KETOKT ON THE ASTEROIDEA. 307 



ences from that typo presented by the example under notice are considerable, and scarcely 

 fall within the range of what might be expected in the young phases of the species referred 

 to, I have considered it the safest course to regard the form in question as a distinct 

 species, at least until other specimens are obtained which may throw light upon the 

 relationship of these nearly allied forms. The following characters may be noticed in 

 comparison with those of Nymphaster protentus as given in the foregoing description. 



In the armature of the adambulacral plates, the furrow series accord in their manner of 

 disposition with those of Nymphaster protentus, but are not compressed ; the number on the 

 plate is also rather less, five or six being present on those near the mouth, and eight or nine 

 about the middle of the ray. The adambulacral plates are comparatively broader, espe- 

 cially near the mouth, and there is no external series of spinelets as in Nymphaster protentus, 

 their place being occupied by a row of granules similar in all respects to those which stand 

 on the outer part of the plate. There are thus three longitudinal series of equal-sized 

 granules behind the furrow series on the inner half of the ray (reduced to two near the 

 middle), and there are usually five or six granules in each series. In harmony with these 

 modifications in the appendages of the adambulacral plates, the armature of the mouth- 

 plates is less echinulate than in the larger form Nymphaster protentus. 



Within the abactinal paxillar area, the plates of the median radial line, and the two 

 parallel series on each side are oval or subcircular in outline instead of being hexagonal ; 

 this shape at least being clearly defined by the marginal series of granules ; the latter are 

 semiglobular and surround a central group of five or six granules on the middle of the 

 tabulum. This small number forms a marked contrast to the closely crowded granulation 

 of the plates in Nymphaster protentus; and, what is still more striking, the granules 

 themselves are comparatively larger in Nymphaster albidus, which is the smaller 

 form. 



The supero-marginal plates in the inner part of the interbrachial arc are very tumid on 

 the lateral wall of the disk, and extend beyond, and appear to overhang, the infero-marginal 

 series. When the lateral wall of the disk is placed in the direct line of view, the line of 

 suture which separates the superior and inferior series of marginal plates is seen to take a 

 bold and well-defined curve, with the convexity downwards, in the inner part of the arc, 

 — the supero-marginal plates thus encroaching on the inferior series, the height of the 

 former being considerably greater. This formation does not, however, affect the general 

 thickness of the marginal wall taken as a whole. 



On the actinal area of the disk there is a considerable amount of inflation and the surface 

 is convex ; there is also a slight depression of triangular form on the outer part of the 

 actinal interradial area in the neighbourhood of the interradial line and bounded by the 

 inner margin of the infero-marginal plates. Both the superior and the inferior series of 

 marginal plates along the ray have their greatest dimension in the length, whilst in Nym- 

 phaster protentus above described the supero-marginal plates are broader than long until 



