126 BULLETIN 100, UNITED STATES NATIONAL MUSEUM. 



were irregularly echinulated and relatively short. But all these 

 characters are those which H. L. Clark attributes to 0. oedidisca. 

 This naturalist notes that between the successive under arm plates 

 there exists a more or less marked depression, and that it is difficult 

 to decide whether the side arm plates do or do not penetrate into 

 this depression. I have determined on certain specimens that the 

 part of the ventral surface of the arms which separates the successive 

 arm plates in the median line is slightly depressed, but on individuals 

 dried and then treated with glycerinated alcohol, I can easily show 

 that the side arm plates penetrate as far as the median line of the 

 arm, along which they are in contact, and that they separate com- 

 pletely the successive under arm plates. This may be readily seen 

 on the specimen of which I give a photograph here (fig. 4) . 



Furthermore, there can not be the least confusion between O. nor- 

 mani and O. relictus. I have included (pi. 9, figs. 5, 6) photographs 

 of a specimen of 0. normani collected by the Albatross, of which 

 I spoke in 1904 (Kcehler '04, p. 107), and it may be seen that the 

 two species differ absolutely from each other in the characters of the 

 arm spines and in the form of the mouth shields, as well as in that 

 of the under arm plates; the disk of O. normani is flattened, while 

 it is thick, with a very convex dorsal surface, in O. relictus, as the 

 photograph reproduced in figure 2 on plate 9 shows. 



Ophiophthalmus relictus was found by the Siboga at different 

 stations between 0-10 S. latitude and 116-131 E. longitude, at 

 depths varying between 538 and 1,624 meters (310 to 887 fathoms). 



H. L. Clark's Ophiacantha oedidisca was dredged in the Japanese 

 seas between 405 and 578 fathoms. 



The preceding lines were written in 1914, and consequently a 

 long time before the appearance of Matsumoto's memoir; I believe 

 there is no necessity for the least modification of them, and I still 

 consider Ophiacantha oedidisca a synonym of 0. relictus. In his 

 memoir of 1917 Matsumoto not only maintains the first as a dis- 

 tinct species, but he even places it in a different and new genus to 

 which he gives the name of Ophiosemnotes and of which the type 

 is H. L. Clark's Ophiolebes tylota. Besides these two species, 

 Matsumoto assigns to the genus OpJiwsemnotes four others placed 

 by H. L. Clark in the genus Ophiolebes 0. pachybactra, O. dia- 

 phora, 0. paucispina, and 0. ~brevispina and in addition Ophiactis 

 clavigera Ljungman, placed by Lyman in the genus Ophiolebes. 



In mentioning 0. oedidisca, Matsumoto ('17, p. 137) adds that al- 

 though this species resembles certain forms of the genus Ophioph- 

 thalmus, it differs from them in its disk, which is very convex. 

 But we have seen that one of the principal characters of O. relictus, 

 and one which I have emphasized, is this same height of the disk. 

 H. L. Clark has already relied on this same character as well as 



