Hill/.' IST< iM.K CKPHKA. 651 



There are no scapulets upon tin- mouth-arms. The radial-muscles are powerfully and the 

 ring muscles weakly developed. A description ot rlu- genera follows: 



(.> l>lii-<i PKKON AM> List M K, iSou. Exumbrella with a central area bearing wart-shaped projections. 



I'.ut \lmln ,i I.. \' i/, 1862. Exumbrella with a smooth central dome without wart-like elevations. Radial-canals 



of tin- 1-11 ;ill similar each to each. 

 Polyrliiza L. AGASSI/-, 1862. Exumbrella with a central concavity and with radiating furrows. 



Genus CEPHEA Pe>on and Lesueur, 1809. 



Cefhra, I'tRu.N > i l.tM-t.uR, 1809, Annal du Mus. Hist. Nat., tome 14, p. 360. ESCHSCHOLTZ, 1829, Sys,t. dc-r Acalcphcn, p. 55. 

 I.ISSON, 1843, Hist. Zooph. Ai-al., p. 410. AGASSIZ, 1862, Cont. Nat. Hist. U. S., vol. 4, p. 155. -H*i< kM. 1880, 

 Svit. der Medusen, p. ^73. --VAMIOH i N, 1888, Bibliotheca Zoologica, Heft. 3, p. 39; 1902, \Visscn. Ergcb. dcutsch. 

 Tiefsee ExpeJ. I'aldhia, BJ. 3, Lfg. i, p. 45. 



Ccf>hca+ Netrostoma, MAAS, 1905, Scypliomedusen der Siboga Expect., Monog. 1 1, pp. 31, 32, 35, 81, 89. 



\etrosloma, SCHUI.TZE, L. S., 1898, Denkschr. Med. Nat. Gesell. Jena., Bd. 8, p. 4^7. 



Micrtstylus+Ptrirhixa, KIMIINIII-VK, 1902, Journal College Sci. Tokyo, Japan, vol. 17, Art. 7, pp. 11, 13. 



Xtvlorlih,!, HAM k'H,, 1880, Syst. der Mcilus,en, p. 6l2. 



Halipetasus, SCHULTZK, 1898, Uenksclir. Med. N.it. Gesell. Jena, Bd. 8, p. 458. 



The oldest known species is "Medusa octost\ln" of Forskal. and this may serve as the 

 i \ pi- ot tin- genus Cephea. 



GENERIC CHARACTERS. 



Rliizrist'initita Jiehit'jtiin in which the 8 mouth-arms tork once dichotomously and each 

 fork gives rise to short dichotomous or dendritic branches. Solid, wart-shaped tubercles at 

 the center of the exumbrella. The central stomach gives rise to 8 rhopalar and numerous 

 inter-rhopalai radial-canals, all of which connect with a network of anastomosing vessels in 

 a wide zone near the margin. Rhopalia without ocelli and without exumbrella, sensory pits. 

 There is no definite ring-canal. Development unknown. 



The described species of <,Y/>//,w are all found in the tropical Indian Ocean and Pacific 

 region. Cephea cephea (Medusa ccplica, Forskal) is apparently widely distributed over the 

 Indo-Pacific region and is distinguished by its numerous, long, tapering, conical, pointed 

 filaments; its deep rhopalar clefts in the bell-margin; oval velar lappets fused one to another 

 by a thin web, and its brown coloration. C. octostyla of the Red Sea Malay Archipelago 

 is distinguished by its very low exumbrella dome with very small tubercles. The marginal 

 lappets are indistinct. Also in C. conifera, C. Jumokuroa, and C. ccerulca the lappets are so 

 indistinct that the bell-margin is entire, save for the deep niches of the 8 rhopalia. In C. Jumo- 

 kuroa and C. ccerulescens the central dome bears warts only on its sides, leaving its apex bare. 

 C. ccerulea has only 16 long filaments, whereas C. conifera has more than 100 and C. dumo- 

 kuron none. In Cephea ccerulescens we find very small tubular and somewhat large spindle- 

 shaped filaments between the mouths, and the subgenital porticus is only partially differ- 

 entiated. In some quadrants the primitive genital sacs may have fused and the fused wall 

 broken down to form an opening, so that one may pass a probe into one subgenital ostia and 

 out through another without pcnci rating any tissue, the passage being continuous and actually 

 a part of the outside world. In other quadrants, however, the gonads may be quite separate one 

 from another or merely fused without any break in the area of fusion. In C. corn/era and 

 C. Jumokuroa, on the other hand, the subgenital porticus is unitary and cruciform, whereas 

 in C. ii-tniu-liitina the 4 genital sacs are fused along their inner walls, but the walls remain 

 unbroken. In Cephea t\phl'>J,-nJrnini the filaments are small and spindle-shaped, and con- 

 fined to the arm-disk. 



Ir is evident that we have in the Red Sea, Indian Ocean, and western parts of the tropical 

 Pacific a large number of closely related forms of Cephea displaying many local variations. 

 I think there are only 2 well-marked forms and these are but the extremes of an intergrading 

 series: C. octostyla with low dome or flat exumbrella and small warts, and C. cephea with a 

 high dome and large warts. 



L. S. Schultz, 1898, proposes a genus Netrostoma to include R/uzostoma Jichotoma with 

 mouth-arms laterally compressed and several times dichotomously branched; no large fila- 

 ments between the mouths on the mouth-arms, although filaments may be found on the 

 mouth-arm-disk. 



