NO. 1846. ON CERTAIN ELEUTHEROZOIC PELMATOZOA—KIRK. 37 



crinoid, is probably a secondary one, being caused by a deposition of 

 calcareous matter within the reentrant angles of the coluninals. In 

 most cases the original form of the ossicles is indicated by the more or 

 less distinct pentagonal section of the column. In many species, 

 however, truly pentalobate columnals are to be found in the adult 

 stages even in the most distal portion of the stem as preserved, 

 proving conclusively that a considerable portion of the column has been 

 lost. 



If we grant the dropping off of columnals during the lives of the 

 animals, a feature that is directly comparable to the condition found 

 in Millericrinus, hereafter described, w^e may readily account for the 

 structure of all the stems known to us among the Pentacrinidse. In 

 very young stages we should find, as we do, considerable portions of 

 the original column of the individual. Gradually more and more of 

 this drops off until it entirely disappears. It is probable that this 

 dropping of columnals extends well into the pentalobate portion of the 

 column in some species. 



It is possible that the gradual loss of the distal portion of the column 

 may be ascribed to periods of attachment followed by disruption of the 

 crinoid from its base. In such cases detachment may occur immedi- 

 ately below the most distal nodal or at some higher point. The uni- 

 form termination of the columns by nodals among such types as Endoxo- 

 crinus suggests that detachment was had at such points. It may be, 

 however, that the column if broken within an internode would 

 subsequently drop off a sufficient number of ossicles so as to bring a 

 nodal at the end of the column. The advantage of this arrangement 

 is obvious, as the presence at the distal extremity of the column of a 

 verticil of cirri as shown by Carpenter (1884) in Isocrinus hlakei 

 (PL 31, fig. 1), in E. wyville-tliomsoni (PI. 19, fig. 1), and in other 

 species, would enable the animals readily to grasp a projection upon 

 the bottom, and thus temporarily anchor themselves. 



Factors governing length of coluTnn. — The degree to which the col- 

 umn is shortened probably depends to a great extent either upon the 

 physical conditions under which the animal lives or upon the structure 

 of the organism itself. The main factor, however, is perhaps the 

 abundance of cirriferous nodals in a given length of column. This 

 in turn depends upon the length of the internodals. This structural 

 feature probably varies in different species, and may perhaps vary 

 within any given species. A case of shortening of the internodes is 

 shown in Isocrinus parrse (maclearanus) (PL 4, fig. 1), This species 

 exhibits perhaps the highest degree of specialization known among 

 the recent Pentacrinidas in this regard. It will be noted in this form 

 that each internode consists of but a single plate. Even greater 

 compression of the nodals would give us essentially the structure to 

 be observed in Group II. In most cases it will be observed that 



