10 [January, 



of the difficulty I used to feel when I was trying to arrange my 

 captures of these insects according to the tables of Thomson, Andre, 

 Cameron, &c. It is now twenty years since the first volume of the 

 latter writer's well-known Monograph appeared, and during the whole 

 of that time, year by year and almost month by month, Herr Konow 

 has been continually publishing fresh works, descriptive, systematic, 

 etc., on various branches of the subject, and has not only thrown 

 much new light on the difficulties of its synonymy, but has greatly 

 altered and (to my thinking at least) improved upon the classifications 

 of the Sub-Order adopted by previous writers. 



These important contributions to our knowledge of the present 

 subject have unha[)pily not as yet been collected into a single volume ; 

 they are dispersed through various numbers of foreign scientific 

 periodicals and Transactions of Societies, French, German, Austrian, 

 Russian, &c. But (in many cases by gift from the author) T have 

 gathered a nearly complete collection of them ; and they include 

 thorough systematic Hevisions of many groups among the Saw^-flies, 

 and materials for the revision of others (new characterizations of 

 genera and species, corrections in synonymy, &c.), which will cer- 

 tainly have to be taken into account whenever a new^ British 

 Monograph or Kevision of our List is attempted. But for this I 

 think the time is not yet ripe — there is much work to be done pre- 

 viously, both in the collection of specimens from British localities 

 which have never been properly worked, and in re-examination of 

 existing collections and investigation of ancient records, many of 

 which latter, I am persuaded, will prove to be based on errors. And 

 to correct these errors will be no easy matter, for our old collectors 

 were not always careful to indicate the " provenance" of their speci- 

 mens, and our English methods of "setting" insects (short pins, 

 extended wings, " carding," &c.) make it often quite impossible to 

 examine properly the characters which must be examined if a specimen 

 is to be determined "for certain." Now and then an insect may be 

 identified by simple inspection of the dorsal surface, which is all that 

 can be seen in many carded specimens ; but this is quite the excep- 

 tion. Sometimes examination of tin; clnivs, or the pleura', or the 

 ventral segments of the abdomen, is indispensabk;, and without it all 

 attempts to name the specimen are mej*e guess work-. Our tnclhods 

 cci'iainly display the wiiios well, and their ncnration, &c., gives many 

 v(!ry useful characters ; but to rely sohdy upon this is extremely 

 dangerous, and has been the source of cmintless difficulties and 

 confusiuns. 



