HABITS OF THE SPHECES. 177 



cations of the head and face, which give it a cruciated 

 appearance. The males vary considerably, from a short 

 ovate form, similar to Bcmbex, to a long and cylindrical 

 shape. Elis, only, is found in Europe ; the others that 

 are not Australian occur in America. 



(157.) We now enter the normal group of Spheces, 

 which are all winged insects. The point of contact 

 seems naturally to be the ScoUadee. With the excep- 

 tion of the few parasites that occur in it, these insects 

 are all predaceous ; but this term, as regards them, 

 must receive explanation, as it would imply that they 

 themselves prey upon and devour other insects. This 

 is not the case ; at least, no instance of it is yet known 

 to us; and it is better thus to modify the assertion ; for 

 we constantly and hourly detect that Nature, in her 

 discursive progress, will not be tied to the rules we lay 

 down from the observation of a few facts. From what 

 has been observed, it appears that these insects seek 

 their prey merely to provision their cells with nutri- 

 ment for their young; and they themselves, for their 

 food, visit flowers only. In the perfect state, they seem 

 to require but little nourishment ; for, excepting a few 

 genera, they are rarely found upon flowers; and this is 

 the more remarkable, as their mouths are of a very 

 highly organised structure. Parasites, amongst them, 

 have not yet been proved, and have been assumed 

 from the circumstance of certain species being divested 

 of the prevailing characteristic of the fossorial tribes, 

 which consists in a long external fringe of setiE to the 

 anterior tarsi, generally coincident with posterior tibiae 

 armed externally either with a longitudinal serration 

 or succession of spines. That this is not invariably 

 the case in the predacious tribes among them, we have 

 already fully shown *, from direct observation. Nor is 

 it the diagnostic exclusively of those which burrow in 

 wood, which we once thought might be the case ; and 

 the theory appeared plausible, as it of course could 



* See Shuckard's Essay on the Fossorial Hynjenoptera, pastrm. 



