OPHIUKANS OF THE PHILIPPINE AND ADJACENT WATERS. 175 



two proximal sides are slightly longer than the two others, and they 

 are united by an acute angle, while the distal sides are united by a 

 very broadly rounded angle. The medium-sized adoral plates are 

 narrowed inwardly, but they are in contact in the interradial median 

 line by a narrow border. They have a straight proximal side and a 

 slightly excavated distal side; they are broadened outwardly, and 

 separate the mouth shield from the first side arm plate. The oral 

 plates are triangular, rather low, a little broader than high. The 

 terminal mouth papilla, which is of medium size, is conical, almost 

 as long as broad; the second is very small, conical, and pointed; 

 the third is a little larger and rounded; the fourth, or external, is 

 large, squamiform, broader than long with the free border rounded. 



The upper arm plates are large and cover a large part of the 

 dorsal surface of the arm ; they are almost semicircular, with a very 

 convex proximal border, and a slightly rounded distal border show- 

 ing in the middle a small and very slightly accentuated lobe. These 

 plates are half again as broad as long. They are barely in contact 

 at the base of the arms in the large specimen, and beyond are sepa- 

 rated by a narrow interval; in the smaller they are separated from 

 the bases of the arms outward. 



The first under arm plate is pentagonal, as long as broad, or a 

 little longer than broad, with an obtuse proximal angle. The follow- 

 ing plates are pentagonal, markedly broader than long, with an 

 obtuse proximal angle, straight sides, and an almost straight or 

 very slightly convex distal border; they are separated beyond the 

 second. In the small specimen the second and third plates are still 

 triangular in form, and become pentagonal from the fourth outward. 



The slightly projecting side arm plates carry three poorly devel- 

 oped arm spines; the central spine is a little longer and a little 

 stouter than the two others, its length equalling almost that of a 

 segment. These spines are conical, with the point blunted, and their 

 surface is smooth. 



The two tentacle scales are small, placed side by side, parallel to 

 the distal border of the side arm plate ; the proximal scale is a little 

 longer than the other, which is inserted on the under arm plate, and 

 which is rounded. 



Although Ljungman's description is very short and somewhat 

 lacking in precision, I do not think I am wrong in assigning the two 

 Albatross specimens to his Amphipholis impressa. All the charac- 

 ters which I observe are in perfect accord with those given by that 

 author. I translate the word "impressis," which he applies to the 

 dorsal plates of the disk, by "gravees," and I suppose that Ljung- 

 man intended by this word to call attention to the sinuous lines which 

 give the more or less deep " impressions " on the surface of the plates. 



