera. From this point on, how- 

 ever, the problem of finding the types ancestral to these forms in 

 turn, becomes much more difficult. The ancestors of the Derm- 

 aptera were undoubtedly extremely closely related to the Ple- 

 coptera with the Embiida; and the Zoraptera are unquestionably 

 intermediate between the Isoptera and Plecoptera so far as their 

 morphological details are concerned. On the other hand, the 

 Isoptera exhibit certain features clearly indicative of affinities 

 with the Protoblattida, in addition to their undoubted relation- 

 ship to the Dermaptera, Kmhiida and Plecoptera. This ilual 

 relationship of the common ancestors of the Dermaptera and 



