SYSTEM OF INSECTS. 388 



the same effect. With regard to their larva, the re- 

 semblance between the case-worms and the pseudo-ca- 

 terpillars of the saw-flies seems to me very distant, and 

 the numerous prolegs of the latter have scarcely a 

 legitimate representative in the former. The larvae of 

 the genus Lyda lose the prolegs intirely, and in one 

 species, which much resembles the vermiform larvae of 

 Hympnoptera, the real legs are so extremely short as to 

 be scarcely discernible 3 ; so that it requires no great 

 stretch of faith to believe that saw-flies or Sirices may 

 exist in whose larvae the legs disappear b . But it is this 

 very tribe, whose larvae thus approach to those of the 

 other Hymenoptera.1 in which Mr. MacLeay finds the 

 greatest external resemblance to the Trichoptcra c . In 

 fact the difference between the saw-flies and Siricidte, 

 and the remainder of the Hymenoptera, amounts to little 

 more than what takes place in the Diptera Order be- 

 tween the Tipulidtf, Asilidce, Muscidtf, &c., in which 

 also the metamorphosis differs. 



Another argument upon which Mr. MacLeay seems 

 to lay some stress, is taken from the number of parts 

 into which the ovipositor of the saw-flies is resolvable, 

 which he finds to consist of Jour pieces ; while in what he 

 considers as the genuine Hymenoptera, it is formed only 

 of three d : but in fact, in these last there are two spiculae, 

 answering to the two saws of TeMhredo, so that the va- 



a De Geer ii. 1035. b Since this was written, 



Mr. Stephens has showed me a remarkable Hyrnenopterous insect 

 taken by him in Hertfordshire, which appears to have the antennae 

 of one of the Ichneumonidce and the wings and abdomen of a Ten- 

 thredo L., so as to form a link connecting the two tribes or suborders. 

 This may probably have a vermiform larva. 



e Hor. Entomolog. 431. d Hor. Enlomolog. 429. 



