ASTEROIDEA 225 



apart to accommodate an unusual development of muscles {macroccphaliis Koehler) ; 

 together with a less modified form (pirocephafus), and conventional crossed pedicellariae. 

 Dorsal skeleton and spines essentially as in less modified Coscinasteriinae, consisting of 

 large, cruciform typically monacanthid carinals (alternating sometimes with secondary 

 smaller plates) which connect with similar superomarginals by short transverse arcs of a 

 few plates ; marginals large, cruciform, monacanthid ; i series of monacanthid actinals ; 

 papulae in 2 dorsolateral, i intermarginal, and, in one species, i actinal series ; adam- 

 bulacral plates predominantly diplacanthid, or monacanthid, or a combination of the 

 two, the spines without attached pedicellariae; major pedicellariae attached to spine 

 sheath at base of spine, or aggregated in wreaths on collar-like thickenings of the sheath, 

 or scattered over dorsal surface ; gonads in one species (armata) open ventrally. 



Remarks. In the light of the Discovery material it is not possible to maintain Aut- 

 osterias on the basis of the number of adambulacral spines. Its type species {pedicellaris) 

 is probably immature. Aiitasterias bongraini, as shown below, is not exclusively mon- 

 acanthid ; and it more closely resembles the diplacanthid Notasterias stolophora than it 

 does the type of Autasterias. Both Autasterias pedicellaris and Notasterias haswelli have 

 the major pedicellariae attached to plates rather than to the spine sheaths, as have the 

 other species. The one has predominantly monacanthid adambulacrals, the other 

 diplacanthid. It is not at all clear that these two constitute a natural generic group. 



There is nothing in the structure of the skeleton and spines which justifies the removal 

 of Notasterias from the Coscinasteriinae. The shape of the plates (Fig. K, i) and 

 distribution of the single spines are as in typical Coscinasteriinae — even to occurrence 

 of spines on alternate superomarginals and carinals. The macrocephalous pedicellariae 

 (which are sometimes absent) have not diverged further from the normal crossed type 

 than have the peculiar falcate straight pedicellariae of Pisaster from the conventional 

 straight pedicellaria. I have therefore placed the Notasteriinae in the synonymy of the 

 genus. 



Notasterias armata Koehler 



(Fig. J, za, 4-4^*; Fig. K, i) 



Notasterias armata Koehler, 191 1, p. 39, pi. 5, figs. 6-1 1 ; pi. 6, figs. 1-8. — 1912, p. 25, pi. i, fig. 8. — 

 1920, p. 58, pi. 7, figs. 7-9; pi. 8, figs. 2-8; pi. 9, figs. 1-6; pi. 10, figs. 1-9; pi. 60, fig. i; pi. 61, 

 fig. I. — Fisher, 1930, p. 244. 

 St. 1658. Ross Sea, 76° 09-6' S, 168° 40' E, 520 m., i specimen. 

 St. 1660. Ross Sea, 74° 46-4' S, 178° 23-4' E, 351 m., 3 specimens. 



Koehler has described and figured this species in great detail. It is remarkable for the 

 large size of the macrocephalous pedicellariae, of which i, and rather seldom 2 or 3, are 

 attached to the base of the carinal and superomarginal spines (R 66 mm., St. 1658). The 

 inferomarginal pedicellariae are much smaller and simpler. Two or three are attached to the 

 spine sheath above the base, while on the actinal spines similar ones are at about mid- 

 height. It is probable that even the giant pedicellariae can be elevated to end of the 

 spine. The largest macrocephalous pedicellariae measure 3 mm. in length and 2 mm. 

 broad at base. I find no minor crossed pedicellariae, nor spatulate straight pedicellariae. 



