son^c Bees from British Guiana. 443 



red ; vertex -with black liair, clieeks with long white liair, a 

 fi'inge of white hair around clyjjeus above and laterally, 

 a tuft of bright red liair at apex of clypeus medially, hair of 

 front mainly black ; thorax Avith short brownish-black hair 

 above, beneath with pale hair (but much black on meso- 

 pleura), at sides posteriorly with red hair; teguhi} black. 

 \Vings dusky, suffused with reddish; femora with light liair, 

 tibia3 and tarsi with dark hair on outer side and red on 

 inner. Abdomen with pale reddish hair on red part of first 

 segment, and black on metallic ; segments 2 to 4 with linear 

 but distinct apical yellowish bands; apex with copious sooty 

 hair; venter with pale red hair basally and dark a])ically; 

 a patch of long yellowish-white hair on each side ventrally 

 beyond the middle ; apical plate broad and black. 



At lights, Hills Estate, U. Massaruni_, 15 Dec, 1917 

 (G. E. Bodkin), 



As to the occurrence o£ Ptiloglossa at light, see Schrottky 

 (under Megacilissa), Smithsonian Misc. Coll. xlviii. (1907) 

 ]). 259. The ocelli are very large, as with other nocturnal 

 Hymenoptera. The type of Ptiloglossa ducalis, Smith, is at 

 Oxford, and is labelled " Chili ?^' Smith said the locality 

 was unknown. Friese formerly thought it identical with the 

 Mexican P. eximia (Smith), but Schrottky in 1914 showed 

 that this was not the case, and stated that he had ducalis from 

 La Rioja, Argentina. The present insect is certainly very 

 near to ducalis, differing by the largely red first abdominal 

 segment and other minor details. Smith's figure of the 

 venation is incorrect, as his description shows. The Brazilian 

 P. obscura (Schrottky) has the base of the abdomen dark red, 

 and is evidently closely allied. Other related forms are the 

 Brazilian P.jjretiosa (Fritse) and the Mexican P. tlioracica 

 (Fox), described under Megacilissa. It seems quite probable 

 that all these forms are local races of one, P. ducalis. On 

 the other hand, P. ducalis, var. buchwaldiy Friese, is a large 

 red insect resembling Xenoglossa fulva in appearance, and 

 must, I think, stand as a distinct species, P. buchivuldi. 



Megalopta idalia, Smith. 



1 2 , Issororo, N.W.D., June 1916 {Bodkin). 



As at present understood, this species is widely distributed, 

 and it may be capable of subdivision when more material is 

 available ■^. 



* Schrottky thinks that what I described as idalia in Proc. Acad. Nat. 

 Sci. Phila., 1900, p. 356, is really ocJirias (Vachal). My insect was a 2 ; 

 Vachal described only the cS . L)ucke holds that ochrias is a variety of 

 idalia. 



29* 



