64 THE CRINOIDEA CAMERATA OF NORTH AMERICA. 



299, we asserted that not only the majority of Stalked Crinoids, but also the 

 ComatiilaB, were built upon the dicyclic plan, and probably had infrabasals 

 in their larval state. 



The centro-dorsal at the Pentacrinoid stage of the Comatulse, as may be 

 seen by examining Plate YI. Figs. 18 and 19, agrees closely with that of the 

 Apiocrinidae. It is interradially disposed at the proximal face, and also at 

 the distal face, so that its angles correspond with the angles of the basals, as 

 in those dicyclic Crinoids whose infrabasals are hidden by the column; and 

 even in its free floating stage the centro-dorsal retains its interradial 

 position. 



So strongly were we impressed with the conviction that the Comatulse 

 are dicyclic Crinoids, — although no traces of infrabasals had ever been 

 found by Wyville Thomson, the two Carpenters, Goette, and others who had 

 extensively studied the embryology of Antedon, — that we urged EurojDean 

 investigators to make fresh search for these plates. 



It w^as therefore with no little satisfaction that we learned from Dr. 

 Carpenter in July, 1887, that infrabasals, w^hose existence we had predicated 

 upon palgeontological evidence, had actually been found by Mr. H. Bury. 

 This important discovery was announced at the Manchester meeting of the 

 British Association, and communicated by Mr. Bury in a paper on '' The 

 early stages in the Development of Antedon rosacear * The plates appear in 

 the ciliated larva, and consist of three unequal pieces, which in the Penta- 

 crinoid stage are fused together with the top-stem-joint (centro-dorsal), so as 

 to form together with the latter one large plate, wdth five angles, radial in 

 position. One of the plates is smaller, and has but one salient angle, the two 

 larger ones having two, — a structure which suggests that the two latter may 

 represent a pair of anchylosed plates. Three similar plates, but with a some- 

 what different orientation, persist throughout life in the Ichthyocrinidse, but 

 the smaller plate of Antedon rosacea^ if we understand Mr. Bury, is placed 

 anteriorly; while that of the Ichthyocrinidse is directed posteriorly (Table 

 A, Figs. 14 and 15).t 



It is of considerable importance that the infrabasals of all Carboniferous 

 Ichthyocriniclae are coalesced with the top-stem-joint (centro-dorsal) similar 

 to those of Antedon. In specimens in which the column has been detached 

 from the crown, the small infrabasals generally adhere to the top-stem-joint 



* PliilosopMcal Transactions of the Eoyal Society of London, 1888, pp. 357-300. 



f The. orientation of this plate was incorrectly given by us in the Revision, Part III. Plate 6, l^ig. 23. 

 Since then we have found out that it is located in all Ichthyocrinidse to the right of the posterior basal. 



