1864.] 553 



that " Grall-gnats cannot be recognizably described from single dried 

 specimens, unless they are distinguished by some striking peculiarities; " 

 {Dipt. N. A. p. 187;) and Osten Sacken observes as follows: — 



It is a peculiarity of the family of Cecidomyidoe, that its natural history has 

 always been studied in close connection with its classification. This is owing 

 chiefly to the fact that the gall, the produce of the insect in its first stage of life, 

 is generally a more striking object in nature than the insect itself. The latter, 

 small, tiny, difficult to preserve on account of their extreme delicacy, still more 

 difficult to distinguish from their congeners on account of the uniformity of their 

 appearance and coloring, would affijrd a very unsatisfactory object of study, 

 unless in connection with the varied deformations which their larvae produce 

 on plants. {Dipt. N. A. p. 173.) 



I find it utterly impossible in one case to distinguish from each 

 other the dried 9 imagos of two undoubtedly distinct species, which 

 form distinct galls of a perfectly distinct structure on different parts of 

 the same Willow, and the pupal integuments of which are structurally 

 very distinct, viz. Crc. s. rhodoides n. sp. and CW-. s. siliqua n. sp. ? 

 I had hoped that, by taking descriptions of numerous specimens of re- 

 cent Cecido}ni/ia, and especially of the abdomen which often loses its 

 coloration almost entirely when dried, some sharply-defined distinctive 

 characters might be arrived at. But T have found from these descrip- 

 tions that the same species, and even the same living individual of the 

 same species, varies greatly in the coloration of the abdomen according 

 to the degree of its maturity, and that what was at an early period in 

 its existence yellowish or reddish, gradually becomes, in the course of 

 a day or two, and sometimes even in the course of a few hours, brown 

 or blackish. I have even repeatedly placed the recently killed 9 O 

 produced from the above two galls side by side, and have found myself 

 utterly unable to discover any constant distinctive character whatever 

 though it is barely possible that the structure of the % antennae may 

 differ. In solitary individuals indeed it is easy enough sometimes to point 

 out distinctive characters ; but on comparing many individuals belono-ino- 

 to the same species, such characters are very generally found to be in- 

 constant and worthless. Lest it should be assumed that the characters 

 in my specimens might have been changed by chemicals, such as chlo- 

 roform, &c., used to deprive them of life, it is proper to state here, that 

 I kill all tlies by simply immersing the vial or bottle, in which they are 

 confined, into hot water up to the cork. 



