ASTEROIDEA 187 



the ray the plates usually carry a much smaller, fourth, furrow spinelet. The first 

 adambulacral plate is much shorter than the second, and is wedged between the second 

 plate and the mouth-plate. However, it carries 4 spines, but the inner subambulacral is 

 forced to the furrow margin, so that there are apparently 2 furrow spines (the distal 

 much the smaller). This accounts for the 2 furrow spines of the first plate, as described 

 by Koehler. 



If the actinal integument is gradually removed by means of sodium hypochlorite 

 (strong alcohol being used to arrest the disintegration) it is possible to define clearly the 

 boundaries of the adambulacral plates and their relation to the inferomarginals. The 

 outer end of the adambulacrals overlaps, as a rule, the inner half of the ventral surface of 

 a pair of inferomarginals. This alternation of the plates of the 2 series is not absolutely 

 regular. 



The narrow, obviously specialized, interradial sulcus which this genus has in common 

 with Korethraster is bordered and sometimes concealed by 2 regular rows of paxillae. 

 The groove is paved by 2 series of modified flattish abactinal plates closely in contact 

 and of course devoid of tabulum and spines. These form a sort of fold, the edge of which 

 is turned toward the coelom. They actually constitute a strong interradial arch, super- 

 ficial in position, the lower end of which is joined to the odontophore by a pair of rather 

 elongate flattish plates with rounded ends. These may represent the first inferomarginals. 

 A true interbrachial septum of plates is lacking. The actinostome is wide and flexible, 

 the frame plates being elongate. The gonads are interradial. 



The abactinal plates have four well-developed lobes by which the plates imbricate 

 regularly. In the smaller interspaces there is a single papula; in the larger, i or 2. 



A highly characteristic feature of this genus is to be found in the curious scoop- 

 shaped abactinal spines which form hollow funnel-shaped fascicles on each plate. Each 

 fascicle is enclosed in a common sacculus, concave at the summit of the group. These 

 have been well described by Koehler, who speaks of the longitudinal elements of the 

 spines as "petits piquants elementaires ". This is satisfactory as a figure of speech; but 

 there is no evidence whatever that broadened spines are compound in the sense of being 

 composed of several ankylosed spinelets. Each very small spine is so modified that the 

 elements are spread out in two dimensions, rather than crowded into the three dimensions 

 of the ordinary cylindrical spine. 



This type of spine occurs on the marginal and adambulacral plates of Korethraster 

 but not on the abactinal plates. 



Koehler intimates that the abactinal plates are not regularly arranged, but this idea is 

 due in part to the slightly dishevelled condition of his material. The key to the arrange- 

 ment is the specialized interradial sulcus. The plates are in fairly regular series parallel 

 to this. For each ray therefore the plates are in chevrons, the odd, apical, plate of which 

 is a carinal plate. The first chevron is formed by the plates bordering the sulcus of two 

 neighbouring interradii and the apical plate is the primary radial or first carinal. The 

 5 primary radials are close to the anus (which is covered by 6 or 7 converging blunt 

 calcareous valves or modified spines). Those spines of the paxilla on the side of the plate 



