OPHIURANS OF THE PHILIPPINE AND ADJACENT WATERS. 157 



three times as long as broad; the external, inserted on the adoral 

 plate, is conical with the point blunted, and is not very well devel- 

 oped. 



The upper arm plates are extremely broad, two and a half times as 

 broad as long, triangular, with a very obtuse and rounded proximal 

 angle bordered by straight sides which pass over very rounded angles 

 to the convex distal border. The side arm plates penetrate between 

 these last as far as the neighborhood of the median line, but the 

 upper arm plates remain in contact by the apex of their proximal 

 angle. 



The first under arm plate is very small, trapezoidal, very much 

 longer than broad, narrowing in its distal region which is inclosed 

 between the corresponding adoral plates. The following plates are 

 pentagonal, a little broader than long, with a truncated and very 

 rounded proximal angle, and the sides and distal border gently con- 

 cave. These plates are in contact. 



The side arm plates, which are only slightly projecting, carry five 

 spines on the segments immediately following the disk, but this num- 

 ber very soon falls to four; these spines are subequal, cylindrical, 

 with the tip rounded, and their surface is rugose ; their length almost 

 equals that of the segment. 



The single tentacle scale is small, and only slightly longer than 

 broad, with the tip rounded. A few pores of the first pair sometimes 

 show two scales in contact inserted side by side on the side arm plate. 



The color of the specimen in alcohol is grayish white. 



Affinities and distinctive features. Amphiura dami-ssa is easily 

 distinguished from all the other species of the genus Amphiura (in 

 the restricted sense) with both surfaces of the disk provided with 

 scales and with only a single tentacle scale. Its central mouth papilla, 

 which is low and broadened, recalls the form seen in A. capensis 

 Lyman, but except for this character there is not the slightest re- 

 semblance between the two species. 



If the description which I have just given be compared with that 

 of A. fusco-alba published by Brock, numerous points of resemblance 

 will be noticed, but it does not seem to me possible to confuse these 

 two species, which, moreover, come from very different depths. Brock 

 says that the radial shields of A. fusco-alba are very large and that 

 their length equals half the radius of the disk, that the plates of the 

 dorsal surface of the disk are grouped regularly around a central 

 plate, while the plates of the ventral surface are extremely small, and, 

 as they are rather elevated, they take on the form of granules; the 

 internal mouth papilla is very large ; the mouth shields are triangu- 

 lar, with an external lobe which causes them to appear cordif orm ; the 

 upper arm plates are twice as broad as long and are separated, and 



