104 AVYVILLE THOMSON, ON ASTERACANTUION VIOLACEUS. 



tisli, tliough now only about ouce and a half the size of the 

 peduncle, has asserted distinctly its echinoderm character. 



The angles of the pentagon project still further, foiining 

 the rudimentary rays. The transparent external layer 

 becomes thicker, and its scattered oil-cells and endoplasts 

 more numerous and distinct. The inner organized granular 

 layer increases in thickness till only a small central space is 

 filled with granular fluid, while between it and the external 

 layer, or in the external layer itself, small plates of the cha- 

 racteristic calcified areolar tissue are irx*egularly scattered. 

 The star-like form now becomes still more distinct, a regular 

 series of calcareous plates are developed on the dorsal surface, 

 one large and rapidly expanding plate at the end of each ray, 

 and a smaller one at each of the re-entering angles. On the 

 lower surface, a pair of plates, each with a concave edge 

 towards the point of the ray, and a convex one towards the 

 centre of the star, are formed at the base of each arm, so that 

 the two plates of a pair unite in the centre of the ray, while 

 their free ends meet the free ends of the adjacent plates of 

 the next pairs, forming a calcareous inter-radial angle, pro- 

 jecting into the central space. These plates are rapidly 

 followed by a double row of almost linear plates with double 

 concave edges ; which extends towards the point of the ray, 

 leaving between every two pairs, two opposite apertures for 

 the passage of the pedal vesicles. 



While these plates are being developed, a tubercle appears 

 on the oral surface at the base of each arm, and a delicate 

 circular vessel forms a slightly raised ring round the centre. 

 This ring, in one part of its course, passes under or blends 

 with and is lost in the base of the peduncle. 



The tubercle at the base of the ray now takes a crescentic 

 form, and shortlv the crescent resolves itself into three 

 tubercles, two opposite and occupying either side of the 

 median line of the ray, the other in the centre of the ray and 

 connected with the circular ring by a delicate straight tube. 

 This central tubercle next becomes slightly crescentic, and 

 resolves itself into three tubercles, which arrange themselves 

 like the first three, and in this way a central vessel, proceeding 

 from the ring, follows the development of each ray, with a 

 row of tubercles on either side. These tubercles are shortly 

 developed into suckers like those of the tubular feet of the 

 peduncle, and supported by precisely similar transparent 

 contractile tubes, filled Avith the same fluid, in which chyle- 

 globules revolve and circulate in exactly the same way. 

 During these changes the peduncle remains unaltered. The 

 embryo stands upon its three feet like a miniature three- 



