MONOGRAPH OF THE EXISTING CRIXOIDS. 



229 



appears to be merely an exaggerated concavity of the dorsal pole, exactly compar- 

 able to the conditions found in many recent species for instance, in Comantlius 

 Icnnctti and in C. pinguis (figs. 171-174, p. 231). There is no reason whatever for 

 supposing that the centrodorsal in any fossil species was open at the dorsal pole any 



FIG. 165. 



FIG. 166. 



FIG. 168. 



FIG. 169. 



FIGS. 165-170. 165, THE CENTRODORSAL AND RADIALS OF A SPECIMEN or COMANTHINA SCHLEGELII FROM THE PHILIPPINE ISLANDS 

 (AFTER P. H. CARPENTER). 166, THE CENTRODORSAL AND RADIALS OF A SPECIMEN OF COMANTHINA SCHLEGELII FROM TIIE 

 PHILIPPINE ISLANDS (AFTER P. H. CARPENTER). 167, THE CENTRODORSAL, RADIALS, AND IBs SERIES OF A SPECIMEN OF 

 COMANTIIERIA ALTERNANS FROM THE PmLLTPINE ISLANDS WITH TWO ATROPHIED CIRRI REMAINING. 168, THE CENTRODORSAL 

 AND RADIALS OF A SPECIMEN OF COMANTHINA SCHLEGELII FROM THE PHILIPPINE ISLANDS (AFTER P. H. CARPENTER). 169, 

 THE CENTRODORSAL AND RADIALS OF A SPECIMEN OF COMANTIHNA SCHLEGELII FROM THE PHILIPPINE ISLANDS (AFTER P. H. 

 CARPENTER). 170, THE CENTRODORSAL AND RADIALS OF A SPECIMEN OF COMANTHINA SCHLEGELLI FROM THE PHILIPPINE 

 ISLANDS (AFTER P. n. CARPENTER). 



more than it is in any recent species, and there is no evidence wliich undeniably 

 supports such a view. 



The comatulid centrodorsal varies in shape from a small stellate or pentagonal 

 plate, smooth at the edges and sunk beneath the dorsal surface of the radial penta- 

 gon (figs. 82, p. 135, 153-159, p. 221, 162, p. 223, 164, p. 227, and 16S-170, p. 229), 

 or a thin disk more or less concave dorsally with a single, often partially deficient, 



