HYM E NOPTE K A . 1 85 



tain similarity in the wings, which, though not extending to details. 

 appears to be significant. The coarse punctures on a shining ground, 

 and to some extent the form of the thorax of Scotia are indicated in 

 the bee genus Temnosoma, though not in Protoxcea. The eyes of 

 Protoxcea are not emarginate, but they are so in many Halictine 

 bees, while the Myzinid wasps have them so in the male, but not in 

 the female.* 



The mouth of Temnosoma could be derived from that of the Sco- 

 lia-Protoxcea type by the shortening of the tongue, paraglossse and 

 first joint of labial palpus; the tongue, as in Hatictvs and Cilissa, 

 preserves the peculiar tapering form of Prctoxcea. The galea in 

 Protoxcea shows no sign of the apical division seen in Scotia, but in 

 Nomia and Hatictus it is plainly indicated, and Nomia (which Mr. 

 Vachal considers nearer to Hatictus than to Andrena) has also the 

 tapering tongue, broad basally and filiform apieally. A very inter- 

 esting genus is Meroglossa Smith, which assuredly does not belong 

 to the Prosopidse. The arrangement of its tongue and paraglossse 

 is quite suggestive of Protoxcea, but the maxillary palpus is very 

 much longer than in that genus. The venation is quite different. 



The present conclusion is, that the whole series of Halictine bees, 

 at any rate, came from an ancestor not far removed from the Scolii- 

 dve. Protoxcea is of course not an Halictine, but it is from the same 

 general stock, and apparently nearer to the Scoliids than is Hatictus. 

 We seem to have divergent rather than successive types, but further 

 study will no doubt make the true relationships much clearer. Ir 

 may be added that Myzine and the Tiphiids have the tongue short 

 and rounded, not in the least as in Scotia. The mouth of Scut in i.- 

 in most respects far more like that of Protoxcea than it is like that 

 of the Myzinids or Tiphiids, though the latter have the divided 

 galea f of Scotia. 



I do not think the Colletidse have any bee-ancestry in common 

 with the Halictines. So far as the mouth-parts go, Colletes shows 

 the closest possible resemblance to Tachytes, the resemblance extend 



* Allied to the Seoliidse are the Thynnidae and Corynura Spinola, as Mr. Vachal 

 remarks, was based on a 9 Thynnid and a % Halietid ; a fact indicative of close 

 resemblance. 



f Kellogg (Am. Nat., Sept., 1902) calls the part here designated the galea, tin 

 maxillary lobe, and says it consists of the galea and lacina fused. So I suppose 

 that the apical part in the Scoliids, etc., is the true galea, and the rest the lacina 



TRANS. AM. ENT. SOC, XXIX. (24) MAY, 190.'i. 



