GOLD MADREPORE. 403 



If any additional evidence were wanting to sliow 

 that this species approaches much nearer the Actiniae 

 than C. Smith a does, it would he found in the stony 

 skeleton. This is very difierent in appearance from 

 that of the kindred species, and is manifestly rudi- 

 mentary. When the soft parts have heen carefully 

 removed by several days' maceration in fresh water, 

 and the gelatinous matter all cleared away fi'om the 

 stony plates by a slender stream of water allowed to run 

 upon it from a height, a vertical view shows the following 

 arrangement : — First, at the very margin there is a 

 narrow circle of white calcareous plates, small and 

 very irregularly anastomosing, so as to resemble in 

 miniature the honev-combed limestone rock that we 

 find around Torquay and elsewhere. In the centre of 

 the cavity, there is another loose spongy mass of 

 similar irregular plates. Eighteen perpendicular radi- 

 ating plates extend between the marginal circle and 

 the central mass, arranged in six threes, so as to make 

 a six-raved star. The order of each trine series is as 

 follows : the middle one is the thickest and shortest, 

 reaching scarcely more than half-way from the cir- 

 cumference to the centre. On each side of this there 

 is a longer thinner plate, neither parallel nor converg- 

 ing towards the centre, but diverging at a small angle, 

 so that each of these lateral plates meets the lateral 

 plate of the next trine series, at a point consider- 

 ably short of the centre, whence a plate sometimes 

 goes to the central mass. The arrangement will be 

 better understood by a reference to Plate XXVI, fig. 

 6, which represents a quadrant of the circle, much 

 magnified. 



