MR. CURTIS ON A NEW SPECIES OF SAW-FLY. 41 



wings are entirely black, with the costa and stigma thickened and darker, as well as the 

 nervures, the surface being iridescent ; the superior (fig. 8) have two marginal and four 

 submarginal cells ; the first minute, the second twice as large, the other two very large, 

 the third receiving the transverse nervures which divide the marginal and discoidal cells ; 

 the inferior have only one discoidal cell (fig. 8*) : the legs are moderately stout and 

 pubescent ; the tibise are spurred at the apex, the spur of the anterior pair notched at the 

 apex ; tarsi 5-jointed, the first four lobed beneath ; the last joint terminated by two bifid 

 testaceous claws and simple pulvilli (fig. 9). 



Although the elongated antennae of this Selandria resemble those of Nematus, and still 

 more those of Cladius, this species is not only distinguished from those genera by the 

 divided marginal cell, but the heavy habit of the females especially shows at once the 

 groups to which it is naturally allied, and these affinities are supported by its trophi, which 

 are intermediate between Atlialia\ and Tenthredo\. I may observe that the number of 

 discoidal cells in the inferior wings varies in the species of Selandria, a character hitherto 

 unnoticed, but which may supply admirable distinctions for reducing the genus into sec- 

 tions. 1st, Those with two discoidal cells, the marginal cell receiving one transverse 

 nervure, of which 8. serva, Fab., is an example (fig. 13). 2ndly, S. stramineipes, Klug, 

 in which both transverse nervures are united with the marginal one (fig. 14). 3rdly, Those 

 with one discoidal cell, as shown in S. Robinsoni (fig. 8*) ; and 4thly, Those having no 

 discoidal cell, as in S.fuliginosa, Schr. (fig. 15). The variations in the position of the 

 nervures and the magnitude of the cells will also be found very useful in identifying the 

 species ; and although occasionally the nervures are not symmetrical, and occasionally the 

 recurrent ones are wanting, such exceptions will not invalidate the divisions I have traced, 

 but will, I trust, lead to a more careful investigation of this fine and interesting family. 



EXPLANATION OE THE PLATE. 



Tab. V. 



[Obs. Those figures with a * attached are magnified.] 



Fig. 1 . A portion of the stem of Convallaria multiflora, as eaten by the larvae of Selandria Robinsoni. 



Fig. 2, 2. The larvae feeding in two different skins. 



Fig. 3. One of the skins cast off" and sticking to a leaf. 



Fig. 4. The cocoon, with the end opened by the fly when it hatched. 



Fig. 5*. Head of the male viewed above. 



Fig. 6*. The trophi or mouth. 



Fig. a*. The labrum or upper lip. 



Fig. b, b*. The two mandibles or jaws. 



Fig. c, c*. The two maxilla. 



t Curtis's Brit. Ent. fol. & pi. 617. X MM. fol. & pi. 692. 



VOL. XXI. G 



