268 NEW STARFISHES AND OPHIUEANS—rERRILL. voL.xvn. 



subfamilies. These are mostly small groups of geuera that have more 

 or less close relations to each other, but the distinctions between some 

 of them seem to me too slight for even subfamily groups. Every new 

 genus discovered is likely to break down some of the distinctions made 

 between such groups. Moreover, some of the distinctive characters 

 given by Mr. Sladen do not hold good for the genera classified by him. 

 Thus, the subfamily Pararchasterinte is said to have the papulfe "con- 

 fined to a limited area at the base of the rays," while the subfamily 

 Plutonasterinjie is said to have them "distributed over the whole abac- 

 tinal area." But, as a matter of fact, scarcely any of the genera 

 referred to either of the subfamilies have the papulae so distributed, 

 and in many of the genera they can be best described as confined to 

 the central part of the disk and to the median or radial areas of the 

 basal part of the rays and disk. They are almost always lacking on 

 the distal and submarginal parts of the rays, and on more or less exten- 

 sive dorsal interradial areas of the disk. This is the case in Plutonas- 

 ter, DytasteTy Pseudarchaster, etc., and is also the usual arrangement in 

 the Pentagonasteridoe . 



The genus Farar chaster, Siladen=Benthopecten, Verrill has essentially 

 this same arrangement of papnlfe, only they are absent from a somewhat 

 greater portion of the distal part of the ray, but different specimens of 

 the same species vary widely in this respect according to their age. In 

 fact, there is nothing very peculiar in their arrangement in this genus, 

 as compared with various other species formerly included in the genus 

 "iLrc/tas/er," so that when the genus Benthopecten was first briefly 

 described by me I did not consider it necessary to refer to this feature, 

 there being various other characters of much greater value. 



The special arrangement of the ijapulne in Pontaster is, however, a 

 character of importance. But there is surely no very close affinity 

 shown between PontaHier and Benthopecten by the arrangement of the 

 papuhe. 



My own view is that Benthopecten may be more closely allied to some 

 of the genera referred to the PentagonasteridiB by Mr. Sladen, for it 

 has neither paxillae nor fascioles, but it does have large, odd, interra- 

 dial marginal plates, a feature found in some of the other genera of 

 the latter family. Probably there should be a special subfamily, Ben- 

 thopectininw, established for it. 



On the other hand, a special subfamily, Pontasterhuv, may well be es- 

 tablished for the genus Potitaster and allied genera, which are evidently 

 closely related to the more typical genera of Archasteridee. 



A very remarakable new genus of this group, and apj)arently closely 

 allied to Pontaster, though it has large papular areas, exists on the 

 Pacific coast. It has the following characters : 



AcANTHARCHASTER, new genus. 



Eays usually five, long, angular, tapered. Disk small; actinal inter- 

 radial plates very few, spinous, confined to the disk; marginal plates 



