THE BOSTRICHID^. 303 



the geniculation of the organ, and this process is dilated 

 at its apex into a fornicate appendage with a long fringe, 

 and, in repose, the two combined meet over the forehead 

 and cover it like a cap. In Tomicus, the elytra are fre- 

 quently retuse at their apex. The little Scolytus is no- 

 torious for the devastation it has caused among the elms 

 in the vicinity of the metropolis, but it is a point not yet 

 quite satisfactorily decided, if it is the insect which is 

 the original depredator, or whether, from some other 

 cause, the tree having fallen into a sickly condition, 

 these insects then visit it, and greatly accelerate its total 

 destruction. The majority of this family are sub-cor- 

 tical dwellers, comparatively few of them piercing into 

 the solid wood. Almost every species has a different 

 mode of forming its gangways, and in some cases the 

 channels they elaborate are extremely elegant. Between 

 these, which have the closest affinity to our next family, 

 and Bostric/ius, Apaf.e, Sec, the chief and most marked 

 differences consist in the clava of the antennse being 

 formed of three distinct and separate joints : there is a 

 great difference also in their larvas; those of the latter, 

 where, indeed, the structure of the antenna has a la- 

 mellated indication, being considerably like those of the 

 Lamellicornes, and have long legs, whereas the former 

 are apods, or have but warty tubercles. Those of Apate, 

 &c.also more frequently bore into the solid wood of trees. 

 In Anohium, and its immediate allies, we observe the an- 

 tennsE becoming filiform, the only intinnation of itspreced- 

 ing structure consisting in the elongation of the three ter- 

 minal joints, which sometimes, asin Dryophilus, exceed the 

 length of the remainder of the organ. We thus detect an 

 indication of the proximity of Ftinus, in which, however, 

 there is a departure from the xylophagous habits of the 

 group, pointing evidently towards Durmestes, whence 

 we doubtlessly come into close contact with the circle 

 of the SilphidcB. By means of Nemosoma, and from 

 it through Trogo-sita (for it is here we expect the 

 latter must come, proved by that affinity), we find col- 

 lateral affinities impinging clearly upon the Cucujidcs in 



