NARCOMEDUSAE 8i 



Almost all the specimens of medium size have 14-16 tentacles. The only specimen with 10 tentacles 

 was taken at St. 1371 ; it is 31 mm. in diameter, and besides the ten fully developed tentacles a tiny 

 one is situated in the middle of one of the marginal lappets; two other specimens from the same 

 station, both 28 mm. wide, have about 1 1 tentacles. The large specimen, 43 mm. wide with 16 tentacles, 

 was taken at St. 2059. As a rule the rectangular gastric pouches are as long as broad or a little longer. 

 In the only two specimens with more than 16 tentacles the number is 20, and their gastric pouches 

 are comparatively short and broad and of somewhat different size, some of them almost twice as broad 

 as long, and the corners are slightly rounded. They are, however, quite different from the pouches in 

 S. incisa, and in spite of the large number of tentacles in these two specimens I refer them without 

 doubt to S. marshaUi. They were taken at Stns 284 and 1606, in hauls 300-0 m. and 190 o m. respec- 

 tively. 



Distribution. Most of the previous records of this species are from the tropical Pacific, but under 

 the name Solmaris flavescens it was recorded by Vanhoffen (1908, p. 58) from several localities in the 

 tropical Indian and Atlantic Oceans and probably again, under the same name, in 1912a (p. 394) 

 from three localities off the west coast of Africa, between the Cape Verde Islands and 20° S. The 

 record by Ranson (1936, p. 208) from the Mediterranean, near the Balearic Isles, is pecuHar, and one 

 cannot but suppose that the only specimen belonged to S. albescens. Ranson, it is true, describes 

 the gastric pouches as more pronouncedly rectangular than in S. albescens; the pentagonal outline 

 generally seen in this latter species is, however, not quite constant, but may sometimes approach the 

 rectangular shape of the pouches in S. marshalli. 



Genus Cunina Eschscholtz 



Cuninidae with otoporpae; with or without peripheral canal system. 



Attempts at a revision of the numerous species of Cunina (incl. Cunoctantha) were made by Bigelow 

 (1909, pp. 51 and 55; 1918, p. 392) and Thiel (1936, pp. 77 and 82). Within the species with more 

 than eight tentacles {Cunina sensu stricto), Bigelow recognized 5 species: globosa Eschscholtz 1829, 

 lativentris Gegenbaur 1856, proboscidea MetschnikoflF and Metschnikoff 1871, duplicata Maas 1893, 

 and peregrina Bigelow 1909. I have not seen lativentris and proboscidea myself, but the three others 

 are represented in the Discovery collections. To Bigelow's revision I shall only add the fact that a 

 gelatinous peduncle, which is emphasized as characteristic of proboscidea, was also found in the present 

 specimens oi globosa, but since this species has a well-developed peripheral canal system, whereas the 

 canal system is degenerate in proboscidea, these two species are evidently distinct. Bigelow is inclined 

 to regard C. duplicata Maas as identical with lativentris; the present collection, however, contains 

 several specimens of a Cunina which I believe must be referred to duplicata, for it is impossible to 

 identify them with lativentris as described and figured by Gegenbaur. On the other hand, there is 

 the possibility that lativentris is the same as globosa (see below). Provisionally I think we should 

 retain the five species mentioned by Bigelow but no more of this group of Cunina, abandoning 

 mucilaginosa Chamisso & Eysenhardt 1821, campanulata Eschscholtz 1829, vitrea Gegenbaur 1856, 

 and oligotis Haeckel 1879. According to Bigelow (1914, p. 33), C. discoides Fewkes 1881 b is probably 

 a young Solmissus. Cunina prolifera Gegenbaur and rhododactyla Haeckel are identical with rubiginosa 

 KoUiker and should be transferred to the genus Pegantha (see above). I entirely disagree with Thiel 

 (1936) who would reduce the number of species of Cunina to one or two {mucilaginosa and rubiginosa). 



Thiel retains the genus Cunoctantha, but unites all the species under the name octonaria. I thmk 

 we may state that there are four distinct species, and owing to their variable number of antimeres 

 (usually 8, but varying from 7 to 10 or 11) I refer them all to Cunina. C. octonaria McCrady 1857 

 (including kollikeri Miiller 1861 and parasitica Metschnikoff 1881) has no peripheral canals; its 



