ANDRENIDM. 205 



resorting to flowers, &e., till they die ; this accounts for the 

 very small proportion of females which appear in tho 

 autumn. In and round the burrows Mr. Perkins says 

 they are abundant, and probably, if anything, outnumber 

 the males. According to Smith, several Nomadai are 

 parasitic on the species of this genus, but as Mr. E. C. L. 

 Perkins remarks, these inquilines appear in July and 

 August, when the Halict! are hatching out, and as the new 

 $ Halictl hibernate like Bomhus, &c., in the impregnated 

 state and make no cells till the spring, it seems clear that 

 it would be useless for the Nomadie to lay their eggs in 

 the foodless burrows. I have never personally taken any 

 species of Nomada with Halictus, and I strongly suspect 

 that there has been some error of observation. 



Some of the species are attacked by one of the Stylopidaa, 

 Halicfophagini, which lives in the body of the bee, tho 

 head protruding as a rule between the segments of the 

 abdomen. 



From the difficulty of the subject the following table is 

 very imperfect, but anyone who knows how closely allied 

 some of the species are will appreciate the difficulties which 

 have had to be contended with. Nearly all the species 

 frequent yellow composites such as Crepis, &c., although 

 they do not all confine themselves to such, 



(50) 1. Species not bronzy. 



(7) 2. Apices of the abdominal segments 

 with white pubescent bands or 

 spots. 



(6) 3. Abdominal bands entire, or nearly 

 so. 



(o) 4. cJ, mandibles simple ; ?, posterior 



tibife clear testaceous . . . Bvnicrxnus. 



(4) 0. cJ, mandibles dilated at the base 



beneath ; ? , posterior tibi;o black qrADRicrNCirs. 



(H) 6. Abdominal segmentg with only a 



lateral white spot . . . MiCULATLS. 



(2) 7. Pubescent bands or spots when pre- 

 sent situated at the base of the 

 segments, their apices unhanded. 

 (21) 8. Large or medium sized species, ab- 

 domen deep black, with conspicu- 



