FAMILY CYNIPIDJ!. 141 



them in an entirely opposite sense, may be granted ; 

 but, on the other hand, it was equally arbitrary to 

 apply the name of " Cynips " to a group of insects of 

 which only few species were included in Cynips by 

 Linne, the great majority (and certainly the first on 

 the list, which ought, according to the usual custom, 

 to be considered the type) of which were true gall- 

 makers OynipidsB in the modern sense. Moreover, I 

 cannot look on Geoffrey's names as " genera " in the 

 Linnsean and modern meaning. In his work he did 

 not use the binomial system of nomenclature, and I 

 therefore do not think his names should be adopted 

 at all unless they have been clearly defined and used 

 by his predecessors defined by them as part of the 

 binomial system of nomenclature. It is true they 

 were at first adopted by Fabricius, but seeing that he 

 afterwards reverted to the name which Linne obviously 

 meant to apply to the gall-makers, and as that name 

 has been applied and become firmly attached to them, 

 I do not feel disposed to follow Karsch in changing 

 the name of the gall-flies to Diplolepis. To do so would 

 only lead to needless confusion, without any adequate 

 advantage being gained. Nor would the confusion 

 end here, for Cynips would have to be, on the same 

 principle, applied to some Chalcid, or if " Cynips" 

 Geof. Fab. (of Sys. Ins.), could not be fixed definitely, 

 dropped entirely. 



The same reasoning applies to Crabro versus Cimbex. 

 Crabro was first applied to the saw-fly, now called 

 Cimbex, by Geoffroy, but was afterwards applied by 

 Fabricius to an Aculeate ; so that, if we are slavishly 

 to stick to the rule of priority, Cimbex must now be 

 called Crabro, and another name be found for the latter 

 as now understood ; a course which would mean end- 

 less confusion, without any scientific object being 

 gained. 



The Cynipidx belong, in contradistinction to the 

 Tenthredinidde, to the division of Hymenoptera having 

 the abdomen petiolated, i. e. not attached by its entire 



