AXDRENW.E. 191 



tings; (Frixbij). Near ]\Ianchestei' ; {Gardner). Norfolk j 

 {Bridgmaii) . 



ACUTILINGUES. 



Tins is a very large section, in which the tongue varies 

 from a short pointed and more or less cordate, to an 

 elongate linear shape. The generic characters are chiefly 

 drawn from the mouth parts ; a good divisional character 

 may be found in the form of the labial palpi, as given 

 below. 



The only exception is found in the genus Bophitcs, but 

 its other characters are so like those of the Andrenidsc that 

 I prefer to place it in that family. 



(2) 1. All the ioints of the labial palpi cylindrical AndrexidjB. 

 (1) 2. Basal joints of the labial palpi sheath-like . Apid.e. 



ANDRENID^. 



In the family of Andrenidm thus defined will be found 

 a few genei'a, including Nomada, which have been con- 

 sidered by most authors as members of the Apidx, but I 

 feel little doubt that this is their proper position, although 

 I know that in the case of Nmnada I am taking a view in 

 opposition to that of Prof. Perez of Bordeaux, whose know- 

 ledge of the Hymenoptera is far superior to mine. I base 

 its position in the Andrenidie on its general preference for 

 species of that family as its hosts, on the depressed andreni- 

 form abdomen narrowly rounded at its base, the transverse 

 labrum, and the cylindrical basal joints of the labial palpi. 

 This combination of characters seems to me to remove it 

 from the neighbourhood of Megachile, where Prof. Perez 

 places it, although the ^ armature would certainly favour 

 that position, and to place it more naturally hore than else- 

 where, especially as, according to Packard, " t he pupa 

 differs from th at of an y other genus known to us except 

 Andrena by having^ three conspicuous spines on the upper 

 aoJ^poBterior edge of the orbit." The other characteristics 



