526 MEDUSAE OF THE WORLD. 



This medusa is distinguished by its 6-rayed structure, whereas in S. inabai the peduncle 

 is 4-sided and there are 8 adradial tentacle-clusters instead of 12 as in 5. hexaradiatus. It is 

 possible, as Broch states, that S. hexaradiatus may be the type of a new genus, but we must 

 await results of future studies before deciding this point. Unfortunately there is only one 

 specimen, obtained by the From off Fosheims Peak, Arctic Ocean, on the second expedition. 



A closely allied form from the Kurile Islands is described by Kishinouye, 1909, under 

 the name Thaumatoscyphus Jistinctus. (See Appendix to this volume). 



Genus LUCERNARIA 0. F. Miiller, 1776. 



Lucernaria, MULLER, O.F., 1776, Prodromus Zoolog. Dan., p. 227. SARS, 1846, Fauna littoral. Norveg., fasc. i,p.io. AGASSIZ, 

 L., 1861, Cont. Nat. Hist. U. S., vol. 4, p. 175. CLARK,' H. J., 1863, Journ. Boston Soc. Nat. Hist., vol. 7, p. 551. 

 AGASSIZ, A., 1865, North Amer. Acal., p. 62. FOL, 1873, Jena. Zeit. fur Natunvissen., Bd-7, p. 487. HAECKEL, l88c, Syst. 

 derMeduscn,p. 389; 1 88 1, Deep-Sea Medusa;, Challenger Report, Zool.,vol. 4, p. 53. HORN ELL, 1893, Natural Science, vol. 

 3, p. 208. HURST, Ibid., p. 209. 



Lucernosa, HAECKEL, 1881, Ibid., p. 62. 



Lucernaria, KOWALEVSKY, 1884, Zool. Anzeiger, Bd. 7, p. 712. KASSIANOW, 1901, Zeit. fur wissen. Zool., Bd. 69, p. 287, 1 1 

 fign., taf. 22-25. HAAS, 1906, ArUtischen Medusen, Fauna Arctica. Bd. 4, Lfg. 3, p. 499. 



Lucfrnaria+ Lucernosa, ANTIPA, 1892, Zoolog. Jahrb., Abth. Syst., Bd. 6, pp. 378, 391. 



The type species of this genus is Lucernaria quadncorius from the northern Atlantic 

 coast of Europe, Greenland, and America. 



GENERIC CHARACTERS. 



Stauromedusae with 4 simple perradial stomach-pouches as in Eleutherocarpidx. With- 

 out marginal anchors or marginal papillae. Peduncle single-chambered with 4 separate 

 taeniola or partial septa. 8 adradial lolies which bear tentacles. 



Antipa, 1892, would restrict Lucernaria to include forms with gonads of simple structure, 

 whereas "Lucernosa" he would institute to define forms having complex gonads made up of 

 numerous more or less separated sacs set side by side in a series of follicular ridges. In this 

 respect he follows the suggestion of Haeckel, 1881, p. 62. 



In all young medusae of Lucernaria the gonads are simple, flat, leaf-like expansions in the 

 entoderm of the subumbrella, and in many of the species they develop transverse ridges, and 

 these may become so marked as to appear as deep foldings across the gonad, thus dividing it into 

 a series of more or less sac-like corrugated pouches. This is, however, a relative matter and can 

 not be safely used as a generic distinction; moreover, I think we should, if possible, avoid found- 

 ing genera upon derails of internal anatomy requiring dissection to determine their condition. 



Kassianow, 1901, finds that in Lucernaria, Craterlophus, and Halicl\stits there is a plexus 

 of bipolar ganglion cells in the ectoderm of the exumbrella. These ganglion cells have each 

 two nuclei. There is a motor center at the point of each arm, and the nervous epithelium 

 spreads out from the bases of the tentacles. The perradial and interradial anchors are not 

 motor centers as one would expect them to be were they derived from rhopalia. 



The early development of Lucernaria has been studied by Kowalevsky, 1884 (Zool. 

 Anzeiger, Jahrg. 7, p. 712), who found that the eggs and sperm were discharged in the Bay 

 of Sebastopol near evening in August. The segmentation is total and equal, and the entoderm 

 is apparently formed by delamination. A single, central, entoderm cell was seen, however, 

 with a slender prolongation extending from between the ectoderm cells, and thus it is possible 

 that the entoderm may originate from one of the ectoderm cells which withdraws into the 

 interior. The larva elongates, the entoderm becoming a single linear row of cells, and the 

 ectoderm becoming very thin at the ends. The ectoderm is not ciliated, but the larva creeps 

 about. On the fourth day the larvae attach themselves and become flat and rounded, and 

 the entoderm forms a mass instead of a single layer, as previously. The larvae then encyst 

 themselves in a hard, secreted covering, within which they remain for about two weeks; after 

 escaping they became lost to observation. Fol, 1873, found also that in Lucernaria the seg- 

 mentation is complete and results in the formation of a single-layered blastosphere, which 

 elongates, becomes ciliated and 2-layered. After this it becomes attached. 8 small, tentacle- 

 like, marginal bodies, 4 perradial and 4 interradial, appear, but soon degenerate and disappear. 

 Thus apparently Lucernaria is descended from a Haliclystus-like ancestor. Hornell, 1893, 

 states that 8 marginal anchors are found in the young medusa, but disappear in the adult. 



