NO. 1846. ON CERTAIN ELEUTHEROZOIC PELMATOZOA—KIRK. 51 



canal and surface features of the inferior face of this ossicle, such as 

 are shown in figure 5, have been filled and smoothed over by a sec- 

 ondary deposit of calcareous matter. The modified channel as thus 

 formed sunulates a centrale to a remarkable degree. It may not be 

 styled a centrale, however, for it is a modified columnal, wliich a cen- 

 trale is not, as will subsequently be shown. The proxunale of 

 Millericrinus and such stumps of columns as are illustrated by fig- 

 ures 1 and 3 are directly comparable with the centrodorsal of the 

 comatulids, however, and are the result of practically an identical 

 process worldng on columns of different yet not greatly unrelated 

 t3rpes. The dift'erences in the two cases are simply those of degree 

 of specialization of the columns involved. 



The whole tendency to be noted within this species seems to be 

 toward the elimination of the column. It is of ver}'' great mterest, 

 then, to observe in practically all of the specimens figures by Car- 

 penter that even in the shortened column there is a distinct counter- 

 tendency toward a lengthening of the stem by the intercalation of 

 new column als. In some of the specimens this addition of new 

 ossicles is quite marked, as in figure 1, Plate 1, of Carpenter's paper. 

 Relative to the intercalation of new ossicles in the stem, there are 

 features of no small interest to be noted in connection wdth the 

 structure of the proximal columnal. 



Structure of the "proximalc^^ in AfUlencrinus. — Wachsmuth and 

 Springer (1897) and Bather (1900) have placed the Apiocrinidce in 

 the Flexibilia Pinnata, doubtless on the supposition that there is a 

 proximale. According to the figures of Carpenter (1882), it does not 

 seem that there is such a persistent columnal in Millericrinus prattii. 

 On the contrar}'', it would appear, as Carpenter himself notes, that 

 there frequently is to be observed the formation of a new ossicle 

 between the basals and what was formerly the proximal columnal. 

 Figures 11, 205, and 17, as given by Carpenter, may well indicate the 

 successive stages by which the external appearance of such a 

 columnal is marked. At first between and beneath the basals appear 

 small subtriangular plate-like patches. These, as Carpenter says, 

 are certainly not infi'abasals. In figure 20h, as shown in the right- 

 hand portion of the figure, it would appear that the radial patches 

 had run together beneath the basals in certain portions of the 

 peripheiy. In figure 17 there seems to be a continuous plate formed. 

 In this figure it is interesting to note that those portions of the plate 

 lying between the basals are of considerably greater height than the 

 intermediate connecting portions lying directly beneath them. 

 This clearly indicates the formation of the columnal by such a series 

 of stages as has been outlined. It would appear, then, that at least 

 during certain stages of the growth of Millericrinus prattii new 

 columnals were formed immediately beneath the basals. 



