264 COLEOPTERA OF NORTH AMERICA. 



Fam. LIV— spondylidae. 



We would unite under this name all the aberrant Cerambycidge 

 of Lacordaire, whether classed with the Prionidffi or Ceramby- 

 cidse. By Mr. Thomson they have been in part separated as 

 distinct families, under the general name Subcerambycidge : he 

 has, however, excluded Spondylis from them and retained it with 

 Scaphinus among the Cerambycidae. 



It seems a more natural view to regard them as sub-families 

 (or tribes, as the case may be), having the same relation to each 

 other as the sub-families and tribes of the Cerambycidae, and 

 representing in the modern fauna the last remnants of the pro- 

 phetic, synthetic, or undifferentiated* types of a former geological 

 age. They are, therefore, few in number, without very obvious 

 relations Mnth each other, or with the numerous forms of Ceram- 

 byeidas, with which they cannot be intercalated, without inter- 

 rupting the obvious series of relationships. 



They may be briefly descriljed as extraordinary forms, differing 

 not only in appearance from other Longicorns, but also by the 

 tarsi being all deprived of the brush of hair beneath; the 3d joint 

 not bilobed, entire or feebly eniarginate,'the 4th joint frequently 

 well-developed ; the antennae are short, with the scape very short, 

 much constricted at base, inserted at the side of the head near 

 the base of the mandibles, under a more or less developed ridge; 

 2d joint rather large, though smaller than the 3d. In our two 

 sub-families the poriferous system of the antennae is contained in 

 deep fovea?, differing in shape according to the genus. The other 

 characters vary, as may be seen by the table in Thomson, Syst. 

 Cerambyc, 312. 



Two sub-families exist in our fauna: — 



Frotliorax margined ; labrnm connate. ParandriD/E. 



Prothorax not margined ; labrnm free. SpoNDYLiDi?!. 



* These three appellations will be acceptable according to the meta- 

 physical school to which the reader may belong. We write not to sustain 

 a theory, but merely to present facts in such relation with other facts, as 

 enables .them to be most conveniently classified. The result is the same 

 whatever hypothesis be adopted. 



