NO. 1S46. ON CERTAIN ELEUTHEROZOIC PELMATOZOA—EIRK. 49 



English specimens. In a specimen from France referred to this 

 species Carpenter gives the column a length of somewhat more than 

 60 mm. In any case the column at its maximum is remarkably 

 short and probably scarcely exceeds the length of the crown. In the 

 majority of figures given by Carpenter there is no sign of basal 

 attachment. In the case of one column, however (1882, Pi. 1, fig. 5), 

 there seems to have been cementation or possibly attachment by 

 lateral root-like processes. It is probable that in all cases observed 

 other than this the animals were detached. In one specimen figured 

 by Carpenter there is a slight flexure of the distal end of the column 

 suggesting that this portion was prehensile and possibly employed 

 for looping about some object as in forms elsewhere noted. The 

 column in this individual is comparatively small. The most striking 

 feature in connection with the stems of this species is their extremely 

 variable length. This varies from a column having about 70 ossicles 

 to one consisting of but a single columnal. Various intermediate 

 stages are represented. From the facts as we have them it seems 

 probable that the shortening of the column as showTi in these forms 

 is due to the dropping ofi^ of some of the distal columnals accompanied 

 by more or less resorption. It does not necessarily follow, however, 

 that this resorption is as a rule other than purely local in its action 

 and is effective merely in rounding off the terminal ossicles. When 

 the stump of the column is unusually short, resorption seems to affect 

 aU the ossicles to the extent of forming them into a subconical knob 

 (PI. 6, fig. 1). The probable steps in the process by which this short- 

 ening of the column is brought about will be given later. 



The process by wliicli freedom was attained by Millericrinus 

 prattii, and the changes through which the animals subsequently 

 passed as regards the structure of the column, hold no small interest 

 for us, for in tliis process we probably see outlined the phylogenic 

 history of the assumption of an eleutherozoic habit by the Comatulse. 

 Indeed, were Millericrinus to have possessed cirri, there is small 

 doubt but that this very species would have formed the radicle of a 

 fine essentially comatulid in habit, and perhaps of considerable vigor. 

 Detachment from the cemented base probably did not take place 

 very early in the life of Millericrinus prattii: perhaps at a not much 

 younger stage than is figured by Carpenter (1882, PI. 1, figs. 6 and 8). 

 It should be noted, by the way, that in the case of all the fairly long 

 columns figured by Carpenter the stems are transversely fractured 

 other than at the distal extremity, and the parts somewhat thrown 

 out of line. This separation of the column comes at various altitudes, 

 in one instance (1882, PI. 1, fig. 8) being quite near the crown. It 

 is evident that the union between the columnals is not a strong one 

 at best. Furthermore, disruption may apparently take place at, 

 94428°— Proc.N.M.vol.41— 11 4 



