494 Dr. R. H. Traquair on Harpacantlius. 



That this spine was new there can be no doubt ; its refer- 

 ence to Agassiz's genus Tristychius is another matter. 



Tristychius arciiatus, the tjpe of tlie genus, was described 

 and figured by Agassiz in the ' Poissons Fossiles/ vol. iii. 

 p. 21, Atlas, vol. iii. pi. i. a.figs.9, 10, 11; and the very beautiful 

 original specimen, from the Carboniferous-Limestone series of 

 the neighbourhood of Glasgow, is now in the museum of 

 Anderson's College in that city. I have carefully examined 

 that specimen, as well as a large number of others from 

 various Scotch Lower-Carboniferous beds, and may therefore 

 sum up the essential characters of this spine as follows : — ■ 



Specimens occur from 1 inch to nearly 5 inches in length : 

 in shape the spine is gently and gracefully curved backwards, 

 tapering to a point ; and it must be noted that some examples 

 are more strongly curved towards the apex than others. The 

 extremity is longitudinally sharply sulcated or ridged, the 

 ridges mostly soon disappearing until, at a distance of from 1 

 to 1^ inch from the point, usually three only remain — one 

 median and two lateral, which pass down beyond the others 

 along the anterior aspect of the spine towards the base. The 

 inserted portion is not, as in CtenacanthuSj Uyhodus, or Gyra- 

 canthus, distinctly marked off from the exserted,and the former, 

 as well as the non-ridged part of the exposed surface, is closely 

 covered with minute and delicate longitudinal furrows mingled 

 with pores. The base is hollow, with rather thin walls, 

 which are always crushed in by pressure ; posteriorly this 

 hollow is widely open by the usual sulcus, superiorly it passes 

 up into the narrow pulp-cavity of the closed portion of the 

 spine. Above the closure of the sulcus the posterior aspect 

 shows a rather narrow concave area, bounded by two prominent 

 edges, immediately within which latter there is on each side 

 a row of strong recurved hooks or denticles ; towards the 

 apex the denticles of each row are very close together and 

 alternate in their arrangement ; and, as I have already em- 

 phasized in my notes on Gyracanthus *, the young spines are 

 not miniatui'es of the larger ones, but represent only their 

 distal portions, so that the proportion of the surface covered 

 by longitudinal ridges varies according to the size of the 

 specimen. In very small ones the entire exserted surface 

 may appear fluted, while in one very large specimen, in which 

 the apical portion is loorn away, only the three long ridges 

 remain. 



Although Agassiz did not enter quite so much into detail 

 in his description, yet the leading characters of the genus were 



• Ann. & Mag. Nat. Hist. (5) xiii. p. 41. 



