340 PROCEEDINGS OF THE ACADEMY OF [1885. 



which numerous detached spines lie upon the surface of the plates 

 close to the tubercles from which they had been detached. That 

 these spines, to some extent, were movable, is more than proba- 

 ble. They were evidently connected with the plates by elastic 

 ligament, so as to 3-ield when accidentally brought in contact with 

 other objects, like the joints in a column, but we doubt if beyond 

 this the}' represent, either functionally or structurally, the spines 

 of the Echini. 



These views differ somewhat from those held by Williams, who 

 thinks it " not improbable that the original plates of Lepidocen- 

 trus eifelianus, described and figured by Johannes Miiller, which 

 were detached plates, associated with spines similar in nature to 

 those just described and borne upon similar tubercles, were plates 

 from the vault of a true Crinoid like Arthroacantha." And he 

 remarks further, " we have here a possible clue to a relationship 

 between true Crinoids and Perischcechinidie. 



There is in our opinion not the slightest doubt but that Mid- 

 ler's figures represent Echinoid plates, and that the spines which 

 were found associated with them had the same functions as those 

 of the true Urchins of later epochs ; but we think that the spines 

 of Arthroacantha form component parts of the plates taken sepa- 

 rately, and as such we regard them as representing in a modified 

 way the ordinary undivided spiniferous plates of other Crinoids. 

 For this reason we cannot regard the movable spines of Arthroa- 

 cantha of much more than of specific importance, but as the spe- 

 cies also possess an additional primary radial, it ma} r be well to 

 separate them generically from species of Hexacrinus which do 

 not have them. We allude to this more particularly, as Williams 

 and also Hinde, was inclined to regard Arthroacantha as the type 

 of a distinct family, a distinction, which, in our opinion, gives to 

 the movability of the spines a degree of importance which it does 

 from a morphological standpoint not deserve. 



Wo also doubt if (?) Arthroacantha Carpenteri had whorls of 

 cirrhi throughout the column, as supposed by Hinde. The col- 

 li in uar fragments which he figured on PI. 4 — if they belong to 

 this species at all — evidently formed the lower [tortious of the 

 stem, as shown l»3 r the size and the irregular arrangement of their 

 branches, and as such are regarded b} 7 us merely as radicular 

 cirrhi. 



Generic Diagnosis. — In form and arrangement of plates closely 



