CARYBDEID.E CARYBDEA. 507 



from another. The most characteristic features for specific distinction are the shape and size 

 of the pedalia and the number and character (branched or unbranched) of the velar canals. 



Carybdea marsupialis, C. rastonii, and C. xaymacana are very closely related, if not mere 

 varieties, of one and the same medusa. They are found in the Mediterranean, tropical Atlantic, 

 and Pacific. 



C. pyramis, which appears to be identical with Haeckel's C. obeliscus and C. alata is 

 distinguished by its long, simitar-shaped pedalia. It comes from the tropical Atlantic. 



C. grandis and C. moscri are closely related tropical Pacific forms and may be mere local 

 varieties, one of the other, indeed Bigelow, 1909, believes them to be mere growth stages of 

 one and the same medusa, C. moseri being the younger. C. grandis is possibly identical with 

 Bursarius cytherece Lesson ( = Tamoya bursaria Haeckel) but the descriptions of the last- 

 named medusa are so vague that it must, I think, be dropped from our lists. 



C. murrayana Haeckel is distinguished by its large number of profusely branched velar 

 canals. In other respects it closely resembles C. marsupialis, of which form it may indeed 

 be a mere variety. 



Semper's Philippine Island Carybdea, called C. philippina by Haeckel, may be identi- 

 cal with C. moseri, but is too vaguely mentioned to be recognizable. 



C. verrucosa Hargitt, 1903, is a very young form, too immature for identification. 



C. aurifera Mayer is distinguished by its rosin-colored bell, being far darker in color 

 than in any other species. 



Haeckel's genus ProcharybJis appears to be composed of immature or imperfectly known 

 young specimens of Carybdea. It is not improbable also that future studies will show that 

 Haeckel's Procharagma is actually Carybdea. 



In some of the species of Carybdea the gastric cirri at the interradial comers of the 

 stomach are in the form of branched tufts. The mature eggs are set free from the gonads and 

 float in the gastric pouches, where they undergo a part of their development; but the larval 

 stages remain practically unstudied. 



A few physiological observations are reported of C. xnyiiiucdiiii In- Hc-rger and are discussed 

 under the description of this species. 



Carybdea marsupialis Pe'ron and Lesueur. 

 Plate 58, figs, i to 5. 



Medusa marsupialis, LINN, 1758, Systcma Naturx, Ed. 10, p. 660; 1788 (Gmclin) tomus i, pars 6, p. 31^4. 

 Carybdea marsupialis, PRON IT LFSUEUR, 1809, Anna], du Mus. Hist. Nat. Paris, tome 14, p. ;-,;. 



Charybdea marsufialis, CLAUS, 1878, Arbeit. Zool. Inst. Wicn, Bd. i, p. 221, 5 taf., fign. 1-48 (anatomy, histology). HArcsri, 

 1880, Syst. dcr Meduscn, p. 442 (references to literature). 



Bell prismatic, 4-sided, somewhat constricted near aboral apex, and with flat top. It 

 is 40 mm. high and 30 mm. wide at level of velarium. The apex of bell, the pedalia. and 

 the interradial ridges ot the exumbrella are besprinkled with numerous, quite regularly spaced, 

 oval nematocyst-warts ot dull ocher-yellow color. The 4 interradial pedalia are each 17 mm. 

 long and 10.5 mm. wide at their widest parts. The flexible, lash-like, filamentous parts of 

 the tentacles are very contractile and range in length from 2 to 12 times as long as bell-height. 

 These tentacles are hollow and regularly ringed with nematocysts. 



The 4 perradially placed sense-clubs are each set in a niche almost covered by a gelat- 

 inous flap. These sense-clubs are 5 mm. above the velar margin of the bell. Each sense- 

 club has a median pair of large eyes with a doubly convex lens, and also 4 small, lateral ocelli, 

 which lack lenses. All 6 eyes are directed inward so as to look into the bell-cavity, and are 

 of ectodermal structure. Besides the 6 eyes there is a large terminal mass of entodermal 

 cystalline concretions. 



The velarium is supported by 4 bracket-shaped buttresses (frenulx) which bind it to the 

 subumbrella in the radii of the sense-clubs. There are usually 3 (occasionally 4) branched, 

 blindly-ending centripetal vessels in each octant of the velarium. These vessels branch quite 

 complexly and irregularly, but do not anastomose. The base of the stomach is wide and 

 flask-shaped, but the throat-tube is \erv narrow and with 4 simple, lanceolate lips. Altogether 

 the manubrium is not more than one-third as long as the depth of the hell-ca\itv. There 

 are 4 minute clusters of gastric cirri, one in each interradial corner of the stomach. These 



