MORPHOLOGICAL PART. 55 



(Plate II. Fig. 4 h), being in some cases perfectly fused with that plate (Plate 

 yi. Fig. 11). The condition is the same as in the two species of Millencrmus 

 in which infrabasals have been recognized, and we conclude from analogy 

 that a fusion of those plates eventually took place in all groups in w^hich the 

 new stem joints are not formed directly beneath the calyx. 



The case is different among the Pentacrinidse, in which the new stem 

 joints constitute the upper part of the stem. Of the principal genera which 

 have been referred to this family, one — Extracrinus — has small infrabasals 

 persistent through life ; while in the other two — Pentacrinus and Metacrinus 

 — no trace of them can be found in the adult. The dicyclic nature of 

 Peniacmiis and Metacrinus is indicated by the orientation of the stem and 

 cirri, the angles of the stem in both of them being interradially disposed, 

 and the cirri radially. But what became of their infrabasals? That they 

 fused with the upper stem joint, like those of the Ichthyocrinidee, need not 

 be considered in this family, as that would necessarily prevent the formation 

 of new joints at the top. They may have been resorbed in the growing 

 animal ; but it appears to us more probable, from palaeontological evidence, 

 that the plates gradually diminished in size, and finally disappeared alto- 

 gether in the group. The structure of the Pentacrinidee is very different 

 from that of the Comatul^ and Apiocrinidse, and it appears to us that Cri- 

 noids in which the upper joint of the stem is the youngest, cannot be derived 

 from those in which the top of the stem is fused with the infrabasals, and the 

 two groups should be widely separated. The Apiocrinidae and Comatul^ 

 which have a centro-dorsal, — /. ^., in which the infrabasals are fused with 

 the upper stem joint, — should be placed together with, or close to, the 

 Ichthyocrinid^ ; while the Pentacrinidoe, which in many points agree wdth 

 the structure of the Poteriocrinidae and Encrinidae, we think might be safely 

 referred to the Inadunata Fistulata. 



It is very singular that while in Extracrinus and Metacrinus the projections 

 of the open space within the basal ring, and the axial canal of the stem, are 

 radially disposed (Plate YI. Fig. 9), both are decidedly interradial in Penta- 

 crinus (Plate VI. Fig. 8). This anomaly, if we may so call it, w^as regarded by 

 Carpenter * as a proof that our generalization upon the stem is not applicable 

 to the Neocrinoids. This, indeed, seemed at first to be indicated also by the 

 orientation of the stem ; but Bury's discovery of infrabasals in the Comatula 

 larva changed the whole aspect of the case. The discovery of these plates, 



* Bibliogr. Notices, Ann. and Mag. Nat. Hist., March, 1886, p. 287. 



9 



