Dr Traquair on the Genus Gyracanthus {Agassiz). 93 



lutely necessary to re-examine the specimens in the Atthey 

 Collection, now in the museum at Newcastle-on-Tyne. And 

 having recently visited that city, I must here express my 

 cordial thanks to my friends Mr W. Dinning, Secretary of 

 the Newcastle Natural-History Society, Mr E. Howse, 

 Curator of the Museum, and Mr J. Hancock, member of com- 

 mittee, for the kind and liberal manner in which they afforded 

 me every facility for examining the specimens in that remark- 

 able collection of Coal-measure vertebrate remains. 



Although I have not seen the original type of Agassiz's 

 Gyracanthus ticherculatus, I have no hesitation in referring to 

 it the great majority of the specimens from Newsliam in the 

 Atthey Collection, and they form, indeed, a most beautiful and 

 instructive series. And as no systematic description has been 

 given of this form since the time of Agassiz, who had only a 

 drawing of a mere fragment to go upon, it will not be out of 

 place to enter somewhat into detail as to the configuration of 

 these spines. 



Proceeding first to the consideration of those labelled 

 " pectoral" in the Atthey Collection, one very fine example is 

 15i inches in length by 2 J in diameter at its widest part near 

 the base ; its distal extremity is obliquely truncated or worn 

 off on the anterior aspect, and the whole spine, when looked 

 at from the front, displays a well-marked lateral curvature or 

 bend, which enables us to distinguish a convex and a concave 

 side. It will also be observed that the lateral surface is more 

 gibbous or rounded on the convex aspect of the spine, flatter 

 on the opposite, so that for purposes of description we may 

 distinguish the two sides as "gibbous" and "subgibbous" re- 

 spectively. Still, regarding it from the front, it will be seen 

 that the sculptured surface ends proximally in an acute angle; 

 but the apparent middle line on which the tuberculated or 

 " gyrating " ridges meet does not Usect this angle, but divides 

 it so that the sculptured part is larger on the gibbous side. 

 ISTow, turning the spine over so as to look at it from behind, 

 we observe that the longitudinal cleft or sulcus leading into 

 the central cavity is not in the middle of the non-sculptured 

 inserted part, but is placed more towards the subgibbous side, 

 so that we have here from the very beginning a marked 



