568 [December 



the pupal envelop of all Cecidomyla is formed in the same way, and 

 that the resinous envelop of C.pini inojjis 0. S. and of the Cecidomyia 

 referred to by Winnertz is strictly homologous with the ''flax-seed" 

 envelop of the Hessian fly, and both of them strictly homologous with 

 the smooth lining of the cell-walls of C. s. batatas n. sp. and the thin. 

 filmy cocoon of the Wheat-midge, ( Cec. tn'tu-ij and of several of the 

 G-all-gnats of the Willow, I have no manner of doubt. Now we know 

 that in the first case the pupal cocoon is exuded. Whence it is but ra- 

 tional to believe, in opposition to the theories of Harris and Fitch, that 

 in all the other cases the pupal envelop is likewise exuded, and not spun 

 nor formed out of the moulted integument of the larva. 



As to the Natural History of the Wheat-midge, when that insect, as 

 is occasionally the case, transforms to pupa in the ear of the wheat, it 

 forms a thin, filmy cocoon and generally transforms to imago the same 

 season. (Mai-sham and Kirby, quoted Harris InJ. Ins. p. 589.) Those 

 that go underground to transform must undoubtedly also form a cocoon ; 

 and from the analogy of the Willow Gr'all-gnats we may conclude, that 

 they ordinarily lie in the cocoon in the larva state all through the win- 

 ter, and at least until the commencement of the following spring, the 

 imago appearing in June and July, and the imago of most of the Wil- 

 low Gall-gnats appearing as early as April and May. There is a simi- 

 lar variation in the habits of the European Willow Grall-gnat, G. termi- 

 nalis Lw., which, according to Winnertz, " sometimes goes under ground, 

 and sometimes transforms within the willow leaves deformed by it." 

 {Dipt. X. A. p. 184.) Harris, singularly enough, while he holds that 

 the thin, delicate cocoon of the only Willow Gall-gnat known to him 

 is spun by that insect, maintains, contrary to the opinion of Kirby and 

 my departed botanical friend. Prof Henslow of Cambridge, England, 

 that the similarly thin and delicate cocoon of the Wheat-midge is, 

 equally with the dense, leathery cocoon of the Hessian Fly, composed 

 of " the outer skin of the larva." (//y. Ins. pp. 590, 596.) He appears 

 to have been led into this belief, in regard both to the Hessian Fly and 

 the Wheat-midge, from observing in the cocoon of both of them faint 

 indications of the same transverse sutures that we see in the coarctate 

 pupa of Stratlomi/s and Musca. {Ibid. pp. 576, 595.) Such pheno- 

 mena are easily explainable on the theory of the cocoon being exuded, 

 but he very justly considered that they were opposed to the theory of 



