98 BULLETIN 84, UNITED STATES NATIONAL MUSEUM. 



The latter are not protruding and each bears five spines, which remain applied 

 on the lateral faces of the arms. The first under spine, which is cylindrical with a 

 blunt point, is a little longer than the article, but the length does not grow much 

 up to the last one. The following spines are flattened and they may bear on their 

 edges, and more especially on their upper edge, extremely strong and protruding 

 teeth, which are broadly conical and sharp ; these teeth are often few, but then they 

 are generally widely set apart; they are more numerous on the spines of the first 

 articles. Moreover, it seems to me that the spines appear especially flattened and 

 strongly denticulated on the first brachial articles only, while they assume beyond 

 that point a more cylindrical shape at the same time as the teeth disappear, but as 

 I can study their characters only on very short arm pieces, I can not possibly assert 

 anything on the subject. 



The tentacular pores are rounded and extremely large; their diameter exceeds 

 half the length of the article. I have previously said that each of the tentacular 

 oral pores carried on its interradial edge two very long and thick spines; they are, 

 in fact, as long as one and a half brachial articles; on the contrary, the radial edge 

 of these pores is completely unarmed. The following two or three pairs of pores 

 are absolutely deprived of spines and not until the third or fourth pore does appear, 

 near the middle of the external side of the brachial under plate, a pretty thin spine, 

 which is very long, as long in fact as two-thirds of the article; this spine represents 

 the single tentacular scale of each brachial pore. The two spines of the same article 

 are parallel and pretty regularly directed toward the extremity of the arm. I have 

 not succeeded in finding any indication either of a second spine or of a second 

 tentacular scale, although, in spite of their thinness, most of the spines are preserved, 

 and where they have been torn away the small tubercles on which they were inserted 

 can easily be recognized. If there were a second spine, a trace of it could certainly 

 be found, so that I think I am justified in stating that each tentacular brachial pore 

 has a single scale in the shape of a long and thin spine, inserted on the radial edge 

 and lacking on the first articles. 



The color of the sample in alcohol is white. 



Connections and differences. 0. permixta is evidently closely allied to 0. cervi- 

 cornis described by Lyman (83, p. 257) from specimens found by the Blake in the 

 Caribbean Sea, between 208 and 573 fathoms, and which Verrill classified in his 

 genus Ophiopristis (99 a, pp. 336, 337, 338, and 347). But the new species is clearly 

 distinct from the latter, owing to the spines of the tentacular oral pore being 

 only two, owing to there being one spine only instead of two on the radial 

 edge of the brachial tentacular pores, and lastly, owing to the very different 

 shape of the mouth shields, which display a very wide distal lobe. I can not com- 

 pare the two species regarding the disk armature and the number of the brachial 

 spines. In fact, Lyman says ' ' disk densely beset with short rounded pointed spines, ' ' 

 and he indicates five brachial spines (83, p. 258), while Verrill writes (99 a, p. 347) : 

 "Disk and radial shields covered with fine granules and small acute spines," and 

 at the same time he indicates six brachial spines, while in the same work, a few 

 pages before (p. 336), he mentions only five. 



