MEDUSAE OF THE WORLD

VOLUME I

THE HYDROMEDUSAE

BY

ALFRED GOLDSBOROUGH MAYER

WASHINGTON, D. C.

PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON

1910

CARNEGIE INSTITUTION OF WASHINGTON PUBLICATION No. 109, VOL. I

Copies of this Book were first issuec*

AUG9 1910

CONTENTS.

VOLUME I.

Introduction i~4

Synopsis of Genera of the Hydromedusas 5-16

Medusae Milleporinae 16

Anthomedusae 17-196

Genera of the Anthomedusae:

Pachycordyle 21 Zancleopsis 91 Podocoryne 135

Amaithasa 22 Pteronema 92 Turritopsis 143

Pennaria 23 Eleutheria 93 Oceania 146

Trichorhiza 28 Mnestra 96 Stylactis 149

Steenstrupia 29 Ctenaria 98 Thamnostylus 151

Hybocodon 37 Cladonema 98 Thamnitis 152

Microcampana 44 Dendronema 102 Lymnorea 153

Dicodonium 44 Protiara 105 Bougainvillia 155

Sarsia 47 Heterotiara 107 Nemopsis 172

Stnuridiosarsia 64 Stomotoca 108 Rathkea • • '75

llydrichthys 66 Dissonema 115 Chiarella 182

Eucodonium 68 Pandea 116 Bythotiara ..185

Ectopltura 68 Turns I2O Sibogita 186, vol. 2 4QI

Corynitis 71 Conis 130 Niobia 187

Slabberia 73 Calycopsis 130, vol. 2 491 Proboscidactyla. . . . . . 188

Margelopsis 80 Cytsis ... ,:. .-...- ... 132 Willsia . . 193

Zanclea 85

Leptomedusae '..'.' • .196-230

Genera of the Leptomedusae:

Thaumantias 198. Staurodiscus 213 Dichotomia m

Laodicea 201 1'tvchogena 214 Dipleurosoma 224

Melicertum 207 I'olyorchis 218 Toxorchis 228

Melicertissa 209 Spirocodon 219 Netocertoides. . . .229

Orchistoma 21 r Cannota 221 Monobrachium. . --230

Timoides . . ..212 Cuvieria 221

MEDUSAE OF THE WORLD.

THE HYDROMEDUSAE.

INTRODUCTION.

This work was commenced in 1892 at the suggestion of Dr. Alexander Agassiz while I was a student in his marine laboratory at Newport. Dr. Agassiz's plan was that we produce conjointly a work upon the Medusae, Siphonophora, and Ctenophorae of the Atlantic coast of North America. In pursuance of this plan, Dr. Agassiz sent me upon expeditions to Halifax, Nova Scotia; Eastport, Maine; Charleston, South Carolina; and Tortugas, Florida. It was also my privilege to accompany him as his assistant upon his expeditions to the Bahamas, and to the coral regions of the tropical Pacific. During these years the majority of our American species were captured and drawings of them made from life.

The description of all of the then known American Atlantic forms was com- pleted by me in 1900, but unfortunately the pressure of other and more important work prevented the revision of the manuscript by Dr. Agassiz, and thus it remained in the Museum of Comparative Zoology at Harvard University until 1904, when Dr. Agassiz generously returned it to me with permission to publish it in any manner whatsoever. Thus the original plan was reluctantly abandoned.

During the four years that elapsed while the manuscript lay unstudied at Harvard, new forms had been discovered along our coast; and Nutting and Hargitt had published their papers upon the hydroids and medusae of the Woods Hole region.

It was necessary to thoroughly revise the manuscript, and in order to render it of greater service, I have attempted to extend the original work to include descrip- tions of all known forms of medusae from all parts of the world. This extension was made possible through the generous establishment by the Carnegie Institution of Washington of a Marine Research Laboratory at Tortugas, Florida. Many forms were collected upon a cruise of the Carnegie Institution yacht Physalia from Boothbay Harbor, Maine, to Tortugas, Florida; and new or interesting medusae have been obtained each season upon excursions over the tropical Gulf Stream, and among the Bahamas. My official position in connection with the laboratory has afforded me every possible facility in time and opportunity tor tilt- prosecution of these studies, and I can not too kindly express my sense of gratitude to the executive officers of the Carnegie Institution tor their generous support.

To Geheimrath Prof. Dr. Anton Dohrn, and to his able corps of associates at the Stazione Zoologica, I am indebted for numerous kindnesses shown to me during my visit to the laboratory at Naples from November to February, 1907 and 1908.

I have also enjoyed full privileges of study in the libraries of the Museum of Comparative Zoology at Harvard University; in the Boston Society of Natural

1

2 MEDUSvE OF THE WORLD.

History; the American Museum of Natural History; Columbia University, the National Museum at Washington, and the Museum of the Brooklyn Institute of Arts and Sciences. Through these facilities I have been enabled to review nearly all of the published works upon medusae, but the review of literature can not pre- tend to completeness for 1907 and 1908, although all papers of those years which the author could discover are recorded.

Moreover, Profs. William K. Brooks and Louis Murbach have been so kind as to lend some of their original drawings, which are reproduced in this work, and the following gentlemen have generously granted permission for the reproduction of figures from their published works, thus enabling us to present text-figures of many forms which would otherwise have been represented merely by descriptions: Dr. Alexander Agassiz, Director of the Museum of Comparative Zoology at Harvard University; Dr. Henry B. Bigelow, of Harvard; Dr. R. P. Bigelow, of the Massa- chusetts Institute of Technology; Prof. Edward T. Browne, of the University of London; Prof. Dr. Carl Chun, of Leipzig; Prof. Dr. S. Goto of Tokyo; Geheimrath Prof. Dr. Ernst Haeckel, of Jena; Prof. C. W. Hargitt, of Syracuse University; Prof. Dr. Cl. Hartlaub, of Helgoland; Prof. W. C. M'Intosh, of Aber- deen University; Prof. Dr. Otto Maas, of Miinchen; Prof. C. C. Nutting, of Iowa; Prof. Henry F. Perkins, of Vermont; and Prof. Dr. Ernst Vanhoffen, of Kiel.

I have always felt that each working naturalist owes it as a duty to science to produce some general systematic work, and this has been an actuating motive in the production of this book. But chiefly have I been moved to the task through respect for the wishes of my generous friend and master in science, Alexander Agassiz. Nor can one remain insensible to the rare grace of form and delicate beauty of color of these creatures of the sea, associated as their study is with mem- orials of the labors of a host of distinguished naturalists. Dry though these pages must be to the reader, to the writer they are replete with memories of the ocean in many moods, of the palm-edged lagoons of coral islands sparkling in the tropic sun, of the cold, gray waters of the northern sea bestrewn with floating ice, of days of withering calm in the heat of the torrid zone, and of adventure in the hurricane ; all centering around the absorbing study of the medusa?. Love, not logic, impels the naturalist to his work.

This book attempts to present a new classification of the medusae. With every respect for Haeckel's great work, it has appeared to me that its subdivisions are often too precise to be convenient, and too artificial to accord with nature. More- over, many of Haeckel's genera are founded upon intergrading characters, and are thus imperfectly separated. The young of many medusae appear in one genus, and the adults in another. The aim of Haeckel's system is to emphasize distinctions, whereas my aim is to indicate relationships. I therefore attempt to separate genera upon positive, not upon relative, distinctions. For example, if we define one genus as having a narrow manubrium (Margelis), and another as having a wide manu- brium (Boiigamvillta), we must either institute a third genus for newly-discovered medusae with manubria of moderate width or place them doubtfully in one or the other of the genera of the extreme members of the series.

On the other hand, if we define one genus as having eight tentacles, and another as having nine or more tentacles, there can be no confusion between them, for the difference, although slight, is positive and numerical, not qualitative and intergrading.

I have not described hydroids which do not produce free-swimming medusae, although I grant it is wholly illogical to admit Podocoryne into a system which excludes Hydractmia, or to include only those species of Stylactis which produce

INTRODUCTION.

medusae and to exclude those which produce sessile gonophores. Nevertheless this should be clearly understood and must be accepted as an artificial limitation of the work.

I have thus attempted to describe only such hydroids as are known to produce medusae, and have endeavored to bring the systematic arrangement of the medusa more nearly into accord with that of the hydroids. A strictly natural system includ- ing both hydroids and medusae can not be constructed, for many of the hydroids remain undetermined. Moreover, dissimilar hydroids (Svm-tjrv>if, StanriJinim may give rise to similar medusae (Sarsia); or the reverse may be the case, as in the medusae of Bongamvillia and Nemopsis, or that of the two sorts of medusa- (Sarsia and Cladoncrna] which may arise from hydroids of Stauridia.

These and many other cases of a similar nature interpose a barrier to our attempts to invent a natural system which includes all hydroids and medusa" within its embrace. At present, I believe, we must content ourselves with a compromise between a natural and an artificial arrangement, confiding in the belief that as more and more of the hydroids are discovered it will become correspondingly more possible to arrange the medusae in a natural system. After consultation with Prof. C. C. Nutting we have mutually decided that the promulgation of such ;i system is at present inadvisable. Such a system has, indeed, been proposed by von Lendenfeld, 1884 (Zool. Anzeiger, Bd. 7), but has gained no acceptance.

Much confusion has been introduced through the habit, in vogue among marine expeditions, of sending all of the medusa" to one specialist and the hydroids to another. Thus the sessile and the reproductive stages of the same animals are worked upon independently from different view-points by different men.

I am inclined to regard the Trachymedusae and Narcomedusa" as being transformed actinutae, for they commonly develop through an actinula larva in which the bell grows out as a collar-like, or intertentacular lappeted expansion from the sides of the body after the tentacles have appeared, and the tentacles of the actinula become those of the medusa. The medusa of the Anthomedusas and Leptomedusae is formed upon a different plan, for the tentacles grow outward from the bell-margin after the bell has developed. I believe, therefore, that the bell of the Trachymedusae and Narcomedusae is not homologous with that of the Antho- medusa and Leptomedusae. It is evident that the entodermal otoliths of the Trachymedusa" and Narcomedusae are not homologous with the ectodermal otoliths of Leptomedusae. Budding and alternation of generations occur in both classes of veiled medusae.

I believe that the medusa-shape has been acquired independently in the Trachyhna and Leptolma forms of veiled medusae.

The colored plates contained in this volume consist of drawings, from life, of medusae of the Atlantic coast of the United States. The text-figures, on the other hand, are chiefly outline tracings from the illustrations of many authors; and are presented in order to spare the reader the trouble of consulting numerous scattered works of reference. These outline copies of previously-published drawings of medusae were carefully traced from the originals by Mr. Carl Kellner, artist of the Tortugas Marine Laboratory of the Carnegie Institution.

Other outline figures are from life, and the majority of these were drawn by the author while studying at Mousehole, Cornwall, England, and at the Naples Zoological Station during the autumn and winter of 1907 and 1908.

This book aims to be something more than an old-fashioned systematic treatise, for it attempts to record, if not to review, all works upon the embryology, cytology, oecology, physiology, etc., of all forms coming within the scope of the text.

MEDUSAE OF THE WORLD.

Many of the medusae are modified more or less profoundly by environmental conditions, and this gives rise to numerous local races, the determination of the relationships between which is all but impossible in the present state of our knowl- edge. Experimental work in this field is in its infancy, yet the few observations of Hallez on Bongainvillia and of Browne in his experiments in maintaining various hydroids under different conditions in aquaria suggest that the results of more extended studies will be of great benefit to the systematist in determining the natural limitations of species. A striking example of the profound effect of environ- mental changes is seen in the experiments of Goldfarb, 1906, upon Eudendrium, wherein he showed that after the regenerative process had entirely ceased in the dark, it could be recalled into activity by an exposure of only five seconds to the daylight.

At present the natural classification of the species of such genera as Obclia, Phialidnnn, Etttnna, Bougamvillia, etc., is impossible, and the Linnean system is inadequate to the task of expressing their actual relationships. Indeed, with the exception of the sponges and corals, there is no phylum of the animal kingdom more difficult to classify than the medusae.

In a work of this magnitude there must needs be both errors and omissions, and for these I can but present my apologies in advance of their discovery, trusting that all such will soon be discovered and announced, and that science may be more advanced than hindered through the publication of this book. I trust that none of my esteemed contemporaries will feel aggrieved at such criticisms of their labors as may appear in the following pages, for should anyone find cause for such offense, he may console himself in the fact that I am obliged to confess to having found more to criticize and amend in my own previously-published papers than in those of any other living naturalist.

ANTHOMEDUS^:.

HYDROMEDUS/E OR VEILED MEDUS/E.

Medusae with a velum, or diaphragm, which partially closes the marginal opening of the bell. When ripe the sexual products are found in the ectoderm.

With a double marginal nerve-ring, one above and one below the velum.

Without gastric filaments. Development either direct from actinula larvae or through alternation of generations from hydroids.

Order 1. ANTHOMEDUS/E Haeckel, 1879.

Hydromedusae with gonads in the ectoderm of the manubrium. Without otocysts. The hydroids are of the Tubularian order.

Family No. i, CODONID.&.

Anthomedusae with ring-like gonad encircling the manubrium. Four to six simple, unbranched radial-canals. Simple, unbranched tentacles.

Subfamily No. i, SARSIANJE. Some or all of the tentacles arise singly, not in clusters, from bell-margin.

PACHYCORDYLE, Weismann, i8%$ = Parvanernus ( ?), Mayer, I9°4-

Degenerate medusae without tentacles, radial or circular canals. Hydroid: Pachy- cordyle Weismann.

AMALTH^EA, Schmidt, 1854.

Four rudimentary tentacles, four radial-canals, and a ring-canal. Hydroid: Corymorpha.

PENNARIA, Oken, 1815; Goldfuss, i%2O = Globiceps, Haeckel, 1879. Similar to Amalthaa, but the hydroid is Pennana.

TRICHORHIZA, Russell, 1906.

Medusa resembles Pennaria, but the hydroid is Tnchorhiza.

STEENSTRUPIA, Forbes, iS^.6 = Eup/i\s(i, Forbes, i$4.% = Eupliysora, Maas, 1905.

Four radially situated tentacles, one of which is long, the other three short. Bell radially

symmetrical. Hydroid: Corymorpha. HYBOCODON, L. Agassiz, \%()2 = HybocodonJt-Amphicodon, Haeckel, 1879.

One or more well-developed tentacles arise from base of one of the four radial-canals. With rudimentary tentacles at the bases of the three other radial-canals. Bell asym- metrical. Hydroid: H ybocodon Agassiz.

MICROCAMPANA, Fewkes, 1889.

With one long and five short tentacles, arising at ends of six radial-canals.

DlcoDONlUM = Dicodonium + Dincmti, Haeckel, 1879.

Two well-developed and two rudimentary tentacles. No meridional lines of nettle-cells

over exumbrella. SARSIA, Lesson, i^^.^ = CoJontutn + Sarsia + S'vnJii'tyon, Haeckel, 1879.

Four equally developed tentacles with abaxial, ectodermal ocelli. No meridional lines

of nettle-cells over the exumbrella. Hydroid: Syncoryne. StauriJiosarsia new subgenus. Medusa similar to Sarsia, but the hydroid is Stuundia.

HYDRICHTHYS, Fewkes, 1888.

Medusa resembles Sarsia, but there are no ectodermal ocelli upon the tentacle-bulbs. Hydroid: H vJrichthys, Fewkes; commensal or parasitic upon fish.

EUCODONIUM, Hartlaub, 1907.

Medusa resembles Sarsia, but stomach is mounted upon a gelatinous peduncle. The four tentacles terminate each in a knob-shaped extremity.

ECTOPI.EURA, L. Agassiz, 1862 (sens. ampl.).

With two or four tentacles. Eight meridional lines of nettle-cells over the exumbrella. Hydroid: Ectopleura.

6 MEDUS.E OF THE WORLD.

CORYNITIS, McCrady, 1857.

Four knobbed tentacles. Manubrium cruciform in cross-section. Ocelli upon tentacle- bulbs. Hydroid: Syncoryne (Hargitt).

SLABBERIA, Forbes, lB^6 = Slabberia + Dif>urena + BathycoJon, Haeckel, 1879.

Four knobbed tentacles. Manubrium tubular and encircled by two or more ring-like gonads. Hydroid: Syncoryne.

Subfamily No. 2, MARGELOPSIN^E.

Four radially placed clusters ot marginal tentacles. No oral tentacles. A ring-like gonad encircles the stomach. Four simple, unbranched radial-canals. Hydroids are pelagic Tubularians, and the medusae arise by budding from their sides.

MARGELOPSIS, Hartlaub, 1897.

Characters of the medusa are those of the subfamily. Hydroid: Margelopsis, in which the tentacles are disposed in definite circlets.

PELAGOHYDRA, Dendy, 1903.

Medusa similar to Margelopsis. Hydroid: PelagohyJra. Its tentacles arise irregularly from sides of the hydranth, and are not disposed in definite circlets.

Family No. 2, CLADONEMID^E.

Tentacles branch dichotomously or complexly, or give rise to a linear series of nematocyst- bearing filaments along their abaxial sides. Gonads ring-like, or segregated upon the inter- radial and adradial sides of the manubrium.

Subfamily No. i, PTERONEMINJE.

Manubrium without oral tentacles.

ZANCLEA, Gegenbaur, l8^6 = Zanclea + Gernmaria, Haeckel, 1879.

With two or four tentacles, each of which gives rise to an abaxial row of nematocyst- bearing branches. With meridional rows of nettle-cells upon the exumbrella. No ocelli upon the tentacle-bulbs. No brood-pouch above the stomach. Hydroid: Gemmaria McCrady; Allman.

ZANCLEOPSIS, Hartlaub, 1907.

Similar to Znticlea, but without meridional lines of nettle-cells over the exumbrella. With ectodermal ocelli upon the outer sides of the tentacle-bulbs.

PTERONEMA, Haeckel, 1879.

Similar to Zanclea, but with a brood-sac above the stomach. Four tentacles.

ELEUTHERIA, Ouatrefages, 1842.

Four to six bifurcated tentacles, and an equal number of simple unbranched radial- canals. There is an ectodermal brood-sac above, but not connected with the stomach. The medusa is hermaphroditic and the germ-cells develop in the brood-sac. Hydroid: Clavatella Hincks.

MNESTRA, Krohn, 1853; Giinther, 1903.

Degenerate medusa parasitic upon Phylhrrhoe. Four to no tentacles, each with an abaxial line of nettle-warts. Four perradial meridional lines of nettle-cells over the exumbrella. Four radial-canals. Throat is blocked by a cavernated mass of entoderm.

Subfamily No. 2, DENDRONEMIN^).

Manubrium with oral tentacles.

CTENARIA, Haeckel, 1879.

Two marginal tentacles with abaxial filaments. Four bifurcated radial-canals. Simple oral tentacles. Brood-sac above the stomach.

ANTHOMEDUS.E. 7

CLADONEMA, Dujardin, 1843.

Four or five bifurcated or eight to ten simple radial-canals. Branched marginal tentacles. Simple oral tentacles. No brood-sac above the stomach. Hydroid: Statin Jin Dujardin.

DENDRONEMA, Haeckel, 1879.

Similar to ClaJonema, but with branched oral tentacles and with brood-sac above stomach.

Family No. 3, OCEANID.S, sensu Vanhoffen.

Anthomedusae in which the gonads are segregated and developed upon the interradial or adradial sides of the manubrium. With unbranched marginal tentacles. Mouth with four lips.

Subfamily No. i, TIARIN.E.

Unbranched radial-canals. Marginal tentacles separate; not grouped into clusters. No oral tentacles. Tentacles hollow. When present the ectodermal ocelli are upon the abaxial sides of the tentacle-bulbs. With the exception of Calycopsis all of the genera have four radial-canals.

PROTIARA, Haeckel, i%jq = Halitiara, Fewkes, 1882.

Four radially placed, well-developed tentacles. Four interradial gonads with smooth outer surfaces. With or without marginal cirri. External surfaces of gonads smooth. Four cruciform, simple lips. No ocelli on the velar sides of the tentacles.

HETEROTIARA, Maas, 1905.

Eight or more tentacles. The ring-canal gives rise to blindly-ending centripetal vessels.

STOMOTOCA, L. Agassiz, \%()2 = Amf>hinema-\-Stomotoca + Codonorchis, Haeckel, 1879.

Two well-developed and many rudimentary tentacles. External surfaces of the adradial gonads are thrown into transverse folds. Hydroid: Pengonimus.

DISSONEMA, Haeckel, 1879.

Similar to Stomotoca, but the gonads finally migrate outward along the four radial-canals.

PANDEA, Lesson, 1843.

Four or more tentacles. Gonads four interradial, folded ridges on the sides of the stomach but these gonads are not completely separated in the four principal radii. Hydroid: Dcndroclava ( ? ?).

TURRIS, Lesson, 184.3 = Tiara + Turris + Catablema, Haeckel, 1879.

Four or more tentacles. Four interradial horseshoe-shaped gonads on the stomach wall. These are composed of more or less fused ridges or network-like swelling. They are completely separated in the four principal radii. Hydroid: Clirculu Wright.

CONIS, Brandt, 1834; Haeckel, 1879.

Similar to Pandea, but the tentacle-bulbs give rise to abaxially-placed clubs which bear ocelli.

CALYCOPSIS, Fewkes, 1882.

Sixteen simple, separate radial-canals. Eight transversely folded, adradial gonads. Ring-canal simple.

Subfamily No. 2, MARGELINjE.

With four unbranched radial-canals. With oral tentacles, or nematocyst-knobs, upon the lips. Tentacles solid. When present the ectodermal ocelli are upon the inner (velar) sides of the tentacles.

CYTVEIS, Eschscholtz, iS2() = Cytxis + Cubogastcr, Haeckel, 1879.

Four simple marginal tentacles. With simple, unbranched, oral tentacles.

PODOCORYNE, Sars, \?>Jtb = Dysmorf>hosa-\-Cytteandra, Haeckel, 1879.

Eight or more simple marginal tentacles. With simple, unbranched, oral tentacles. Hydroid: Podocoryne. When present the peduncle above the stomach is solid and gelatinous.

MEDUSAE OF THE WORLD.

TURRITOPSIS, McCrady, 1856.

Eight or more simple marginal tentacles. The entodermal walls of the radial-canals above the stomach are composed ot vacuolated cells forming a peduncle-like base for the stomach. The mouth is studded with a row of nematocyst-bearing knobs. With ectodermal ocelli on the velar sides of the tentacles near their bases. Hydroid: DenJroclava (Brooks).

OCEANIA sensu Kolliker, 1853; Gegenbaur (in part) 1856.

Medusa similar to Turritopsis, but with solid gelatinous, non-vacuolated peduncle above the stomach. Hydroid: C lava-like.

STYLACTIS, Allman, 1864.

Degenerate medusae, with four to eight rudimentary marginal tentacles and no oral tentacles. Hydroid: Stylactis.

THAMNOSTYLUS, Haeckel, 1879.

With two simple, marginal tentacles, and with branched oral tentacles.

THAMNITIS, Haeckel, 1879.

Four radially placed, simple, marginal tentacles, and branched oral tentacles.

LYMNOREA, Peron and Lesueur, iSog — Limnorea + Thamnostoma, Haeckel, 1879. Eight or more simple, marginal tentacles. Branched oral tentacles.

BOUGAINVILLIA, Lesson, \$4.T> = Margelis + Lizusa + Hif>pocrene, Haeckel, 1879.

With branched oral tentacles. The marginal tentacles are grouped in four radial clusters. All of the tentacles are filiform. Hydroid: Bougainvillia.

NEMOPSIS, L. Agassiz, 1849.

Similar to Bougainvillia, but each cluster of marginal tentacles consists of a median pair of clavate tentacles flanked by filiform tentacles. Hydroid: Bougainvillia.

RATHKEA, Brandt, lS^y = Lizzia + Lizella + Rathkea + Margellium, Haeckel, 1879.

With eight clusters of marginal tentacles. Simple or branched oral tentacles. Ring- canal simple.

CHIARELLA, Maas, 1897.

Sixteen (eight double) clusters of marginal tentacles. The ring-canal gives rise to centrip- etal vessels. Branched oral tentacles.

Subfamily No. 3, DENDROSTAURINjE.

With branched radial-canals. No oral tentacles. Marginal tentacles arise singly, and are not grouped into clusters. Tentacles hollow. No cirri or marginal clubs.

BYTHOTIARA, Giinther, 1903.

Four bifurcated radial-canals and a ring-canal. Four interradial gonads.

SIBOGITA, Maas, 1904.

Four principal radial-canals, which branch complexly. Four interradial gonads. Ring- canal present.

NIOBIA, Mayer, 1900.

Four principal radial-canals, two of which bifurcate so that six canals reach the circular vessel. Four interradial gonads. The marginal tentacles develop into medusae, and are cast off.

PROBOSCIDACTYLA, Brandt, \%Tfi=Dyscannota + Dicranocanna+ Willeta

-\-Proboscidactyla, Haeckel, 1879.

The four primary radial-canals give rise to simple or branched side branches. No ring- canal. With intertentacular lines of nematocysts upon the exumbrella above the bell-margin. Gonads on the adradial sides of the stomach extending outward along the sides of the four main radial-canals.

LEPTOMEDUS.E.

9

WILLSIA, Forbes, 1846.

Similar to ProbosciJactyla, but with six or more primary radial-canals. The hydroid belongs to the genus Lar Gosse.

Order 2. LEPTOMEDUSjE Haeckel, 1886.

Hydromedusae with gonads upon the radial-canals. When present the otoliths are of ectodermal origin. The medusae arise through alternation of generations from Campanu- larian hydroids.

Family No. i, THAUMANTIAD^).

Leptomedusae without lithocysts.

Subfamily No. i, MELICERTIN^.

With simple, unbranched radial-canals and an equal number of lips, without oral ten- tacles. Cirri or marginal clubs may or may not be present.

THAUMANTIAS, Eschscholtz, \%29 = Tetrtinfmti + T/iiiii>tiiinti(is, Haeckel, 1879.

Four or more tentacles. Four radial-canals. No marginal clubs or cirri. Hydroid: T haumantias.

LAODICEA, Lesson, i$4$ = Octonema + Laodice, Haeckel, 1879.

Four or more tentacles. Four radial-canals, with marginal clubs or cirri. Hydroid: Cuspidella.

MELICERTUM, Oken, 1815; sensu L. Agassiz, i%ta=Melicertella+Melicertum, Haeckel, 1879. With eight or more tentacles. Eight radial-canals. No marginal clubs or cirri. Hydroid: Melicerturn Agassiz.

M.ELlCER.TlSSA = Mclicertissa + MflicfrtiJnitn, Haeckel, 1879.

Similar to Melicertum, but with marginal clubs or cirri between the tentacles.

ORCHISTOMA, Haeckel, 1879.

More than eight radial-canals. With or without marginal clubs or cirri.

TIMOIDES, H. B. Bigelow, 1904.

Four radial-canals. The ring-canal gives rise to blindly-ending centripetal diverticula. Numerous tentacles and cirri. Four gonads on the four radial-canals. Stomach mounted upon a peduncle.

Subfamily No. 2, POLYORCHIN.E.

The radial-canals give rise to branches which end blinJlv and do not connect with the marginal ring-canal.

STAURODISCUS, Haeckel, 1879.

With four radial-canals, each of which gives rise to two side branches, which end blindly.

PTYCHOGENA, A. Agassiz, 1862, 1865.

With four radial-canals which give rise to numerous blindly-ending side branches. Gonads leaf-shaped and developed upon the side branches of the radial-canals.

POLYORCHIS, A. Agassiz, 1862 to 1865.

With four radial-canals which give rise to numerous blindly-ending side branches. Numerous sac-like, sausage-shaped gonads attached to the radial-canals and to their side branches. Ring-canal simple. Bell-margin not cleft into lappets.

SPIROCODON, Haeckel, lS^g = Goniomeam/rus, Kirkpatrick, 1903.

Similar to Polyorchis, but the ring-canal gives rise to blindly-ending, centripetal branches, and bell-margin is cleft into lappets.

10 MEDUS/E OF THE WORLD.

Subfamily No. 3, BERENICINJE. The radial-canals give rise to branches which connect with the ring-canal.

CANNOTA, Haeckel, 1879.

With four radial-canals, each of which gives rise to two side branches which join the ring-canal.

CuviERIA, Peron, 1807 = Berenice, Haeckel, 1879.

With four main radial-canals which give rise to numerous non-dichotomous side branches. Gonads on the terminal ramuli of the canals.

DICHOTOMIA, Brooks, 1903.

With four main radial-canals which divide dichotomously two or more times. The gonads extend outward from the stomach over the canals.

DIPLEUROSOMA, Axel Boeck, i$66 = Tetracannota, Mayer, 1900.

With three or more main radial-canals which give rise to non-dichotomous side branches. Gonads on the canals adjacent to the stomach. Hydroid: Cuspidella (?).

ToxoRCHiS = Toxorchis + ClaJocanna, Haeckel, 1879.

Four or more main radial-canals which branch dichotomously one or more times. Gonads on the outer branches of the canals near the ring-canal.

NETOCERTOIDES, Mayer, 1900.

Eight main radial-canals which branch dichotomously. Gonads extend outward from the sides of the stomach along the radial-canals.

Family No. 2, EUCOPID^.

Leptomedusaewith lithocysts, and with less than eight radial-canals upon which the gonads are developed.

Subfamily No. i, OBELIN^E. With eight adradial lithocysts. Four radial-canals. Stomach without a peduncle.

EUCOPELLA, von Lendenfeld, 1883.

Degenerate medusae. No tentacles. No manubrium. Branched radial-canals. Hydroid:

Eucopella.

AGASTRA, Hartlaub, 1897.

Degenerate medusae. No manubrium. Simple, unbranched radial-canals. Hydroid: Campanularia ( ?).

EUCOPE, Gegenbaur, 1856.

Basal bulbs of the tentacles are simple and hollow, and do not project inward into the gelatinous substance of bell. Lithocysts on bell-margin. Hydroid: Campanulana.

OBELIA, Peron and Lesueur, 1809.

Entodermal cores of tentacles project inward into the gelatinous substance of the bell. Otocysts on bases of tentacles. Hydroid : Obelia.

TIAROPSIS, L. Agassiz, 1849.

An ocellus with entodermal pigment above each lithocyst. Tentacle-bulbs simple and hollow.

Subfamily No. 2, PHIALIN^.

With more or less than eight lithocysts. Four to five radial-canals. Stomach without a peduncle.

CLYTIA, Lamouroux, \%i2 = Ef>enthesis, McCrady, 1857.

Sixteen tentacles alternating with sixteen lithocysts. Four radial-canals. No cirri. Hydroid: Clytia.

LEPTOMEDUS.E. 11

PHIALIDIUM, Leuckart, i%$() = 0<:eania, Agassiz, 1862, 1865.

Sixteen or more tentacles. More than sixteen lithocysts. Four radial-canals. No rudi- mentary tentacles. No cirri. Hydroid: Campanulina.

PHIALUCIUM, Maas, 1905.

Similar to Phialidium, but with permanently rudimentary tentacle-bulbs. Hydroid un- known.

BLACKFORDIA, Mayer.

Numerous tentacles and lithocysts. Entodermal cores of some or all of the tentacles project inward into the gelatinous substance. Four radial-canals. No cirri. No permanently rudimentary tentacles.

PSEUDOCLYTIA, Mayer, 1900.

Five radial-canals, five lips, five gonads. Numerous tentacles and lithocysts. No cirri.

GASTROBLASTA, Keller, \?>?>^=Multioralis, Mayer, 1900.

Two or more manubria. No cirri. The medusa propagates by fission.

EUCHEILOTA, McCrady, 1857.

Four or more closed vesicular lithocysts. Four radial-canals. Marginal or lateral cirri. Hydroid: Campanulina ( ?).

MnROCOMA = Phialis + Mitrocoma + Mitrocomiurn + Mitrocomella, Haeckel, 1879.

Similar to Eucheilota, but the lithocysts are contained in open folds of the velum. No entodermal ocelli such as are found in Tiaropsis.

STAUROPHORA, Brandt, lS^ = Staurostoma + Staurof>hora, Haeckel, 1879.

Mouth an open, cruciform, gutter-like slit extending down the four radial-canals.

Subfamily No. 3, EUTIMINJE.

With eight adradial lithocysts. Stomach mounted upon a gelatinous peduncle. Four radial-canals.

SAPHENIA, Eschscholtz, 1829.

Two tentacles. Numerous cirri.

EUTIMA, McCrady, 1 857 = Eutima + Eutimeta + Octorchis + Octorchandra

-\-Eutnnalphcs, Haeckel, 1879.

Four or more tentacles. Numerous cirri or marginal warts. Four or eight gonads upon the four radial-canals. Hydroid: Campanopsis (Claus, Brooks).

JLwTIMlUM = Eutimium + Octorchi(lium, Haeckel, 1879.

Similar to Eutima, but without cirri. Hydroid: Campanulina ( ?)

Subfamily No. 4, EIRENINvE.

With more than eight lithocysts. Stomach mounted upon a gelatinous peduncle. Four or six radial-canals.

PHORTIS, McCrady, 1857.

Four or more tentacles. No cirri. Four radial-canals. Hydroid: P/iortis Brooks.

IRENOPSIS, Goette, 1886, non Ireniopsis, Mayer, 1894.

Six or more tentacles. Six radial-canals. Six lips. Six gonads.

EIRENE, Eschscholtz, i%2g= Irene +Irenittm, Haeckel, 1879.

Four or more tentacles. Numerous marginal warts or cirri. Four or eight gonads developed upon limited parts only of the four radial-canals. Hydroid: Campan- ulina ( ?)

TIMA, Eschscholtz, 1829.

Similar to Eirene, but with gonads upon the entire lengths of the four radial-canals. Hydroid: Campanulina ( ?)

12 MEDUS.E OF THE WORLD.

Family No. 3, ^EQUORIDjE.

Leptomedusae with otocysts, and with eight or more radial-canals.

OCTOCANNA, Haeckel, 1879.

Eight radial-canals, 45° apart. Eight lips. No ocelli. (Is this a young jEquorea ?)

OCTOGONADE, Zoja, 1896.

Similar to Octocanna, but the marginal sense-organs have ocelli as well as lithocysts.

STOMOBRACHIUM, Brandt, lS^^ = Stomobrachium + Staurobrac/iium, Haeckel, 1879.

Eight or more simple, unbranched radial-canals, which arise at equal intervals from the margin of the stomach. Four lips.

HALOPSIS, A. Agassiz, 1863, 1865.

Radial-canals arise in four groups from the four perradial corners of stomach. Four lips.

^EQUOREA, Peron and Lesueur, i$og = &(juorea + R/iegmatoJes + Mesonema

-\-Polycanna, Haeckel, 1879.

More than eight simple, unbranched radial-canals which arise separately from the mar- gin of the stomach. More than four lips. Subumbrella smooth, without gelatinous papilla-like protuberances. Hydroid: Campanulina.

ZYGODACTYLA, Brandt, 1835; sensu Agassiz, 1862.

Similar to jEquorea, but with interradial rows of papilla-like, gelatinous protuberances upon the subumbrella.

, Haeckel, 1879.

With bifurcated or branched radial-canals which arise at equal intervals from the margin of the stomach. More than four lips. No peduncle.

ZYGOCANNULA, Haeckel, 1879.

Similar to Zygocanna, but the stomach is mounted upon a gelatinous peduncle.

Order No. 3. TRACHYMEDUS.E Haeckel, 1866.

Medusae with a marginal velum, and with lithocyst concretions of entodermal origin. With simple uncleft bell-margin.

Family No. i, OLINDIADJE.

Some or all of the tentacles project from the sides of the bell, above the margin, and have adhesive disks. Gonads linear, sac-like, or folded, and developed upon the four or six radial- canals. The tentacles arise separately and are not grouped in clusters.

GONIONEMUS, A. Agassiz, 1862, 1865.

All of the tentacles project from sides of bell in a zone slightly above bell-margin. All have adhesive disks. Four radial-canals. Lithocysts external, on bell-margin. No centripetal canals. Development through an attached hydra stage. (Perkins.)

CUBAIA, Mayer, 1894.

Similar to Gonionemus, but with two sets of tentacles, one arising from the bell-margin and the other set projecting from the sides of the bell, as in Gonionemus.

VALLENTINIA, Browne, 1902.

Similar to Cubaia, but with lithocysts inclosed within the gelatinous substance of the bell, adjacent to the ring-canal, and on the inner side above the velum. (Is this a young Olindias])

OLINDIAS, F. Miiller, 1861.

Similar to / allentmia, but with blindly-ending, centripetal diverticula from the ring-canal.

TRACHYMEDUS.K. 13

OLINDIOIDES, Goto, 1903.

Similar to OlinJias, but with six radial-canals (two bifurcated and two simple). Six gonads. Four lips. The exumbrella tentacles project at various levels from the sides of the bell.

Family No. 2, PETASIDjE Haeckel, 1879.

Trachymedusae with tour radial-canals upon which the linear or sac-like gonads are developed. Tentacles without adhesive disks. Four lips.

PET ASUS = Pftasus + Dif>etasus + Petas(ita + Pftacli>: uni, Haeckel, 1879.

Tentacles arise at equal intervals, not grouped into clusters. No centripetal canals. Free marginal hthocyst-clubs.

AGLAUROPSIS, F. Miiller, 1865.

Similar to Petasus, but the lithocysts are vesicular, and project from the bell-margin between the tentacles.

CRASPEDACUSTA, Lankester, i$%o=LimnocoJium, Allman, 1880.

Tentacles arise singly as in Pi-tnsus and Aglauropsis, but the lithocyst concretions are each inclosed in a cavity within the gelatinous substance of the velum on the inner (centripetal) side of the ring-canal. (The medusa lives in fresh water among water- lilies.) The hydroid is devoid of tentacles.

MlCROHYDRA, PottS, 1885.

Is possibly identical with Limnocodiutn, but the mature medusa is unknown. The young medusa has no lithocysts, and it arises by budding from a minute hydroid which has no tentacles.

MJ^OTIAS, OstroumofF, 1896.

Tentacles arise at equal intervals, not in clusters. Numerous centripetal, blindly-ending canals arise from the ring-canal.

GOSSEA, L. Agassiz, 1862.

Tentacles grouped into clusters. No centripetal canals. Lithocyst concretions free or inclosed.

Family No. 3, LIMNOCNIDIDjE.

Numerous hollow tentacles which project singly, not in clusters, from the sides of the bell in a zone slightly above the margin. Tentacles without adhesive disks. Numerous inclosed lithocysts on the exumbrella side of the velum. Mouth a round opening. Gonads developed diffusely in the ectoderm of the stomach-wall. Four (occasionally five or six) radial- canals. Medusa-buds arise from the sides of the stomach, and are set free.

LIMNOCNIDA, Giinther, 1893.

Generic characters are those of the family. The only known species is /,. tanganjicd from the fresh-water lakes of Central Africa, and the Niger river.

Family No. 4, PTYCHOGASTRIDJE.

Numerous more or less isolated clusters of tentacles, some of which bear adhesive disks. Numerous free lithocyst-clubs. Eight radial-canals. Four lips. Stomach eight-lobed. These stomach-lobes are in the radii of the radial-canals, and are bound to the subumbrella by means of eight mesenterial partitions. The gonads are upon the eight stomach-lobes, and each is more or less divided by the mesentery so there may be eight double (sixteen) gonads.

PTYCHOGASTRIA, Allman, iSfi^Pcctyllis + Pectis + Pectanthis, Haeckel, 1879. The generic characters are those of the family.

Family No. 5, TRACHYNEMID^E.

Trachymeduss with eight simple radial-canals upon which the gonads are developed. No mesenterial partitions in the subumbrella. Tentacles without adhesive disks. Ring- canal simple without centripetal branches.

14 MEDUSA OF THE WORLD.

Subfamily No. i, RHOPALONEMINjE.

Trachynemidae in which the stomach lacks a peduncle. RHOPALONEMA, Gegenbaur, iS^6=Trachynema (young medusa), Gegenbaur, 1854

= Trachynema + Rhopalonema + Marmanema, Haeckel, 1879.

With eight well-developed radial tentacles, and eight or more small cirrus-like or club- shaped interradial tentacles. All tentacles arise in a single row. Eight gonads localized on the eight radial-canals. Four lips. SMINTHEA, Gegenbaur, 1856.

Similar to Rhopalonema, but with only eight tentacles, one at the foot of each of the

eight radial-canals.

HOMCEONEMA, Maas, i8g3 = Colobonema, Vanhoffen, igo2 = lsonema (in part), Maas, 1906. Similar to Rhopalonema, but the tentacles are all of one sort. No small club-shaped or

cirrus-like tentacles. Four lips. PANTACHOGON, Maas, 1893 (sens. ampl.).

Gonads not localized as in Homceonema and Rhopalonema, but developed diffusely over

the radial-canals. Four lips. HALICREAS, Fewkes, i$82 = Halicreas + Haliscera, Vanhoffen, 1902

= Isonema (in part), Maas, 1906.

The mouth is a simple round opening, without four lips. (In all known species the radial-canals and ring-canals are very broad and flat.) Wart-like protuberances may be present upon the sides of the exumbrella. Radial tentacles large, interradial, small. Tentacles arise in a single row. BOTRYNEMA, Browne, 1908.

Similar to Halicreas, but the tentacles are grouped in linear clusters in a single row

around the bell-margin. CROSSOTA, Vanhoffen, 1902.

The tentacles arise in several rows from the bell-margin. Mouth with four lips.

Subfamily No. 2, AGLAURINjE.

Stomach mounted upon a peduncle. AGLAURA, Peron and Lesueur, 1809.

Eight gonads upon the peduncle above the stomach. Sexes separate. Development direct. AGLAmHA = /fglantha + 4glisi:ra, Haeckel, 1879.

Eight gonads upon the subumbrella, or at the turning points of the eight radial-canals

between the peduncle and the subumbrella. Sexes separate. AMPHOGONA, Browne, 1904.

Similar to Aglantha, but medusa is bisexual, four of gonads being male, and four female. STAURAGLAURA, Haeckel, 1879.

Four gonads, one upon each alternate radial-canal. PERSA, McCrady, 1857.

Two gonads on two of the radial-canals, 180° apart. The six other radial-canals are sterile.

Family No. 6, GERYONIDjE.

Trachymedusae with four or six radial-canals upon which the flat, expanded, leaf-like gonads are developed. Stomach mounted upon a gelatinous peduncle. The ring-canal gives rise to blindly-ending centripetal canals. LIRIOPE, Lesson, i%^ = Linantha -\-Lirwpc + Glossocodon -\-Glossoconus, Haeckel, 1879.

Four radial-canals. Four gonads. Four lips. With four primitive, solid, radial, and four solid interradial, and four hollow, flexible, radially-placed tentacles; all twelve of which may be found upon the medusa at one and the same time. Development direct through a free-floating, actinula-like larva. GERYONIA, Peron and Lesueur, i8og = Geryones + Geryonia + Carmaris

+ Carmarina, Haeckel, 1879. Similar to Liriope, but with six radial-canals, six gonads, six lips, etc.; instead of four.

NARCOMEDUS.E. 15

Order No. 4. NARCOMEDUS^ Haeckel, 1879.

Veiled medusae with bell-margin cleft into intertentacular lappets. With free lithocyst- clubs, containing concretions of entodermal origin. These medusx develop from actmula larvae either directly or by budding. The bell grows outward from the sides ot the body of the actinula, or the medusa-bud, leaving the tentacles stranded in the partially closed-over clefts between the lappets of the bell. The Narcomedusae are thus medusiform, actinula-like animals, the bell of which is not homologous with that of the Anthomedusae or Leptomedusae.

Family No. i, SOLMARIDjC.

Narcomedusae in which the outer margin of the stomach is plain, entire, and without peripheral stomach-pouches. Saccules may, however, arise from the subumbrella floor of the stomach.

o/)'fo//>a (young) + Sohnaris, Haeckel, 1879. Without subumbrella saccules. Gonad is a simple annulus in ectoderm of subumbrella

floor of stomach. PEGAJXTBA=Pegasia+Polyxenia+Pegantha+SoImoneta (in part), Haeckel, 1879.

With out-pocketings on the subumbrella floor of the stomach. The gonads are developed in these subumbrella saccules.

Family No. 2, jEGINID^E, sens. ampl.

Narcomedusae in which the central stomach gives rise to simple or cleft marginal out- pocketings in the radii of the tentacles.

CUNANTHA, Haeckel, 1879.

Four tentacles. Four peronial strands in the tentacular radii. Four simple, uncleft, peripheral stomach-pouches in the radii of the tentacles. This "genus" is prob- ably only a young stage of JEgina. JEciNA, Eschscholtz, i$2g = Cunarcha + £gina + SolmunJus, Haeckel, 1879.

Four tentacles. Four peronial strands. Four cleft ( = eight peripheral) stomach-pouches, outer margins of which may be still further divided.

SoLMUNDELl.A=dlgi>iella + SolmunJella, Haeckel, 1879.

Two tentacles. Four peronial strands. Four cleft ( = eight peripheral) stomach-pouches. An apical (exumbrella) sense-organ is present in larva, but does not persist in adult. Derived from jEgina by the disappearance of half ot its tentacles.

HYDROCTENA, DawydofF, 1903.

Two tentacles. No peronial strands. Two simple, uncleft stomach-pouches in the tentacular radii. There is an apical (exumbrella) sense-organ consisting in a ciliated pit containing two lithocyst-clubs. A median axial canal extends upward from the stomach to the bottom of the sensory pit.

CUNOCTANTHA, Haeckel, 1879.

Eight tentacles. Eight peronial strands. Eight simple, uncleft stomach-pouches in the tentacular radii.

unoctona + ^Eginura, Haeckel, 1879.

Eight tentacles. Eight peronial strands. Eight cleft ( = sixteen peripheral) stomach- pouches. The outer margins of these pouches may be still further divided so as to give thirty-two marginal pouches.

S, Brandt, 1835.

Four tentacles. Eight peronial strands. Eight cleft ( = sixteen peripheral) stomach- pouches. Derived from jEginura by the disappearance of half of its tentacles.

CUNINA, Eschscholtz, 1829.

Nine or more tentacles, and an equal number of peronial strands. Peripheral stomach- pouches simple, uncleft and equal in number to the tentacles, in the radii of which they are developed. With otoporpae above the sense-clubs.

16 MEDUS.E OF THE WORLD.

SOLMISSUS, Haeckel, 1879.

Similar to Cunina, but without otoporpae.

CvNlSSA = Cnnissa + ^EginoJorus, Haeckel, 1879.

Nine or more tentacles, and the same number of peronial strands. Peripheral stomach- lobes twice as numerous as the tentacles, being cleft in the tentacular radii.

jEciNODiscus, Haeckel, 1879.

Eight tentacles, sixteen peronial strands. Sixteen cleft ( = thirty-two peripheral) stomach- pouches.

MEDUS/E MILLEPORIN^;.

Under this heading we may place the degenerate, free-swimming medusae of Millepora. They have no velum and are thus separated from the veiled medusae or Craspedotae. Not only is the velum absent, but the medusa is also devoid of a peripheral canal system and of marginal tentacles.

The medusae Milleponna? and Craspedotae are doubtless derived from a common ancestral phylum, but have departed widely, one from the other, so that the Craspedotae are constantly characterized by a diaphragm-like membrane, or velum, which partially closes the opening of the bell-cavity at the tentacular margin; whereas this structure is absent in the Milleporinae.

The only known forms of Medusae Milleporinae are those of Millepora.

Millepora alcicornis (Medusa).

Millepora (medusa*), HICKSON, 1900, Proc. Roy. Soc. London, vol. 66, p. 3, figs. i-io. — DUERDEN, 1899, Journal of the Institute of Jamaica, March, 1899. — HICKSON, 1906, Cambridge Natural History, vol. I, Coelenterata, p. 259, fig. 129.

Each medusa lies in an ampulla, or cavity, of the corallum, and is attached by a narrow stalk to the innermost part of the wall of the cavity. The bell is 0.4 to 0.6 mm. in diameter and is devoid of radial or circular canals or velum. It consists of a median lamella of entoderm covered on the exumbrella and subumbrella sides by an ectodermal epithelium. There are no tentacles, but instead there are 4 or 5 swollen masses of nematocysts 90° or 72° apart, near the bell-margin, but projecting from the sides of the exumbrella. The manubrium is greatly swollen, and fills the greater part of the bell-cavity. A mouth may (or may not ?) be present. There are I to 5, usually 3 or 4, large ova in the ectoderm ( ?) of the manubrium. The central entodermal cavity of the manubrium consists of an axial chamber which often gives rise to 4 perradial pouches which project into the entoderm of the manubrium, and end blindly. In some medusae these pouches do not exist, in other specimens there are but 2 or 3, but probably the most common condition is that of 4 pouches 90° apart. The fully-developed ova occupy positions alternating with the pouches. Dr. Duerden observed that these medusae are set free and swim slowly about with infrequent feeble pulsations. Soon after liberation the ova begin to be discharged into the water and the medusa dies after 5 or 6 hours of life, having discharged all of its eggs.

The Pacific millepores also produce medusae (see Hickson, S. J., 1891, Quart. Journ. Micros. Sci., 1898; Proc. Zool. Soc. London; and Philosoph. Trans. Roy. Soc. London, B, vol. 179).

OKOKIf A.Vmn.MKDI S.K. 17

VEILED MEDUS/E. MEDUSAE CRASPEDOT^E Gegenbaur, 1856.

Cr\ftacar/iir, ESC'HSCHOLTZ, 1829, Syst. drr Acalc'phac.

Gymnopthalmeet FORBES, 184^, British Naked-eyed MeJuvr.

CnispeJala, GEC-.ENBAI'R, 1856, Zeit. fiir wisscn. Zool., Bd. X, p. 217.

Craspcdotf, HAECKEL, 1879, Syst. der Medusen. — MAAS, 1X9}, Ergi-h. ilcr Plankton-Expedition] Bd. 2, K.c.

CHARACTERS OK THK VKII.ED MEDUSJE.

Medusae with a velum or diaphragm which partially closes the marginal opening of the hell-cavity; with ectodermal gonads (i. e., the sexual products u'lifn ripe are commonly found in the ectoderm). Without gastric filaments, \\irh a douhle marginal nerve-ring, one ahmi- and one below the velum. Development either direct or through alternation of generations from hydroids.

Order ANTHOMEDUS^E Haeckel, 1879.

Oceanitit (in part), GEUENBAUR, 1856, Zeit. fur wissen. Zool., Bd. 8, pp. 218, 219.

Tubularix (in part), ACASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, p. 337.

AnthomeJustr, HAECKEI., 1879, System der Medusen, p. 3. — VANHOFFEN, 1891, Zoologischer An/eiRtr, Bd. 14, pp. 439, 442.— MAAS, 1905, Craspedoten Medusen der Siboga Expedition, p. 5; 1897, Mem. M. C. Z. at Harvard Coll., vol. 23, p. 9. — HARTLAUB, 1892, Nachrichten kgl. Gesell. \Vissenschaftcn Gottingen, p. 17; 1897, Nordisches Plankton, Nr. 12, p. 4. — HARGITT, 190^, Bull. U. S. Bureau of Fisheries, vol. 24, p. 29.

CHARACTERS OF THE ANTHOMEDUSyE.

Hvdromedusae in which the gonads are contained within the ectoderm of the manubnum. There are no marginal otocysts or sensory clubs. The hydroids are of the Tubulanan order. We may distinguish three families:

(1) CodoniJif with simple tentacles, and with the gonad in the form of one or more rings encircling the manubrium. No oral tentacles.

(2) CladonemiJtr with feathered or branched marginal tentacles. Gonads ring-like, or more or less separated. With or without oral tentacles.

(3) OceaniJir with gonads confined to the interradial or adradial sides of the manuhnum and separated one from another in the meridians of the principal radii. With or without oral tentacles. With unhranched marginal tentacles.

The bell in the Anthomedusae is without marginal lappets, and usually dome-shaped. The tentacles arise from the bell-margin, not from the sides of the bell. The velum is simple and annular, and is provided with circular muscles serving to produce the periodic con- tractions of the bell. The hell-margin is simple and entire, and there are no otocysts, otohth- clubs, or club-shaped sense-organs. Marginal cirri rarely present. In the young medusa the exumbrella is more or less besprinkled with nematocysts, but these usually disappear or become less conspicuous in the adult; although definite nematocyst tracts often persist over the exumbrella, this being far more commonly seen in Anthomedusas than in other orders of Hydromedusae.

The tentacles usually arise singly from the bell-margin, and are usually simple and unbranched. Ectodermal ocelli are often found upon the tentacle-bulbs, and Romanes showed that in Sarsia these subserved a visual function, and that the medusa was strongly attracted by rays between the red and violet. The main shafts of the tentacles are thickly covered with nematocysts, which may be mounted upon filaments or converted into adhesive organs, as in the Cladonemidae. The tentacle-bulbs are hollow, and connected with the gastrovascular system ot the medusa.

The radial-canals are usually simple, but in the Tiarinae they often give rise to short, blindly-ending diverticula, which may be glandular in function, (ilandular swellings are found in the walls of the radial-canals of L\tnn<ir,-ii iil,-\iinJri, and probably in those of /)y- inorpliosa dubia and Sliikbi-ria hnltcrtita. A ring-canal is present in all genera excepting

18 MEDUS.E OF THE WORLD.

Pachycordyle and the Williadi, where it has become filled with a more or less solid core of entoderm cells. The ring-canal is usually simple, and rarely gives rise to blindly-ending diverticula.

The proboscis, or manubrium, is usually flask-shaped, and the mouth in most of the Codonidae is a simple round, or cruciform, opening, but in the Oceamdae it is surrounded by lips which are more or less folded or crenated. Oral tentacles are found in the Margelmae, and in some Cladonemidae. The stomach is often mounted upon a solid, conical peduncle, but in Turritopsis the peduncle is made up of large, highly-vacuolated, entodermal cells which constitute the walls of the proximal parts of the 4 radial-canals. The medusae are carnivorous, feeding upon Crustacea, fishes, and other medusae or Siphonophorae.

In the Codonidae the gonad encircles the manubrium in a ring-like manner, the mature genital products being found in the ectoderm on the sides of the stomach. In the genus Slabbena there are 2 or more of these genital rings. In the Oceamdae, however, the gonads are restricted to the interradial, or adradial, sides of the stomach. Haeckel, 1879, believed that they were radial in position, corresponding with the radial-canals, but VanhofFen, 1891, showed that this was an error, for they are commonly interradial, or on both sides of each radial line. In Eleutheria, according to Hartlaub, and possibly in other Cladonemidae, the genital products are developed in the ectoderm of a peculiar brood-pouch, which is not connected with the stomach, but is invaginated from the general ectodermal wall of the subumbrella. In the Williadi and in Nemopsis the radial corners of the stomach extend outward along the radial-canals, and the gonads develop upon the sides of these pouches. Eleutheria is successively hermaphroditic, either sex preceding, but in all other Anthomedusae the sexes are separate. The mature genital products are usually found in the ectoderm, although they often originate in the entoderm. The eggs are cast out into the water by the breaking down of the ectodermal walls ot the manubrium, but in some species of Bougatnvillia, or in Margelopsis, or Hybocodon, the larvae may be retained until they have passed into the planula or even into the actinula stage. None of the Anthomedusae are known to develop directly from the egg into medusae, but the hydroid stages of many genera remain unknown. Wherever the sexual development is known it is through hydroids of the Tubulanan order, in which the medusae bud out separately and are not protected in special capsules or sporangia.

Asexual development of medusae is found in Codonidae, in medusae of the genera Hv/>o- codon, Slabberia, and Sarsia, and in the Oceanidae in the genera Cytieis, Dysmorphosa, Bou- gainvillia, Rtithkea, and in the Williadi. In some of the Williadi the medusa-buds are borne upon stolons which arise either from the radial corners of the stomach, or from the forks of the radial-canals. In the case of Sarsia and Hybocodon the budding medusae develop as in hydroids, the entoderm of the parent medusa forming the entoderm of the budded medusa, and the ectoderm of the bud being formed from the ectoderm of the parent. In Rathkea, Chun, 1895, discovered that the budding medusa is formed out of the ectoderm of the parent, although a connection is finally established between the entoderm of the bud and that of the parent a short time before the hud is set free. In Bougainvillia tuobe, on the other hand, I find that the budding medusa is formed entirely from the ectoderm of the parent, no con- nection being established between the entoderm of the bud and that of the parent. In the genus Niobia the tentacle-bulbs develop into medusae and are set free.

The majority of the Anthomedusae undergo a considerable development while swimming freely in the water. In some cases, however, the medusae may, at times, become mature and even discharge their genital products while still attached to the hydroid. This is seen in Pennaria, Podocoryne, Sarsia, and Stylactis. Among the most short-lived medusae are those of Pachycordyle, which have neither tentacles, marginal sense-organs, nor radial nor circular canals.

All of the Anthomedusae are inhabitants of salt water. They are rare in the open ocean far from land, but are abundant along coasts, especially along continental shores, and many of them thrive in harbors where the water may be more or less brackish. They appear to be the simplest, and phylogenetically the oldest, of the Hydromedusae.

ORDER ANTHOMEDlS.i:. 19

Family CODONIDjC Haeckel, 1879, sens, amend.

Sarsiadat (in part), Founts, 1848, British Naked-eyed Medusa;, p. 55.

Sarsiada+ Tubularidtf, McCRADY, 1857, Gymn. Charleston Harbor, p. 21.

Sarsiadtf (in part)-t- Tubularidx+ Pennaridj: (in part), AGASSIZ, L., 1862, Cont. Nat. Hist. (-'. S., vol. 4, pp. ^9, 542.

SarsiadtE+ Orthocorynidx + Tubulands + Pennanda, AGASSIZ, A., 1865, North Amer. Acalephx, pp. 175, 183, |S6, 189.

Codonidz, HAECKEL, 1879, Syst. der Medusen, p. 9. — VANHOFFEN, 1891, Zool. Anzeiger, Bd. 14^.442. — HARGITT, 1904, Bull.

Bureau of Fisheries U.S., vol. 24, p. 29. — HARTLAL-B, 1907, Nordisches Plankton, Nr. 12, pp. 5, 6. Codonina+ Cor\nid<z + Tubularidtt, VON LENDENFELD, 1884, Zool. Anzeiger, Bd. 7, pp. 446, 447.

FAMILY CHARACTERS.

Anthomedusae in which the gonad is ring-like and encircles the manubrium. Mouth with- out oral tentacles or prominent lips. The 4 to 6 radial-canals are simple and unhranched. \Yhen ocelli are present they are found upon the outer sides of the tentacle bulbs. There are no marginal otocysts. The tentacles are neither branched nor feathered.

Medusae of the genera Sarsia, Steenstrupia, Ectopleura, Pennaria, Amalthtfa, Trichorhiza, Hyhocodon, H yJnc/ithys, Coryrutts, and Margt-lopsis are known to develop asexually through alternation of generations from Tubulanan hydroids. Direct development of medusae from the egg is unknown in this family, although medusae are budded asexually from the manubrium of some species of Sarsia, from the basal bulbs of the tentacles of HybocoJon, and from those of Sarsia. Actinula larvae are set free from the manubrium of Margelopsis and Hybocodon.

Haeckel, 1879, considered the presence or absence of an apical projection upon the bell, and the presence or absence of an axial canal extending upward from the stomach into this apex, to be a criterion for the separation of genera. It should be borne in mind, however, as was pointed out by VanhofFen, 1891 (Zool. Anzeiger, pp. 439-446), that the young medusae often lack an apical projection and axial canal, whereas the mature individuals possess these characters. They are also highly variable in development. We have therefore considered these characters to be of specific but not generic value.

A natural classification of the Codonidae can not be based upon the characters of the medusae alone, for the medusae of Pcnnana and Amaltheea are similar, while their hydroids are distinct, the former being Pcnnana and the latter Carymorpha, and an equally remarkable condition is presented by hydroids belonging to the two genera Syncoryne and StaunJi/i, both ot which give rise to medusae belonging to the genus Sarsia. This peculiar case is still further complicated by the fact that at least one species of StauriJia hydroid gives rise to a ClaJonema medusa.

In HyJrichthys we meet with an extraordinary case of parasitism or commensalism of the hydroid, and in Margelopsis we find a free-floating hydranth, recalling the hypothetical ancestral form of the Siphonophorae. Margclnpsis is also interesting in that its medusae form, apparently, a connecting link between the Codonidae and the Bougamvilleae, although they are more closely related to the Codonidae than to the last-named family. Indeed their resemblances to the Bougamvilleae are probably due to mere parallelism and not to blood- relationship.

The Codonidae may conveniently be divided into two subfamilies, as follows:

(1) Srirsi/inn, with simple, unbranchtd, marginal tentacles, which arise singly from the bell-margin.

One or all of the tentacles may he degenerate or absent.

(2) Margelopstna, with marginal tentacles grouped into clusters.

20 MEDUS.E OF THE WORLD.

The following table defines the characters of the genera ot Codomdae.

SUBFAMILY No. i. SARSIAN^E.

With simple marginal tentacles which arise singly from bell-margin. One or all of the tentacles may be rudimentary or absent.

Tribe I. ANEMIDI.

Without tentacles or chymifcrous canals.

Pachycordyle, WEISMANN, 1883; HARCJITT, \^o^=Parvanenius ( ?) MAYER, 1904.

Medusa without tentacles, marginal sense-organs, radial-canals or circular canal. Hydroid: Pachycordyle ', WEISMANN.

Tribe II. AMALTHIADI.

The hydroid is Corymorpha. Medusae have 4 radially placed, rudimentary tentacles and 4 radial-canals. Amalthtza, SCHMIDT, 1854. M.edus*= Amafthaa, HAECKEL.

With 4 rudimentary tentacles, 4 radial-canals, and ring-canal. Genital products on manubrium. Hydroid: Corvmorpha.

Tribe III. PENNARIDI.

Medusae similar to those of Tribe II, but hydroid is Pcnnaria. Pennaria, OKEN, 1815; GOLDFUSS, 1820.

With 4 rudimentary tentacles, 4 radial-canals, and ring-canal. Genital products within manubrium. Hydroid: Pennaria.

Tribe IV. PSEUDOPENNARIDI.

Medusa similar to those of Tribes II and III, but hydroid is Trichorhiza. Trick or kiza, RUSSELL, 1906.

Medusa similar to Pcnnaria, but hydroid is Trichorhiza.

Tribe V. EUPHYSIDI.

With small or rudimentary tentacles situated at bases of all but one radial-canal, and with one or more well-developed tentacles situated at base of remaining radial-canal. Bell may or may not possess an apical projection, or an axial canal extending into it from stomach-cavity. Sleenstrupia, FORBES, 1846= Euph \sa, FORBES, 1848; Eufthysora, MAAS, 1905.

The bell is radially symmetrical and the 4 radial-canals are all of equal length, i long and 3 short tentacles. Hydroid:

Corymorpha. ffybocodon, L. AGASSIZ, 1862= Amphicodon, HAECKEL, 1879.

The bell is asymmetrical. One of the 4 radial-canals is long, the one opposed to it short, and the two other canals of inter- mediate length. One long tentacle arises from the base of the long radial-canal, while the other 3 tentacles are short In old medusae of Hybocodon one is apt to find 2 or more tentacles at the base of the long radial-canal, and the medusa, is in the Amphicodon stage. Hydroid: Hybocodon. Microcampanay FEWKES, 1889.

With 6 radial-canals and 6 tentacles, one of which is well-developed and 5 arc rudimentary. Hydroid unknown.

Tribe VI. DICODONIDI.

With 2 well-developed, diametrically opposite tentacles, and 2 rudimentary tentacles. With or without an apical projection

and with or without axial canal extending upward from the stomach into apex of bell. Dicoeloniums HAECKEL, 1879.

With 4 simple radial-canals and 4 radially placed tentacles; 2 of the tentacles are long, and 2 short. Gonad is ring like and surrounds stomach. No lines of nettling cells over exumbrella.

Tribe VII. SARSIADI.

Bell radially symmetrical. With 4 equally-developed, unbranched, marginal tentacles. With or without an apical projection, and with or without an axial canal extending upward from stomach into apex of bell, (i) The manubrium is encircled by a single ring-like gonad. (2) With 2 or more ring-like gonads upon the manubrium. Sarsia, LESSON, i$4$ = S\ndict\ont A. AGASSIZ, 1862; Codoniumt HAECKEL, 1879.

Manubrium slender and tubular. Each tentacle-bulb bears an abaxial, ectodermal ocellus. Hydroid: S \ncoryne.

Also Staiindta. Hydrichthys, FEWKES, 1888.

Medusa similar to Sarsia, but without ocelli upon the tentacle-bulbs. Hydroid: Hvdrichthys. Eneodonium, HARTLAUB, 1907.

Medusa similar to Sarsia, but stomach is mounted upon a gelatinous peduncle, and tentacles end each in a knob. Ectopleura, L. AGASSIZ, 1862.

8 longitudinal lines of nematocysts extend from bases of tentacles to apex of bell. Hydroid: Ecto pleura. Corynitis, McCRAov, 1857.

Manubrium cruciform in cross-section, and bound to the 4 radial-canals by hollow mesenteries. 4 radially placed marginal tentacles which are covered with wart-like clusters of nematocysts. Their basal bulbs bear ocelli. Hydroid: Syncoryne. Slabber ia, FORBES, 1846= Difwrena, McCRADY, 1857. Dif>urena + Bathvcodon, HAECKEL.

Similar to Sarsia, but with two or more ring-like gonads upon the manubrium. Hydroid: S\ncor\ne.

SUBFAMILY No. 2. MARGELOPSINJE.

With 4 radiallv-placed clusters of marginal tentacles. No oral tentacles. Gonads ring-like, and encircling the manubrium. MargelopsiSy HARTLAUB, 1897.

Marginal tentacles are grouped in 4 radially-situated clusters. Gonad ring-like, and surrounds manubrium. No oral tentacles, mouth is a simple round opening. Medusae develop by budding from a free-floating hydranth, A/drgr/o/fj/i. This interesting genus recalls the Bougainvillidi in its radially situated clusters of marginal tentacles. It resembles the Codonidts in its ring-like gonad, and in the absence of prominent lips and oral tentacles. Pelagohydra, DENDY, 1903.

Medusa similar to Margelopsi* but hydroid is Pelagohydra, and differs from hydroid of Margelopsn in that tentacles arise at irregular intervals from the sides of the hydranth and are not arranged in definite verticils as in Margelopsis. Hydroid is pelagic.

PLATE i.

Fig. I. PaehycorJyle degeneratus, male. Nassau Harbor, Bahama Islands, July 19, 1903.

Fig. 2. Budding hydranth of Pennaria tiarella with female medusa-bud still attached. Agassiz Laboratory, Newport, Rhode Island, July 5, 1895. Found attached to eel grass in shallow water.

Fig. 3. A recently liberated male medusa of Pennaria tiarella. Agassiz Laboratory, Newport, Rhode Island, July, 1895.

Fig. 4. Hydroid stock of Pennaria tiarella, natural size. Agassiz Labora- tory, Newport, Rhode Island, July, 1895.

Fig. 5. Pennaria tiarella. From the coral reef at Tortugas, Florida, May 7, 1905.

fig. 6. Steenstrupia vtrgulata. From a drawing made from life by Dr. Alexander Agassiz at Nahant, Massachusetts, August 21, 1862.

Fig. 7. Steenstrupia rubra, male. Oregon Inlet, Pamlico Sound, North Carolina, November 12, 1904.

Fig. 8. HyhoeoJon forbesii. Tortugas, Florida, May 25, 1905.

Fig. 6, from a drawing by Dr. Alexander Agassiz Figs, i to 5, 7, 8, from life, by the author.

AXTHOMKIHS K I'A( IIV( -iililiVLK. '_' I

DESCRIPTION OF GENERA AND SPECIES OF ANTHOMEDUSjE.

Genus PACHYCORDYLE Weismann, 1883.

Pachycordylt, WEISMANN, 1883, Entstehung Seiualzellen Hydromedusen, pp. 87, 217. HAKI.II r, 1404, Mitth. Zool. Station

Neapel, BJ. |6, p. 553. Paminrmus, MAYER, 1904, Mem. Nat. Sci. Museum Brooklyn Institute Arts and Sti., v.il. i, No. i, p. 6.

GENERIC CHARACTERS.

Codonidae without tentacles, radial-canals, or circular vessel. Manubrium surrounded by a ring-like gonad. The hydroid-stock is Pachycordyle.

Except the medusas of Millepora, these are the most degenerate and short-lived of the free-swimming Hydromedusae. They are even more degenerate than are the medusae of Corymorpha and Pennaria, and may be compared with Eucopflla (R. von Lendenfeld, 1883, Zeit. fur wissen. Zool., Bd. 38, p. 497) and Agastra (C. Hartlaub, 1897, Wissen. Meeresunter- such. Biologisch. Anstalt auf Helgoland, Neue Folge, Bd. 2, Heft i, Abt. 2, p. 504, taf. 22, figs. 5, 8 10). In these last-named medusae we find neither manubrium nor marginal tentacles, hut there are 8 otocysts and a velum, and the radial and circular vessels are well developed.

Pachycordyle weismanni Hargitt.

Pachvcorii\lf lutnmannt, HAROITT, 1904, Mittli. Zool. Station Neapel, Bd. 16, Heft 4, p. 553, plate 21, figs. 1-8.

( ?) Pachyconlyle napolilana, WEISMANN, 1883, Die Entstehung der Seiualzellen bei den Hvdromedusen, Jena, pp. 87, 217.

Medusa pyriform, 2 mm. high, 1.3 mm. wide. Tentacles and marginal sense-organs lacking. No radial-canals. Ring-canal a mere fissure with vestiges of an entodermal linin» near the margin. Velum narrow, with a small opening. Manubrium large, conical, and with- out a peduncle. Ripe ova are in the entoderm, and are discharged very soon after medusa is liberated. Mouth lacking. Manubrium orange or dark-brown, other parts colorless. The medusa swims with a short, jerky motion, but lives only one or two hours.

The hydroid is found in the Bay of Naples growing upon the shell of Fusus rostratits. The colony arises from a delicate, reticulated hydrorhiza. Hydrocaulus sparingly branched, 3 to 8 mm. high. Perisarc dull yellowish-brown, not extending beyond base of the hydranth. Hydranths club-shaped, with subconical hypostome. 8 to 16 irregularly arranged filiform ten- tacles, delicate and thread-like when expanded. Body of hydranth orange or reddish, hypostome whitish. Not more than 2 or 3 medusa-buds are borne on the side of the stem of each budding polyp. Occasionally the medusa-buds develop on the side branches, more commonly on the main stems. The ova originate in the entoderm where they remain until discharged into the water from the manubrium of the medusa. This species may possibly be the female torm of VVeismann's Pachycordyle napolitana. On Weismann's hydroid, how- ever, there were no free medusae, only sessile gonophores. All of Weismann's specimens were males, while Hargitt's were females, and it is possible, as Hargitt states, that the females only give rise to free-swimming medusae.

Pachycordyle degeneratus.

Plate I, fig. I.

Paroanemu* degtneretuit MAVFR, 1904, Memoir-. N.it. Sn. Museum Rrooklvn Institute of Art^. and Sci., vol. i. No. i, p. fi, plate ',. "H- "•

Bell thin-walled, about 0.75 mm. high and 0.3 mm. in diameter. Bi-ll-walls quite rigid, velum powerful and well developed. There are neither tentacles, radial-canals, circular vessel, nor marginal sense-organs. Manubrium spindle-shaped, and about a third as lon» as height of bell. Fluids within the stomach-cavity are maintained in rapid motion, appar- ently through the action of cilia. Near aboral end of bell is a deep conical cicatrice which apparently marks the place of last connection between the medusa and its hydroid stock. Bell is translucent and milky in color, while manubrium is cream-colored. Only 5 specimens, all males, were found in Nassau Harbor, Bahamas, on the nights of July 18 and 19, 1903.

22 MEDUS/E OF THE WORLD.

They swam actively in arcs of circles, but all died early in the morning although maintained in large glass dishes filled with pure sea-water. They appeared to be mature, for sperm was discharged constantly from the sides of the manubrium.

The hydroid is unknown, and possibly the medusa may not belong to the genus Pachy- corJyle, though in the medusa stage it conforms with the generic character of the medusae which are set free from Pachycordyle iveismanni of Naples.

Genus AMALTH^EA Schmidt, 1854.

Corvinorplia (hydroid), SARS, M., 1835, Beskriv og Jagttagelser, p. 7.

Corymorpha (medusa and hydroid), STEENSTRUP, 1854, Vidensk. Meddel. Nat. For. Kjobenhavn, p. 46.

Amahhxa (medusa), SCHMIDT, 1854, Handatlas vergl. Anatomie, p. 13. — HAECKEL, 1879, Syst. der Medusen, p. 38.

Amallhiza, ALLMAN, 1872, Monog. Tubularian Hydroids, p. 393. — LOMAN, 1889, Tijdschrift, Nederland. Dierk. Vereen, Ser. 1,

Deel. 2, p. 270. Corymorpha, HARTLAUB, 1905, Zoolog. Jahrb., Abth. Syst., Suppl. 6, p. 543; 1907, Nordisches Plankton, Nr. 12, p. 75.

GENERIC CHARACTERS OF THE FREE MEDUSA.

Codonidae with 4 rudimentary tentacle-bulbs, 4 radial-canals, and a ring-canal. Hydroid is Corymorpha, but distinguished by the character of its medusae. The majority of species of Corymorpha hydroids do not give rise to free-swimming medusae. Hydroid of Arnalthtta might therefore be called Amalthaa. The medusae of Amalthcra resemble those of Pennaria, but the manubrium is usually longer, and extends beyond the velar opening.

Hartlaub, 1907, includes the medusae of Steenstrufia, Euphysa, and Amalthaa in the genus Corymorpha, for they all arise from identical hydroids; while the medusae differ as follows: Steenstrufia, 3 short, and I long tentacle, and with an apical projection and axial canal above the stomach. Euphysa, similar to Steenstrufia, but without an apical projec- tion or axial canal. Amalthxa with 4 rudimentary tentacles.

Amalthrea sarsii Allman (Medusa only).

Corymorpha sarsii, STEENSTRUP, 1854, Vidensk. Meddel. Naturhist. For. Kjobenhavn, p. 48. — BONNEVIK, 1898, Zeit.fiir wissen.

Zool., Bd. 43, p. 476. — HARTLAUR, 1907, Nordisches Plankton, Nr. 12, p. 86, fig. 82.

Amatthtea sarsi:, ALLMAN, 1871, Monog. Tubularian Hydroids, p. 393. — HAECKEL, 1879, Syst. der Medusen, p. 38. ( ?) Arnaltluea amtebigera, HAFCKEL, Ibid., p. 38, taf. I, figs. 10, ll.

It is not certain that this hydroid gives rise to free-swimming medusae, though when arti- ficially set free the medusae swim about actively.

The bell of the attached medusa-bud is elongate with a subconical basal apex and with fairly thick bell-walls. It is 4 mm. high. There are usually 4 equally developed rudimentary tentacles, but occasionally one of the tentacles is longer than the others. Velum well developed. 4 straight, slender radial-canals. Manubrium spindle-shaped. Mouth a round opening studded with nematocysts. In the male the manubrium projects one-third its length beyond velar opening, but in the female it is shorter and the mouth is at the level of velar opening. Ova are few in number. When immature they are amoeboid, but when older they become spherical and project over the surface of the gonad, attached by short pedicels. The gonad encircles the entire manubrium. Manubrium straw-yellow, tentacle-bulbs light-red. Found oft coast of Norway. Hydroid: Corymorpha sarsn.

Amalthaea uvifera Schmidt.

Amalthira uvifera, SCHMIDT, 1854, Nordisches Plankton, Nr. 12, p. 88, fig. 83.

Corymorpha uvifera, HARTLAUB, 1907, Handatlas der vergleich. Anatomie, p. 13, taf. 9, fig. 2.

This medusiform gonophore is not known to be set free. It is borne upon branched stolons which arise from the sides of the polypite above the basal circlet of long tentacles. Each stolon is tree-like and gives rise to 30 to 40 grape-like medusa-buds, so that each hydroid gives rise to more than 100 buds in various stages of development.

In the medusa-bud (when about to be set free ?) there are 4 large globular, rudimentary tentacle-bulbs. The bell is higher than a hemisphere, with walls thin at apex. Manubrium

ANTHOMKDUS.K— AMAI.TII.KA, I'KN \.\HI.\. -'.',

spindle-shaped, as long as the depth of the bell-cavity. The hydroid is C.or\in<n [>hn uvifern, found at Loppen Island, ahout 10 miles from Hammerfest, at a depth of about I fathom.

Amalthaea vardoensis Loman. Amahh<ta vardiitnsis, LOMAN, 1889, Tijdschr., Nederland. Dierk. Ver., Ser. 2, Dcel. 2, p. 271, teit-fig. 5, taf. 13, figs. 1-9, 15.

Hydroid about 50 mm. high. The thin, transparent perisarc extends only over lower halt of stem. The terminal polypite is sharply set off from the stem by a constriction at its base. Nearly 50 large tentacles in basal circlet at wide base of polypite. Over 100 very short tapering oral tentacles in 7 to 9 circlets. 16 to 20 peduncles arise in a circlet from the sides of the polypite between the basal and oral tentacles. Each peduncle bears a number of medusa- buds, which have 4 radial-canals and 4 equally developed, very short tentacle-bulbs. The hydroid is translucent rose-colored. Found in Busse Sound at Vardri, 71° N. lat., Norway. For details of histology, see Loman.

Amalthaea (?) Hybocodon(?) januarii Steenstrup.

Corymorpha januarii, STEENSTRUP, 1854, Vidensk. Meddcl. Nat. For. Kjbbenhavn, p. 46.— SARS, 1861, Annals and Mag. Nat.

Hist., vol. 8, p. 356. Amalthtfa januarii, ALLMAN, 1871, Monog. Tubularian Hydroids, p. 394. — HAECKEL, 1879, Syst. der Mi-JuM-n, p. 39.

Described by Steenstrup from a single imperfect hydroid found in the harbor of Rio Janeiro, Brazil. The hydrocaulus is about 150 mm. long and 8 mm. wide. About 80 tentacles in the proximal circlet, and these are about 50 mm. long. The oral circlet was imperfect and can not be described. There were about 40 branched peduncles above the basal circlet of tentacles. These bear numerous medusa-buds having 4 equal tentacle-bulbs, but oblique margins. The free medusae are unknown.

Genus PENNARIA Oken, 1815.

Pennaria (in part), OKEN, 1815, Lehrbuch der Naturgesch., Bd. I, p. 94.

Pennaria (hydroid), GOLDFUSS, 1820, Handbuch der Zoologic, p. 89.

Pennarin (medusa), MI/CRADY, 1857, Gymn. Charleston Harbor, p. 50.

Pennaria + Halocordyle, ALLMAN, 1871, Monog. Tubul. Hydr., pp. 363, 368.

Pennaria, AOASSIZ, L., 1861, Cont. Nat. Hist. U. S., vol. 4, p. 278. — AC;ASSIZ, A., 1865, North Amer. Acal., p. 187.— HARTLAUB,

1907, Nordisches Plankton, Nr. 12, p. 72. — PEEBLES, 1903, Archiv. Entwick.-mech., Bd. [4, p. 55 (regeneration). Pennaria, AGASSIZ and MAYER, 1899, Bull. Mus. Comp. Zool. at Harvard Coll., vol. 32, p. 161 .--CLARKE, 1907, Mem.

Museum Comp. Zool. Harvard Coll., vol. 35, p. 6. GlobicfpSf AC.ASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, p. 344. — VANHOF* EN, 1891, Zool. An/eiger, p. 443. — HAECKF.I , 1879,

Syst. der Medusen, p. 39.

Globicfps (hydroid), AVERS, 1852, Proc. Boston Soc. Nat. Hist., vol. 4, p. 193. Rucorvne (hydroid), LEIDV, 1855, Journ. Acad. Nat. Sci. Phila., ser. 2, vol. 3, part. 2, p. I VI- plate 10, figs. 1-5.

The type-species ot this genus is Pennaria ilistielia ot Goldfuss, 1820, from the Mediter- ranean. Oken's "Pennaria" is wholly indefinite, including as it does Plumularia, Aglao- phenia, etc.

(JKNERIC CHARACTERS.

Codonidae with 4 permanently rudimentary tentacles, which are reduced to mere basal bulbs. The hydroid stock is a Pennaria.

Medusae of the genus Pennaria may become mature, and discharge their genital products while still attached to the hydroid, and at the same time other individual medusae from the same stock may be set free in an immature state. Generally, however, the genital products are discharged a few hours after the medusae are set free into the water.

There is no generic difference between the medusae of Pennaria and medusae belonging to the genus Amalthtsa, but their hydroids are different. The hydroid stock of Amalthaa is Corvnifjrpha, and is closely related to the hvdroids of the medusa genera Hybocodon, St., n strupia, and Ectopleura. The hydroid of the medusa genus Pennariti McCrady is Pennaria Goldfuss (Globiceps Ayers). On account of this decided difference in their hydroids we have separated Pennaria from Amaltheea = (Corymorpha\ for their apparent similarity is only a case of parallelism.

24

MEDUSAE OF THE WORLD.

Pennaria disticha Goldfuss (European).

( ?) Pennaria marina, IMPERATO, 1599, Dell' historia naturale libri ventoto, Napoli, p. 747.

Pennaria cavoiini, SCHNEIDER, 1892, Jena. Zeitsch. Naturw., Bd. 20, p. 435, taf. 14, figs. 49-54 (histology).— Du PLESSIS, 188 Mittheil. Zool. Sta. Neapel, Bd. 2, p. 147. — PHTET, 1893, Revue Suisse Zool., tome i, p. 12, plate I, figs. 7-9.— WEI

, 1881,

is-

1903, colonies).

colonies). Pennaria disticha) BEDOT, 1901, Revue Suisse Zool., tome 9, p. 459; tome 13, 1905, p. 96 (citation of papers to 1850). — GOLD-

fucc iSii-i T-Ian^hiirh t\e*r 7.rwil . 11 Hn

FUSS, 1820, Handbuch der Zool., p. 89. Pennaria symmetric a, CLARKE, 1879, Bull, Mus. Comp. Zool. at Harvard College, vol. 5, p. 240, plate I, figs. 2, 3.

Pennaria , PEEBLES, 1903, Archiv. Entwickelungs-mech., Bd. 14, p. 55, fig. (regeneration).

Sertularia fiennaria, DELLE, CHIAJE, 1841-44, Animali senza vertebra del Regno di Napnh, vol. 5, pp. 5, 17, plate 157, figs. 3,

14, 15; Ibid., 1822, tav. 43.

(tySertuIaria pennaria, LINN^US, 1758, Systema Naturae, edit. 10, p. 813. Seriolara ftennaria, CAVOLINI, 1785, Mem. Polipi marini, Napoli, p. 134, plate 5. Globiceps globator, HAECKEL, 1879, Syst. der Medusen, p. 40. ( ?) Medusa, Euphysa globaior, LEITKART, 1856, Archiv. fur Naturgesch., Bd. 22, p. 28, taf. 2, fig. 4.

Fin. i . — Pennaria disticha, from nature, by the author, from specimens collected by Dr. Lobianco in the Bay of Naples. Ay hydrocaulus, natural size. B, terminal branches showing law of growth. C, terminal polypite. D and £, hydranths showing completely-annul.ited pedicels.

ANTHOMKDUS.E — PENNARIA. I'.")

This hydroid is one of the earliest forms known, yet reliable information from the system- atic standpoint is still incomplete. One of the best modern descriptions is that <>t Allman, 1872. I believe the American " Penmn-in tinn-lln" to be closely related to P. Jiiliilm. The only difference appears to be that the terminal ramuli in P. disticha are ringed through- out, whereas in P. tiarella they are usually ringed only at base and summit, although I have seen an occasional one ringed throughout in the American hydroid. (See VV. S. Wallace, 1908, Year Book of the Carnegie Institution.)

In the European hydroid, P. Jisticha, the stems become about 150 to 175 mm. high. The main stem is slightly zigzag and with a uniform growth-curve from base to summit. There are about forty side branches, regularly alternate. Perisarc annulated at the origins of the branches and on the ultimate hydranth-bearing ramuli. Hydranths flask-shaped, with a single verticil of about 10 to 13 basal filiform tentacles each about I to 2 times as long as the body of the hydranth, and each ending in a blunt, slightly swollen tip. About 20 short, stiff, knobbed tentacles, irregularly arranged in 3 verticils, arise from the sides of the hydranth above the basal verticil of tentacles. The medusa-buds are similar to those of Pennaria tiarella and are without ectodermal ocelli upon their tentacle-bulbs. The medusa? usually wither upon the hydroid without being set free, but this often occurs also in P. tiarella, especially in the warm water of Florida.

Weismann, 1883, found that the germ-cells originate in the ectoderm of the inner layer of the bell-nucleus and do not wander from their place of origin, becoming mature in the free medusa, or when the medusa-bud is ripe.

Very elaborate studies upon regeneration, regulation, and restitution in injured colonies were carried out by Cast and Godewski, 1903.

This hydroid is found in the Mediterranean. Pictet, 1893, found it at Amboina, Malay Archipelago. It occurs at Naples, Italy, from May to November.

Pennaria tiarella McCrady (American).

Plate I, figs. 2-5. REFERENCES TO THE AMERICAN HYDROID.

G/obiceps tiarella, AOASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, p. 344. — AVER, 1852, Proc. Boston Soc. Nat. Hist., vol. 4, p. 193. — HAECKEL, 1879, Syst. der Mejusen, p. 39.

Pennarin tiarella, AGASSIZ, A., 1865, North. Amer. Acal.,p. 187, figs. 311-315. — NUTTING, 1901, Bull. U. S. Fish Commission for 1899, vol. 19, pp. 337, 374, figs. 14, 83. — HARGITT, 1900, Amer. Naturalist, vol. 34, p. 387, plates 1-4, 36 figs.; Bull, of the Bureau of Fisheries U. S., for 1904, vol. 24, p. 32, plate 3; 1904, Archiv. fur entwickrlungs-mech. organ- ism, Bd. 18, p. 453, taf. 24-28; 1902, Amer. Naturalist, vol. 35, p. 31 1, figs. 8, 9; p. 597, fig. 36; 1899, Biol. Bulletin Woods Hole, vol. i, p. 35-40, 6 figs, (grafting eiperiments); 1900, Science, New series, vol. 12, p. 340; and 1901, Biol. Bulletin Woods Hole, vol. 2, p. 223 (variations of hydroids and medus.T.) — MCCRADY, 1857, Gymn. Charleston Harbor, p. 51. — SMALLWOOD, 1899, American Naturalist, vol. 33, p. 861, 7 figs, (histology). — HARTLACB, 1907, Nor- disches Plankton, Nr. 12, p. 72, figs. 68-70. — GOLDFARB, 1906, Journ. Experimental Zool., vol. 3, p. 148 (regeneration).

Pennaria liarclla=(P. svmmetrica, CLARKE), CONGDON, 1907, Proc. American Acad. Arts and Sci., vol. 42, p. 464.

Pennaria, j/>., THACHER, 1903. Biol. Bulletin, vol. 4, p. 96.

I'tnritiria gibbosa, AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 278, 344; vol. 3, plate 15, figs. I, 2.

Halcord\Ie tiarella, ALLMAN, i87i,Monog. Tubul. Hydroiils, p. 369.

Kuforyne elegant, LEIDV, 1855, Marine Invert. N. J. and R. I., p. 4, plate 10, figs. 1-5.

THE AMERICAN PENNARIA.

Medusa. — The bell is about 2 mm. in height, and is ellipsoidal, being higher than it is broad. Bell-walls very thin, flexible, and much distorted by the remarkably large ova, which are situated within the ectoderm of the manubrium and often fill the entire cavity of the bell. There are 4 small, rudimentary tentacle-bulbs, without ocelli, I at the base of each radial- canal. Radial-canals straight and quite broad. Velum wide and powerful. Manubrium of male more slender than that of female, the latter being usually greatly distended with 4 or 5 large ova. Several ova are often set free into the water before the medusa is liberated from the hydroid stock. The medusa is commonly mature at the time of its liberation, and it is probable that it does not usually survive for more than a few hours, though Dr. A. Agassiz has maintained them alive for several weeks. The entoderm of the manubrium is rose-pink, and there are a number of deep-pink blotches in the entoderm of each radial-canal. The entoderm of the tentacle-bulbs is pearly-white, as are also the eggs within the manubrium. These colors vary considerably in hue and may be almost white.

2fi

MEDTS.E (>K THE WOULD.

Hydroid. — The hydroid is very abundant upon eel-grass, seaweeds, stones, or wharves, below low tide. The stems arise from a ramifying stolon. They attain a height of 100 to 125 mm. The main stems branch alternately, the longest and oldest side branches being found nearest the base of the stock. The side branches give rise to simple or slightly branched

ramuli from their upper sides. These ultimate ramuli are annulated at base with 5 to 6 rings, and the main stem and primary branches are annulated beyond each branch. The stems are covered with a chitinous, horny perisarc. The polyp-heads are flask-shaped, and the mouth is situated at the ex- tremity of a long, slender, conical throat-tube. There are 2 or 3 verticils of oral tentacles, each row being composed of 5 to 7 short, stiff tentacles, each terminating in a knob-shaped cluster of nem- atocysts. Besides these there is a single circlet of 12 to 16 long, flexible tentacles near the proximal base of the polypite. The medusa-buds develop upon the sides of the polypites between the oral and basal rows of tentacles. There are usually not more than 2 or 3 buds upon each polypite at the same time. The entoderm of this hydroid is white to rose-colored, the ectoderm silvery and translu- cent, and the perisarc horny-yellow to black.

This hydroid ranges from the West Indies and Bermudas to the coast of Maine. Pentiaria gibbosa L. Agassiz, of the Florida coast and West Indies, is probably identical with P. tiarella, but the medusae of the southern form wither on the stem as do those of P. Jisticha. Indeed Agassiz's figure in 1862, shows the ultimate pedicels of his P. gibbosa, from Key West, Florida, ringed throughout as in the Mediterranean Pennaria disticha.

Hargitt, 1900, has made a study of the life-history and development of P. tiarella. Early in the summer the hydroids are found growing on rock-weed, piles of docks, etc.; while late in the summer they take to eel-grass. The early summer brood is not so bright in color, and the medusae mature more slowly than in the late brood. The individual hydranths of the late brood are, however, smaller than those of the early summer brood. The medusae dis- charge their ova soon after liberation, and then die rapidly. The ova are 0.4 to 0.5 mm. in diameter and are heavily laden with yolk. They are creamy-white to orange in color. The cleavage is found by Hargitt to be subject to much individual variability, so that an extra- ordinarily irregular, loosely connected mass of cells is formed, resembling the condition described by Rittenhouse, 1907, in Turritopsis nutricula. No matter how irregularly shaped the embryo may be, it results finally in the formation of a spherical solid morula, and this soon changes into a pear-shaped, ciliated, planula larva. In 5 to 7 days after the beginning of development, the larvae settle down and then develop into small monogastric hydroids covered with ringed perisarc, and provided with whorls of tentacles as in the adult. Hargitt finds that the removal of small parts of the segmenting eggs does not alter the future history of development. The eggs may also be divided at the first or second cleavage and will still develop into normal larvae. Moreover, Hargitt finds that single eggs sometimes separate normally into two masses, each of which gives rise to a normal embryo. The irregularities in cleavage observed by Hargitt have been seen also by Miss Cora J. Beckwith. She finds that the segmentation is mytotic, not amitotic; the nuclear division constantly outnumbering the cytoplasmic so that a syncytium is formed.

Hargitt, 1901, states that hydroid stocks of Pennaria which grow upon eel-grass in shallow water are more pinnatified in their branching, and more highly colored than are those which grow upon stones, etc., in deep water. He also finds that the medusae of the deep-water hydroids are less active than are those developed upon the hydroids which grow in shallow

Flo. 2. — Southern variety of Pennaria tiarella. From life, by the author, Tortugas, Florida, May 27, 1908.

Branch of hydroid and enlarged view of a polypite showing pedicels ringed only at their bases.

ANTHOMEDfS.E — I'ENNAHI \. 27

water, and are often retained until they wither upon the stems after having cast out their genital products.

I have found only the pale form of Pennarin on the Florida Reefs, and its medusae appear to cast out their genital products before being set free, as is described by Hargitt for the deep- water forms of Woods Hole, Massachusetts. In Florida, Pomona grows in tuft-like clusters upon gorgonians. This tropical form is called Pennana gtbbosa by L. Agassiz, but those seen by me can not be separated from the Pennaria of the coast of New England.

Hargitt, 1899, carried out a series of grafting experiments upon Pennaria and other hydroids. He finds that pieces of hydroids of the same species may readily be grafted one upon the other, both in oral or aboral relations, there being little or no evidence of polarity in the regenerative process. Pieces of opposite sex but of the same species will readily unite in any manner, but pieces belonging to hydroids of different genera would not unite. Elaborate studies of this character were also carried out by Gast and Godewski, 1903, on P. Jisticha.

In 1900 and 1901, Hargitt studied the variations of the medusx and hydroids of Pen- naria. The medusa sometimes displays ectodermal blister-like protrusions on its exumbrella, and is variable in other respects.

According to H. Miiller, 1907 fZeit. fur wissen. Zool., Bd. 89), the eggs of Pennaria develop at the expense of the weaker egg-cells of the ovary, which they engulf as food to form the yolk-granules.

Thacher, 1903, shows that in Pennaria, Campaniilana, and EiiJenJnurn, the hydranths, when they degenerate, are absorbed not by liquefaction of their protoplasm, or by the with- drawal of the polyps as a whole; but absorption takes place by the degenerating cells of both ectoderm and entoderm being turned into the digestive tract of the hydroid.

Goldfarb, 1906, finds that light is absolutely essential for the normal growth, develop- ment, and regeneration of Pennaria. He finds, however, that this is true in a sense also for Eudendrium, but not to the same degree, for Eudendrnim ramositm colonies, kept in the dark until all the influence of their previous illumination has been lost, will not then regenerate new hydranths unless they be again exposed to light; but the surprisingly short exposure to light of only 5 seconds will suffice to restart the regenerative process.

Pennaria rosea von Lendenfeld.

Pennaria rosea, VON LENDENFELD, 1884, Proc. Linnean Soc. New South Wales, ser. I, vol. 9, p. 51)4, plate 24, figs. 40, 42. — BALE, 1888, Ibitl., ser. 2, vol. 3, p. 747 — 7*. australii, BALE, 1884, in Cat. Australian Hydroid Zoophytes, p. 45.

Main axial stems 80 mm. long, with about 20 alternately arranged, pinnate side branches. 4 to 6 hydranths on each branch. Hydrorhiza and main stems intensely black and opaque; outer half of each stem light-yellow, proximal half intense black. Hvdranths rose-colored, with 9 to 14 oral, and 7 to 12 filiform, basal tentacles. Medusas are produced on the proximal hydranths. The medusa bell is slender, oval, 2 mm. high, I mm. wide. 4 large rudimentary tentacle-bulbs with a minute external ocellus upon each. Manubrium with sperm or ova fills the entire cavity of subumbrella. Ova discharged after medusa is set free. Color, intense rose.

Coast of New South Wales, Australia. Mature in May.

This species is distinguished from the European and American Pennaria by its tentacu- lar ocelli.

Pennaria adamsia von Lendenfeld.

Pennaria aaamsia, VON LENDENFF.LD, 1884, Proc. Linnean Soc. New South Wales, vnl. 9, p. 595, plate 25, figs. 45-48; plate 26, fig- 49-

MeJusir. — The male medusae are 1.5 mm. long and only 0.7 mm. wide; while the female medusas are 1.5 mm. long and as broad as they are high. The male medusas have 4 radially situated marginal tentacles, about as long as the width of the bell. These tentacles have well-developed basal bulbs with minute ocelli, and the tentacles move about freely. The spermatozoa fill the space between the manubrium and the sides of the subumbrella, and are discharged within an hour after the medusa has been set free. The female medusas are broad, and the tentacles are mere rudiments without ocelli. The bell-cavity is filled with ova, which are soon discharged.

28 MEDUS.E OF THE WORLD.

H ydroid. — The stems are 60 to 80 mm. long, zigzag, and with 15 to 20 alternate branches. The longest branches are nearest the base and are about 12 mm. long. There are about 8 hydranths, 1.7 mm. long, on each branch. These hydranths are slender, the proximal ones only about half as large as the distal primary one on each branch. There are 2 verticils of oral tentacles, each with 4 tentacles alternating in position with those of the other row. These oral tentacles are short and each ends in a nematocyst-knob. The uppermost row of 4 oral tentacles is in the 4 perradii, and the lower row in the 4 interradii. There are 24 long filiform tentacles in a single row at the broad base of the polypite. These are about as long as the polypite itself. The perisarc of the hydrorhiza and main stems is opaque and black, and the side branches are yellow. The hydranths are white.

Coast of New South Wales, Australia. Medusae are produced in March.

This species is distinguished by the minute ectodermal ocelli upon the tentacle-bulbs of the male medusae, and by the length of its tentacles. There are also but 2 verticils of oral tentacles.

Pennaria pacifica Clarke.

Pennaria pacifica, CLARKE, 1907, Mem. Museum Comp. Zool. Harvard Coll., vol. 35, No. l, p. 6, plate I, figs. 1-6.

Hydrocaulus 20 to 35 mm. high. Internodes of the main stem without annulations at their distal ends, and with only I or 2 rings at their proximal ends. Branches alternate. Peduncles taper slightly to the base where there are 2 or 3 annulations. Hydranths with 12 to 14 filiments, and about 16 capitate tentacles somewhat irregularly arranged.

Pinco Island, Gulf of Panama, tropical Pacific.

Distinguished only by its few annulations at the internodes of the main stem. It is probably only a local variety of P. tiarella.

Pennaria ? vitrea Agassiz & Mayer.

Pennaria vitrea, AGASSIZ and MAYER, Bull. Mus. Comp. Zool. at Harvard Coll., vol. 32, p. 161, plate I, figs, l, 2.

Bell is 3 mm. in height; walls thick and rigid. There are 4 rudimentary tentacle-bulbs. Velum not prominent. The 4 radial-canals are straight and narrow. The manubrium in the female specimen was retracted within the cavity of the bell, but it was flask-shaped in the male and projected for a short distance beyond the velar opening. These conditions, how- ever, may be merely different states of contraction and not constant differences; but as we observed only two individuals, one a male and the other a female, we can not be certain upon this point. The ova are large and pyriform, and are grouped in 4 interradially arranged clusters within the manubrium. The mouth-opening is simple, and there are no fimbriations or appendages.

Prominent circular muscles were observed in the ectoderm ot the cavity of the bell in the female, but these were not seen in the case of the male. It is possible, however, that they become apparent only during certain states of contraction. In the female the ova and tentac- ular bulbs were flesh-colored, and the entoderm of the mouth was green. In the male the tentacular bulbs were green, the entoderm of the manubrium pink, and the lips green.

Found among the Fiji Islands, off Kimbombo Island, November 25, and ofF Mbatiki Island, December 5, 1897.

Not having seen the hydroid I am very doubtful concerning the generic identity of this medusa with Pennaria.

Genus TRICHORHIZA Russell, 1906.

Trichorhiza, RUSSELL, 1906, Proc. Zool. Soc. London, p. 99, plate 5.

GENERIC CHARACTERS.

The medusa is similar to that of Pennaria, but with one tentacle-bulb somewhat better developed than the 3 others. The hydranth is solitary, with a branched, filiform hydrorhiza. The perisarc forms a cup into the cavity of which the hydranth may be partially retracted. There are 2 verticils of tentacles, an oral and a basal; and the medusa-buds arise from the sides of the hydranth between these 2 verticils. The type-species is Trichorhiza brunnea Russell, from the Firth of Clyde, Scotland.

ANTHOMEDUS.E — TRICHORHIZA, STEENSTRUPIA.

29

Trichorhiza brunnea Russell.

Trichorhiza brunnta, RUSSKLL, 1906, Proc. Zool. Soc. London, p. 99, plate 5, figs. 1-2; Abstract, Pruc. Zool. Soc. London, No. 26, p. 6, Feb. 13, 1906.

Mature medusa unknown. When newly set free from the hydroid the medusa is pyriform, with a thin-walled bell, well-developed velum, and 4 rudimentary tentacle-bulbs, one of which is larger than the others. The manubrium when extended is as long as the depth of the bell-cavity, and it is cylindrical, narrow, and provided with a simple, circular mouth- opening which is surrounded by nematocysts. There are 4 simple, straight radial-canals, but no gonads. The manubrium and tentacle-bulbs are yellow, the former being faint and the latter golden in color.

This medusa is liberated by a solitary hydroid which was found on June 29, 1905, at a depth of 17 fathoms in Ethick Bay, Bute, Firth of Clyde, Scotland. The hydrorhiza of the

Trichorhiza was entwined among and around the tentacles of Coryrnorpha nutans. The hydroid is 1 1 mm. in total length, the hydranth itself being only 1.5 mm. long. The hydrorhiza is long, filiform, and sparingly branched, with about 6 simple branches which arise from its lower (aboral) half. The coenosarc does not appear to extend over this branched lower half of the hydrorhiza. Above the upper end of the hydrorhiza the perisarc extends to form a cup-like hydrotheca, which bears 4 transverse grooves. Immedi- ately below this hydrotheca the peri- sarc displays several longitudinal lines. Hydranth conical, 1.5 mm. long and 0.8 mm. wide, with an oral circlet of 7 very short tentacles having knob- shaped ends. There is also a basal circlet of 12 filiform tentacles, which are ringed with nematocysts and are tapering, and fully as long as the hy- dranth itself. 8 or 10 medusa-buds arise at various levels from the sides of the polypite between the basal and oral circlets of tentacles.

The perisarc of the hydrorhiza is straw-colored, and that ot the hydro- theca chocolate-colored. Body of hy- dranth pale reddish-brown, tentacles translucent white.

This hydroid bears some resemblance to Blastothela of Verrill, 1878. Amer. Jour. Sci., ser. 3, vol. 1 6, p. 374.

FK.. 3. — Trichorhiza brunnea, after Russell, Proc. Zool. Soc., London, 1900. Hydroid, and enlarged view of recently liberated medusa.

Genus STEENSTRUPIA Forbes, 1846.

Sleenslrufia, FOR BBS, 1846, Annals and Mag. Nat. Hist., vol. 1 8, p. 287.— LKI:CKART, 1856, Archiv. fiir Naturgesch., Bd.22, p. 29. Slrrnstrupia + Euf>h\!a, HAF.CKEL, 1879, Syst. der Mcdusen, pp. 29, 3 i .— BROWN t, 1895, Trans. Liverpool Biol. Association,

vol. 9, pp. 247, 248. Eu/ihysa ±Sticnitrufia, AI.ASSIZ, L., 1862, Com. Nat. Hist. I'. S., vol. 4, pp. 542, \^\. HAM KM , 1X64, Jena. Zeitsrli. fur

Naturw., Bd. I, pp. 338, 339. -FoRin.s, \X^, Hntisll N,iked-eved Medusa-, pp. 71, 72. Euph\ia, A(.ASSIZ, A., 1865, North Amer. Acal., p. 1*0. UK. MOW. i'jc>4, Bull. Mus. Comp. Zool. at Harvard Coll., vol. VI.

p. 251. — VANHOKFFN, 1891, Zool. An/eiger, lid. 14, p. 443. Ku/>h\sora, MAAS, 1905, Hydromeduscn der Siboga Kipediiion, p. 6. Heterosirf>hanus + Stcenslrupia+ Eaf>h\sa, HARTLALB, 1907, Nordischcs Plankton, Nr. 12, pp. 74, 76, 80.

30

MEDUS.E OF THE WORLD.

The type-species ot this genus is Steenftrupia rubra Forbes, 1846, from the Atlantic coasts of Europe.

GENERIC CHARACTERS.

Codonids with radially symmetrical bell, and with 4 radially situated tentacles, I of which is well developed while the 3 others are short or rudimentary.

This genus is separated from Hybocodon by its radially symmetrical bell. In Hybocodon i radial-canal is long, I short, and 2 of medium length; but in Steenstrupia the 4 radial -canals are all of the same length.

Synopsis of the Characters of MeJuste of the Genus Steenstrupia.

	S. rubra Forbes=	S. aurata= Euphy-	S. tetrabrachia=E.	S. bigelowi Maas.	S. australis von
	S.flaveola Forbes	sa aurata Forbes,	tetrabrachiaH.B.		Lendenfeld.
	= S. galanthus	= E. aurata+E.	Bigelow.
	Haeckel =(?)£.	mediterranea
	gracilis Brooks=	Haeckel = (?) E.
	( ?) S. cranoides	virgulata A.
	Haeckel=S. 1m-	Agassiz.
	eata Leuckart.
Shape and size	Bell has well-devel-	Dome-like apex.	Pyriform. 4 high,	Pyriform. 10 high,	Half-egg-shaped,
of bell in mm.	oped apical pro-	Cylindrical sides.	2.5 wide.	5 wide.	2.5 high, 1.7
	jection. 5 high,	6 high, 4.5 wide.			wide.
	3 wide.	Bell-walls and
		apex thick.
Condition of ten-	3 long, narrow ten-	3 small, similar,	3 short tentacles,	3 short tentacles	3 tentacle-bulbs,
tacles. Length	tacle-bulbs with-	rudimentary ten-	each= r long, i	about 2 r long	and i long tenta-
in terms of bell-	out tentacles, all	tacle-bulbs, i	large tentacle	and not ringed.	cle, ringed with
radius (r).	similar, i long	long tentacle.	about 7 T long.	One longtentacle	nematocysts.
	tentacle ringed	Length = 2 r+ .	All tentacles	4 T long, with	Long tentacle
	with nematocysts.	No ocelli. No	sparsely ringed.	many prominent	about 8 r long.
	Length = 7 r.	definite rings of		partial rings of
		nematocysts on		nematocysts.
		tentacles.
Axial-canal above	Axial canal always	No axial-canal.	No axial-canal.	Present. Small,	No axial-canal.
stomach.	present.			slender.
Length of manu-	1 .5 r. Stomach	1.25 r. No ped-	2. 5 r long. Mounted	1.5 r long. Mouth	1.5 T long. Cylin-
brium in terms	mounted on short	uncle.	on very short ped-	not quite reach-	drical.
of bell-radius	peduncle.		uncle ? Mouth	ing level of velar
(<•).			extends slightly	opening.
			beyond velar
			opening.
Gonads.	Single gonad en-	Single gonad en-	Eggs developed in 4 wide interradial,	Gonad encircles
	circles stomach,	circles stomach.	8 adradial, longi-	ectodermal swell-	manubrium.
	leaving mouth and		tudinal rows, in	ings on sides of
	peduncle free.		ectoderm on sides	mid-region of man-
			of stomach.	ubrium. Separated
				by 4 narrow, per-
				radial spaces.
Color.	Tentacle-bulbs and	Variable. Tenta-	Gonads brownish-	Entoderm of man-	Mouth deep violet.
	stomach light-	cle-bulbs may be	yellow. Mouth	ubrium, canals,	4 brown patches
	pink to reddish-	yellow, orange,	pink. Tentacle-	and tentacles dull	on gonad. Tenta-
	brown. Bell apex	red, or colorless.	bulbs and rings	yellow .	cle-bulbs and long
	sometimes faint	Manubrium red-	pink.		tentacle brown
	pink .	dish to yellow or			with violet spots.
		faint purple.
Where found.	Atlantic coasts of	Mediterranean and	Suvadiva Atoll,	Malay Archipelago.	Harbor of Sydney,
	Norway, Ger-	Atlantic coasts of	Maldive Islands,		New South
	many, England,	Europe.	Indian Ocean in		Wales, Australia.
	Scotland, and Ire-		January.
	land. (Mediter-
	ranean?) (South-
	ern United States,
	Atlantic coast?)
Hydroid.	Corymorpha nu-	Corymorpha nana,	Unknown.	This medusa is pro-	Unknown.
	tans, Sars, 1835;	Alder, 1857.		bably identical
	Hincks, 1868;			with S. tetrabra-
	Allman, 1871.			chia. Hydroid
				unknown.

ANTHOMEDUS.E — STEENSTRUPIA. 31

The hydroid generation is Corymorpha Sars, i$3$ = Halatractus of Allman, 1871.

Haeckel, 1879, restricts Steenstrupia to describe medusae having characters as above, but with an apical projection upon the bell, and with an axial canal extending upward into this projection from the stomach. Euphysa he would restrict to include forms which lack an apical projection and an axial canal. As Vanhoffen, 1891, rightly states, an apical projection is always lacking in young medusae and is highly variable in its development even in mature individuals of the same species. The same is true of an axial canal. Moreover, among species discovered since Haeckel wrote his "System der Medusen," Euphysa tetrtibrachin, Bigelow, 1904 (Bull. Mus. Comp. Zool. at Harvard Coll., vol. 39, p. 251, plate I, fig. i), has a well-developed apical projection, and no axial canal, and would therefore have no place in Haeckel's system.

Maas, 1905, would institute a genus Euphysora to describe Codonidae having I large, and 3 well-developed but still considerably smaller tentacles. Here again, however, we meet with a condition of degree only. When, for example, are the 3 small tentacles large enough to be Euphysara or small enough to be Steenstrupia ? In order to avoid confusion, it appears best to combine all of these forms under one generic name. Vanhb'ffen, 1891, suggests Euphysa as the proper generic name to include all of these forms, but Steenstrupia takes precedence over Euphysa, for it was used by Forbes in 1846, while Euphysa was first used by him in 1848.

Bigelow finds that the eggs of S. tetrabrachia are arranged in 8 adradial longitudinal lines in the ectoderm of the stomach, and Maas states that in S. bigelou'i the gonads are interradial and separated by very narrow linear interspaces in the 4 principal radii. These conditions are interesting, for they foreshadow those characteristic of the Oceanidae where the gonads are interradial or adradial, and are often quite widely separated in the principal radii.

Hartlaub, 1907, defines Heterostephanus (Allman, 1871) as a Steenstruf>ia-\\ke medusa-bud, not known to be set free from its Coryrnorpha-\\\ie hydroid. Medusa with a single tentacle, ringed at its base, and terminating distally in a knob. The only known species is H . annuli- cornis, Allman, l^l=Hybocodon annulicornis Haeckel. This was first described by M. Sars, 1859; as Corymorpha ? annulicornis.

Steenstrupia rubra Forbes.

Plate I, fig. 7.

Corymorpha nutans (hvdroid), SARS, M., 1835, Beskriv. og Jagtt., p. 6, plate i, fig. 3; SARS, O., 1877, Fauna Littor. Norveg.,

tome 3, p. 2, taf. 2, figs. 25-28.

Corymorpha nutans (hvdroid), BEDOT, 1905, Revue Suisse de Zool., tome 13, p. 63 (literature to 1850). Corymorpha nutans (hydroid and medusa), HARTLATB, 1907, Nordisches Plankton, Nr. 12, p. 76, figs. 72-75 (complete li

authors, localities, and best modern description of hydroid and medusa). Corymorpha nutans, HINCKS, 1868, Hist. British Hvdroid Zoophytes, p. 127, plate 22. — AI.LMAN, 1871, Monog. Tubularian

Hydroids, p. 388, plate 19. Corymorpha nuians = S. galanthus (Haeckel)=.S. faveola, Forbes = S. rubra, Forbes, BROWNE, 1896, Proc. Zool. Soc. London,

pp. 463, 464, plate 16, fig. i (review of literature). Steenstrupia galanthus, BEDOT, 1905, Revue Suisse de Zool., tome 13, p. 148 (literature to 1850). — HAECKEL, 1879, Syst. ilrr

Medusen, p. 31. Hybocodon ntttans, Ibid., p. 34. — MULLER, 1908, Zeit. fur wissen. Zool., Bd. 89, p. 52, taf. 5, figs. 20-24

(origin and structure of the eggs). Steenstrupia rubra, HARTLAUB, 1904, Wissen. Meeresuntersuch. Kommiss. Meerc Kiel, Abth. Helgoland, Neue Folgc, Bd. 5,

p. 105, fig. 3.

Steenstrupia rubra (Forbes), BROWNE, 1895, Trans. Liverpool Biol. Soc., vol. 9, p. 247.

Steenstrupia rubra (medusa)+S. faveola, FORBES, 1848, British Naked-eyed Medusa, pp. 73, 74, plate 13, figs, i, 2. Steenstrupia cranoides, HAECKEL, 1879, Syst. dcr Medusen, p. 30, taf. 2, figs. 10-14. Steenstrupia gracilts, BROOKS, 1882, Studies Johns Hopkins Vniv. Biul. Laboratory, vol. 2, p. 144. — MAYER, 1900, Bull. Mus.

Comp. Zoo!, at Harvard Coll., vol. 37, p. 29, plate 16, figs. 36, 37. Steenstrupia lintata, LEUCKART, 1856, Archiv. fur Naturgesch., Bd. 22, p. 29, taf. 2, fig. 6. — SPACNOLINI, 1876, Catali-i:" \

Mediterraneo, p. 17, taf. I, figs. 1-4. — HAECKEL, 1879, Syst. der Medusen, p. 30. — DU PLESSIS, 1888, Recueil /.<«.!. s,n re,

tome 4, p. 543.

This medusa is found on the coasts of Europe from Norway southward to the Mediter- ranean. S. cranoides Haeckel = .V. lineata Leuckart, from the Mediterranean, appears to be identical with S. rubra. S. gracilis Brooks, of the Atlantic coast of the United States south of Virginia, is probably identical with S. rubra of Europe, but the hydroid of the American form remains unknown. I believe that 5. Un<-ntn, S. rrtinoiJfs, and .V. i>r/ii-ilis are identical, and that they are probably identical with .V. ruhrn Forbes.

For description of i'. rubra of Europe, see synoptic table of species of Steenstrupia. 3

32

MEDUSAE OF THE WORLD.

In the American form the development of nematocyst-rings upon the long tentacle is subject to great variability. The Mediterranean S. cranoides Haeckel lacks such rings, whereas they appear to be better developed in northern Atlantic specimens from the coasts of Europe. The same difference appears to be exhibited by our American specimens, those from Tortugas, Florida, being unnnged or only slightly ringed, while those from North Car- olina are often profusely ringed. The apex of the bell often bristles with nematocysts, but in some medusae it may be smooth.

Browne, 1896 (pi. 16, fig. i), gives a figure of S. rubra derived from specimens found by him at Valencia Island, off the Irish coast. He shows a narrow conical peduncle above the stomach, the peduncle being about one-fifth as long as the gastric portion of the manubrium. No such peduncle has been observed in the American S. " gracilis," when the bell is expanded, although when it is somewhat contracted the appearance of a well-developed peduncle is often produced. On the other hand, when the stomach is widely distended with food and the bell expanded no peduncle may be seen. Haeckel's series of figures (taf. 2, figs. 10-12) will

3>

FIG. 4. — Corymorpha nulans, after Hartlaub, in Nordisches Plankton. (Young hydroids and mature medusa.)

serve to illustrate the formation of a peduncle-like body of gelatinous substance above the stomach through contraction of the bell. I have frequently seen the same phenomenon in our American S. " gracilis" = S. cranoides Haeckel. I have also observed this peduncle in living medusae of S. rubra taken off the coast of Cornwall, England.

Hartlaub, 1907, gives a list of the bibliography and of localities for this species, and his description of the medusa and the young hydroid are the best yet produced (see fig. 4).

The egg is amoeboid as in Amalth&a. The young hydroid has a single circlet of 4 short, knobbed, oral tentacles, and another circlet of 5 to 8 simple, flexible, filiform basal tentacles. H. Miiller, 1908, finds that the full-grown eggs are very few in number, having developed at the expense of other weaker egg-cells in the ovary, which they devour. The exoplasma is quite wide and is separated from the germinal vesicle. The ooplasma is a network of delicate fibers of wide mesh, and the exoplasma and endoplasma are distinct, one from another. The egg contains numerous pseudo-cells in advanced stages of degeneration and also yolk-granules.

ANTHOMEDUS.E STEENSTRUPIA.

33

The following description of the medusa is derived from a study of specimens found by the author off the coast of the United States:

The bell is 5 mm. high and surmounted by a slender conical projection about 2 mm. long. There are 4 tentacles. One of these is about 10 mm. long, and is ringed at irregular intervals by prominent swellings, between which there are small rings at fairly regular inter-

FIG. 5. — Corymorfha nuians, hydroid and medusa, after Allman, in Ray Society, 1871-72. A' Detail showing manner in which medusa1 bud off from hydranlh.

34

MEDUSAE OF THE WORLD.

vals. The tentacle which is diametrically opposite to the long tentacle is tapering, and about 0.25 mm. long; while the 2 other tentacles are mere bulbs. The velum is well developed. There are 4 narrow radial-canals, and a slender ring-canal. A long, slender, sinuous canal extends from the aboral apex of the stomach upward into the apical projection of the bell. In mature medusae the manubrium extends a short distance beyond the velar opening. Ordinarily the mouth is a simple, round opening at the tapering extremity of the manubrium, but when widely open, as in our figure, it shows 4 thick but not prominent lips. The genital products are found in the manubrium, and in the female the eggs project from the surface of the ectoderm. Entoderm of manubrium intense yellow-green and rose-color. Apical canal intense yellow, often flecked with rose-color. The entoderm of the tentacle-bases is rose- color and yellow, while the entoderm of the large annular swellings is rose-color. When young the apical projection is not very high, the tentacles short, and without nematocyst- nngs, and the manubrium short and tapering; not extending beyond the velar opening as in the mature medusa.

6.

FIGS. 6 and 7. — Sieenslrufia rubra.

6. — From life, by the author. Off Mousehole, Cornwall, England, October 23, 1907. 7. — A. After Leuckart ("S. lineata"), 1856, Archiv. fur Naturgesch., Bd. 22. B. After Spagnolini ("S. lineata"*), 1876, Catalogo Acalefi Mediterraneo.

Found at Oregon Inlet, Pamlico Sound, North Carolina, in November, and at Beau- fort, North Carolina, and Tortugas, Florida, in summer. It is apparently identical with S. cranoides and S. lineata of the Mediterranean.

Definite rings of nematocysts are not found upon the tentacles of the young medusa, and are very variable in their development in mature specimens, some being profusely ringed and some entirely unringed. Haeckel describes only unringed individuals from the Medi- terranean, his specimens being similar to those found by me at Tortugas, Florida.

I am inclined to believe that this Mediterranean and tropical American medusa will prove to be identical with, or only a variety of, S. rubra of the Atlantic coasts of western Europe. Certainly no differences, other than those well within the limits of common vari- ability, can be detected between the medusae of S. rubra and S. lineata = S. cranoides; but a

ANTHOMEDUS.E — STEENSTRUPIA. 35

careful comparative study of the hydroids of these forms must be made before we may safely assert either that they are identical or separate species.

The hydroid of S. rubra is Corymorpha nutans of Sars, and is common on sandy bottoms, off the northern coasts of Europe, at moderate depths. The stems of the hydroid are solitary, and are about 50 to 75 mm. high, and about 4 mm. wide at the widest part. The whole stem is corrugated by numerous narrow longitudinal bands. The widest part of the stem is usually at a short distance above the lower end. This lower end is bent sharply at right angles to the main part of the stem and bears long, hair-like filaments which serve to anchor the hydroid. There are also blunt, papilla-like processes which arise from the sides of the stem above the bent portion. The polypite is large and flask-shaped, and has a basal zone of 30 or more long, tapering tentacles, about as long as the polypite itself. In addi- tion to these tentacles there are 6 to 7 closely crowded verticils of oral tentacles, which are much shorter and thinner than the proximal. 15 to 20 branched peduncles arise from the sides of the polypite close to the bases of the proximal circlet of tentacles, and bear the medusa-buds. The hydranth is light-red, the stem being paler than the polypite. On the English coast the medusa-buds are set free during the summer.

Allman gives a good description of the hydroid.

Steenstrupia aurata.

Euphysa aurata, (medusa) FORBES, 1848, British Naked-eyed Meduss, p. 71, plate 13, fig. 3.

Euphysa aurata (medusa)+E. mcditcrranea, HAECKEL, 1879, Syst. der Medusen, p. 32, taf. 2, figs. 8, 9.

Euphysa medilrrranca, DU PLESSIS, 1888, Recueil Zool. Suisse, tome 4, p. 543.

Euphvsa aurata (Forbes), BROWNE, 1895, Trans. Liverpool Biol. Soc., vol. 9, p. 248; 1896, Proc. Zool. Soc. London, p. 474.

Euphvsa aurata, BROWNE, 1905, Proc. Roy. Soc. Edinburgh, vol. 25, p. 749. — BEDOT, 1905, Revue Suisse de Zool., tome 13,

p. 134. Corymorpha nana, (hydroid) ALDER, 1857, Cat. Zooph. Northumberland and Durham, p. ill, plate 7, figs. 7, 8. — HINCKS, 1868,

Hist. British Hydroid Zooph., p. 130, plate 22, fig. 3.

Corymorpha nana, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 81, figs. 76-78 (list of authors and localities). (?) Euphysa virgulata, AGASSIZ, A., 1865, North Amer. Acal., p. 189, figs. 316-319.

This medusa is found off" the Atlantic coasts of western Europe and in the Mediter- ranean. Steenstrupia virgulata of Massachusetts Bay is probably identical with S. aurata. For description of the European form, see synoptic table of characters of the species of Steen- strupia. The European S. aurata appears to be smaller than the American S. virgulata.

Steenstrupia virgulata = (?) S. aurata Forbes. Plate i, fig. 6.

(?) Steenstrupia aurata, FORBES, 1848, British Naked-eyed Medusae, p. 71, plate 13, fig. 3.

Euphysa virgulata, AGASSIZ, A., 1865, North Amer. Acal., p. 189, figs. 316-319. — HAECKEL, 1879, Syst. der Medusen, p. 33. — NUTTING, 1901, Bull. U.S. Fish Com m., vol. 19^.370. — HARGITT, 1904, Bull. U. S. Bureau of Fisheries, vol. 24, p. 33. Corymorpha virgu/aia, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 84, fig. 79.

Adult medusa. — The bell is pyriform, with a broad, dome-shaped apex. It is 5 to 12 mm. in height, and 4.5 to 9 mm. in diameter. Surface of exumbrella smooth and without rows of nematocysts. There are 4 tentacles, one at the base of each radial-canal; 3 of these are mere rudimentary bulbs, but the fourth is large and conical. Its surface is thickly covered with nematocyst-cells of large size. There are 4 simple, straight radial-canals, and a slender, circular vessel. The velum is wide, with an indented, free edge. The manuhrium is cylin- drical without a peduncle, and extends about half the distance from the inner apex of the bell-cavity to the level of the velar opening. No apical, axial canal. The mouth is a simple, round opening without prominent lips. The genital products are contained within the ectoderm of the manubrium. Manubrium light-yellow. The entoderm of the radial- canals near the bases of the tentacles is intense pink, and the ectoderm of the tentacles is milky-white. This species was found by Dr. Alexander Agassiz, in Massachusetts Bay, at Nahant, and is recorded from Woods Hole, Massachusetts, by Nutting and Har^itt.

The figure here shown is reproduced by his kind permission from Dr. Alexander Agassiz's drawing of the medusa. There is no difference between this medusa and S. aurata of Europe except that the American form appears to be larger; but the hydroid of the American medusa is unknown, and until this is discovered it will be impossible to determine the identity of the American form.

36

MEDUSAE OF THE WORLD.

Steenstrupia tetrabrachia.

Euphysa telrabrachia, BIGELOW, H. B., 1904, Bull. Mus. Comp. Zool. at Harvard College, vol. 39, p. 251, plate I, fig. I. (?) Euphysora bigelowi, MAAS, 1905, Craspedoten Medusen der Siboga Expedition, Monog. 10, p. 7, taf. I, figs. 1-3.

Bell 4mm. high, 2. 5 mm. wide. Pyriform,with solid apical projection. One long sparsely- ringed tentacle about 4 times as long as the bell is high. 3 other, smaller tentacles, each about one-third as long as the bell-height. These bear each about 3 rings of nematocysts, while the long tentacle bears about 6 such rings. Velum well developed. Manubrium large, spindle- shaped, with mouth projecting beyond the velar opening. Gonads on sides of stomach. Eggs arranged in 8 fairly distinct rows. Bell colorless. Gonads brownish-yellow. Manubrium pinkish. Tentacle-bulbs and rings on tentacles rose-pink. Suvadiva Atoll, Maldive Islands, Indian Ocean; in January.

This medusa is probably redescribed by Maas as E. bigelowi.

Steenstrupia bigelowi.

Euphysora bigelowi, MAAS, 1905, Craspedoten Medusen der Siboga Expedition, Monog. 10, p. 7, taf. i, figs. 1-3; 1906, Revue

Suisse de Zool., tome 14, p. 84, pi. 2, figs, i, 2. — MULLER, 1908, Zeit. fur wissen. Zool., Bd. 89, p. 59. (?)Euphysora tetrabrachia, BIGELOW, 1904, Bull. Mus. Comp. Zool. at Harvard College, vol. 39, p. 251, plate i, fig. I.

Fir,. 8. — Steenstrupia tetrabrachia, after Bigelow, in Bull. Museum Comp. Zool. at Harvard College.

FIG. 9. — Steenstrupia bigelowi, after Maas, in Hydromedusen Siboga Expedition.

Bell 13 mm. high; more than twice as high as it is wide, and with a well-developed apical projection. Bell-cavity 10 mm. deep. Apical projection 3 mm. high. Side walls thin. 4 tentacles at the bases of the 4 radial-canals. One of these tentacles is longer than the bell- height and bears about 30 swollen nematocyst-warts. The 3 other tentacles are each one- third to half as high as the bell and are tapering, without nematocyst-warts, but covered with diffuse nettling cells. Manubrium spindle-shaped, not extending beyond the velar opening.

ANTHOMEDUS.E — STEENSTIU'I'IA, HYBOCODON. 37

In most specimens there is a blindly-ending axial canal extending into the gelatinous substance of the apical projection, but this is not constantly present. The gonads are developed upon the sides of the stomach, leaving only the basal and mouth ends of the manubrium free. A cross-section shows that they are separated by 4 minute, perradial, longitudinal lines. There are therefore 4 interradial gonads. There are no medusa-buds produced by the medusa.

The entoderm of the manubrium, radial-canals, and tentacle-bulbs is filled with yellow pigment granules. The nematocyst clusters on all 4 tentacles are red.

Found quite commonly in the Malay Archipelago, at Sulu, Ternate, Damar, Manifa, Saleyer, and Amboina.

This form differs from the closely allied S. letrabrachia Bigelow, from the Maldive Islands, in that in the Maldive species there are a few rings of nematocysts upon the ten- tacles, whereas in S. bigelou'i there are only warts, not inclosed rings. Also there appears to be no axial canal in S. tetrabrachia, whereas this is usually seen in 5. bigelowt. The size constitutes a disparity in the two medusae 4 mm. high in S. tetrabrachia and about 13 mm. in S. bi^elowi. Future studies will probably show that these distinctions are not of specific value, but merely changes due to growth and variation, and that the two medusae are identical and should be called S. tetrabrachia. "Euphysa" tentaculata, Linko, 1905 (Zool. Anzeiger, Bd. 28, p. 214), from Barents Sea has also 3 well-developed tentacles, and is 5 mm. high, with orange-colored manubrium and oval bell. It may be Hybocodon pcndula.

Steenstrupia australis. Euphysa ausiralis, VON LENDENFELD, 1884, Proc. Linnean Soc. New South Wales, vol. 9, p. 586, plate 21, fig. 33.

Bell is 2.5 mm. high, 1.7 mm. wide. Half-egg-shaped and symmetrical. No lines of nematocysts over the exumbrella. One very long retractile tentacle, 2 to 3 times as long as the bell-height. This tentacle has a large basal bulb, and is covered with rings of nemato- cysts. The other 3 tentacles are mere basal bulbs terminating in a knob-shaped cluster of nematocysts. Velum well developed. 4 straight radial-canals. Manubrium arises from the center of the umbrella cavity, and is cylindrical and about half as long as the bell-height. The gonad encircles the manubrium. Mouth deep violet. 4 brown patches upon the gonad, and a few brown spots on manubrium near its base. Tentacle-bulbs and the large tentacle brown with violet spots. Port Jackson, New South Wales, in May and June. Rare.

Hydroid unknown.

Genus HYBOCODON L. Agassir, 1862.

Htbocodon, AGASSIZ, L., i86z, Cont. Nat. Hist. U. S., vol. 4, p. 243.— AGASSIZ, A., 1 865, North Amcr. Acal., p. 193.— VANHOFFEN, 1891, Zool. Anzeiger, Bd. 14, p. 443. — BROWNE, 1896, Proc. Zool. Soc. London, p. 466. — HARTLAI B, 1905, Zoolog. Jahr- buchern, Suppl. 6, p. 544; 1907, Nordisches Plankton, Nr. 12, p. 96.

HybocoJon + ArnphicoJon, HAECKF.L, 1879, Syst. der Medusen, pp. 33, 35.

Amphirodon, BROWNE, 1901, Annals and Mag. Nat. Hist., ser. 7, vol. 9, p. 275.

Diplura, ALLMAN, 1871, Monog. Tubularian Hydroids, p. 326.

Corymorpha, AGASSIZ, A., 1865, /feiW., p. 192.

This genus was established in 1862 by L. Agassiz for Hybocodon prohfer, a medusa which arises by budding from a Corymor^/ja-like hydroid on the New England coast. Accord- ing to Browne, 1896, this medusa is also found off the northern coast of Europe. It is prob- able that the same medusa was described by Steenstrup, 1842, from Iceland, as Corymorpha fritillaria, but the hydroid from which Steenstrup supposed this medusa to be derived is certainly not HvbocoJon, but may be an Arnalthira or Diplura. Steenstrup does not figure a basal circlet of tentacles upon the polypites, which have only an oral circlet, and below this a circlet of medusiform gonophores, each with a 4-sided bell, and 4 equally developed rudimentary tentacles. It is probable that the hydroids of Hybocodon differ more among themselves than do the medusae, and no final classification of the medusae can be attempted until all of the hydroids have been discovered.

GENERIC CHARACTERS.

Codonidae with asymmetrical bell. One of the 4 radial-canals is long. I short, and 2 of medium length. There are I or more long tentacles at the foot of the long radial-canal; and 3 small or rudimentary tentacles, I at the foot of each of the 3 other radial-canals. The hydroid is HybocoJon.

38

MEDUS.E OF THE WORLD.

Haeckel, 1874, instituted the genus Amphicodon to include medusae with 3 rudimentary tentacles, and a cluster of 2 or more long tentacles at the foot of the longest radial-canal. Vanhoffen, 1891, Browne, 1896, and Hargitt, 1901, have pointed out, however, that the young medusae commonly have but I long tentacle, and that others appear, and develop from the side of the basal bulbs of this original tentacle; thus the genus "Amphicodon" is only a mature Hybocodon. These secondary tentacles may appear before or after the medusa- buds begin to develop upon the tentacle-bulbs. According to Hargitt, 1902, 1904, Perkins, 1904, and Linko, 1905, the sexual products of the manubrium become mature while medusa-buds are still being produced upon the tentacle-bulbs.

Synopsis of the Species of Hybocodon.*

	H. prolifer L. Agassiz.	H. pendula Haeckel = Corymorpha pendula L. Agassiz.	H. forbesii Mayer.	H. unicus Browne.
Shape and size of bell	Dome-shaped. 2.5 high,	Quite similar to H. pro-	Ellipsoidal, asymmetri-	Bell-shaped. 3 high,
in mm.	2.2 wide, i radial-	lifer. 5 high, 3.5	cal. 2.5 high, 2.1	2 wide.
	canal long, 2 inter-	wide.	wide.
	mediate length, and i
	short. Long and short
	are 180° apart.
Number of longitudi-	2 extend upward from	As in H. prolifer.	None.	?
nal lines of nemato-	base of well-developed
cysts on exumbrella	tentacle, and I from
	base of each of 3 rudi-
	mentary tentacle-bulbs.
Condition of well-de-	i to 3 well-developed ten-	Only i well-developed	Only i well-developed	Only i well-developed
veloped tentacle, or	tacles at base of longest	tentacle. No medusa-	tentacle. No medusa-	tentacle "between 2
cluster of tentacles,	radial-canal. Medusa-	buds.	buds.	rudimentary basal
at base of longest	buds are produced at			bulbs." No medusa-
radial-canal.	bases of these tentacles.			buds.
	Often only i well-
	developed tentacle pro-
	duced.
Condition of 3 rudi-	Mere basal bulbs.	i mere basal bulb. 2	i short conical tentacle.	3 mere basal bulbs.
mentary tentacles.		small tentacles. Small	2 mere basal bulbs.
		tentacles 90° and	Short conical tentacle
		basal bulb 180° from	1 80° and basal bulbs
		large tentacle.	90° from long tentacle.
Gonads.	Developed over ectoderm	On ectodermal sides of	On manubrium. No	?
	of manubrium. Actinula	manubnuni. No ac-	actinulx attached to
	larvs develop upon sur-	tmulie seen attached	gonad.
	face of gonad in ecto-	to gonad.
	derm of manubrium.
Color.	Entoderm of tentacle-	Pink granules in tenta-	Entoderm of tentacles	?
	bulbs, radial-canals,	cle-bulbs. Entoderm	and manubrium yel-
	and stomach orange to	of manubrium pink	low, flecked with red.
	deep blood-red.	and lilac.
Where found.	North Atlantic coasts of	Coast of New England,	Bahama Islands, and	Stanley Harbor, Falk-
	Europe, Iceland, and	United States.	Florida coast. Com-	land Islands.
	America.		mon in spring months,
			at surface.
Hydroid .	Hydroid described by L.	Hydroid is Hybocodon	Unknown.	Hydroid may be H.
	Agassiz, 1862, as H.	pendula= Corymorpha		chilensis Hartlaub,
	prolifer.	pendula L. Agassiz.		1905.

*For description of H. chilensis Hartlaub, H. chrislinf Hartlaub, H. pu/cher Hartlaub, and H. (?) januarii, see text.

Hybocodon prolifer L. Agassiz. Plate 2, fig. i; plate 3, fig. 3.

(?) Corymorpha fritillaria, STEF.NSTRUP, 1842, Generations-wecksel., p. 20, taf. I, figs. 41-46 (the medusa only may be identical

with //. prolifer; the hydroid appears to be an Amaltheea or Diplura). ( ? ) Stcenstrupia globosa, SARS, 1859, Christiania Vidensk. Selsk. Forhandl., p. 101. — KOREN and DAWELSSEN, 1877, in Sars's

Fauna Littoralis Novegiie, tome 3, p. 20, taf. i, figs. 1-6. (?) Corymorpha annulicornis (young medusa), KOREN and DANIELSSEN, 1877, Ibid., p. 8, taf. i, figs. 7-13.

PLATE 2.

Fig. I. Hybocodon prolifer. Woods Hole, Massachusetts, March 4, 1907. This drawing shows one of the normal aspects of the medusa, with the bell somewhat contracted. When the bell is extended the mouth comes to the level of the velar opening.

Fig. 2. Hybocodon pendula. Newport, Rhode Island, April 23, 1897.

Fig. 3. Hybocodon forbesii. Tortugas, Florida, June 26, 1906.

Fig. 4. Dicodoniurn jeffersoni. Tortugas, Florida.

Fig. 5. Dicodoniurn floriJana. Tortugas, Florida, June 17, 1897.

Drawn from life, by the author.

PLATE 2

ANTHOMEDUSJE — HYBOCODON. 39

Hybocodon prolijer (hydroid and medusa), AOASSIZ, L., 1861, Cent. Nat. Hist. U. S., vol. 4, pp. 143, 343, plate 133, figs. 10,

11; plate 25, 19 figs. Hybocodon proltffr, AGASSIZ, A., 1865, North Amer. Acal., p. 193, figs. 325-328. — ALLMAN, 1871, Monog. Tubul. Hydroids,

p. 422. — VERRILL, 1873, Invert. Animals, Vineyard Sound, p. 736, plate 38, fig. 282. — BOHM, 1878, Jena. Xritschr. fiir

Naturw., Bd. 12, p. 195, taf . 7j figs. 7-9. — NUTTING, 1901, Bull. U.S. Fish Commission, vol. 19, p. 341, fig. 76. — HARGITT,

1904, Bulletin Bureau of Fisheries U. S., vol. 24, p. 33, plate 2, fig. 2; 1901, American Naturalist, vol. 35, p. 580, fig. 39;

Ibid., 1902, vol. 36, p. 552; 1901, Biol. Bulletin Woods Hole, vol. 2, p. 222. — PERKINS, 1904, American Naturalist 38,

p. 516 (simultaneous sexual and asexual reproduction). (?) Hybocodon prolijer, BROWNE, 1896, Proc. Zool. Soc. London, p. 466. Hybocodon prolifer + H.annulicornis + ? Amphicodon fritillaria+ ? A. giobosus + A.amphipleurus, HAKCKEL, 1879, Syst. JerMcdu-

sen, pp. 33, 35, 36, 37, taf. I, figs. 7-9.

(?) Amphicodon fntillaria, BROWNE, 1895, Trans. Liverpool Biol. Soc., vol. 9.

(?) Amphicodon grtn°idum, LINKO, 1905, Zool. Anzeiger, Bd. 28, p. 215 (simultaneous sexual and asexual reproduction). (?) Hybocodon prolifer, BROWNE, 1905, Proc. Roy. Soc. Edinburgh, vol. 25, p. 752. Non Hybocodon prolijer, BONNEVIE, 1899, The Norwegian North Atlantic Expedition, 1876-1878, vol. 26, Hvdroida, p. 28, plate

i, fig. 6. Hybocodon prolijer, HARTLAUB, 1907, Nordisches Plankton, Mr. 12, p. 98, figs. 94-97. — MULLER, H., 1908, Zeit. fur wissen.

Zool., Bd. 89, pp. 62, 73 (origin and structure of the eggs). ( ?) Hybocodon gravidum + H. islandicus + H . amphipleurus, HARTLAVB, 1907, Nordisches Plankton, Nr. 12, pp. 104, 106,

figs. 99, 100.

The following description is derived from studies made of medusae from the southern coast of New England, United States:

Adult medusa. — Bell about 2.5 mm. high and 2.2 mm. wide. It is asymmetrical, the side bearing long tentacles being longer than the other sides; or as Browne aptly describes it, the margin is not at right angles to the longitudinal axis of the bell, but slopes toward the side bearing the cluster of long tentacles. The 4 radial-canals are of lengths corresponding to the sides of the bell. The canal leading to the cluster of long tentacles is the longest; while the canal diametrically opposite to this is the shortest, the 2 other canals being of inter- mediate length. There are 3 small, rudimentary tentacle-bulbs, I at the foot of the shortest canal, and I at the foot of each of the intermediate canals. The cluster of tentacles at the foot of the long radial-canal has wide, hollow, tapering basal bulbs. The mam shaft of each tentacle is, however, slender, cylindrical, and contractile, and is annulated at regular intervals by well-developed clusters of nematocysts. Young medusae commonly have but a single long tentacle, but as growth proceeds they sometimes develop another and finally a third; and all 3 grow to be of equal length, and form a conspicuous cluster. A number of medusa-buds in various stages of development arise from the sides of the hollow base of the one or more long tentacles, near the bell-margin. These medusa-buds themselves develop a single long asymmetrical tentacle even before the bud is mature. When ready to be set free they resemble the parent medusa in that they are sometimes observed to be developing a third generation of medusae upon their tentacle-bulbs. 5 longitudinal lines of nematocyst-cells extend from the bases of the tentacles to near the apex of the bell. 3 of these rows arise from the bases of the 3 rudimentary tentacles, and extend up over the surface of the exumbrella immedi- ately over the radial-canals. In addition to these there are 2 rows which start from the base of the well-developed tentacles, and extend upward over the exumbrella surface on both sides of the long radial-canal. The velum is well developed. The radial-canals are narrow and straight. The manubrium is a simple tube, which usually extends about two-thirds the distance from the inner apex of the bell-cavity to the velar opening. The mouth has 4 short lips with their edges surrounded by nematocysts. The mature eggs are found in the ectoderm of the stomach, and there they develop into actinula larvie before being set free. The entoderm of the tentacle-bulbs is intense orange. The rows of nematocysts upon the exumbrella often display an orange tinge, as does also the entoderm of the stomach.

Hydroid. — The hydroid was found in Massachusetts Bay. growing in tide-pools where the water was very pure. Stems about 50 mm. in height. They grow singly, or in small clusters, and do not branch. Each stem terminates distally in a single large polypite. The stems are not more than I mm. in diameter at the base, gradually enlarging toward upper end, and are about 3 mm. in diameter at base of polypite. They are covered with a delicate sheath of chitinous perisarc, which widens and displays several well-developed annulations near the base of the polypite. The polypite is flask-shaped with a very broad base; the mouth is situated at the extremity of a narrow cylindrical neck, which is capable of much distension. There are 2 oral verticils, each composed of about 1 6 tentacles. The tentacles of the row

40

MEDUS.E OF THE WORLD.

near the mouth are only about half as long as those of the lower circlet. In addition to these there are about 25 long, tapering, hollow tentacles in a zone surrounding the base of the polypite. During the breeding season, which occurs from January until May, great numbers of medusae are developed upon the sides of the polypite immediately above the circlet of basal tentacles. The budding medusae arise singly from the sides of the polypite and are not produced in clusters upon peduncles as in Hybocodon christina, H. chilensis, and H. pulcher. Longitudinal bands of orange pigment extend up the stem of the hydroid. The entoderm

FIG. 10. — Hydroid, young, and budding medusse of Hybocodon prolifcr, after L. Agassiz, in Con. Nat. Hist. U. S., vol. 4.

(A.) Mature hydranth. (B.) Oral eltremity of hydranth. (C.) Dissection showing intertentacular zone of medusa-buds.

Remaining figures are of medusa-buds and young medusae.

of the polypite is orange. L. Agassiz has shown that the entoderm of the stem is thrown into longitudinal ridges which form partial septa projecting into the cavity of the stem. The cavity, however, is continuous, and the septa do not fuse as in some other Hyboconidae.

This species is found upon the New England coast. According to Browne, 1896, it is found upon the British, and also on the Irish coast, at Valencia, and off Iceland (Steen-

ANTHOMEDUS.E — HYBOCODON. 41

strup) and Norway (Sars). It is apparently widely distributed over the North Atlantic, along the shores of continents and islands. It is rarely taken far from some coast. We can not be certain that the American and European forms are identical until the hydroids of both are discovered.

Hargitt, 1904, states that the egg-cleavage is closely similar to that of Pennaria. Also Browne, 1895, and Hargitt, 1902, 1904, find that the ova begin their development within the walls of the manubrium of the medusa and are set free as actinulae. Muiler, 1908, and Hargitt, 1904, find that the developing embryos within the walls of the manubrium absorb their fellow ova, as has been observed by Doflein in Tuhularia mesembryanthcmurn, and by Allen in T. crocea. Developing actinulae and budding medusae are abundant at Woods Hole, Massachusetts, during the spring months.

Hargitt, 1902, and Perkins, 1904, find that actinula larvae develop upon the manubrium at the same time that medusa-buds are being set free from the tentacle-bulbs. When set free the actinulae have 10 tentacles. The mouth and oral zone of tentacles develop only after the actinula is set free; and appear at the pole which was adjacent to the parent medusa during the attached period. Linko, 1905, also observed this simultaneous process of development of medusa-buds and of actinula larvae in his " Amphicodon graviJum," which develops actinulae with 1 1 tentacles. It is probably identical with H . prolifer. A single specimen was found in Barents Sea, north Russia.

H. Miiller, 1908, finds that the ova are large and amoeboid. Only about 2 eggs survive to maturity in the ovary; the others having been devoured by the successful eggs. The ooplasma is a network, the exoplasma being narrow-meshed and the endoplasma wider. There are numerous pseudo-cells in degenerate stages, sometimes dividing amitotically.

Hybocodon pendula Haeckel. Plate i, fig. 2.

Corymorpha nutans (hydroid), STIMPSON, 1853, Marine Invert. Grand Manan, p. 9.

Corymorpha pendula (hydroid), AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 276, 343, plate 26, 6gs. 7-17. — VERRILL,

1873, Invert. Anim., Vineyard Sound, pp. 510, 736, plate 36, fig. 273. Corymorpha pendula, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 85, fig. 81. — NUTTING, 1 901, Bull. U.S. Fish Commission,

vol. 19, pp. 337, 370, fig. 15. — MAY, 1903, American Naturalist, vol. 37, p. 579, n figs, (histology and embryology). Corymorpha pendula (medusa), AGASSIZ, A., 1865, North Amer. Acal., p. 192, fig. 324. Hybocodon pendula, HARGITT, 1904, Bulletin Bureau of Fisheries U. S., vol. 24, p. 34, plate 2, fig. 3. Hybofodon pcndulus (medusa), HAECKEL, 1879, Syst. der Medusen, p. 34. Monocautus pendulus, ALLMAN, 1871, Monograph Tubularian Hydroids, p. 397.

Adult medusa. — Bell pyriform and about 5 mm. in height. It is relatively higher than the bell of H. prolifer, and the gelatinous substance at the apex is much thicker. There are 5 rows of nematocysts upon the exumbrella, as in H. prolifer. The basal bulb of the well-developed tentacle is much smaller than in H. prolifer, and no medusa-buds have been observed to arise from it. The well-developed tentacle is 2 to 3 times the length of the bell-height. Its surface is studded with large, swollen rings of nematocysts, which give it a heavy appearance. The tentacles at the bases of the 2 intermediate radial-canals are quite well developed, and this is not the case in H. prolifer. The velum is wide and thin. The radial-canals are narrow and straight. The manubrium is longer than in H. prolifer, and extends a short distance beyond the velar opening. There is a small peduncle. The lips are thickly covered with nematocysts. Pink pigment-granules are found in the entoderm of the tentacle-bulbs. The entoderm of the manubrium is pink and lilac, and contains also some pink pigment-granules.

Hydroid. — The hydroid (Corymorpha[Hybocodori\ pendula)\s found in depths of IO to IOO fathoms off the New England coast, with its base buried in the sand. It is 80 to 125 mm. in height, and 6 mm. in diameter at the widest part. It always grows singly and is never branched. The mid-region of the stem is very thick and is covered with a canaliculated coenosarc, but the basal end narrows considerably, as does also the region near the free upper extremity, which is long, slender, and pendulous. The stem is anchored by a number of root-like, tubular, fleshy processes. The perisarc exists only as a thin delicate film. The head of the polypite is large and highly contractile. There is a single verticil of long, hollow tentacles at base of polypite. The mouth is situated at the extremity of a large flask-shaped proboscis,

42 MEDUSAE OF THE WORLD.

and is surrounded by a couple of rows of numerous irregularly arranged tentacles. These oral tentacles are highly contractile, and are much smaller than those at the lower base of the polypite. The medusae are borne upon branched stolon-like diverticula of the side walls of the polypite, immediately above the zone of basal tentacles. Fully-developed medusae have not been seen to be set free from the hydroid, but the similarity of the most advanced medusa- buds observed to the free medusa found in the ocean leaves but little doubt concerning this point. May, 1903, has studied the histology and embryology, and concludes that the medusa- buds may at times become free, but usually mature while still attached to the hydranth. This species has been found from Vineyard Sound to the mouth of the St. Lawrence River. The medusa appears upon the southern New England coast in April and May, but is not seen during the summer months, although the hydroid is abundant at this time.

It is possible that Euphysa tentaculata Linko, 1905, is identical with H. pendula (see Zool. Anzeiger, Bd. 28, p. 214). Linko's medusa is from Barents Sea, north of Russia. I have referred to this medusa in the description of Steenstrupia bigelowt.

Hybocodon forbesii Mayer. Plate I, fig. 8; plate 2, fig. 3.

Hybocodon jorbesii, MAVF.R, 1894, Bull. Mus. Comp. Zool. at Harvard College, vol. 25, No. II, p. 236, plate I, fig. l; 1904, Memoirs Nat. Sci. Museum Brooklyn Institute, vol. i, No. i, p. 8, plate 2, fig. 13.

Medusa. — Bell asymmetrical, about 2.5 mm. in height and ellipsoidal in shape, being slightly higher than broad. The gelatinous substance is of uniform thinness. There is a single well-developed tentacle situated at the base of the longest radial-canal. A short, conical tentacle is found at the base of the shortest radial-canal, and two smaller tentacle-bulbs are situated one at the base of each of the intermediate canals. The well-developed tentacle is about as long as the bell-diameter. Its base is small, and hardly greater in diameter than the shaft of the tentacle. Its free extremity is fusiform, and covered with prominent nemato- cyst-cells. No medusa-buds have ever been observed. The velum is narrow. The 4 radial- canals are straight and slender and the circular canal is narrow. There are no rows or clusters of nematocysts upon the exumbrella. Manubrium is spindle-shaped and swollen, and the mouth is a simple, round opening situated at the extremity of a narrow tubular neck, which extends beyond the velar opening. The entoderm of the terminal swelling of the large ten- tacle is yellow streaked with red. Entoderm of manubrium yellow with red flecks.

This species is found in Nassau Harbor, New Providence Island, Bahamas, and at Tortugas, Florida, in March to May. It is an abundant surface form. I have captured many hundreds of specimens, but have never found them producing either medusa-buds or actinula larvae. It is distinguished by its decided yellow and orange color, and the absence of meridional lines of nettle-cells over the exumbrella.

"Hybocodon unicus."

Amphicodon unicus, BROWNE, 1902, Annals and Mag. Nat. Hist., ser. 7, vol. 9, p. 276.

(?) Hybocodon chilensis (hydroid), HARTLAUB, 1905, Zoolog. Jahrbuchern, Suppl. 6, p. 545, fig. W.

Bell 3 mm. high, 2 mm. wide, bell-shaped. I solitary tentacle between 2 rudimentary basal bulbs; 3 perradially situated bulbs without tentacles. Medusa-buds ( ?) Manubrium cylindrical, nearly as long as the umbrella cavity. Color ( ?) Asymmetry of bell ( ?) One specimen was found by Vallentin, and briefly mentioned without figures by Browne, from Stanley Harbor, Falkland Islands. Hybocodon chilensis Hartlaub, from the coast of Chile, may prove to be the hydroid of this medusa ( ?) It will be impossible to identify the medusa from the brief mention of it given by Browne, unless, indeed, it be rediscovered in Stanley Harbor.

Hybocodon chilensis Hartlaub.

Hybocodon chilensis, HARTLAUB, 1905, Zoolog. Jahrbuchern, Suppl. 6, p. 545, fig. W.

(?) Amphicodon unicus (medusa), BROWNE, 1902, Annals and Mag. Nat. Hist., ser. 7, vol. 9, p. 276.

(?) Steenstrupia occidentalis (medusa), FEWKES, 1889, Bulletin Essex Institute, Salem, vol. 21, No. 7, p. 107, plate 3, fig. I.

Hybocodon occidenlalis, HARTLAUB, 1905, Zoolog. Jahrbuchern, Suppl. 6, p. 545.

As Hartlaub states, this hydroid may be the stock of Amphicodon (Hybocodon) unicus Browne, from the Falkland Islands.

ANTHOMEDUS.E — HYBOCODON. 43

The hydrocaulus is very thick and massive, about 50 mm. long and unbranched save for the presence of" its roots and stolons. It is covered with a stiff layer of chitin, which is not expanded at the base of the polypite. Polypite large, with 17 to 20 proximal tentacles, each about 6 mm. long. There are also about 27 oral tentacles arranged in several rows. Above the bases of the proximal tentacles there are 8 long, thick, medusa-bearing stolons, which are thickly covered with numerous clusters of medusa-buds. Each medusa-bud has a single very large tentacle. The stem of the hydroid is rusty yellow and the polypite light rose-color. Found at Calbuco, Chile, South America.

This Hybocodon is closely related to, or possibly identical with, the form from Norway described by Bonnevie under the name Hybocodon prolifer, but it differs from H. prolifer Agassiz in having large, specialized, medusa-bearing stolons. Hartlaub proposes to call this Norwegian hydroid Hybocodon christincc.

Fewkes, 1889 (Bull. Essex Inst., and also Amer. Naturalist, vol. 32, p. 597), gives a brief description of a medusa from the coast of California which he calls by two names, Steen- strupia occidentahs and S. californica; and which may be derived from Hybocodon chilensis. This medusa is described as follows: Size (?) Bell ovoid without an apical prominence. 4 (?) 5 ( ?) rows of meridional lasso-cells extend upward from the 4 tentacle-bulbs, over the exumbrella toward the bell-apex. I long tentacle and 3 rudimentary tentacle-bulbs at the bases of the 4 radial-canals. The long tentacle is ringed at regular intervals and has a large pigmented basal bulb from which there arise numerous medusa-buds. Color ( ?) Velum well developed. 4 straight, narrow radial-canals. Manubrium shorter than the depth of the bell-cavity. No axial canal. Coast of California, United States.

Hartlaub proposes to call this medusa Hybocodon occidentalis. I find nothing in Fewkes's description to distinguish it from H. prolifer L. Agassiz, but apparently there is less difference between the medusas of the various forms of Hybocodon than between their hydroids.

Hybocodon christinae Hartlaub.

Tubular ia prolifer^ Hybocodon prolifer, BONNEVIE, 1899, The Norwegian North Atlantic Eipedition, 1876-79,70!. z6, Hydroida,

p. z8, plate I, fig. 6. Hybocodon chrislin,t, HARTLAUB, 1905, Zoolog. Jahrbuchern, Suppl. 6, p. 546; 1907, Nordisches Plankton, Nr. iz, p. loz, fig. 98.

The medusa attributed by Bonnevie to this hydroid has a single well-developed tentacle with a basal cluster of large medusa-buds, each bud bearing a single tentacle. The medusa- buds resemble H. prolifer, but the hydroid is distinguished by bearing its medusae upon 8 branched peduncles.

Hydrocaulus unramified, tubular, springing from a ramified hydrorhiza; occurrence solitary; longitudinal striping, no collar; height about 50 mm. The hydranth has 14 proxi- mal tentacles and 2 distinct circles of (oral) distal tentacles. The oral tentacles are shorter and more numerous than the proximal. There are 8 blastostyles in a circle about midway between the oral and basal tentacles, and these bear numerous medusae upon short pedicels. The medusa-buds have 4 very wide radial-canals, and I highly developed tentacle which exhibits at its swollen base the bud-rudiments of 4 new medusae even before the first has become detached. This species is distinguished by its well-developed, branched, medusa- bearing stolons. Found off Bodo, Norway. It is closely related to H. chilensis Hartlaub, of the northern coast of Chile.

The medusa-buds in H. christinie appear to be confined to the under side and the sides of the base of the well-developed tentacle.

Hartlaub, 1907, finds that the medusa becomes 4 mm. high and 3 mm. wide, with thin bell-walls and an evenly rounded apex. The 4 radial-canals and ring-canal are band-like, and wider than in other species of Hybocodon.

Hybocodon pulcher Hartlaub.

Auliscus pulcher, SAMUNDSSON, 1899, Vid. Meddcl. Nat. For. Kjobenhavn, Ser. 6, Aarg. i, p. 4Z5, taf. 4, 7 figs.

Hybocodon pulcher, HARTLAUB, 1905, Zool. Jahrbuchern, Suppl. 6, p. 545 ; 1907, Nordisches Plankton, Nr. iz, p. 96, figs. 9Z, 93.

Hydroid 40 to 50 mm. high with an oral circlet of 30 short tentacles, and another circlet of 24 to 30 long tentacles around widest part of body of the hydranth. The only distinctive characters of this species are the well-developed medusa-bearing stolons, the symmetrical

44

MEDUSAE OF THE WORLD.

bell, and 2 large principal tentacles of the budding medusa. In H. prolifer, on the other hand, the medusae are usually set free with but one well-developed tentacle. The bell of H. pulcher may become asymmetrical in later life ( ?) The hydrorhiza is thin and branching. There is a flexible collar-like pensarc at the base of the hydranth. Medusas are developed upon stolons from the hydranth above the circlet of long tentacles. When set free the medusa is 1.5 to 2 mm. high, of symmetrical form. 5 longitudinal lines of nematocysts extend up the sides of the bell to the apex. 3 small tentacle-bulbs. 2 equally well-developed tentacles arise side by side, from the base of one of the radial-canals. Entoderm of manubrium and tentacles red. Found off Iceland. I believe this form is probably identical with H. prolifer.

Genus MICROCAMPANA Fewkes, 1889.

Microcampana, FEWKES, 1889, Amer. Naturalist, vol. 32, p. 595; Bull. Essex Inst., Salem, vol. 21, No. 7, p. in.

The type-species is Microcampana conica Fewkes, from Santa Cruz Island, off the coast of California.

GENERIC CHARACTERS.

Anthomedusae with 6 radial-canals, and 6 radially placed marginal tentacles. One of these tentacles is well developed, and the other 5 are rudimentary.

It is possible that the vaguely described Rhabdoon singulare, of Keferstein und Ehlers, (1861, Zoolog. Beitrage, p. 86, taf. 13, figs. 6, 7), from Messina, Mediterranean, is a form of Microcampana, but there are apparently 12 longitudinal lines of nettling cells over the exumbrella, and it is uncertain whether there are 4 or 6 radial-canals. The bell is 1.5 mm. high and oval with uniformly thin walls. It is possibly an abnormal medusa of P elella.

Microcampana conica Fewkes.

Microcampana conica, FEWKES, 1889, Bull. Essex Inst., Salem, vol. 21, No. 7, p. in, plate 4, fig. 8; American Naturalist, vol. 32, p. 595, fig.

Size ( ?) Bell conical with a well-developed, elongate, conical apex; slightly asymmetrical. Exumbrella smooth, without meridional rows of nematocysts. 6 marginal tentacles, 60° apart. 5 of these are rudimentary, but the sixth is club-shaped, and about half as long as the bell-height. 6 straight, narrow radial-canals and a ring-canal. Manubrium conical to spindle-shaped, about as long as the depth of the bell-cavity. There is a long, slender, straight, axial canal above the stomach. Bell pink, tentacle-bulbs bright-red, manubrium yellow. Found off Santa Cruz Island, California; under the cliffs of Punta Diablo.

Genus DICODONIUM Haeckel, 1879, sens. ampl.

Dieodonium+Dinemas HAECKEL, 1879, Syst. der Medusen, pp. 27, 28.

ATorc Dinema, VAN BEN ED EN, 1867, Mem. Acad.Roy.des Sci.Belgique, 10111.36, art. 2, pp. 127, 130.

Dicodonium, VANHOFFEN, 1891, Zool. Anzeiger, Bd. 14, p. 443.

Sarsiella, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 66.

The type-species of this genus is DicoJonium cornutum Haeckel, 1879, of the Red Sea. Dinema Van Beneden is a medusa which arises by budding from a Perigotiirnus-\\ke hydroid and therefore belongs to the Tiannae. We use the term Dicodonium in the sense proposed by Vanhoffen, 1891.

GENERIC CHARACTERS.

Codonidae with 2 well-developed and 2 rudimentary tentacles. No meridional lines of nettle-cells upon the exumbrella. An apical projection to the bell and an axial canal pro- jecting upward from the stomach, may or may not be present.

Some of the so-called "species" of Dicodonium are probably only abnormal specimens of Sarsia with 2 tentacles instead of the normal 4.

FIG. II. — Microcampana after Fewkes in Amer- ican Naturalist.

PLATE 3.

Fig. I. Dicodonium Jefferson:. Tortugas, Florida, June 15, 1897.

Fig. 2. Sarsia rnirabilis, young medusa. Nahant, Massachusetts, March

26, 1897. Fig. 3. Hybocodon prolijer. t\ primary; t", secondary tentacle beginning to

develop; m, young medusa-buds. Agassiz Laboratory, Newport,

Rhode Island, September 17, 1895. Fig. 4. Sarsia mirabilis, mature male. Nahant, Massachusetts, May 7,

1897. Fig. 5. Hydroid of Sarsia mirabilis with ripe male medusa-bud attached to

the hydranth. Swallows Cave, Nahant, Massachusetts, May 8,

1897.

Drawn from life, by the author.

PLATE 3

ANTHOMKDUS.E — DICODONIUM.

45

Tabular Description of the Medusa of DicoJonium.

	D. cornutum Haeckel.	D. dissonema Haeckel.	D. ocellatum = Dinema ocellatum Haeckel.	D. floridana Mayer.	D. jeffersoni Mayer.	D.adriaticum Graeffe.	D. dinema = Sarsiella dinema Hartlaub.
Shape and size	With bulging	Bell-shaped to	Bell-shaped	Cylindrical with	Dome-like	Bell-shaped	Half-egg-
of bell in	sides, and	egg-shaped	to hemi-	dome-like	0.75 high,	with a small,	shaped. 3
mm.	pointed	with blunt	spherical. 5	apex. 4	0.5 wide.	conical, api-	high, 2 wide.
	apex. 4	conical	high, 4 to 5	high, 3 wide.		cal projec-
	high, 4 wide.	a prx.	wide. Ex-			tion. 4 high,
			umbrella			3.5 wide.
			besprinkled
			with tufts of
			nettle-cells.
Length of 2	Curled up-	6 r ± lone,.	2 r lun^. v. irli	2 r long with a	r long. Basal	8 r long. A	2 r + long.
long tenta-	ward. 3 r	Verv large	fairly large	knob-like	bulbs bear	large car-	Basal bulb-
cles m	long. A	basal bulbs	tentacle-	swelling near	ocelli.	mine ocellus	small; with
terms of	row of bns-	and with an	bulbs, each	tip. Extreme		on each ten-	.'i .'Hi.
bell-radius	thngclusters	abaxial ocel-	with an ab-	tip ends in		tacle-bulb.
«.	of nettle-	lus.	axial ocellus.	thin lash. No
	cells on their			ocelli.
	abaxial
	sides. Tips
	club-shaped.
	Basal bulbs
	small.
Condition of	Not present.	>	Mere bulbs.	Small, taper-	Mere basal	Men- basal	Absent.
2 rudimen-				ing, rudimen-	bulbs with	bulbs with
tary tenta-				tary.	abaxial	small car-
cle-bulbs.					ocelli.	mini- "1 rill.
Axial canal	Well devel-	Present.	Not present.	Not present.	A small axial	Not present.
above stom-	oped .				canal is
ach.					presrnt.
Shape of man-	Spindle-shap-	Spin. He-shap-	Club-shaped.	Flask-shaped,	Cylindrical,	Short, thick,	Spindle-
ubrium,	ed. 1.5 r	ed. 2 to 3	5 to 7 r	widest above,	with a nar-	and four-	shaped,
and length	long. Mouth	times as	long.	but near mid-	row tubular	sided.	6r long.
in terms of	a simple	wide in mid-		dle.	mouth-end.
bell-radius	round open-	dle as at
«•	ing sur-	either end.
	rounded bv
	nettle-cells.
Gonail.	Ring-like, swol-	Large swollen	Gonad en-	Encircling	Large gonad	In stomach	Encircles en-
	len, encir-	gonad en-	circles stom-	stomach.	encircles	wall.	tire stom-
	cling middle	circles stom-	ach, thickest		stomach		ach.
	third of	ach.	near mouth.		from base to
	manubrium				near mouth.
Color.	Stomach, ten-	)	Bell-margin	Entoderm of	Entoderm of	p	Entoderm
	tacles, and		rose-re, 1.	stomach and	stomach		brownish-
	nettle-ring		Nematocyst	tentacles yel-	creamy		yellow.
	of margin		clusters of	low. Knoh-	pink; of
	dark purple-		exumbrella	like ends of	tentacle-
	red.		black.	tentacles	bulbs deli-
			Ocelli of ten-	flecked with	cate green
			tacle-bulbs	red to orange.	or pink.
			black.		Ocelli red.
Where found.	Red Sea, Gulf	Coast of Aus-	Trieste, Adria-	Tortugas, Flor-	Tortugas, Fla.	TruMr, Adria-	Normandy,
	of Suez.	tralia.	tic Sea.	ida. On sur-	M.iv to	tic Sea.	coast of
				face in June	June, com-		France.
				to July, rare.	mon.		Mediterran-
							ean ?
Remarks.	Development	Development	Development	Development	Development	1 ' • 'pment
	unknown.	unknown.	unknown.	unkri"u n.	unknown.	unknown.
			Is this an	Occasionally	•	This species
			abnormal	rudimentary		is distin-
			Sarsia with	tentacles de-		guished by
			only 2 ten-	velop so as to		stiff sensory
			tacies ?	be almost as		hairs which
				long as the		border ocelli
				larger pair.

46 MEDUSAE OF THE WORLD.

Dicodonium cornutum Haeckel.

Dicodonium cornutum, HAECKEL, 1879, Syst. der Medusen, p. 27, taf. I, fig. 6.

Haeckel found this species at Tur, near Sinai, in the Red Sea. See tabular description of medusae of Dicodonium. It has no ectodermal ocelli upon the bulbs of the 2 large tentacles.

Haeckel presents a beautiful figure of the medusa, drawn from life.

Dicodonium dissonema Haeckel. Dicodonium dissonema, HAECKEL, 1879, Syst. der Medusen, p. 27.

Haeckel describes this from a preserved specimen from the coast of Australia. See tabular description of the medusae of Dicodonium.

"Dicodonium ocellatum."

FIG. 12. — Dicodonium cornutum, after Sarsia ocellata, BUSCH, 1851, Beobach. wirbellos. Seeth., p. 16, taf. 2, figs. 1-3. Haeckel, 1879. Dinema ocellatum, HAECKEL, 1879, Syst. der Medusen, p. 29.

This medusa is described by Busch from Trieste, Adriatic, and it is probably an abnormal Sarsia with only 2 of its 4 marginal tentacles developed. See tabular description of the medusae of Dicodonium.

Dicodonium floridana Mayer. Plate 2, fig. 5.

Bell about 4 mm. high and 3 mm. wide, with thin, uniform, vertical walls and a slight apical projection. There are 2 equally-developed, diametrically opposed tentacles, each about three-fourths as long as the bell-height. Near the outer end of each of these tentacles there is a large, knob-like, swollen region, which terminates in a thin, tapering, nematocyst- bearing lash. The knob-shaped swelling is hollow and its cavity is connected with the general gastrovascular system of the medusa by means of a slender tube which extends through the entodermal core of the tentacle. The basal bulbs of the tentacles are not large, and there are no ocelli.

In addition to the 2 long tentacles, there are 2 small, tapering, rudimentary tentacle- bulbs 90° from the large tentacles. The velum is well developed. There are 4 straight, narrow radial-canals and a simple, narrow circular canal. The manubrium is flask-shaped, being narrower at its base than at its middle point. The mouth projects slightly beyond the velar opening, and is a simple, round opening at the extremity of a long, tapering neck. The gonads are within the wall of the manubrium. The entoderm of the stomach is yellow, and that of the distal bulbs of the tentacles yellow flecked with orange. The entoderm of the basal bulbs of the tentacles is also tinged with yellow.

This medusa is occasionally found at Tortugas, Florida, in June. Occasionally a speci- men is taken in which the normally rudimentary tentacles have developed so as to be nearly as long as the pair of large tentacles, thus illustrating the imperfect line of separation between Dicodonium and Sarsia.

Dicodonium jeffersoni Mayer.

Plate 2, fig. 4; plate 3, fig. I. Dinema jeflersoni, MAYER, 1900, Bull. Mus. Comp. Zool. at Harvard College, vol. 37, p. 30, plate 37, fig. 126.

The bell is dome-shaped, higher than a hemisphere, and about 0.75 mm. high. The exumbrella surface is sparsely sprinkled with nematocysts. There are 2 short marginal tentacles, and 2 rudimentary tentacle-bulbs. The tentacles are radially situated, and are covered with numerous small wart-like clusters of nematocysts. There are 4 ectodermal ocelli, I upon the outer side of each of the 4 tentacle-bulbs. The velum is well developed. There are 4 straight, narrow radial-canals and a narrow circular vessel. The manubrium is about as long as the depth of the bell-cavity. It is simple, cylindrical, and tube-like, and the mouth is a round opening at the extremity of a short, cylindrical neck. A simple canal

ANTHOMEDUS^E — DICODONIUM, SARSIA. 47

projects upward from the stomach into the gelatinous substance of the apex of the bell. This is probably only the remnant ot the connection between the medusa and its hydroid stock. The gonad is ring-like, and encircles the stomach, leaving the short proboscis free.

The entoderm of the tentacles and tentacle-bulbs is of a delicate green or pink. The entoderm of the stomach is creamy pink. The ocelli are bright-red, and all other parts are colorless.

This medusa is quite common at the Tortugas, Florida, in May and early June. Although small, it appears to be mature, for sperm is often given off from the gonad of the males.

Dicodonium adriaticum Graeffe. Dicodonium adriaticnm, GRAEFFE, 1884, Arbeit. Zool. Inst. \Vien, Bd. 5, p. 551.

Bell 4 mm. high, 3.5 mm. wide, bell-shaped, with a small, conical, apical projection. 2 long, radially placed tentacles, more than 12 mm. long, with large, thick, basal bulbs, each with a large carmine ocellus upon the abaxial side of the bulb. 2 tentacle-bulbs at the bases of the 2 radial-canals 90° away from the large tentacles. These basal bulbs have small red ocelli. Each of these 4 ocelli are bordered by a ring of stiff sensory hairs. There are 4 small, interradial tentacle-bulbs without ocelli. Thus there are 2 long tentacles, and 6 rudimentary tentacle-bulbs. 4 radial-canals. Stomach short, thick, and 4-sided, with the gonads in the stomach-wall. Mouth simple, with 4 lips. Found at Trieste. Adriatic Sea, in October.

Dicodonium dinema. Sarsiclla dintma, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 67, fig. 63.

Bell oval, half-egg-shaped, 3 mm. high, 2 mm. wide. Exumbrella thickly besprinkled with nematocysts. Only 2 tentacles, 1 80° apart. These are longer than the bell-diameter. They have small basal bulbs, each with a reddish-brown ocellus. No trace of tentacle-bulb or tentacles 90° apart from the well-developed tentacles. Manubrium about 2 times as long as the bell-height. Spindle-shaped and encircled throughout by the gonad. Manubrium, tentacles, and 4 radial-canals brownish-yellow. Found off the coast of Norway and in the Mediterranean ( ?) Is this an abnormal young Sarsia with only 2 tentacles ? Hydroid unknown. Medusa rare.

Genus SARSIA Lesson, 1843.

Sarsia, LESSON, 1843, Hist. Zooph. Acal., p. 333. — AGASSIZ, L., 1849, Mem. Amer. Acad., New Scries, vol. 4, p. 224; 1862, Cont. Nat. Hist. U. S., vol. 4, p. 211. — WAGNER, 1885, Wirbellosen des Weissen Meeres, p. 76. — VANHOFFEN, 1891, Zool. Anzeiger, Bd. 14, p. 442. — HARTLAUB, 1897, Hydromedusen Helgolands, p. 454; 1907, Nordisches Plankton, Nr. 12, p. 7. — VON LENDENFELD, 1884, Proc. Linnean Soc. New South Wales, vol. 9, p. 582. — CHUN, 1895, Bibliothcca Zoologica, Bd. i, Heft. 19, p. 4. — GOETTE, 1904, Zool. Anzeiger, Jahrg. 27, p. 473.

Coryne + Syndictyon, AGASSIZ, A., 1865, North Amer. Acal., p. 175.

Coryne (hydroid), CALKINS, 1899, Proc. Boston Soc. Nat. Hist., vol. 28, p. 336.

Codonium + Sarsia + Svndictvon, HAECKEL, 1879, Syst. der Medusen, pp. 13, 16, 20.

Syncoryne, WEISMANN, 1883, Entsteh. Sexualzellen Hydromedusen, pp. 56, 216.

Syncoryne (hydroid), HARTLAUB, 1905, Zool. Jahrbuchern, Suppl. 6, p. 524. — NUTTING, 1901, Proc. Washington Acad. Sci., vol. 3, p. 165.

Syncoryne (medusa), HARGITT, 1904, Bull. U. S. Bureau of Fisheries, vol. 24, pp. 29, 30.

Syncoryne, BEDOT, 1905, Revue Suisse de Zool., tome 13, p. 119 (citation of all references to 1850).

SynJiclon, AGASSIZ, A., 1862, in L.Agassiz's Cont. Nat. Hist. U. S., vol. 4, p. 340. — MAYER, 1904, Mem. Nat. Sci. Museum Brook- lyn Institute, vol. i, p. 7.

Stenyo, DUJARDIN, 1845, Annales Sci. Nat., ser. 3, tome 4, p. 257.

The type-species of this genus is Sarsia tubulosa of the northern coasts of Europe. This medusa was first described by Lesson, 1843. The hydroid form was first described by Gartner, 1774, in Pallas's Eleunch. Zooph., under the name of Coryne. Ehrenberg, Sars, and Allman introduced the name Syncoryne. Staurtdia producta also gives rise to a medusa which can not be distinguished from Sarsia.

GENERIC CHARACTERS.

Codonidae with 4 long, simple, equally developed tentacles, I at the foot of each radial- canal. The manubrium is tubular and surrounded by a ring-like gonad. There is an ectodermal ocellus upon the outer side of each tentacle-bulb. There are no meridional nematocyst-tracts upon the exumbrella.

48 MEDUSA OF THE WORLD.

An apical projection of the bell may or may not be present; and there may or may not be an axial canal extending upward from the stomach into this projection. The hydroid is Syncoryne or Staundia.

We use the name Sarsia in the sense defined by Vanhoffen, 1891 (Zool. Anzeiger, p. 442).

In 1862, 1865, A. Agassiz described, under the generic name Syndictyon, a Sarsia having reticulate nematocyst-cells upon its exumbrella, and clusters of such cells upon its tentacles. These are, however, only characters of immaturity and largely disappear in the full-grown medusa, which is a true Sarsia in all respects. In 1879, Haeckel formed the genus Codonium to include medusae resembling Sarsia but distinguished by the possession of an apex upon the bell into which a blindly-ending axial canal extends from the stomach. A bell-apex and axial canal are characters which are acquired during growth in varying degrees by almost all species of Sarsia, and are therefore not of generic value.

Two European and one American species of Sarsia produce medusae by asexual budding from the tentacle-bulbs or from the walls of the manubrium. Chun, 1895 (Bibliotheca Zoo- logica, Heft 19, fig. 2), showed that both ectoderm and entoderm of the manubrium take part in the formation of these proliferating medusae; the entoderm of the manubrium forming the entoderm of the daughter medusa, and the same being true of the ectoderm. Sarsia prolifera Forbes, described and beautifully figured by Haeckel, 1879, under the name Codonium codonophorum, produces medusa-buds upon its tentacle-bulbs.

The majority of Sarsia medusae are probably produced asexually by hydroids of the genus Syncoryne, but at least one medusa identical with Sarsia is derived from the hydroid called Staundia Dujardin, 1843. Such medusae may conveniently be placed in a subgenus Staundwsarsia. A medusa which appears to be closely related to Sarsia is produced by the remarkable parasitic hydroid Hydrichth vs.

The generic name Syncoryne was restricted by Allman, 1871-1872, to designate the hydroid which produces the medusa Sarsia. The name Syncoryna was first proposed by Ehrenberg, who applied it to hydroids now known as Clava, Coryne, etc., and in this old sense it does not apply exclusively to the hydroid of Sarsia. By general consent, Allman's name has been accepted in this restricted sense, Calkins, 1899, being almost alone in main- taining that the generic name of the hydroid should be Coryne.

The commonly accepted arrangement is to retain the old name Coryne to include hydroids in which the reproductive elements are produced in fixed sporosacs growing upon the hydranth, while Syncoryne applies to like hydroids which, however, produce free medusae.

Weismann, 1883, found that the germ-cells of both sexes of Syncoryne sarsii originate in the ectoderm of the budding medusa, and do not wander from their place of origin, but become mature in the free medusa. Goette, 1904, finds, however, that in Sarsia the sperm originates and remains in the peripheral ectoderm of the manubrium of the medusa, but the pgg-cells, contrary to Weismann's contention, originate in the entoderm of the medusa-bud while it is still attached to the hydroid, but afterwards they migrate into the ectoderm of the manubrium, where they mature.

Many of the species of Sarsia display considerable individual variability, the colors of the manubrium and tentacle-bulbs ranging from green or yellow to red. An apical pro- jection and an axial vessel above the stomach may or may not be developed, and the length of the manubrium at maturity is subject to much variability. Moreover, the hydroids may form densely or sparingly branched colonies in accordance with environmental conditions, and as is well known in S. mirabihs, free medusae are produced in early spring, whereas late in the season the medusae mature while still attached to the hydroids. Much contusion has been introduced into the synonymy of the genus, and different stages of the same medusa have occasionally received different specific names.

Sarsia " nodosa" Busch, 1851, appears to be the young of some European Sarsia. The exumbrella of very young medusae of Sarsia are usually besprinkled with nettling cells and their tentacles bear prominent nematocysts, and Busch's medusa displays both of these characters. (See Hartlaub, 1907, Nordisches Plankton, Nr. 12, p. 68, fig. 64.)

ANTIKIMKDrs.K — SAKSIA.

49

Tubular Description of the Medusa of Sarsia.

	S.tubulosa Lesson, i843*.	S. mirabilis L. Agassiz •»49-t	S. eximia = Syncor- yne eximia Hincks, 1868.	S.radiata von Lrndenfeld.	S. conica= Codon- ium conicum Haeckd.
Shape and size of	Egg-shaped. Half-	As in S. tubulosa, 7 to	Oval with quite	Semi-ovate. 3	Barrel-shaped with
bell in mm.	ellipsoidal. loto	10 high, 4 to 8 wide.	thick walls. 3	high, 2. 5 wide. long p<
	12 high, 6 to 8		high. 2 wide.		apex. 12 high, 4
	wide. With or				wide.
	without apical
	projection.
Character of 4 ten-	Basal bulbs small;	As in S. tubulosa.	Basal bulbs large,	As in S. eximia	7 r long, with small
tacles. Length	each with abaxial		with abaxial ocel-		basal bulbs.
in terms of boll-	ocellus. Tentacles		lus on each. Ten-
radius (r).	about 4 to 6 r		tacles about 4 r
	long, and without		long-
	prominent clusters
	of nettle-cells.
Shape and length	Narrow, cylindrical	Narrow tubular base.	Cylindrical; wide	As in S. eximia,	Spindle-shaped,
of manubnum	at its base. Mid-	Mid-region swollen and	above, with nar-	i .5 r long or	only 1.5 to 2 r
in terms of bell-	dle part cylindri-	cylindrical. Line of	row, short, tubular	less.	long.
radius (r).	cal, swollen by	demarcation between	neck above mouth
	gonad. Stomach	narrow, cylindrical,	Length about 2 r
	small, spindle-	basal, and wide middle	or less.
	shaped. Mouth at	part very sharp.
	end of short coni-	Stomach swollen and
	cal neck, as in	spindle-shaped and
	S. mirabilis.	near outer end of tin-
	Length 4 r.	manubrium. Beyond
		stomach there is a
		short, tapering, narrow
		neck. Mouth is a
		round opening. Total
		length 3 to 6 r. Average
		4 r.
Gonads.	Ring-like and de-	Thick ring encircling	Encircles manu-	As in S. eximia.	Encircles stomach,
	veloped over	manubrium. Begin-	brium from base		leaving both ends
	nearly the whole	ning abruptly at a short	to mouth end,		free.
	length of the	distance from base,	leaving only short
	manubnum, leav-	and extending to upper	neck above the
	ing a short dis-	part of stomach, leav-	mouth free. No
	tance at both ends	ing both ends of man-	medusa-buds.
	free. No medusa-	ubnum free.
	buds.
Color.	Entoderm of stom-	Entoderm of stomach	Entoderm of stom-	Entoderm of	?
	ach green, bluish	usually green, some-	ach yellowibh-	manubrium
	lilac, or red-	times green and red,	red to reddish-	and tentacle-
	Tentacle-bulbs	or red. Tentacle-bulbs	brown or green.	bulbs deep
	green, blue, or red.	green or red. Ocelli	Ocelli black in	brown.
	Ocelli black.	black.	young and ml m
			mature medusa.
			TentacJe-bulbs
			red, or brownish-
			red.
\Vhere found.	North Atlantic	From coast of New Jrr-	Coast of England,	New South	Indian (Kc.m.
	coasts of Europe	scv northward to Arctic	S( ntland, Helgo-	W.ilrs, Aus-
	to Arctic Ocean.	Ocean. Coasts of Nor-	land, Nnnvav,	tralia.
		wav. northern Russia,	Juneau, Alaska,
		Greenland, Alaska to	North Pacific
		northern Chile along	(Nutting).
		Pacific coast to South
		America.
Hydroid.	Syncoryne sarsia	Svnotrvne "miraluli "	Svncoryne eximia,	Sviu i>rvnr r.uli-	1'iikrmwn.
	Loven. Hartlaub,	L. Agassi/,. S. reticu-	Allman.	aba.
	1905, describes	lata is a closely related
	(this hvdroid ?)	variety (see text).
	from Terra del
	Fuego, Antarctic.

*Description based on accounts by European writers.

j-Description based on original observations of medusa: found off southern coast of New England, United States.

50

MEDUS/E OF THE WORLD.

A number of Syncoryne hydroids have been described which probably produce Sarsia medusae, but are not known so to do. Among these may be mentioned Syncoryne crassa Pictet, 1893 (Revue Suisse Zool., tome I, p. 8), a small hydroid only 2 mm. high and rela- tively thicker and shorter than S. pulchella of Allman, 1871 (Monog. Tubularian Hydroids, p. 279, plate 6, fig. 3). Pictet's hydroid comes from Amboina, Malay Archipelago. It has 30 to 40 short, knobbed tentacles, and the medusa-buds arise singly between the tentacles. The hydrorhiza is net-like, and the hydroid is pale orange in color.

Hartlaub, 1907, has made a masterly study of the genus Sarsia, and shows that the species fall conveniently into two groups: the Eximia group, with short manubrium occupied entirely by the gonad, which extends from its base to near the mouth, and the Tubulosa group,

Tabular Description of the Medusa of Sarsia. — Continued.

	S. prolifera Forbes = S. codono- phora Haeckel. (See text.)	S. gemmifera Forb. = S. siphono- phora Haeckel.	S.hargitti = S.pro- ducta Hargitt.	S. flammea Hart- laub, 1907.	Sarsia (Stauridio- sarsia)producta = Stauridiaproducta Wright, Hartlaub.
Shape and size of	Bell-shaped, with	Bell ellipsoidal.	Pyriform. 1.5 high,	Oval. 12 high, 7	Three-fourths-egg-
bell in mm.	or without apical	8 high, 6 wide.	i wide.	wide.	shaped. 10 high,
	projection. 2.5				7 wide.
	to 8 high, 3 to 8
	wide.
Character of 4 ten-	Basal bulbs very	Well-developed	Well-developed	Tentacle tips knob-	Large basal bulbs
tacles. Length	large and wide,	basal bulbs with	basal bulbs with	like. Shafts cover-	with abaxial ocelli.
in terms of bell-	with black or red	abaxial ocelli.	abaxial ocelli.	ed with broken	Tentacles about 3
radius (r).	ocelli. Tentacles	Tentacles4r long.	Tentacles ir long.	(partial) rings.	to 4 r long.
	hollow, tapering.			No ocelli.
	2 to 6 r long;
	with clusters of
	medusa-buds
	upon their bases.
Shape and length	Spindle-shaped with	Very long, tubular,	Spindle-shaped at	Conical — spindle-	Cylindrical, about
of manubrium	narrow neck and	with spindle-	both ends, narrow-	shaped, only two-	3 r long.
in terms of bell-	simple, round	shaped stomach	er and tubular in	thirds as long as
radius (r).	mouth-opening.	near outer end.	middle. 2 to 3 r	the depth of the
	About i .5 r long.	Mouth at extrem-	long. A circlet of	bell-cavity.
		ity of narrow	medusiform
		neck. Manubrium	gonads( = medusa-
		8 r long.	buds ?) near prox-
			imal end of man-
			ubrium.
Gonads.	Encircling stomach.	Gonad ( ?)	Gonad ( ?)	Ring-like, encirc-	Gonad at base of
	Medusae pro-	Medusae produc-	Medusiform buds	ling the manu-	stomach. No me-
	duced asexually	ed asexually by	on stomach in cir-	brium from its	dusa-buds.
	upon the tentacle-	budding from	clet near base.	base to near
	bulbs.	sides of manu-		mouth.
		brium. There may
		be 10 to 22 or
		more of these
		budding medusae
		upon manubrium
		at one and same
		time. Law of bud-
		ding discussed in
		text.
Color.	Entoderm of tenta-	Entoderm of stom-	Basal part of manu-	Entoderm fiery red	Stomach brownish,
	cles and stomach	ach, tentacles, and	brium orange.	or orange.	gonads yellowish-
	yellow to sage-	radial-canals or-	Distal end blue-		white, tentacle-
	green. Mouth or-	ange-yellow to or-	green. Tentacle-		bulbs red. Ocelli
	ange. Ocelli	ange-red. Ocelli	bulbs orange,		black.
	brown. Tentacle-	black.	edged with green.
	bulbs contain red		Ocelli black.
	entodermal pig-
	ment.
Where found.	English Channel to	Atlantic coasts of	No Man's Land,	Arctic Ocean.	Coast of Great
	Mediterranean.	Europe, from	near Woods Hole,		Britain to Helgo-
		Norway south-	Massachusetts,		land.
		ward to Canary	United States. A
		Islands.	single specimen
			was found.
Hydroid.	Unknown.	Unknown.	Unknown.		Stauridia producta.

ANTHOMEDUS.E — SARSIA.

51

with long, tubular manubrium with the gonad confined to a short length near its free outer end. He states that the Eximia group are represented by such forms as Sarsia eximia, brachygaster, flarnmea, barentst, prolijera, angulata, and apiculata. The Tubulosa group are represented by S. tubulosa, Jensa, decipiens, litorea, pulchella, frutescens, mirabilts, reticulata, princeps, rosaria, etc.

The genus Plotocnide Wagner, 1885, is defined by Hartlaub, 1907, as a Sarsia-Kks medusa with nettle-cells upon the exumbrella. The gonad surrounds the manubrium from the base downward. No ocelli. Hydroid unknown. It appears to me that unless it be proven that the hydroid is different from Syncoryne, this genus should be merged with Sarsia.

Weismann, 1881 (Zool. Anzeiger, Bd. 4, p. 61), shows that the circulation of fluids within the gastrovascular cavity of Coryne pusilla is aided by the rhythmical contraction of the walls of the gonophore. The systole and diastole are not always of equal duration, but each ranges from 60 to 75 seconds. Thus the circulation in certain hydroids may be aided by periodic peristaltic contractions as well as by the movement of cilia.

Annandale, 1907, Journal and Proc. Asiatic Society of Bengal, vol. 3, finds Syncoryne filarnentata, sp. nov., developing free medusae and growing in brackish pools of one-third the salinity of sea water at Port Canning, Lower Bengal. The mature medusa is unknown.

Tabular Description of the Medusa of Sarsia. — Continued.

	S. rosaria= Coryne rosaria A. Agassiz, 1865.	S. minima von Lendenfeld.	S. brachygaster Gronberg, 1898.	S. angulata Mayer.	S. gracilis Browne, 1902.	S. princeps Haeckel, 1879.
Shape and	Oval, with small	Ovate. 3 high,	Three-fourths- Half-egg-shap-	Cylindrical,	Conical with thin
size of bell	apical projection.	2.5 wide.	egg-shaped, ed. 3 High,	thick-walled.	walls. 281040
in mm.	15 to 30 high, 10		widest above 2.8 wide. 4-	5 high, 3 wide.	high, 15 to 30
	to 15 wide.		the middle. 15 sided in con-		wide.
			to 18 high, 8 traction.
			to 10 wide.
Character of	Basal bulbs large;	Tentacles about	Basal bulbs	Basal bulbs	Ocelli on basal	Basal bulbs
4 tentacles.	each with abaxial	2.5 r long.	large with a large, with well-	bulbs. Each	elongate, con-
Length in	ocellus. Tentacles		small abaxial developed ab-	tentacle 3 r	ical, and Banked
terms of	3 to 4 r long.		ocellus. Ten-	axial ocellus.	long, termina-	by pad-like clus-
bell-radius	Basal bulbs		tacles 3 to 4	Tentacles z r	ting in a ni-nia-	ters of nemato-
(r).	flanked by pads		rlong.	long-	tocyst-knob.	cysts. Each bulb
	of nettle-cells.					bears small ab-
						axial ocellus.
						Tentacles 8 r
						long.
Shape and	Spindle-shaped,	As in S. rosaria.	As in S. eiimia.	Shaped as in S.	Two-thirds as	Tubular, cylin-
length of	only 1.510 2 r long.			rosaria. r to	long as the	drical, with a
manubrium				one-twelfth r	depth of the	round mouth-
in terms of				long.	bell-cavity.	opening 4 r
bell-radius						long.
(r).
Gonads.	Encircles stomach,	As in S. rosaria.	As in S. eximia.	As in S. rosaria.	CO	As in S. rosaria.
	leaving both ends
	free. No medusa-
	buds.
Color.	Entoderm of stom-	Entoderm	Entoderm of	Entoderm of	(0	Entoderm of ten-
	ach ranges from	brown.	stomach and stomach and		tacle-bulbs and
	yellow to pink or		tentacle-bulbs tentacle-bulbs		manubrium pur-
	reddish-brown to		orange-red. robin-egg		ple. Ocelli
	purple. Tentacle-		Ocelli black. blue; never		black.
	bulbs yellowish-		red. Ocelli
	brown to red.			deep-brown.
Where found.	Pacific coast of	Port Jackson,	Spitzbergen and Bahamas to	Stanlev Harbor,	Greenland and
	North America,	New South	Greenland, in Tortugas,	Falkland Is-	Spitzbergen, in
	southern Califor-	Wales, Aus-	summer. Florida. Win-	lands.	summer.
	nia to Puget	tralia.	ter to mid-	•
	Sound .			summer.
Hydroid.	Syncoryne rosaria	Syncoryne min-	Unknown.	Unknown.	(? 0	Unknown.
	A. Agassiz and	ima.			Syncoryne sar-
	Fewkes.				sii Hartlaub,
					1905(0 1=
					this medusa a
					young Slab-
					beria ?

52

MEDUS.E OF THE WORLD.

Sarsia tubulosa Lesson.

Syncorvna sarsii (hydroid), LOVEN, 1835, K. Vet. Acad. Handl. for Ar., p. 275, plate 8, figs. 7-10.

Oceania tubulosa (medusa), SARS, M., 1835, Beskriv og Jagtt., p. 25, plate 5, fig. 1 1 . Also: Syncorine sarsii, 1846, Fauna littor. Noveg., part. 2, p. 2, tab. i, figs. 1-6.

Syncoryne sarsii, LOVEN, 1837, Archiv. fur Naturgesch., Jahrg. 3, p. 321, taf. 6, fig. 25. — WEISMANN, 1883, Entsteh. Sexuabellen bei Hydromedusen, Jena, pp. 56, 216. — GARSTANG, 1894, Journal Oxford Club, vol. 2, No. 30, p. 7. — CITRON, 1902, Archiv. Naturges., Jahrg. 68, pp. I, 26, taf. i, 2 (sensory cells of tentacles). — HINCKS, 1868, British Hydroid Zoophytes, p. 52, plate 7, fig. 3. — ALLMAN, 1872, Monog. Tubularian Hydroids, p. 275.

Syncoryne sarsi and Sarsia tubulosa, BEDOT, 1905, Revue Suisse de Zool., tome 13, pp. 120, 147 (all literature 1835-1850).

Sarsia tubulosa, LESSON, 1843, Hist. Zooph. Acalephes, p. 333. — SCHI'LZE, 1873, Ueber den Bau von Syncoryne Sarsii, p. 14, taf. 3.— ROMANES, 1885, Jellyfish, Star-fish, and Sea Urchins, etc., International Scientific Series, vol. 49 (reactions to stimuli). — LINKO, 1905, Zool. Anzeiger, Bd. 28, p. 212. — HARTLALIB, 1907, Nordisches Plankton, Nr. 12, p. 19, figs. 10-16. — BROWNE, 1903, Bergens Museums Aarbog, No. 4, p. 9; 1895, Proc. and Trans. Liverpool Biol. Soc., vol. 9, p. 246.

Sarsia tubulosa, SARS, i$$$=Sarsia macrorhyncha, BUSCH, 1851; BROWNE, E. T., 1905, Proc. Royal Soc. Edinburgh, vol. 25, p. 758.

Syncoryne gravata, HINCKS, 1868, Hist. British Hydroid Zooph., p. 53.

Sarsia tubuhsa+S. macrorhyncha, HAECKEL, 1879, Syst. der Medusen, pp. 16, 19.

Sarsia tubulosa + S. pulchella, FORBES, 1848, British Naked-eyed Medusa-, pp. 55, 57, plate 6, figs. ^, 3.

Corynt pusilla, AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, p. 340.

(!) Syncoryne Sarsii, HARTLAUB, 1905, Zoolog. Jahrbiichern, Suppl. 6, p. 525, fig. F (hydroid from Terra del Fuego).

Sarsia macrorhyncha, BUSCH, 1851, Beobach. wirbell. Seeth., p. lo, taf. 3, figs. 7-10; taf. 4, figs. I, 2.

FIG. 13. — "Sarsia litorea," after Hartlaub, in Nordisches Plankton.

FIG. 14. — "Sarsia decipiens," after Hartlaub, in Nordisches Plankton.

FIG. 15. — "Syncoryne gravata," after Hincks, in British Hydroid Zoophytes — probably hydroid of Sarsia

T^l- ' . _J ' r 1 _£T^l__ TT 1' t_ . - _1 ' ^1_ - ! L..^ :^ J' _

mirabilis

This medusa is found off the English coast early in the spring, but it disappears before August. It is evidently an Arctic species, is abundant off the northern coasts of Europe, and is found off Iceland. It is very closely related to, if not identical with, the American S. mirabilis, but may possibly be distinguished by its more slender and higher bell, its very long manubrium, and its somewhat shorter tentacles. All of these characters are, however, very variable in Sarsia mirabilis, and I have become convinced that the American and European forms are at most only varieties, one of the other. For details see the tabular description of medusae of Sarsia.

Romanes, 1885, carried out many interesting physiological experiments upon this species, and showed that its ocelli are organs for the perception of light. The medusa is sensitive only to rays between the red and violet, and is strongly attracted by the light. The smallest part of the bell-margin is capable of initiating and maintaining the rhythm of the bell, but if the margin be entirely removed all pulsations of the bell instantly cease, while the cut-off margin continues to pulsate. Stimulation of the subumbrella of the bell causes contraction of the manubrium (proboscis), and indeed the bell, when deprived of its margin, still responds by contractions to all sorts of stimuli, chemical, thermal, electrical, or mechan- ical, although sustained rhythmical pulsation is never resumed.

The hydroid is Syncoryne sarsii, which is common in shallow water along the coasts from England to Norway. The polypites are spindle-shaped and elongate and have about 12 to

PLATE 4.

Fig. i. Sarsta mirabilts. Abnormal medusa with a tentacle arising from the side of the manubrium. Nahant, Massachusetts, May 8, 1897. Metschnikoff, 1870, observed a similar abnormality in Slabbena catenata.

Fig. 2. Median section of a young medusa of Sarsia mirabilis. The ocellus on the right-hand tentacle is shown as if depigmented in order to illustrate its structure. The eye on the left-hand tentacle-bulb is shown normally pigmented.

Figs. 3 and 4. Sarsia mirabilis var. reticulata, young medusa. Figure 4, one of the nematocyst-cells from the tentacles of figure 3. Agas- siz Laboratory, Newport, Rhode Island, June 13, 1895.

Drawn from nature, by the author.

ANTHOMEDUS.E — SARSIA. 53

16 short tentacles which arise at irregular intervals from their sides. Each of these tentacles ends in a knob-like cluster of nematocysts. The 2 or 3 medusa-buds arise from the sides of the polypite between the tentacles. The stems of the hydroid are quite smooth, sparingly branched, and about 12 to 15 mm. high. The stems are translucent, slightly horny in color, and the polypites are light-red.

I can not determine any well-defined distinctions between S. sarsii and S. mirabtlis, excepting that in S. sarsn the medusa-buds appear always to arise from near the middle of the sides of the polypite between the tentacles, whereas in S. mirabilts the medusa-buds arise from near the base of the polypite at or below the level of the lowest zone of tentacles.

Hartlaub has given excellent figures and descriptions of a number of medusae which are closely related to Sarsia tubiilosa, if not mere varieties of the latter. These are S. pattersoru Haddon, S. frutescens Allman, 5. decipiens Hartlaub, S. litorca Hartlaub, and S. pulchella Forbes. (See Hartlaub, 1907, Nordisches Plankton, Nr. 12, pp. 29, 30, 32, 36; figs. 20-226, 23, 24, 28, 29.) I hesitate to quote these as distinct species, for I have observed the same or nearly the same variations among individuals in swarms of S. mirabilis at Nahant, Woods Hole, and Newport on our coast. A statistical study, or better still, a study of the respective hydroids, is required before we can hope to determine these so-called "species" with certainty.

Garstang, 1894, observes that the hydroid of S. tubulosa gives rise to dimorphic medusae, as does S. mirabilis on the coast of New England, where early in the spring the medusae are set free, whereas in May they mature while still attached to the hydroid.

Sarsia tubulosa Lesson, variety Sarsia mirabilis L. Agassiz.

Plate 3, figs. 2, 4, and 5; plate 4, figs. 1 and 2. LITERATURE RELATING TO THE AMERICAN VARIETY or SARSIA TUBULOSA.

Sarsia mirabilis, AGASSIZ, L., 1849, Mem. Amer. Acad., New Ser., vol. 4, p. 224, plates 4, 5. — STIMPSON, 1853, Marine Invert. Grand Manan, p. u. — HAECKEL, 1879, Syst. der Medusen, p. 17. — FEWKES, 1881, Bull. Mus. Comp. Zool. at Harvard Coll., vol. 8, p. 141, plate 3, figs. 1 1, 12. — LEVINSEN, 1893, Vid. Meddel. Nat. For. Kjobenhavn (5), Bd. 4, p. 143. — BIR- ULA, 1896, Annuaire Musee Zool. Imp. Sci. St. Petersbourg, tome I, p. 332. — HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 37, figs. 30-40. — LINKO, 1900, Mem. Acad. Sci. St. Petersbourg, ser. 8, vol. 10, No. 4, p. u, taf. I, figs. 1-12 (structure of ocelli); 1905, Zool. Anzeiger, Bd. 28, p. 212; 1900, Traveaul Soc. Jmperiale des Nat. St. Petersbourg, torn. 29, p. 151.

Sarsia mirabilis (medusa), AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 21 1-217; vol. 3, Ibid., plate 18, figs. 15-250.

Sarsia glafialis, MORCH, 1857, Beskriv af Gronland, p. 95.

Corync mirabilis (hydroid), AGASSIZ, L., 1862, Cont. Nat. Hist. U. S., vol. 4, pp. 185-21 1, plates 17-19; vol. 3, Jbid., figs. 1-16, plate 17; figs. 1-14, plate 18; figs. 1-27, plate 19.

Corvnt mirabilis, AGASSIZ, A., 1865, North Amer. Acal., p. 175, figs. 282-285. — CALKINS, 1899, Proc. Boston Soc. Nat. Hist., vol. 28, p. 336.

Oceania lubulosa, GOULD, 1841, Invert, of Mass., p. 348.

Syncorvnt mirabilis, ALLMAN, 1871, Monog. Tubul. Hydroids, p. 278. — HARGITT, 1904, Bulletin Bureau of Fisheries U. S., vol. 24, p. 30, plate v, fig. i; 1901, Amer. Naturalist, vol. 35, p. 578, fig. 33; 1903, Science, vol. 16, p. 344 (variations). — HARTLAUB, 1905, Zool. Jahrbuchern, p. 526; Zool. Jahrb., Syst. Abth., 1901, Bd. 14, p. 356. — TORREY, 1902, Univ. California Publications, vol. I, p. 31. — NUTTING, 1901, Bull. U. S. Fish Commission, vol. 19, pp. 328, 372, figs. 3, 81.

( ?) Syndictyon angulatum (young medusa), MURBACH and SHEARER, 1903, Proc. Zool. Soc. London, vol. 2, p. 168.

(})Sarsi barfnisii (young red-colored medusa), LINKO, 1905, Zool. Anzeiger, Bd. 28, p. 214 (north of Russia).

( ?) Euph\sa tentaculata (abnormal medusa with only 3 well-developed tentacles), LINKO, 1905, Zool. Anzeiger, Bd. 28, p. 214 (Barents Sea, North Russia).

Syncoryne densa, HARTLAUB, 1897, Wissen. Meeresuntersuch. Komm. Meere. Kiel, Helgoland, Neue Folge, Bd. 2, p. 452, taf. 166, figs. 4, u ; taf. i6r, figs. 7, 8 (hydroid forms dense tufts). Also: Sarsia sp., HARTLAUB, 1896, Verhandl. Deutsch. Zool.Gesell., Leipzig, Bd.6, Vers., p. 182 (medusa; with branched manubria); 1907, Nordisches Plankton, Nr. 12, p. 26, figs. 17-19, 22a (hydroid and medusa).

The following description is derived from a study of medusae and hydroids obtained off the southern coast of New England, United States.

Adult medusa. — Bell is half-egg-shaped, about 7 mm. in height and 4 mm. in diameter. There is no apical projection, and the gelatinous substance is not very thick at the aboral pole and becomes successively thinner near the margin. There are 4 long, highly contractile tentacles, one at the base of each .radial-canal. The surface of each tentacle is covered with prominent nematocyst-cells, which are clustered especially at the outer end of the tentacle. Each tentacle arises from a well-developed basal bulb which contains a single ectodermal ocellus upon the outer nerve-ring on the abaxial side of the tentacle-bulb. According to Linko, 1900, the ocellus is composed of a cup-shaped invagmation of densely pigmented ectodermal cells between which there are spindle-shaped bipolar nerve-cells. The ento-

54

MEDUSAE OF THE WORLD.

dermal core of the tentacle is hollow, and its lumen is continuous with the gastrovascular system of the medusa. The ectodermal cells of the tentacle-bulb are very thick and are probably nervous in function. The velum is well developed, being wide and thin. There are 4 straight, slender radial-canals, and a simple, narrow, circular tube. A short, blindly- ending tube extends upward from the base of the stomach into the gelatinous substance of the bell. The entodermal cells of this small projection are several layers thick. The manubnum is long and extends far beyond the velar opening. Its proximal part is slender and tubular, but in its outer parts it is much distended by the genital products and consists of a long, wide, cylindrical tube. The line of demarcation between the narrow and the wide part of the manubrium is very sharp. The mouth is a simple opening at the extremity of a short flask-shaped proboscis, and the lips are studded with nematocysts. The mature genital pro- ducts are found in the ectoderm of the distal part of the manubrium. The entoderm of the manubrium is usually green, but occasionally it is red. The entoderm of the tentacle- bulbs is either red or green, and in some individuals the entoderm of the bulb is red while the ectoderm is green.

FIG. 16. — "Syncoryne pulchella."

FIG. 17. — "Syncoryne frutescens."

The above figures are after Allman, in Ray Society, 1871-72.

Hydroid and young medusa. — The hydroid stock is Syncoryne mirabilis. The stems are attached by a creeping stolon. They are about 15 millimeters in height and branch profusely. The main stems and also the side branches terminate each in a single polypite. The stems are incased in an unannulated chitinous perisarc, which terminates sharply at the bases of the polypites. Each polypite is fusiform and has about 12 to 18 tentacles which arise in 3 or more indefinite whorls from the sides of the polypite. These tentacles are not long, but are quite contractile. Each terminates in a knob-shaped cluster of nematocyst- cells. The mouth of the polypite is a simple round opening situated at the extremity of a conical proboscis. Medusa-buds are developed upon the sides of the polypites immediately below the tentacles, near the lower base of the polypite. Each polypite bears I to 4 medusa-buds in various stages of development. In Massachusetts Bay the breeding season begins early in March and lasts until the end of May. During March the medusa-buds

ANTHOMEDUS.E — SARSIA.

55

develop 4 long tentacles, and are set free in an immature state, but during the last half of the breeding season they fail to develop tentacles or give rise to mere short lashes upon their basal bulbs; and they become sexually mature while attached to the hydroid, the manubrium of each bud being greatly distended with the genital products. This observation was first made by L. Agassiz, 1862 (pp. 189, 203), and has been confirmed by us in hydroid stocks obtained in Swallow's Cave, Nahant, Massachusetts. Plate 3, fig. 5, is derived from one of these sexually mature medusa-buds found upon a hydroid on May 8, 1897. It will be observed that the manubrium of the bud is distended with sperm, while the tentacles are not developed. In this connection it is interesting to observe that Pennaria and Podocoryne carnca sometimes give rise to medusae which are sexually mature at their time of liberation, while in other stocks of the same species the medusae are set free in an immature condition. Garstang, 1894, observes the same phenomenon in the European S. sarsii.

This hydroid of our Sarsia is very abundant, from March until May, in Massachusetts Bay, where it appears to grow equally well both in pure sea-water and in the brackish mouths of rivers. The medusae appear in great numbers on the southern coast of New England

FIG. 18. — Hydroid and medusa of "Sarsia Jensa," after Hartlaub, in Nordisches Plankton.

between February and April. They become rare during May, and are not seen during the summer months. The hydroid extends northward to the Greenland coast, but has not been recorded from Beaufort, North Carolina, or farther south. Linko and Birula, 1896, found it in the White Sea, and Linko, 1905, records it from the eastern parts of Barents Sea between Kanin and Kolgujew Islands. Calkins, Torrey, and Hartlaub have found this medusa along the Pacific coast of America as far south as Chile. I believe Syncoryne densa Hartlaub from Helgoland to be an environmental form of S mirabilis.

We have observed an abnormal medusa of Sarsia mirabilis in which a single well- developed tentacle arose from the side of the manubrium at the point of juncture of the long tubular basal region and the gemmiferous part of the manubrium. (See plate 4, fig. i). This abnormal tentacle was studded with clusters of nematocyst-cells. It lacked a basal bulb and had no ocellus. Asexual budding of medusa- from the walls of the manubrium is not known in Sarsia mirabilis. Medusae of Sarsia with branched manubria are described by Hartlaub, 1896, 1907.

Professor Hartlaub finds that in Sarsiti mirabilis the stomach is confined to the distal end of the manubrium and the gonad is confined to the mid-region of the manubrium above the stomach. Both the proximal and distal ends of the manubrium lack the gonad. On the other hand, in S. brachygaster and S. cximia there is no differentiated stomach-region, and the gonad may extend over the whole, or nearly the whole, length of the rnanubrium.

56

MEDUSAE OF THE WORLD.

The chief and possibly only well-marked point of difference between the forms S. mira- bilis and S. tubulosa is that in the hydroid of S, mirabilis the medusa-buds arise from near the base of the polypite, whereas in the hydroid of 5. tubulosa they arise from points higher up on the sides of the polypite, between the tentacles.

20.

FIG. 19. — Sarsia eximia, after Hartlaub, in Nordischcs Plankton.

FIG. 20. — Syncoryne eximia, after Allman, in Ray Society, 1871-72. Hydroid'and young medusa.

PLATE 5.

Fig. I. Sarsia angulata. Nassau Harbor, New Providence Island, Bahamas, July, 1903.

Fig. 2. Corynitis agassizii. Charleston Harbor, South Carolina, September, 1897.

Fig. 3. Ectopleura minerva. Tortugas, Florida.

Fig. 4. Ectopleura dumortieri, young medusa. Agassiz Laboratory, New- port, Rhode Island, June, 1893.

Fig. 5. Ectopleura dumortieri, mature male. Agassiz Laboratory, Rhode Island, July, 1896.

Fig. 6. Sarsia mirabilis var. reticulata. Nahant, Massachusetts, March 25, 1897.

Drawn from life, by the author.

PLATE 5

AXTHI i.MKDTS.E SARSIA. 57

Sarsia mirabilis var. reticulata. Plate 4, figs. 3 and 4; plate 5, fig. 6.

Syndictyon reticulatum, A. AGASSIZ in L. AGASSIZ'S, 1862, Cont. Nat. Hist. U. S., vol. 4, p. 340. — AGASSIZ, A., 1865, North Amer.

Acal., p. 177, figs. 290-300.

Syncor\ne reticulata, ALLMAN, 1871, Monog. Tubul. Hydroids, p. 283. Syndictvon reticulatum, HAECKF.L, 1879, Syst. der Medusen, p. 21.

( ?) Syndictyon raiculalum, MAAS, 1893, Ergeb. der Plankton-Expedition, Bd. 2, K. c., p. 67. (tySarsia turrreula, MrCRADV, 1857, Gymn. Charleston Harbor, p. 36, plate 8, figs. 6-8. S\ncor\ne reticulata, HARGITT, 1904, Bulletin Bureau of Fisheries U. S., vol. 24, p. 30. Sarsia reliculata, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 45, figs. 41-43. (?) Sarsia pulchella, AI.I.MAN, 1871, see Hartlaub, 1907, Nordisches Plankton, Nr. 12, p. 34, fig. 27 (abnormal twin medusa:,

fig. on p. 109). — SPAGNOLINI, 1876, Catalogo Acalefi Mediterranco, p. 18, tav. 2, figs. i. 2.

Adult medusa. — Bell ellipsoidal in shape, being about 4 mm. in height and 3.5 mm. in diameter. No apical projection. Gelatinous substance quite thick at the aboral pole, but thin at the bell-margin. There are 4 long, highly contractile tentacles, I at the base of each radial-canal. Surfaces of these tentacles covered with prominent nematocyst capsules. Basal bulbs of tentacles well developed and each one contains an ectodermal ocellus upon its outer side. Velum wide and thin. There are 4 straight, narrow radial-canals, and a slender circular vessel. Manubrium short and club-shaped, and does not extend far beyond the velar opening. Mouth a simple, round opening. Genital products developed along the greater part of the length of the manubrium. The entoderm of the manubrium and tentacle-bulbs is brick-red.

Hydroid and young medusa. — Smaller than Coryne mirnliilis, being not more than 3 mm. in height. Stems slender and hardly ever branch, excepting in old specimens, which some- times give rise to a single branch near the base of the stem. Polypites large and club-shaped, and having several whorls composed of 8 to 10 short tentacles. The medusae develop among the tentacles near the proximal base of the polypite. When set free the young medusa is remarkably large, being about 1.5 mm. in diameter. The bell is covered \\irh reticulated clusters of nematocyst-cells (plate 4, figs. 3, 4) which are especially numerous near the bell- margin above the circular canal. Some of these nematocyst-cells are large and round, while others are narrow and long. The tentacles are thickly covered with helically arranged clusters of nematocyst-cells. These cells (plate 4, fig. 4) are ellipsoidal in shape and are mounted upon a short basal pedicel. Each cell gives rise to a long, sharp-pointed, sensitive hair. The nematocyst thread lies coiled in a helix within the cavity of the cell. As the medusa becomes mature the reticulated nematocysts disappear from the surface of the exumbrella and the nematocysts upon the tentacles become less prominent. This medusa is found upon the Ne\\ England coast from April until June. It is distinguished from Sm-sm mirabilis only by its nematocyst-covered tentacles and exumbrella and its small hydroid. It appears also to be constantly brick-red, while 5. mirabilis is highly variable in color. It is often impossible to distinguish mature medusae of S. reticulata from those of .V. mirabilis. It is possible that the S. pulchella of Spagnolini, 1876, from Naples, Italy, is identical with .S'. r, n, uLita.

Sarsia eximia Boehm.

Coryne eximia, ALLMAN, 1859, Annals and Mag. Nat. Hist., ser. 3, vol. 4, p. 141; 1864, Ibid., vol. 13, p. 357.

Syncoryne eximia, HINCKS, 1868, Hist. British Hydroid Zooph., p. co, plate 9, fig. 2. — ALLMAN, 1871, Monog. Tubul. Hydroids,

p. 282, plate 5. Sarsia eximia, BOEHM, 1878, Jena. Zeitschrift fur Naturw., Bd. 12, p. 191, taf. 6, figs. 7-26; taf. 7, figs. 1-6. — (In part), HAEC-

KEL, 1879, Syst. der Mcdusen, p. 17. — HARTLAIPII, 1894, \Vissen. Meercsuntcrsuch. Komini^. Mo-rc. Kid, lldi;ii]and,

Ser. 2, Bd. i, p. 187.

Svncorvne eximia (hvdroid). NUTTINC;, 1901, Proc. Washington Acail. Sa., vol. 3, p. 166, plate 14, figs. 3, 4. Sarsia eximia, BROWNE, 1905, Proc. Roy. Soc. Edinburgh, vol. 21;, p. 756 (hydroid and medusa). — BROCB, 1905, Bergcns

Museums Aarbog, No. 11, p. 4.

Sarsia tximia=S. bretonica, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 8, figs, i, 2a, 26 (full list of recent literature). Syncorync eximia, BROWNE, 1907, Journal Marine Biol. Association, vol. 8, p. 37 (growth of the hydroid).

This form is found off the coasts of Great Britain, Helgoland, Shetland Islands, Norway, and Juneau, Alaska (Nutting). For details of the medusa, see tabular description of the medusas of Sarsia.

The hydroid is about 30 mm. high, forming a bush-like cluster of profusely branched stems; the branches are short and simple and arise very irregularly from the main stems, and are usually faintly ringed at their points of origin. The main stem is usually unringed,

58

MEDUSJE OF THE WORLD.

except at its base, and is quite smooth. The polypites are very elongate, spindle-shaped, and have about 1 6 to 24 short, knobbed tentacles arranged in 4 to 6 somewhat irregular verticils. The medusa-buds arise singly upon short peduncles near the bases of the tentacles of the lower verticils. The entoderm is red to reddish-brown and the stems are yellow.

Browne found that confinement in an aquarium under somewhat unnatural conditions caused the hydroid to grow rapidly and to form stolons, these being developed from branches

which touched the glass sides of the aquarium. He also discovered that the medusa becomes mature in from J to 10 days after being set free from the hydroid. Later, in 1907, Browne found that one of these hydroids placed in a glass tube with a constant current of water passing through it grew in length from 14 to 77 mm. in the course of 9 days, and developed branches having a total length of 500 mm. The hydroid was fed upon copepods.

Sarsia radiata von Lendenfeld.

Sarsia radiata, VON LENDENFELD, 1884, Zool. Anzeiger, Jahrg. 7, p. 584; 1884, Proc. Linnean Soc. New South Wales, vol. 9, pp. 583, 635; plate 20, figs. 31, 32; plate 30, figs. 1-4.

Medusa. — Bell semiovate, slightly higher than broad, 3 mm. high, 2.5 mm. wide. 4 ten- tacles, each about 1.5 times as long as bell-height, and with large bulbs about half as wide as the manubrium. Ocelli (?) Velum wide. 4 straight radial-canals. Manubrium cylindrical, half as long as the bell-height. The gonad incases the sides of the manubrium from the inner apex of the bell-cavity to near the mouth. No medusa- buds. Entoderm of manubrium and tentacle- bulbs deep brown. Other parts colorless.

H \droid. — The hydrocauli arise from a creeping hydrorhiza which anastomoses in a very open network. The perisarc terminates with an oblique elliptical margin at the base of each hydranth, and the hydranth is provided with a muscle at this point which enables it to bend downward and "shut up" as if it were the blade of a penknife. The hydranths are spindle- shaped, narrow, and elongate; and are, includ- ing their hydrocauli, 3 to 5 mm. high. They have 6 to 8 verticils, each of 4 tentacles, situated in 4 meridional lines, 90° apart. These tentacles are all knobbed at their ends. The hydranths which produce medusae are shorter than the sterile polypites. The medusae bud out from the lower half of the polypite between the tentacles. The entoderm is intensely brown in color, and the perisarc is bright brownish-yellow. Other parts colorless. Found on the coast of New South Wales, Australia. The medusae are produced in April and May.

Sarsia conica. Codonium eonicum, HAECKEI., 1880, Syst. der Medusen, p. 634.

Bell barrel-shaped with conical apex one-third as long as the sides of bell. 12 mm. high, 4 mm. wide. 4 tentacles longer than bell-height, and with small oval basal bulbs. The manubrium is half as long as the depth of the bell-cavity. The stomach is subspherical and swollen by the encircling gonad. The mouth is at the end of a short, cylindrical throat-tube which is free of gonads. Color ( ?) There is a long axial canal above the stomach. Indian Ocean. Briefly described, without figures, by Haeckel.

FIG. 21. — Sarsia eximia, from life, by the author. Mouse- hole, Cornwall, England, Nov. 14, 1907.

A\THi>MKI>rs.K--SAi;-.I \. .V.I

Sarsia rosaria Haeckel.

Coryne rosaria, AGASSIZ, L., 1862, Cent. Xat. Hist. U. S., vol. 4, p. 340. — A- \ i/. \.. i Sf>;, Virrh Arm-r. Aca!., p. 176, fig 289.

Sarsia rosaria, HAECKEL, 1879, Syst. Jer Mi .i'i CDj p. 18.

Santa rosaria= Syncoryne rosaria, FEWKES, iSSij. Am. r. Vituralist, vol. 23, p. 597, plate 25, fig. 7; tcit-figs. 8, 9 (hydroid?).

Syncoryne occidentals, FEWKKS, 1889, Bull. Essei Inst., Salem, vol. 21, No. 7, p. 99, plate 3, figs. 2, 3.

( ?) Syn diclyon angulatum, Mt RUAI H and SIIK^RV.R, 1903, Proc. Zoo]. Soc. London, vol. 2, p. 168.

Codonium apiculum, MURBACH and SHEARER, 190}, Proc. Zool. Soc. London, vol. 2, p. 165, plate 17, fig. i; plate 22, figs. 4, 5;

1902, Annals and Ma^. Nat. Hist., scr. 7, vol. 9, p. 72. Sarsia apicula+S. rosaria, HARTLAI'B, 1907, Nordischcs Plankton, Nr. 12, pp. 17, 50, figs. 9, 45.

Bell 15 to 30 mm. high, 10 to 15 mm. wide, with fairly thick walls and small apical projection. 4 equally developed, radially placed tentacles 1 .5 to 2 times as long as bell-height. The basal bulbs of these tentacles are large and are flanked on either side by a large nemato- cyst-pad. Each tentacle-bulb bears an abaxial ocellus. There are 4 slender, straight-edged radial-canals and a narrow ring-canal. The velum is well developed. Manubrium short and spindle-shaped, and mouth about at the level of the velar opening. There is a short axial canal above the stomach. The gonad encircles the stomach, leaving both ends tree. N<> medusa-buds. The colors are quite variable as in other species ot Sarsia. The tentacle- bulbs range from yellow through red to brownish-red, and the stomach is yellow, pink, or reddish-violet to purple.

This is the most abundant Sarsia along the Pacific coast of the United States. It occurs in great swarms in San Francisco Harbor in spring; and in Victoria Harbor, Puget Sound, in July.

The hydroid is Syncoryne rosaria found by A. Agassiz and Fewkes in shallow water attached to piles of wharves. Each tentacle terminates in a knob, and the hydroid is a true Syncoryne.

Sarsia minima von Lendenfeld. Sarsia minima, VON LENDENFELD, 1884, Proc. Linnean Soc. New South Wales, vol. 9, pp. 584, 915, plate 21, figs. 34, 35.

Bell of medusa 3 mm. high and 2.5 mm. wide with "a long manubnum like the northern Sarsia;." It is therefore readily distinguished from Sarsia radiata, which has a short manu- brium. S. minima has a spindle-shaped, nearly cylindrical manubnum which extends tor about half its length beyond the velar opening. The 4 marginal tentacles are somewhat longer than the bell-height and are covered with rings of nematocysts. The entoderm ot the stomach is pale brown, other parts colorless.

Hydroid. — The stems are 2 to 3 mm. high and arise from a creeping, slightly branched, non-anastomosing hydrorhiza. The perisarc which invests the hydrorhiza and hydrocauli is irregularly annulated or wavy throughout, and terminates at the bases of the hydranths in a transverse margin. The hydranths are slender, spindle-shaped, 0.6 to 0.8 mm. long, and with 8 to 12 irregularly scattered tentacles, all of which are knobbed at their ends. When they produce buds they become stouter, and are so thickly covered by the budding medusae "that nothing of their bodies remains visible." The perisarc is reddish-brown.

This hydroid is found at Port Jackson, New South Wales, Australia, overgrowing Obclla geniculata, on buoys and submerged ropes. The medusae are produced in April and May. Von Lendenfeld did not obtain any mature medusae.

Sarsia brachygaster Grbnberg.

Sarsia brachygaster, GRONBERO, 1898, Zoolog. Jahrb., Abth. Syst., Bd. 1 1, p. 459, taf. 27, figs. 3, 4. — HARTLAI n, 1907, N'or- disches Plankton, Nr. 12, p. n, fig. 3.

Bell 15 to 18 mm. high and 8 to 10 mm. wide and three-fourths-egg-shaped, the greatest breadth being above the middle. 4 radially situated tentacles, each being about twice as long as the bell-height. Basal bulbs of these tentacles well developed, and a single MI\ small ocellus upon the outer side of each bulb. Manubrium cylindrical, and two-thirds as long as height of bell-cavity. Mouth situated at extremity of a shoit cylindrical neck. The gonad is tubular and surrounds the stomach. No medusa-buds. The manubrium, gonads, tentacle-bulbs, and tentacles are orange-red. The ocelli are black.

60

MEDUS.E OF THE WORLD.

Gronberg found this species at Spitzbergen in summer, and he also identified it among a collection of medusae from Jakobshavn, Greenland, where it appears to be rarer than at Spitzbergen.

Sarsia angulata Hartlaub. Plate 5, fig. i; plate 6, fig. 3.

Syndictyon angulatum, MAYER, 1900, Bull. Mus. Comp. Zool. at Harvard College., vol. 37, p. 5, figs. 6-8, plate 3; 1904,

Memoirs Nat. Sci. Brooklyn Institute Museum, vol. I, No. I, p. 7, plate I, fig. 6. N on Syriilict\on tiHgtihitum, MURBACH and SHEARER, 1903, Proc. Zool. Soc. London, vol. 2, p. 168. Sarsia angulata, HARTLAUB, 1907, Nordisches Plankton, Nr. 12, p. 16.

Bell i mm. high; half-egg-shaped, with moderately thick walls. Becomes almost square in cross-section when contracted. There are 4 slender tentacles with fairly thick spindle- shaped ends. These tentacles are each about as long as bell-height and their distal halves are tapering and are armed with nematocysts. The basal bulbs of the tentacles are not very large, and each one bears an ocellus formed by a cup-like invagmation of ectodermal cells. The velum is large, and the radial-canals and circular vessel are of fine caliber. The manubrium is spindle-shaped with a narrow tubular o?sophagus and without an aboral projection. It is about two-thirds as long as the height of the bell-cavity. The gonad encircles it, extending from the base to near the mouth, leaving the throat-tube free. The entoderm of the tentacle- bulbs and manubrium is robin-egg blue, while the ocelli are deep-brown, almost black. All

other parts are hyaline. This medusa is abundant in the Tongue of the Ocean, Bahama Islands, in June and July, and was found at Turks Island in January. It is rare at Tortugas, Florida.

Sarsia gracilis Browne.

Sarsia gracilis, BROWNE, 1902, Annals Mag. Nat. Hist., ser. 7, vol. 9, p. 275. (? ?) S\ncoryne sarsii (hydroid), HARTLAUB, 1905, Zoolog. Jahrbiichern, Suppl. 6, p. 525, fig. F.

Bell 5 mm. high, 3 mm. wide; cylindrical, with moderately thick walls and quadrangular margin. 4 tentacles, about as long as the bell-height, and end- ing each in a large knob containing nematocysts. An ocellus on the basal bulb of each tentacle. Manu- brium about two-thirds as long as the depth of the umbrella cavity. Color ( ?) Gonads ( ?) Found at Stanley Harbor, Falkland Islands, by Vallentin, and briefly described without figures by Browne. The hydroid, Syncoryne sarsn, described by Hartlaub from southern Terra del Fuego, may be the stock ot this medusa.

The medusa may be a young Slabberia ( ?)

Sarsia princeps Haeckel.

Codonium princeps, HAECKEL, 1879, Syst.derMedusen,p.l3,taf.l,figs. 1,2.

Sarsia princtps, HAECKEL, 1879, Ibid., p. 655. — LINKO, 1905, Zool. An- zeiger, Bd. 28, p. 212. — BROWNE, i9O3,Bergens Museums Aarbog, No. 4, p. 8, plate i, fig. i; plate 3, fig. 4. — HARTLAUB, 1907, Nor- disches Plankton, Nr. 12, p. 47, fig. 44.

Codonium princeps, LEVINSEN, 1893, Vid. Meddel. Nat. For. Kjobenhavn (5), Bd. 4, p. 143. — GRONBERG, 1898, Zoolog. Jahrb., Abth. Syst., Bd. n, p. 458, taf. zj, figs, i, 2.

Bell thin-walled and conical, about 25 to 40 mm. high, and 15 mm. wide. There is a short, conical, apical projection. There are 4 tentacles with long conical basal bulbs. The shafts of these tentacles

1-iG. 22.