NO. 1846. ON CERTAIN ELEUTHEROZOIC PELMATOZOA—KIRK. 75 



bears radicular cirri. These are obviously for purposes of attach- 

 ment. Apparently never more than two cirri are borne by a columnal. 



In another Cretaceous genus, TMolliericrinus (PI. 7, figs. 3-5), we 

 have made a great advance. In this genus detachment had become 

 universally effective, disruption occurring immediately beneath the 

 proximal columnal. Even though detachment took place late in 

 the life of the organism, as is evidenced by the large size of the articular 

 face on the lower surface of the centro-dorsal (PI. 7, fig. 5), the very 

 fact of disruption bemg actually effective at this point, shows clearly 

 that an eleutherozoic habit had long been maintained. This is further 

 indicated by the presence of cutI on the centro-dorsal. As is shown 

 by T. Jierherti (PI. 7, fig. 3), the cirri are small, few m number, and 

 irregularly placed. In T. Jlexuosus (fig. 4 on the same plate) the 

 cirri are comparatively larger, more numerous, and form a practically 

 contmuous rmg about the centro-dorsal. From the shape of the artic- 

 ular face as preserved On the centro-dorsal (PI. 7, fig. 5) it is evident 

 that this genus had a typical Bourgueticrinus column. Such a crinoid 

 as TMolliericrinus might weU serve as a point of inception for a coma- 

 tulid line. If such be the case, it seems probable that in this particular 

 evolutionary line none of the columnals other than the proximal one 

 goes to form the mcipient centro-dorsal. This seems to be indicated 

 from the fact that, owing to the structural pecuHarities of the stems 

 involved, disruption constantly takes place immediately beneath the 

 proximal columnal. From such evidence as is available, it seems 

 possible that in many of the comatuHd lines other columnals fuse 

 with the proximale, however. 



Such fusion of columnals seems clearly to be indicated m the onto- 

 genetic development of Antedon as figured by Carpenter (1888, PI. 

 14). Some of the figures are here reproduced (PI. 7, figs. 6-10). 

 These figures serve equally well to show the development of the cirri 

 on the centro-dorsal. Figures 8 to 10 represent stages in develop- 

 ment of a single species and as such are not subject to the structural 

 variations that apparently obtain in the case of different species.* 



In figure 7, the proximale is shown as the shortest columnal in the 

 stem. There is no indication of the presence of cirri. In figure 8, 

 the proximal columnal is still very short, but bears a verticil of incip- 

 ient cirri. These are five in number and radially disposed. It is 

 to be noted that immediately distad to the proximale there are three 



1 Mr. A. II. Clark says that these three specimens identified by Carpenter as ^. muUispina do not appear 

 to belong to the same species and, indeed, may be referable to diflerent groups. Furthermore he states 

 that the figures are inaccurate. It scarcely seems probable that the draftsman could go far wrong in regard 

 to the columnals, however, and the evidence of fusion having taken place, though by no means conclusive, 

 still seems fairly good. 



Mr. Clark believes that the centro-dorsal never consists of any columnals other than the proximal one. 

 This is unquestionably true in many species, and possibly in many phyletic lines, but such inconclusive 

 evidence as is at present available scarcely warrants a postulate of similar conditions as a universal attribute 

 of the Comatulse. 



