ME. P. H. CARPENTER ON THE GENUS ACTIXOMETRA. 99 



radial furrows (fig. 4, d.i.f), the rays of the hasal star are short and flattened, and do not 

 hy any means reach the angles of the radial pentagon (fig. 4, S). Their dorsal surface is 

 somewhat depressed along the median line, while the depression (fig. 6, s) at the central 

 end of the ventral surface which receives the Mind end of the axial interradial canal is 

 continued outwards in a peripheral direction somewhat further than in Act. pectlnata 

 (PI. V. fig. 8 a, s); and the median ridge which runs from its end to the apes of the ray 

 is less marked than in this species. 



In another specimen of the type, however, in which hoth hasal grooves (PL VI. 

 fig. 8, b.g) and hasal folds (fig. 9, bf) are wide and well marked, the basal rays are 

 stout and thick, with a fairly distinct median ridge on their ventral surface. In fig. 8, 

 three of them are seen occupying their normal position in the hasal furrows on the 

 ventral surface of the centrodorsal piece, with which they are closely connected, while 

 the other two rays have remained in connexion with the rosette and radial pentagon. 



In a third specimen of the type the ray S of the hasal star are very imperfectly 

 developed ; two only extend for any distance towards the angles of the radial pentagon 

 (PL VI. fig. 11, <$'), while of the other three little or nothing is to he seen. In this 

 specimen, as in the one first described, there are no basal folds, and the basal grooves 

 are parallel-sided and only imperfectly developed (fig. 10, b.g). It is also remarkable 

 for the fact that the absorption of the apex and outer margins of each embryonic basal 

 plate seems to have been very incomplete ; for the bars of the basal bridge are so wide, 

 and extend so far towards the centre from the inner margins of the dorsal surfaces of the 

 first radials, with which they are closely united, that they entirely conceal the apertures 

 in the compound basals (PL V. figs. 3, 8; PL VI. figs. 1, 6, 13, 18, 19, 22, 21, x u x,, y x , 

 y 2 , &c.) through which the secondary basal cords pass in order to reach the central 

 canals of the first radials. Consequently nothing is seen of the rosette in a dorsal view 

 of the pentagonal base but its central opening surrounded by a raised rim (PL VI. 

 fig. 11, r.o). In all the other figures, however (PL V. fig. 3 ; PL VI. figs. 1, 13, 21), these 

 apertures are large and distinct, every one being situated between a radial (p) and an 

 interradial process (o) of the rosette. 



In Act. polymorpha, var. 1, both basal grooves (PL VI. fig. 15, b.g) and basal folds 

 (fig. 13, b.f) are well marked and somewhat lancet-shaped in form; the rays of the 

 basal star wdrich occupy the former are much flattened dorsally (fig. 22 b, S), as in one 

 of the specimens of the type (fig. 1, S). They are not, however, so short as in this case, 

 but, like the basal folds at their sides, reach the outer angles of the radial pentagon. 



This is also the case in the other three varieties. In var. 4 the basal folds diverge 

 considerably at about the middle of their course (fig. 21, b.f), so that the dorsal inter- 

 radial furrow is here very wide, and then rapidly narrows towards its peripheral end. 

 In correspondence with this, the basal rays also widen somewhat from their narrow 

 central ends, and then begin to decrease in width as they approach the angles of the 

 radial pentagon (fig. 24, S) ; they are also marked by a slight median furrow along their 

 dorsal surface, as is the case in one specimen of the type (fig. 4, S). In varieties 2 and 3, 

 as in the type, and in var. 1, the basal rays are widest at their central ends (figs. 18, 

 19, 22, S). In both cases the basal grooves (figs. 17, 21, b.g) and basal folds (PL VII. 



SECOND SERIES, — ZOOLOGY VOL. II. 11 



