THE CANADIAN ENTOMOLOGIST. 309 



narrowing rather rapidly in its basal third, and then tapering evenly to 

 apex ; surface scarcely visibly spinose. 



Described from four winoless individuals collected as follows : 



o 



Illinois — Urbana, May 12, 9 and ^ , under dead bark on peach tree (J. 

 D. H.) ; Urbana, April, (J, reared from nymph taken on willow tree (J. 

 J. Davis). Pennsylvania — Harrisburg, March 10, 9. "i^^ burrow of 

 Lepidopterous or Coleopterous larva in dead willow stem." 



This species is very close to C. carbonarius^ Hood, but differs from 

 that species in being much smaller, in lacking the two pairs of prominent 

 bristles in the region of the ocelli, and in the shape of the tube. The 

 prothoracic sclerites differ in size and shape in the two species, and the 

 prothoracic spines do not exactly correspond. 



NOTES ON MOSQUITO WORK. 



BY HARRISON G. DYAR AND FREDERICK KNAB, WASHINGTON, D. C. 



In using the recent literature on mosquitoes, several points have been 

 noted which it seems worth while to record, in order, if possible, to arrest 

 certain errors that have crept into the subject and are being perpetuated. 

 One author copies from another (usually without credit) until the original 

 error becomes almost a classic fact. We wish, therefore, to emphazize 

 these corrections as much as possible in order to counteract this tendency. 

 Some of these errors refer to characters that have been considered of 

 fundamental value in classification, and it is important that they should 

 be pointed out, even if the system, of classification now in vogue be not 

 adhered to. They further illustrate the weakness of this system, which we 

 have on several other occasions attacked. 



The mosquitoes have suffered in their classification from the fact that 

 the earliest generic separations were made upon the relative lengths of the 

 palpi in the sexes. This is a secondary sexual character, as we have 

 pointed out. Nor have the sysiematists who have used the palpi in classifi- 

 cation made any study of their actual structure, except Neveu-Lemaire, 

 whose results were obtained upon the study of too small material, and 

 when extended show only intergradation. Now, while the separation of 

 these genera on such characters may have been allowable (for even yet 

 many systematists found genera on secondary sexual characters), their 

 elevation to higher groups and subfamilies, as has been done by Theobald 

 and his followers, is indefensible. We are unaware of any other case 



September, 1908 



