80 CATALOGUE OF THE BLASTOIDEA. 



canal-system of the recent Echinoderms, and that as the convoluted plates of the 

 former have the same structure and connections as the madreporic sacs and tubes or 

 sand-canals of the latter, they are most probably all the homologues of each other." 

 He pointed out first of all the resemblance between Codaster and Caryocrinus, each with 

 its five groups of hydrospires on the summit 1 ; and he continued : — " In Pentremites, 

 as in Codaster, the five hydrospires are divided into ten equal parts by the five rays. 

 In Codaster these ten parts remain entirely separate from each other ; but in 

 Pentremites they are reunited in pairs, the two in each interradial space being so 

 connected at their inner angles that their internal cavities open out to the exterior 

 through a single orifice or spiracle." He also described how the hydrospires of 

 Pentremites (Granatocrinus) ellipticus " instead of being formed of broad sacs with 

 a number of folds on one side, consist of ten simple cylindrical tubes connected 

 together in five pairs." 



As he believed Codaster to be a Cystid, he was able to say that " between the 

 Cystidea and the Blastoidea the most important changes are that in the latter the 

 hydrospires become connected in pairs, and are also brought into direct communi- 

 cation with the pinnulae. In the palaeozoic Crinoidea (or at least in many of them) 

 concentration is carried one step further forward, the five pairs of hydrospires being 

 here all connected together in the centre." It was at this stage of his demonstration 

 that Billings made a serious mistake. He had distinctly described the hydrospires 

 as grouped into ten interradial pairs, each with its summit-opening between the origins 

 of two ambulacra. But his next step was to suppose that the two hydrospire tubes 

 at the sides of each ambulacrum of a Blastoid became united in the Crinoid into a 

 single tube, situated beneath the ambulacrum, and therefore radial in position, 

 namely, the radial water-vascular trunk. The five ambulacral canals thus formed 

 were supposed by Billings to be attached, not to an oesophageal ring, but to the 

 upper extremity of the convoluted plate, which he regarded as representing the sand- 

 canal of a Starfish. Subsequent researches have shown, however, that this latter 

 view is altogether untenable ; while the supposed concentration of the ten inter- 

 radial hydrospires of Granatocrinus ellipticus into the five radial water-vessels of a 

 Crinoid is at variance with every principle of Echinoderm morphology. Not only is 

 there every reason to believe that the Blastoids possessed ambulacral vessels which 

 were united into a ring round the mouth, but also representatives of the hydrospire- 

 sacs of the Blastoids are to be found in recent Echinoderms (Ophiurids) as pointed 

 (nit by Ludwig 2 . 



1 Billings remarked that two of those groups are incomplete in Codaster, in order to mako room for an 

 opening which he called " the large mouth and vent." It would have hcen more correct if he had grouped 

 the hydrospires into interradial pairs instead of into radial ones (p. 412), and had simply said that one pair 

 was altogether absent (PI. X. fig. 19 ; PI. XII. figs. 1,4; PL XIII. figs. 1, 4). 



2 ' Morphologische btudien an Echiuodermcn,' Leipzig, 1877, p. 282. 



