550 p. H. CARPENTER ON CRETACEOUS COMATUL^. 



of A. paradooca or A. rugosa. On the other hand I have seen no 

 cirrhus-socket in either of these species which shows any trace of 

 an articular ridge surrounding a central opening. A considerable 

 proportion of each socket (fig. 4) is occupied by the large keyhole- 

 shaped pit. The opening of the axial canal was probably at the 

 bottom of this pit, while there are small lateral elevations causing 

 the constrictions of its sides that may be the disoonnected ends of a 

 transverse articular ridge. But I have been unable to satisfy myself 

 that they are so, while it is unlikely that such an appearance would 

 be always artificial. In any case, the appearance of the sockets, as 

 we find them now, forms a good distinction between A. 'paradoxa 

 and A. rugosa, on the one hand, and A. perforata with the remaining 

 Cretaceous Antedons on the other. The latter species also differs 

 from the two former ones in having six large and distinct openings 

 at the dorsal pole, and not a simple five-rayed impression. The 

 presence of the radial ribs on the walls of the axial cavity gives it 

 an intermediate position between them; for while there are no 

 ribs at all in^. rugosa, there are both interradial and radial ones in 

 A. paradooca. The radial ribs, the plaiting of the basal grooves, 

 the depth of the cirrhus-sockets, and the form of their articular 

 surfaces distinguish A. perforata from A. Lundgreni, to which we 

 will now pass on. 



2. Antedon Lundgreni, n. sp. (PI. XXIII. fig. 3.) 



The centrodorsal is irregularly hemispherical, with a rounded 

 pentagonal outline, and is nearly covered by ten vertical rows of 

 cirrhus-sockets (fig. 3, h, c). Each row is separated from its neigh- 

 bours by more or less distinct ridges, and contains two or three 

 sockets which have striated margins and minute central perforations. 

 The precise shape of the articular surfaces is difficult to make out ; 

 it is most like that to be described immediately in A. striata (fig. 5). 

 Bather to one side of the apex is an irregular hole with constricted 

 sides. The ventral surface (fig. 3, a) is flat with indistinct basal 

 grooves, the sides of which show no traces of plication, which would 

 probably have been preserved had it existed, as the striation of the 

 cirrhus-sockets is visible. 



One of the grooves is occupied b^ a small prismatic basal, the outer 

 end of which reaches the margin of the centrodorsal so as to appear 

 externally (fig. 3, c). From its narrow inner end diverge two small 

 bony bars. One of these, the right in the figure (3, a), is followed 

 by two others, but the rest of the margin of the axial opening is 

 rather irregular. "Were it complete with all the basals in situ, it 

 would evidently be decagonal and closely resemble the corresponding 

 part in Goldfuss's figure of A. paradoxal. Just outside the margin of 

 the opening are five small, but double radial pits. Each of these is 

 very shallow and more or less separated into two parts by a tangen- 

 tial division, just as represented in Goldfuss's figure. The central 

 cavity is tolerably deep, without any ribs on its walls. 

 * Petrefacta Grermauias, Taf. li. fig. 1, b. 



