424 PEOF. p. M. DUNCAN ON THE STPvTJCTTJEE OF 



Geological Survey, Jermyn Street, and in most the number of single 

 primary plates near the apical end is very striking, indeed almost 

 leading to the idea that all the parts of the ambulacrum are made up 

 of them. But there are one or two compound plates preserved. In 

 one of these, near the peristome, the triple nature is evident, and the 

 line of suture between the middle and adoral plates is curved, and 

 passes from the adoral pore of the middle pair towards the mamelon 

 of the tubercle and thence to the median or vertical suture, the con- 

 vexity of the curve being directed apically. In another specimen the 

 line of suture between the middle and the aboral plate is seen, and 

 it is curved, with the convexity towards the mamelon. Hence these 

 compound plates are on the Diadematoid type, and it may be presumed 

 from the phenomena presented by the recent species of Diadema that 

 growth-pressure has changed the shape of the original primaries. 

 In one of the specimens in the Museum there are four plates and four 

 pairs of pores in one of the compound plates, and it appears, but 

 not very satisfactorily, that the additional plate is a low primary. 

 This is not without its significance ; for a similar structure is seen 

 in allied genera *. 



Hemipedina marehamensis, Wright, from the Coral Hag, is a fine 

 form belonging to a section of the genus which has numerous 

 primary tubercles in rows on the interradia. 



There is a specimen in the British Museum (no. 75923) which 

 shows the shape of the triplet of plates which combine to form a 

 geometrical plate near the ambitus. The compound plate is broader 

 than high, and there is a space between the tubercle and the median 

 or vertical suture (fig. 3). The direction of the sutures between the 



Pig. 3 (see p. 451). 



three plates indicates their shape, especially as the whole compound 

 plate is contained between an aboral and an adoral transverse 

 suture. 



The pores are rather oblique, and the adoral pore of the first pair 

 is on the suture which unites the first and middle plates. The 

 line of this suture is, from the interradium to the adoral pore, and 

 thence with a curve, convex adorally, up the flank and over the top 

 of the boss of the tubercle apically to the groove at the base of the 

 mamelon, and then down the slope to the edge of the boss to reach 

 the median suture at a short distance from the aboral and inner 

 angle of the compound plate. 



This first plate is therefore a low primary resembling the corre- 



* With regard to Hemipedina BowerhanTcii, Wright (op. cit. p. 145), illus- 

 trated on plate ix. of the Monograph already noticed, it must be observed that 

 the figure 2 b cannot be correct. It represents the pores as if they were turned 

 upside down, and the adoral pore, or that which is furthest from the inter- 

 radium, as aboral to the other in position. 



