144 [June, 1906. 



reallj sexual. 'J'he thorax and abdomen together form a stout cjlindrical mass, hardly 

 narrower towards the head, but tapering rather abruptly near the anal extremity; 

 in colour they are wliite, with the lines, which consist of a moderately broad dorsal 

 one, flanked on either side by a very broad subdorsal, and a rather broad subspiracular 

 one, dull crimson, and conspicuous ; these lines are all continuous, but none of them 

 are of even width throughout, for they display in their course many small projections 

 and irregularities ; they are rather paler, and less well pronounced, on the thorax 

 than on the abdomen. Skin smooth, but not glossy ; segmental divisions clearly 

 defined. In the S the embryo testes show through the back of the fifth abdominal 

 segment as a dark internal blotch. Tuhercles brown, polished, mostly of moderate size. 

 Spiracles minute, circular, white, ringed w^ith black. Anal plate polished, dingy 

 pale ochreous, with the underlying frass in the cloaca showing through the anterior 

 part of the plate as a dark brown blotch. Hairs and bristles brown, mostly short 

 and inconspicuous. Ventral stirf ace, and prolegs, white. Z/*?^.? polished, raw-uraber. 



Other larvae were examined whenever they were found out of 

 their burrows ; these were all more or less well-grown, and the only 

 variation shown was in the colour of the prothoracic, and, to a less 

 degree, of the anal plate. On a careful comparison of Mr. Barrett's 

 description with the living larvae, the most important differences 

 proved to be that he gives the body as "flesh-colour," and the "sub- 

 dorsal lines" (in which term he presumably includes the subspiracular 

 lines as well), as "much broken," whereas, in all the larvae seen, one 

 of which was preserved for referenc(?, the former was most decidedly 

 while, without even a tinge of flesh-colour, and the latter were dis- 

 tinctly continuous, though with very irregular margins. I totally failed 

 to find any trace of the "two orange dots at the edges of the dorsal 

 line " on each segment, which Mr. Barrett mentions. 



The larva, at any rate in its later stages, lives inside, and feeds on, 

 the rootstock of Rumeoc crispus, L., throwing out from the end of a 

 silken gallery, that projects for an appreciable distance beyond the 

 mouth of the burrow, its ochreous-brown or reddish-brown frass, of 

 which a heap is generally seen adhering to the exterior of the gallery. 

 It is spoken of by Mr. Barrett, in a general way, as " mining in the 

 stem, eating the pith and leaving its excrement in the hollowed space," 

 but this can only be true of it in its earliest stages if Mr. Purdey's 

 opinion, which Mr. Barrett himself quotes, is correct, that the larva, 

 on hatching, mines down the stem, and after reaching " the crown of 

 the root" before the winter, continues to feed therein until "becoming 

 full-grown by June." It is, however, not always full-fed so early, for 

 last year the larva was unusually late in feeding up, and Mr. Purdey 

 and I both found that certain individuals did not become full-fed 

 until the very end of July. 



