NO. 1846. 02V CERTAIN ELEUTHEROZOIC PELMATOZOA—KIRK. 67 



leaving no opening, therefore obviating the necessity of any secondary 

 structvires. 



These four types of structure will now be taken up in the order here 

 given, and a more detailed discussion given to each group. 



The Crinoidea constituting this assemblage are entireh^ restricted 

 to Mesozoic and recent forms. In the present seas representatives 

 are very numerous and have an enormous range. It seems highly 

 probable that the Comatulas represent a number of genetic lines, and 

 that even some of the families as defined by Bather are polyphyletic 

 in origin. Wliether sufficient material and information will ever 

 become available to enable one closely to discriminate between these 

 convergent lines it is impossible to say. The component genera after 

 all have an essential unity of structure, and have doubtless been 

 derived from not widely divergent types. Such being the case, and 

 inasmuch as these forms are so well and widely known, no attempt 

 will be made to discuss any given genus in considerable detail. Cer- 

 tain points regarding the structure of the animals are given, however, 

 because of the bearing they have on questions dealt with in the 

 present paper. The probable origin and development of the centro- 

 dorsal is treated at some length, for instance, not only because of its 

 importance in tliis group, but also because of the bearing it has on 

 larger systematic questions. 



During the past two or three years the comatulid genera as recog- 

 nized by Carpenter have been divided and subdivided into numerous 

 genera, chiefly through the efforts of Austin Hobart Clark. Inas- 

 much as the descriptions of these genera are of a more or less pre- 

 liminary nature, one may be pardoned for not making use of the new 

 appelations. For the purposes of the present paper extreme nomen- 

 clatorial refinements are after all in no wise essential. On this ac- 

 coimt very largely, no attempt has been made to depart from the 

 nomenclature of Carpenter, and the generic names of the free- 

 swimming crinoids as here employed are those commonly accepted. 



The structure and develojjment of the centro-dorsal. — In the well- 

 known Antedon, as well as in the other genera of this group, the cen- 

 tral apical portion of the dorsal cup consists of a centro-dorsal. This 

 is made up chiefly of the proximale, which either persists alone, or 

 possibly has a limited number of columnals fused with it. To the 

 proximale also are probably fused in all cases the infrabasals. These, 

 because of their size, are a negligible quantity in a consideration of 

 the general composition of the adult centro-dorsal. The centro- 

 dorsal is cirriferous, except m the adults of certain species, where the 

 cirri have been lost. 



