OPHIURANS OF THE PHILIPPINE AND ADJACENT WATERS. 93 



broader than long, with a rather open proximal angle which is only 

 slightly projecting, and the four sides are almost equal. The proxi- 

 mal angle is truncated for about two-thirds of the length of the 

 arm, becoming sharp in the last third. The plates are then as long 

 as broad, and even, in the terminal portion of the arms, a little 

 longer than broad. They are all in contact. 



The first under arm plate is quadrangular, with a narrow distal 

 border, a broader proximal border, and diverging sides ; it is a little 

 broader than long. The following plates are quadrangular, with a 

 straight proximal border, and the distal border slightly notched in 

 the middle; the sides are slightly excavated by the tentacle pores. 

 These plates are at first as long as broad, then becoming longer than 

 broad beyond the disk, at the same time becoming separated by a 

 narrow interval occupied by the side arm plates; toward the ex- 

 tremity of the arms they are at -least half again as long as broad. 



The side arm plates bear six spines at the base of the arms, and 

 this number is continued for a considerable distance. The first ven- 

 tral spine exceeds a segment and a half, and the length increases to 

 the last dorsal spine which is as long as three segments. These spines 

 are slender, cylindrical, pointed, and smooth. 



I can not discover the least trace of a tentacle scale on the two 

 specimens from station 5450, which are in good condition; but on 

 the third, in which the dorsal surface of the disk is torn away, a 

 certain number of the pores of the eight or nine first segments are 

 provided with a small, short, spiniform and pointed scale. 



The color of the disk is a dark green, while the arms are white. 



The two specimens from station 5460 are very interesting, because 

 they resemble more than the preceding the type collected by the 

 Siboga. One of them, which I have shown in figures 1 and 3 on 

 plate 20, has the dorsal surface of the disk in a great measure torn 

 away; the two surfaces of the disk show very distinct plates, and 

 furthermore there is a tentacle scale which occurs in an inconstant 

 and irregular manner as far as the pores of the tenth or twelfth pair. 

 These scales are very small, slender, conical, and pointed, and they 

 must be very easily lost. The mouth shields are broader than in the 

 two specimens from station 5450, and in their shape they strongly 

 recall that of the Siboga specimen, with a small lobe on the middle of 

 their distal edge ; the proximal angle is so rounded on some of these 

 shields that it grades into the lateral borders, and the shield finally 

 acquires a semicircular shape. The upper arm plates of the two or 

 three first segments are at first lozenge-shaped, as in the specimens 

 from station 5450 ; they then rapidly elongate, and become triangular 

 as in the Siboga specimen, with this difference, however, that they 

 are a little longer than broad, while in the latter they are a little 



