340 Mr. A. S. Woodward on the 



dimensions ; but the posterior (fig. 6) can readily be distin- 

 guished by the evidence it bears of rehitively deep and oblique 

 insertion. Both spines are considerably crushed and abraded, 

 and incomplete distally ; but they appear to have been much 

 compressed laterally, with a sharp anterior border. A very 

 characteristic ornament is also preserved, consisting of small 

 elongated tubercles, more or less arranged in series, and often 

 fused above into delicate continuous ribs, which begin to 

 prevail at a considerable distance from the upper extremity. 

 The posterior face is slightly ridged, but the double series of 

 denticles is almost destroyed. Measured from the extreme 

 proximal limit of the ornament, the base of insertion of the 

 first dorsal spine is almost precisely equal to that of the 

 second ; but the posterior opening of the central cavity is only 

 0*205 in length in the former, while in the latter it mea- 

 sures 0*235. 



Cephalic Spines. — The cephalic spines of Asteracanthus 

 are very similar to those of Hyhodus and Acrodus, originally 

 described by Agassiz under tlie name of Sphenonchus * ; but, 

 as indicated by specimens nos. 1 and 2, these dermal weapons 

 are relatively larger. One of them is shown, from the upper 

 and lateral aspects, two thirds nat. size, in figs. 7 and 8. Two 

 pairs of such spines occur upon the sides of the head of Hy- 

 hodus and Acrodus ; and the fossil now under discussion shows 

 that there was also a paired arrangement in Asteracanthus. 

 The base of insertion of the spine is very robust and some- 

 what saddle-shaped, but with one side-lobe much more deve- 

 loped than the other. The exserted portion is slender, grad- 

 ually arched, rising backwards (or downwards) from the 

 broader anterior (or upper) extremity, and terminating in a 

 barbed point. At the base it is oval in section, being some- 

 what laterally compressed ; and the only keels proceed, one 

 from the point to the large inferior barb^ thence diverging 

 and disappearing, and the other from the point to the small 

 lateral barb, from whence it is continued but also rapidly 

 vanishes. The shining exposed surface is smooth, except 

 superiorly upon the proximal half, where a few large longi- 

 tudinal rugge are to be observed. 



Specific Determination. — On comparing the dorsal fin- 

 spines described above with typical examples of A. orna- 

 tissimus striking differences will at once be observed. If, 

 indeed, these were isolated spines they might well receive a 

 distinct specific name ; and there are some peculiarities in 

 the associated teeth which also might appear to justify the 



* See E. Charlesworth, Mag. Nat. Hist. u. s. vol. iii. (1839), p. 245, 

 with iig. ; also E. C. H. Day, Geol. Mag. vol. ii. (1865), p. 665. 



