104 Dr. E. C. L. Perkins on Aculeate 



collected. Many of the species so described were not true 

 Prosopis, and after the removal of those that belong to 

 Merofflossa and Falaorhiza, the residue contains a complex 

 of species belonging to a number of genera and subgenera. 



I have here separated off two groups of species as distinct 

 genera, but do not care to proceed further until I can obtain 

 a larger number of species for dissection. There is no doubt 

 that the terminal ventral segments and geuital armature of 

 the male are of great importance in the classification of these 

 insects. In some cases the structure of tlie propodeum or 

 median segment furnishes very valuable characters; but I do 

 not think that it can be safely used for the formation of 

 generic characters, unless these are corroborated by those 

 of the male characters. Thus I have not thought it advisable 

 to give any name to an important group of species ■with a 

 remarkable flattened bifurcate process of the eighth ventral 

 segment, since it includes species with very different pro- 

 podeal structure, nor am I satisfied that the X. -American 

 species with this structure are congeneric with Australian. 

 The remarkable group of over fifty species of the genus Neso- 

 prosopis, in the Hawaiian group of islands, show considerable 

 variety of structure in the propodeum ; but the terminal 

 segments, while differing greatly in detail, are extremely 

 uniform in their general features. AVhile the extreme forms 

 of Nesoprusopis might, by themselves, hardly be recognized 

 as belonging to the same genus, the intermediate forms, still 

 existing, show that all belong to a single series, and the male 

 characters amply confirm this. Many neurational characters, 

 considered of importance by somehymenopterists, are of ver}' 

 minor importance in Prosopidae, owing to their instability 

 even in individuals of the same species. 



Gnathoprosopis, gen. uov. 



I propose this name for a group of Australian Prosopis, 

 quite remarkable for the extremely short, wide, flattened 

 mandibles, broadly truncated at the apex in botli sexes, and 

 with quite a difl'ercnt a])pcarance from the comparatively 

 narrow, strongly carinated organs of ordinary species. Most 

 of the species known to mc in the male have either tubercles 

 or processes on the third ventral segment, a peculiar erect 

 abdominal pubescence, and a more or less dilated scape to 

 the antenuai. 



Anterior area of propodeum seen from in front bounded 

 by a raised line at the truncation, the area being rugose ; 

 posteriorly it is marked off l)y an impression and the surface 



