OPHIURANS OF THE PHILIPPINE AND ADJACENT WATERS. 75 



arms in 0. inutilis. In mentioning this latter species in 1911, H. L. 

 Clark added that the identity which I had suspected with O. gratiosa 

 appeared to him probable, but that he did not dare to affirm it ('11, 

 p. 208). 



The study of the very fine series collected by the Albatross at a 

 number of different stations has enabled me to determine a number 

 of variations, affecting the form of the mouth shields, the condition 

 of the under arm plates, which are sometimes in contact for a certain 

 distance along the arms, and sometimes separated from the arm bases 

 outward by a more or less well-marked interval, and the development 

 of the tentacle scales and the covering of the disk. The variations 

 which I have observed have convinced me that a specific differentia- 

 tion can not be based on these characters, and that it is not possible 

 to maintain O. inutilis, which thus becomes from now on a synonym 

 of O. gratiosa. 



Let us review rapidly the variations which may be observed. 



The mouth shields may be very large, almost as long as broad, as in 

 the specimen upon which I established 0. gratiosa, and as I find 

 them in the specimen from station 5300 (pi. 11, fig. 7), which shows 

 a form of mouth shield identical with that which I figured in my 

 drawing of 1896 ('96, pi. 8, fig. 61). These shields may be a little 

 smaller and more elongated, while still remaining of considerable 

 size (specimen from station 5623, pi. 11, fig. 5) ; when their length 

 diminishes while their width remains the same these shields take the 

 form reproduced in figure 3 (station 5622), the proximal angle then 

 becoming very much more obtuse, and may even be replaced by a 

 very convex border. If the length still further diminishes, the mouth 

 shields become very small, twice as broad as long, with the angles 

 usually well marked; this is the condition which I find in a large 

 number of specimens from station 5402 (pi. 11, fig. 4). 



The upper arm plates may be in contact on many segments at the 

 base of the arms (pi. 11, fig. 1, station 5410) or they may be sepa- 

 rated from the base outward (fig. 6, station 5626) ; in the first case 

 the plates are usually very much larger than in the second. The 

 spines which occur on the distal border are sometimes five or six 

 in number, sometimes reduced to two or three only, with considerable 

 variation in their thickness and in their length; these spines are 

 rather easily lost. 



The tentacle scales may be large and broad (figs. 3, 5) or they may 

 be thin and narrow (figs. 4, 7) ; there are sometimes two scales on 

 the tentacular pores of the three first pairs, sometimes either one or 

 two, alternately or irregularly. 



The dorsal surface of the disk is covered with granules, which are 

 sometimes very short (fig. 1). but which may become very much 



