MORPHOLOGICAL PART. 113 



latter were true vault plates, as Carpenter inferred we did. We held that 

 while in the vault of the typical Actinocriniis the interradial dome plates 

 meet over the ambulacra, in Platycrinus these plates opened out so as to 

 expose the covering pieces, and that these were gradually incorporated 

 into the vault. In a typical Platycrinoid, the covering pieces are so 

 modified as to lose almost their original character, being as large and 

 nearly as heavy as the surrounding plates, and they are united with one 

 another, as well as with the latter, by close suture. In some of the later 

 PlatycrinidaB they even may have been separated from the food grooves, 

 for in the internal casts they left no impressions ; while in casts of Acti7io- 

 crinus from the same localities, and in casts of certain Silurian Platycrinidae, 

 the outlines of the ambulacra are generally sharply delineated (Plate LXXY. 

 Fig. 14). 



Carpenter probably supposed the ventral structure of the Melocrinidae 

 and Rhodocrinidse to be in the same condition as that of the Actinocrinidse 

 and Platycrinidae respectively, that is, a disk when the ambulacra are ex- 

 posed, a vault when they are concealed. He alluded to Glyptocrinus in 

 connection with the Reteocrinidae and Ichthyocrinidse, in all of which the 

 ventral pavement is composed of an immense number of very minute, 

 irregularly arranged pieces. In the Ichthyocrinidae these plates are trav- 

 ersed by regular rows of alternating pieces, passing out from the oral centre 

 to the arms ; in the two other families, however, such alternating plates, 

 when present at all in the tegmen, are found only close to the arm bases. 

 Carpenter says respecting these groups,* '' I venture to think that in the case 

 of GlyptocrinuSj Reteocriniis, Xenocrinus, and also of the Ichthyocrinidae, the re- 

 semblance to the Pentacrinidae, Apiocrinidse, and Comatulae, is such as to 

 leave no reasonable doubt that the so-called vault of these Pal^ocrinoids is 

 homologous with the ventral surface of the body in the Neocrinoids." This 

 is true enough as to Taxocrinus and Onychocrinus^ and probably the Ichthyocri- 

 nidae generally, in which mouth and food grooves are exposed, as we have 

 found out from actual observation ; but in the case of Glyptocriniis and Reteo- 

 criniis, there is nothing to prove it beyond a superficial resemblance of the 

 plates. Carpenter's argument loses much of its force, considering that 

 among the Actinocrinidae within the same genus some species have large 

 plates, others very small ones, and the evidence seems rather to prove 

 that either these plates are all disk plates, or none of them are. 



* Chall. Rep. Stalk. Crin, p. 185. 

 15 



