164 BRITISH BEES. 



iiig pollen. JMr. Kirby, however, gives direct testimony 

 in favour of Sphecodes being a burrower, in the case of 

 which bee it ought not to be a matter of much difficulty 

 to determine, for on sandy plateaus I have occasionally 

 found it very abundant, especially where there was rag- 

 wort {Senecio) in flower in the vicinity, to which the 

 males resorted; but being at the time more intent on 

 other matters, I neglected the opportunity. Other ob- 

 servers concur with ^Ir. Kirby as regards Sphecodes, 

 and also say as much for Prosopis (better known as 

 Hylmus). I strongly incline to the opinion enunciated 

 by Latreille and Le Pelletier de St. Fargeau, that they 

 are parasites. Aly opinion is based upon peculiarities 

 in them other than, although strengthened by, the nega- 

 tive characteristic of absence of poUiniferous organs. A 

 negative cannot be proved, it is true, yet what has been 

 positively asserted may as certainly result either from 

 defective observation, or from too strong a desire to find 

 no parasites among the Andrenidoe. '\iy reasons occur 

 elsewhere in this work, and I need not repeat them. It 

 is still an open question, and the young entomologist, if 

 euteriug the arena unprepossessed, might win his spurs 

 in determining it. It would be well worth the trouble 

 of attending to for those who have leisure, and if decided 

 in favour of the independency of these genera^ which 

 must be corroborated by a plurality of observations, and 

 not confined to one locality, they would form strong and 

 remarkable instances of a defective analogy in nature's 

 workmanship, and suggest looking further for the causes 

 of so extraordinary an anomaly, and urge us to endea- 

 vour to trace the equivalent which supersedes it. 



The main subdivision of the Apidce results from the 

 habits of the insectSj which divides them into social 



