12 

 On the genera of Dryinidae allied to Gonatopiis. 



In my former papers (Bull. I Pt. i and lo) I divided the in- 

 sects that are usually assigned to Gonatopiis into a number of new 

 genera. It is with some reluctance that one breaks up a genus 

 of insects so similar in general form and strikingly distinct from 

 any other, but there are very good reasons for this division. In 

 the first place my studies have convinced me that the species 

 assigned to Gonatopus have been derived from winged forms, 

 which would themselves universally be assigned to distinct genera, 

 and that the great superficial resemblance of species differing in 

 most important characters is due to convergence from similarity 

 of habits and does not indicate their true phylogeny. 



Further, it is obvious that the genus Gonatopus (s. 1.) is a 

 difficult one, and the difficulties are such as will be likely to in- 

 crease, as species are added to it. Obviously the species are ex- 

 tremely numerous. In a few limited localities, and these by no 

 means rich, as compared with some now known to us. ]\Ir. Koebele 

 and myself were able in a short time to collect some thirty-five 

 new species. Most are of limited range or local. The species in 

 Sydney, X. S. W. are generally dift'erent from those found at 

 Brisbane and Bundaberg, Queensland, and the species in Arizona 

 are certainly mostly distinct from the Californian. The genus, in 

 Its wide sense, appears to be ubiquitous, and even occurs naturally 

 in such remote oceanic islands as the Hawaiian group, while it 

 is well represented in the Viti islands. 



Recently in an area of a few square miles in Arizona Mr. 

 Koebele bred and collected at least twenty-five species in a short 

 time. These facts, and the consideration that many species have 

 been described by others from caught specimens, lead me to 

 estimate the number of species of Gonatopus ( s. 1. ) at fully one 

 thousand, and probably several times this number may eventually 

 be found. I therefore believe that students of the Dryinidae will 

 sooner or later be very glad to adopt in the main the characters 

 employed by me for the division of the genus Gonatopus. although 

 to examine the mouth-parts accurately requires a considerable 

 amount of labor, and a trifling amount of skill, if the specimens 

 are to be dissected without miaterial injury. 



I have still thought it better at present to neglect the classifica- 

 tion of the males, and have avoided describing numbers of possibly 

 new species in this sex. the names of many of which would have 

 to be eventually sunk as synonyms. Whether it will be easy to 

 find generic characters for the comparatively unspecialized males 



