THE FAMII/T AEBACIAD^. 43 



the third pair, the line of direction being almost vertical, with, as 

 is the case in the suture of the aboral demi-plate, a slight devia- 

 tion towards the median line of the ambulacrum. Thus the 

 sutures of the demi-plates do not unite with any from the central 

 plate so as to cross the internal expansion of the compound plate 

 towards the median line. Assisted by these inner sutural lines, 

 there is no difficulty in observing that the arrangement of the 

 plates externally which combine to form the great tubercle-bear- 

 ing triplets is the same as is seen in the fossil species Coelopleurus 

 Forlesi, d'Archiac. 



The aboral demi-plate comes as far as the groove at the base 

 of the mamelon, at a spot almost directly abactinal to the 

 mamelon. The adoral suture of the demi-plate is in contact with 

 the poriferous part of the central primary plate. 



Again, the adoral demi-plate extends towards the groove at the 

 actinal edge of the mamelon, and its aboral suture is in contact 

 with the adoral edge of the poriferous part of the great plate. 



The suture passes inwards and very slightly aborally up the 

 side of the boss, and it then turns adorally with a rounded angle. 

 The aboral suture starts from the adoral pore of the pair no. 2, 

 and the inner part of it terminates either at the adoral pore of 

 pair 3, or at a spot nearer the median line on the transverse 

 suture between the compound plate and the one immediately 

 below. 



There is no doubt that the poriferous zones of the ambulacra 

 are very narrow, especially in the curved parts. The lines of the 

 -sutures between the zones and the interradia are hardly to be made 

 out externally, but they are very distinct on the inside of the 

 test. There is no rounding-off of the poriferous parts of the 

 plates towards the interradia when they are seen from within, 

 for the lines of the sutures pass direct from the aboral pore of 

 one pair to that of the next below and above (PI. I. hg. 8). 

 The only exception to this is where the line of suture trends 

 towards the interradium between the pairs 1 and 2, to meet the 

 transverse suture of an interradial plate. 



Tlie Amhtclacra hettveen tlie Large Tubercles of the Ambitus and 

 tJie Sittdial Plates. — A well-developed primary tubercle is seen 

 close to the angle made by the pairs of pores below the radial 

 plate, and there are generally three of the same sized tubercles 

 and a larger one above the great tubercles. The distance between 

 the small tubercles increases towards the radial plate, so that the 



