AND ITS RELATIONS TO RECENT COMATUL^. 205 



represented on PI. X. figs. 14, 15. On the other hand, the cen- 

 trodorsal may be exceedingly shallow and the exterior of the 

 radials very low, as in the Woodwardian specimen shown in 

 PI. X. fig. 17a, and in some of the forms from the Swiss Jura 

 figured by de Loriol. In tab. 96. figs. 52-55 of his ' Encriniden,' 

 Quenstedt represents four different calices that are all alike in 

 their general features, but differ in minor points, such as the 

 height of the outer surface of the radials and the relative promi- 

 nence of the basals. The centrodorsal is of much the same size 

 and shape in all of them, in no case reaching the length shown in 

 PI. X. fig. 14, while it is never so small as in the original of 

 PI. X. fig. 17 a. The total height of this specimen is 7 millims., 

 that of the radials 5 millims., and its diameter 8 millims. ; while 

 in the other perfect (Woodwardian) specimen already mentioned, 

 with slightly smaller radials, the depth of the centrodorsal is 

 doubled, viz. 4 millims. instead of only 2 millims. 



A. scrobiculata differs considerably from A. costata in the size 

 and disposition of its basals, which Quenstedt* has well described 

 as follows : — " Man findet auf den untern Kelchflachen (tab. 96. 

 fig. 57) fiinf nach den Ecken strahlende Binnen, welche die Un- 

 terseite der Basalia bilden, die sich urn den grossen JN"ahrungs- 

 kanal zur einer Elache ausbreiten und so eine festere Unterlage 

 der Radialglieder bilden. Die Elache ist bald eben (fig. 57), bald 

 ansehnlich vertieft (fig. 58)." A comparison of the side and dorsal 

 views of the radial pentagon, as represented in figs. 18 a and 18 b, 

 gives a very good idea of the basals as prismatic rods, the dorsal 

 surface of which is almost entirely occupied by a groove with 

 plaited sides. But the actual basal pieces themselves are rather 

 wider than these grooves, which does not appear in Quenstedt's 

 description of them, though it is just traceable in his figures. 

 This is seen still more clearly in the basals of Act. cheltonensis (PI. 

 XI. fig. 20 b), which have far more distinctly plaited grooves than 

 those of Ant. scrobiculata. Both these last-mentioned species differ 

 from Ant. costata in the confluence of the inner ends of the basals, 

 so as completely to separate the radials and centrodorsals for 

 some little w r ay round the opening of the central funnel. In 

 Ant. costata, however, the central ends of the basals do not seem 

 to meet one another at allf, while the same might be said of 

 Ant. comjolanata (PI. IX. fig, 9b). 



* Encriniden, p. 179. 



t Encriniden, tab. 96. figs. 29, 44. 



16* 



