1885.] NATURAL SCIENCES OF PHILADELPHIA. 219 



of organization in these Crinoids, especially as these plates in- 

 crease numerically in the individual by growth. In the Crotalo- 

 crinidae they cover the entire peristome, including the central 

 piece and proximals. In the Reteocrinida? and Glyptoerinidae 

 they extend from the basals to the central piece. In Actino- 

 crinus, Melocrinus and Platycrinus, from the first radial to the 

 proximals, exactly as in the early Cyathocrinus, only that in the 

 latter the interradials consist of a large single plate, in the others 

 of numerous small ones. 



If it were true that the deltoids of the Blastoldea, and their 

 representatives, the interradials of the Cyathocrinid;e, were 

 orals, the first interradials of all Camarata would be oral plates, 

 and all higher orders upward growth of the orals. That this is 

 not the case is clearly shown by the fact that all these plates, 

 from the first to the last, are calyx plates, i. e., abactinal ; while 

 the orals of the Neocrinoidea are actinal, being developed around 

 the left peritoneal tube. 



That the interradials and their associates, the interaxillaries 

 and interbrachials, dorsally and ventrally, are abactinal plates is 

 further shown by the presence of perisomic plates underneath 

 the vault, which, wherever they have been observed subtegmin- 

 all}-, extend from the first interradial to the end of the central 

 piece (PL 2, fig. 8). The disk of the Palreocrinoida 1 , thei-efore, 

 begins from beneath the first interradial, and rests, as in the 

 Neocrinoidea, against the first primary radial, thereby making 

 the first interradial, in the true sense of the word, a vault plate. 1 



According to Carpenter, the Ichthyocrinidae and some of the 

 doubtful Silurian forms, such as Reteocrinus and Xenocrinus, 



1 The term "vault" has been heretofore applied by most writers to all 

 plates of the ventral side. In this sense it is actually a misnomer. If the 

 term is used at all, it should by right include all interradial, interaxillary 

 and interbrachial plates, dorsally and ventrally, and these might be very 

 appropriately designated as vault plates, to distinguish them from the 

 perisomic or disk plates, which are placed beneath the others, and follow 

 their direction. But fearing that the introduction of a new term, or giving 

 a different intei pretation to the same term, might produce confusion, we 

 it t ,i in it as a convenient and short mode of expression for all plates of the 

 ventral side that are not perisomic. It is therefore a merely conventional 

 term. Carpenter applies it to all actinal plates of the dome, with the 

 exception of the perisomic ones, in which he includes all interradials of 

 the ventral side which he takes to be actinal. 



