28 BULLETIN 84, UNITED STATES NATIONAL MUSEUM. 



This plate is as wide as long or a little longer than wide, and is placed close to the 

 first one, though not touching it. On each side, and nearer the proximal angle 

 than the- distal border, is a slightly developed pore covered with a small rounded 

 scale. The third brachial under plate is very small, triangular, and separated from 

 the foregoing by an interval which is equal to about half its length; on either side 

 of that plate is to be seen a very small pore provided with a rudimental scale, but 

 this pore is generally located on the corresponding lateral brachial plate. Further 

 on, there are no more brachial ventral plates. 



The lateral brachial plates alone take up all the inner faces of the arms. They 

 are slightly protruding on the sides, and the successive pairs are separated by a 

 somewhat sinuous furrow, both on the upper and on the under face. Each of them 

 carries five sharp conical spines which are relatively long enough for an Ophio- 

 musium. 



The granules of the upper face of the disk pass uninterruptedly to the upper 

 part of the lateral plates as well as to their sides, but these gramiles grow smaller 

 as they are farther from the basis of the arms and more so on the larger specimen ; 

 they are completely wanting on the under part of the arms. 



Connections and differences. 0. rugosum is allied chiefly to 0. granulosum, 

 Lyman and to 0. relictum Koshler. 0. granulosum, which was taken by the 

 Challenger at a depth of 1,875 fathoms (lat. 34 N.; long. 140" E.), is remarkable 

 owing to the very minute granules which cover the disk plates on both faces and 

 which, instead of being few in number and rounded, are very dense and even grow 

 into little spines on the margins of the plates, including the mouth shields. The 

 oral papillae are altogether indistinct; moreover, the shape of the mouth shields 

 and of the genital plates, as well as the arrangement of the upper plates of the disk, 

 is different from what is observed on 0. rugosum. 



0. relictum, which was dragged up by the Siboga from a depth of 469 meters 

 (lat. 29' S.; long. 130 E.), offers, through the disposition of the upper plates 

 of the disk, a very great likeness to 0. rugosum, but these plates have no tubercles 

 except near the outline of the disk and their margins are thicker, while they are 

 uniformly granulous in 0. rugosum, the granulations passing even from the upper 

 face of the disk over to the dorsal and lateral parts of the lateral brachial plates, 

 at least on a certain length of the arms, which gives to the new species a very different 

 appearance. The brachial spines are also more developed in 0. rugosum than in 

 0. relictum. 



OPHIOMUSIUM SCULPTUM Verrill. 



Plate 1, fig. 9. 

 Ophiormisium sculptum VERRILL (99), p. 16, pi. 2, fig. 2; pi. 8, fig. 2. 



Blake. Two miles east of Havana; 200 fathoms. One specimen. 



The species has been described by Verrill after a specimen from Havana which 

 had been dredged between 110 and 260 fathoms. 



In the example which was handed to me, the diameter of the disk is 7.5 mm., 

 and the arrangement of the plates of the upper face is not quite in accordance with 

 VerruTs description and drawings; still, I believe it is one and the same species, 



