50 [March, 



must ascertain positively what $ $ and ? $ pair together normally, 

 and whether their offspring reproduce exactly the characters of their 

 parents, &c. But all this, even if it is likely to be attempted, will take 

 long years of skilled labour ; and in the meantime we should do well 

 not to feel too confident that our reputed species, varieties, &c, are 

 correctly distinguished and defined. 



It is possible, however, that even in cabinet specimens characters 

 may yet be found, which will bring us a little nearer towards finality 

 than we are at present. Cameron held that clear characters correlated 

 with true specific difference might be found in the " saws " of all 

 Tenthredopsis ? $ , and from my experience of such characters in other 

 genera, I should have expected this to be so. But personally I have 

 not succeeded in detecting such characters in the present genus ; and 

 Cameron, who believed that he had detected them, contrived to 

 separate by means of them in his Vol. I, several forms, which after- 

 wards (in Vol. IV) he considered to be identical ! My impression is 

 that such differences probably exist ; but that we have still to leam 

 what they are, and how we are to look for them. Mere outline- 

 sketches of the jagged under-edge of the saw, I feel certain, will not 

 teach us much. 



For the present, however, we can only deal with this genus by 

 making such use as we can of the differences already ascertained to 

 characterize certain forms, and so dividing them into groups, some of 

 which, probably, may coincide with real species, while others may 

 ultimately rank higher yet (sub-genera) and others probably are at 

 most local races, &c. There seems no reason why we should not 

 provisionally accept some of these groups as species, and make use of 

 the orthodox methods (" Priority," &c.) to furnish them with names. 

 But I fear that a great deal of the labour that is now-a-days expended 

 on this subject will in the end prove to have been wasted; or, at 

 least, survive only in the shape of additional " synonyms." 



If, setting aside all questions of species, varieties, &c, we think of 

 British Tenthredopsis merely as so many "forms " agreeing and dis- 

 agreeing in points which are capable of more or less precise definition, 

 I think we shall come to certain results which may be stated as 

 follows : — 



(a) On a certain character — almost the only quite positive and 

 u u mistakeable character of structure which occurs among them — we 



