INVERTEBRATES. 219 



usually nearly or quite isolated from the vault above ; each 

 occupied by from ten to seventeen or eighteen pieces. Inter- 

 axillary or interbrachial spaces occupied by from two to eight 

 or ten small pieces, the upper of which connect with the false 

 arm-bases and vault. True arms very long, slender, pendu- 

 lous, and more or less branching, supporting minute ascending 

 tentacula. Pseudo-brachial appendages five, or rarely six 

 (from the division of the one on the anal side), composed of two 

 ranges of pieces, with a central cavity in those of each range, 

 or of several ranges inclosing a double central cavity; not 

 furrowed above or provided with any external openings; all 

 extending out from the margins of the dome, at first horizon- 

 tally, and then bifurcating — the divisions spreading, curving 

 downwards, and each tapering to a point. Vault much de- 

 pressed, with usually a more or less marked concavity opposite 

 each false arm; opening sublateral, or about midway between 

 the centre and margin of the dome, not raised upon a proboscis. 



This interesting group, as here defined, includes two sections, distinguished 

 as follows : 



1. GILiBERTSOCRINUS, Phillips (proper).=OZiacrin«s, Cumberland. 



Pseudo-brachial appendages located directly over the interbrachial spaces; anal 



and interradial pieces not quite isolated from the dome. 

 Examples: G. calcaralus, G. bursa and G. mammillaris, Phillips. — (European.) 



2. GONIASTEROIDOCRINUS, Lyon and Casseday,= Trematocrinus, Hall. 



Pseudo-brachial appendages located over the interradial and anal areas, which 

 (excepting sometimes the latter) are entirely separated from the dome. 



Examples: G. tuherosus, Lyon and Casseday ; G. Jiscellus, M. and W., and G. 

 papillatus, G. reticulatus, G. typus, etc., published by Prof. Hall under the 

 name Trematocrinus. — (American.) 



From the foregoing description, it will be seen this genus agrees almost 

 exactly, in the structure of the body below the arms, with Jihodocrinus, 

 Miller, with which it is often confounded. It was, indeed, if we mistake not, 

 also originally separated from that group by Phillips, under a misapprehension 

 in regard to the number of basal pieces in IZhodocrinus, which had been de- 

 scribed as having a tripartite base. After this error in regard to the structure 

 of Rhodocrinus had been pointed out, by Austin and others, it is not at all 

 surprising that nearly all who had only seen the specimens of Gilbertsocrinus 

 figured by Phillips, divested of the appendages of the upper part of the body, 



