28 Z. BOUCEK 



coarsely but not very densely punctured, interspaces partly broader than punctures and slightly 

 dull due to microscopical longitudinal reticulation. Hind tibia with outer ventral carina confined 

 to basal two-fifths. Fore wing : uncus longer than true stigmal vein. 



Gaster fully 1-2 times as long as head plus thorax combined, with median keel distinct only 

 on third, fourth, basal half of fifth tergite and on epipygium. Puncturation in general moder- 

 ately dense and not very fine, on central part of fifth tergite fairly coarse and sparse; first 

 tergite with smooth elongate-triangular basal fovea, posteriorly very distinctly and rather 

 densely punctured, on flanks with a mixture of sparse fine and coarser punctures. Pilosity 

 blackish brown, basally on gaster less conspicuous, on fifth tergite basally very dense and rather 

 uniformly short, more posteriorly double, consisting of shorter and of abundant longer, more 

 erect hairs, all much denser than in P. sphegoides. Fifth tergite 1-7 times as long as the first, 

 about 1 '3 times as long as broad, nearly four times as long as sixth tergite medially; hind margin 

 of fifth tergite slightly angulately emarginate. Epipygium nearly twice as broad as long, 

 slightly compressed from sides, not elevated; rudiments of cerci low but large, long-oval, 

 slightly behind middle of epipygium; ventral outline of epipygium slightly convex, tip of 

 sheaths very slightly exceeding epipygium. Hypopygium reaching three-fifths along gaster. 



<§. Unknown. 



Biology. Unknown. 



Holotype $, Panama: Curiche, Choco, Colon, viii.-ix.1g67, Malaise trap (G. B. 

 Fair child) (EM, East Lansing). 



Superficially P. atrata is very similar to P. sphegoides, but in many morphological 

 characters probably closer to P. nitidiventris Ducke, a species mainly brownish 

 yellow in colour. 



LEUCOSPIS Fabricius 



Leucospis Fabricius, 1775 : 361. Type-species: Leucospis dorsigera Fabricius, by monotypy. 

 Coelogastev Schrank, 1780 : 303-306, pi. 8, fig. 4. Type-species: Leucospis dorsigera Fabricius, 



by subsequent monotypy (first subsequently included species). 

 Leucopsis Olivier, 1792 : 531. [Incorrect subsequent spelling.] 

 Leucopsis Dumeril, 1823 : 167—168. [Unjustified emendation.] 

 Leucaspis Burmeister, 1835 : 47. [Unjustified emendation.] 

 Metallopsis Westwood, 1839 : 264, 265 [as subgenus of Leucospis Fabricius]. Type-species: 



Leucospis cayennensis Westwood, by monotypy. 

 Exochlaenus Shipp, 18946 : 245. Type-species: Leucospis anthidioides Westwood, by original 



designation. 

 Parexoclaenus Girault, 1915 : 355. Type-species: Parexoclaenus vespoides Girault, by original 



designation. 

 Exoclaenoides Girault, 1915 : 356. Type-species: Exoclaenoides uncinctus Girault, by original 



designation. 

 Epexoclaenoides Girault, 1915 : 357. Type-species: Epexoclaenoides bicinctus Girault, by 



original designation. Syn. n. 



Under 'Coelogaster', Schrank (1780) described and figured a Leucospid and his 

 one-word name has been regarded as establishment of a genus without species, 

 although his expression 'Gattung' for the taxon at that time could mean what we 

 call a species nowadays. This suspicion seems to be well-founded by Schrank's 

 subsequent statement (1781 : 307) that his species (Coelogaster) was already described 

 before him by Fabricius as Leucospis dorsigera, although he added some doubts 

 as to the specific identity. The name Coelogaster has, however, been treated as a 



