130 NEW YORK STATE MUSEUM 



circular in outline, with a proximal concavity. The holotype, 

 however, (plate 7, figure i, interradius c) shows a very different 

 outline and again arouses a suspicion that the plesiotype is perhaps 

 specifically distinct from the holotype. The radial diameter here is 

 1.2 times the transverse diameter. 



Of these plates, Schuchert says, in his description of generic 

 characters in Urasterella, 1915, page 174, " In none of the mature 

 specimens have been seen well-developed or larger axillary infra- 

 marginals or interbrachial marginal plates." From the great thick- 

 ening of the first floor plates in two of the specimens under study, 

 it would seem reasonable to assign an approach to maturity in 

 the specimens here under study. 



If the axillary inframarginals were spine-bearing, this spine 

 must have arisen from the oral surface. In plate 10, we find what 

 appears to be a cross section of such a spine. It lies between the 

 second and third pairs of covering plates of interradius a. 



In turning now to the arm marginals we shall describe first of 

 all the new revelation which U . medusa makes concerning the 

 "articular spines and probable paxillae " which Schuchert (1915, 

 page 298) feels he must credit to Urasterella grandis 

 (Meek) though, following a universally accepted opinion, under 

 his definition of paxillae (191 5, page 16) he says " None are known 

 in Paleozoic genera." In U . medusa, however, we have forms 

 preserved which are strikingly like the paxillae of Hymenaster 

 p e r i s s oxi o t u s Fisher (1911, plate 115, figure la), only the 

 plate bases are not so deeply incised and the pedicels are more 

 nearly perpendicular to these bases. In fact,- the three long and 

 slender articulated spinelets borne on the projective pedicels of the 

 inframarginals of U. medusa and the certainty that all the 

 apical plates bore similar pedicels with similar spinelets, almost 

 leads one to infer that this ancient species possessed a nidamental 

 membrane. The paxillae we are about to describe are without 

 doubt homologous with those of recent forms, but they show 

 peculiarities well worth careful study. 



We will first examine plate 4. The fourth marginal of arm B 

 shows two of its spinelets turned distally at an angle of a little 

 more than 45 degrees with the axis of the pedicel. The upper 

 of the two spinelets is slightly separated from the pedicel and 

 shows its concave articular face. Both spinelets are longer than 

 the pedicel and plate thickness combined. The head of the same 

 pedicel shows an articular face from which a spinelet has been 

 lost. The seventh marginal of the same arm shows two spinelets 



