92 \VM. H. ASHMKAI). 



Second recurrent nervure interstitial with the third transverse cubitus ; second 

 cubital cell quadrate, the third a little longer than wide, slightly nar- 

 rowed above; tongue very long, spiculiform ; labial palpi 4-jointed, the 

 first joint nearly as long as joints 2 I united. Megaloptera Smith. 

 6. Subdiscoidal nervure in hind wings distinct, well developed ; temples broad, 

 or at least never very narrow, sometimes in Iltilictus with a tooth 

 below in 9 : labial palpi 4-jointed. 

 Ocelli very large, the lateral ocelli almost touching the eye margin. 



ft|>lic<'o<lognMli-u Ashm. n. g. (Type Parasp. texana Cr.) 

 Ocelli normal, the lateral ocelli very distant from the eye margin. 



IIali<'tiis Latreille. 

 Subdiscoidal nervure in hind wings wanting or subobsolete ; temples very nar- 

 row. Battened ; labial palpi ''-jointed ; abdomen maculate with yellow. 

 Nomiodes Schenck = Lueasius Hours. 



Subfamily III. Sphecodin^. 



This subfamily is scarcely distinguishable from the subfamily 

 Halictinse, agreeing with it in all essential characters, the only diflfer- 

 ences noticeable, and which may be depended upon to separate the 

 species from those in the latter, are that in the 9 the hind coxa 1 and 

 femora are without a flocculus, the pollen-brush on their tibiae and 

 tarsi, although denser on the tarsi, is sparse and thin or nearly 

 wanting, the abdomen being without a distinct ventral scopa, and 

 the hist dorsal segmenl is without a distinct rima. 



1 know of no good character to separate the males from those in 

 the Halictinse. The head, however, viewed from in front, is a little 

 wider than long, the clypeus not being produced anteriorly, while 

 the metathorax, as a rule, is much more coarsely sculptured than in 

 the Halictinse, being usually coarsely rugoso-punctate. 



The genus Parasp hecodes Smith, as I have identified it, from a 

 specimen in the National Museum from Australia, is closely allied 

 to Halictus and belongs to the Halictinse. Parasp hecodes texana 

 Cress, has been wrongly referred to this genus and forms a new 

 genus in the same subfamily. 



Only three genera seem to tall into this group as here denned, 

 and these may be distinguished by the use of the following table: 



Didonia Gribodo is, however, placed here without an examina- 

 tion of specimens. 



Table of Genera. 



First cubital cell, along the cubitus, distinctly longer than the second and third 

 united ; stigma well developed ; subdiscoidal nervure originating at 



TRANS. AM. ENT. SOC. XXVI. JUNE, 1H99. 



