18 CATALOGUE OF THE BLASTOIDEA. 



the divergence of the two strong ridges (1 & 2) in the azygos basal x (which is placed 

 downwards in the figure), as is well shown in the side view (PL XIX. fig. 13). This 

 one angle of the trihedral base thus corresponds to radii A and B ; while its other 

 two angles send ridges (4 & 7) on to C and E. The D radius is formed opposite to 

 the side joining these two angles by the convergence of the adjacent ridges (5 & 6), 

 which start from these angles as described above. They are stronger than ridges 

 3 & 8, which start from the same points, and converge on to radials A and B, where 

 they meet the strong ridges from the azygos basal (1 & 2). The base of the 

 asymmetrical Pentepkyllum (PL XVI. fig. 16) is essentially like that of Triccelocrinus. 

 It has four prominent ridges, two of which are on the azygos basal, while the other 

 two (4 & 7) occupy the median lines of the paired basals. It might be possible to 

 use this criterion for determining the symmetry of the ventral side, as in the case of 

 Eleutherocrinus ; but we have no material for such an investigation except an internal 

 cast. 



The trihedral base of Troostocrinus is bounded by ridges 4 & 7 and the common 

 stem of 1 & 2, i. e. by the median ridges of the three basals, and the same is the 

 case in the more or less three-lobed base of Codaster trilobatus (PL XIII. figs. 2, 13). 

 This character is also distinctly traceable in good specimens of Pentremites and Pen- 

 tremitidea, viz. median ridges on basals y and z, which terminate in the lips of radials 

 C and E respectively, and a divergent ridge on the azygos basal x, the limbs of which 

 end on radials A and B respectively, while the base falls away in the direction of 

 radius D (PL V. figs. 8, 11, 14, 29). 



The radially situated under-basals which occur in so many Crinoids, both Palaeozoic 

 and Neozoic, appear to be altogether unrepresented in the Blastoidea ; though an 

 attempt has been made to establish the existence of a series of plates intervening 

 between the basals and the top stem-joint. Thus Lyon l wrote in 1857 that " Pen- 

 trt mites florealis, (jlobosus, piriformis, and others have severally three small plates 

 or pieces, distinctly separated from the pieces heretofore designated as the " Basal 

 pieces ;" these three pieces form the base of the cup, and as they lie below the pieces 

 heretofore recognized as basal, are true basal pieces, and the others necessarily 

 become first radials." In consequence of this observation Lyon was led to propose a 

 change in the generic formula of Pentremites. The three plates which he believed 

 himself to have discovered were termed basals ; while those which lie immediately 

 above them, and had been previously regarded as the basals, received the name " first 

 radials " ; and the true first radials or fork-pieces were transferred to the category of 

 second radials 2 . It never seems to have occurred to Lyon that there might be two 



1 Palseontological Report, Owen's 3rd Report Geol. Survey Kentucky, 1857, pp. 468, 469. 



'-' This is an excellent instance of the traps which nature lays for those empirical pataontologists who 

 neglect the study of morphology. No one who had mastered Miiller's memoir on Pcntacrinus (then fifteen 

 years old) could possibly have made such a blunder. 



