OPHIURANS OF UNITED STATES NATIONAL MUSEUM. 81 



a classification among the Ophiacanthidse, taking as a basis the characters of the 

 mouth papillae only. In a lot coming from one and the same locality one may 

 find, beside such examples as have three oral papillae, the last of which is widened, 

 some others which have a regular row of four subequal papillae and sometimes 

 even five ; this structure might be due to a splitting of the external papilla, as 

 formerly suggested by Duncan and Sladen. But besides the three or four papillae 

 which form a regular row, there may also be seen either one or a variable number 

 of papillae, which are inserted on a different level from the others and are generally 

 smaller than they are. These supplementary papillae often appear at the junction 

 of the oral and adoral plates, and they may vary in number from one to five ; but 

 they may also appear near the terminal tooth papilla, and thus constitute supple- 

 mentary tooth papillse which are almost as much developed as the normal papillse. 

 All these variations are observed on specimens of equal size; they appear to very 

 variable degrees on the mouth angles of the same specimen, and they are absolutely 

 not due to age. So that, according to which specimen is observed, nay, even accord- 

 ing to which of the mouth angles of the same specimen is observed, one is likely to 

 meet the characters which Verrill took as a basis either to maintain the genus 

 Ophiacantha, s. str., or to establish new genera, such as those called by him Ophiec- 

 todia or Ophientodia, the value of which becomes consequently very doubtful. 



One may see, by the illustrations which I reproduce here, and which represent 

 the under face of the disk of two 0. bidentata from Icelandic waters, how much 

 the mouth papillae may differ either by their number or by their arrangement from 

 the type admitted as being normal (pi. 8, figs. 3 and 4). 



OPHIACANTHA ECHINULATA Lyman. 



Ophiacantha echinulata LYMAN (78a), p. 229, pi. 1, figs. 7-9. 



Ophiacantha echinulata LYMAN (83), p. 262. 



Ophioscalus echinulatus VERRILL (99), p. 39. 



Ophioscalus echinulatus VERKILL (99a), pp. 327, 331, 336, and 338. 



Ophiacantha pectinula VERRILL (99o), pp. 325, 338, 340, and 342. 



Ophiacantha echinulata KCEHLER (07), p. 319. 



Albatross station 2117. Jan. 27, 1884. Lat. 15 24' 40" N.; long. 63 31' 

 30" W.; 683 fathoms; yl. m. fne. s.; temp. 39.75 F. One specimen. 



Albatross station 2629. Mar. 8, 1886. Lat. 16 54' N.; long. 75 10' 40" W.; 

 1,169 fathoms; co. s.; temp. 38.4 F. One specimen. 



Albatross station 2651. Nov. 28, 1887. Lat. 16 54' N.; long. 63 12' W.; 

 687 fathoms; wh. oz.; temp. 73.4 F. One specimen. 



The two specimens from stations 2629 and 2651 are in good condition; the 

 diameter of the disk ranges between 5.5 and 6 mm., and the arms are from 40 to 

 45 mm. long. The third specimen is somewhat larger, the diameter of the disk 

 reaching 7 mm., but the arms are incomplete. 



These specimens are not absolutely in agreement with the type which Lyman 

 described and figured in 1878 from a single specimen, and especially they do not 

 offer the bare widened radial shields which that writer indicated, but it must be 

 noticed that, in the year 1883, after having studied many specimens gathered by 

 the Blake in several stations of the West Indies, Lyman wrote (83, p. 262) : "Some- 



