The oceanic Hydrozoa; a description of the Calycophorid¿ and Physophorid¿ observed during the voyage of H.M.S. "Rattlesnake," in the years 1846-1850

Cero reece tetet heen peanrenchabmg et eee ETM Ora
abalgnteytcnaeesoeek See es otevepeen

ee ie re
machiotehareh nee ene

Ta HtWateieny state poe nets eaenecte ees .
ass Ncichburentenes.netgearhagines tase sashes

Mee

THE

my SO Cr ery:

INSTITUTED MDCCCXLIV.

LONDON:

MDCCOLVIII.

i

ih Y B ,
Vi Y ty LBL F-L; G

na ". s
d 2

THE

OUR ANC FY DR Oso

A DESCRIPTION OF THE

CALYCOPHORIDH AND PHYSOPHORID

OBSERVED DURING THE VOYAGE OF H.M.S. “RATTLESNAKE,”

IN THE YEARS 1846-1850.

WITH A GENERAL INTRODUCTION.

BY

THOMAS HENRY HUXEBY, PRS, 90.8. SHC. Gret:

ACAD. IMP. LEOP. CAR. NAT. CUR.; PROFESSOR OF NATURAL HISTORY,
GOVERNMENT SCHOOL OF MINES.

LONDON :
PRINTED FOR THE RAY SOCIETY.

MDCCCLIX.

MCZ LIBRARY
HARVARD UNIVERSITY
CAMBRIDGE. MA USA

PRINTED BY J, E. ADLARD,
BARTHOLOMEW CLOSE, LONDON.

TO

SIR JOHN RICHARDSON, M.D. F.RB.S.,

INSPECTOR OF HOSPITALS AND FLEETS, R.N.,

ETC., ETC.

My pear Sir Jonn Ricnarpson,

Since the time (now thirteen years ago) when, a subaltern under your orders, I
learned to admire your character and to respect your high scientific attainments, we have met
but rarely; and the arduous and perilsome service into which your sense of duty has led you
in the interval, may well have obliterated from your mind all recollection of the fact that
I owe to you those opportunities for scientific observation which I have enjoyed and have

endeavoured to use.

If I remind you of such a matter, it is only that I may seem less bold in venturing to
dedicate to one so justly honoured as a man and so esteemed as a naturalist, these late first-

fruits of a beginner in zoology.
Tan,
My dear Sir Jonn Ricuarpson,

Yours very faithfully,

THOMAS HENRY HUXLEY.

Tae GoveRNMENT Scuoot or Mines, Jermyn STREET ;
August 11th, 1859.

batt

We
HATO LT \y i : tt FA,

PR TARE in TA Cee * pha naat

be Ah tema y

rol 91 uh ae :

mn re oe
rT

> cobwebs SEY, Glaitss fetal ak de Coste ui iT! even ) oe Sai omit in ie jet ce
- eee Presa i) 0 AOE TU OE OUR tal A txtordyy juin gostei ta) bi!

OPTS ead aah Yih i whi, leben ot elvis shoal sey be wingibes ta ort Daye i yom ‘wal -
Oat) (AOR ddd iy Wolo Ue DATE (Oe WOT deine key ae Hise die dawiptiie with? aaa
“rail Pi hatoute uth b vigelite Wott Oily of. HE PICTON: wr oid HOY. om aves E i
Cy ae a a
ey aiad Ae sdiatieean h ene i ott Gti aba) MAL nla hy: ky nay mbar Lc
Wevcilk Wat's ann? Ya; mara ay hi. Ate) , TH lin i iil if" 1 Py ta ey: Ory: FS ant al
is! i“ % Mola in iti pet Hin

Ian )

Tere weer ree
Ar A ;

| te lh eal Cal ly

“Tt is the opinion of the Lords Commissioners of the Admiralty that it would be to the honour
and advantage of the Navy and conduce to the general interests of Science, if new facilities and
encouragement were given to the collection of information upon scientific subjects by the officers,
and more particularly by the medical officers, of Her Majesty’s Navy, when upon foreign service.

: And it will be for their Lordships to consider whether some pecuniary reward
or promotion may not be given to those who succeed in producing eminently useful results.”?

\

In the autumn of the year 1846, Her Majesty’s Ship “ Rattlesnake” was commissioned
by the late Captain Owen Stanley, who had been charged by the Admiralty with the duty of
surveying the intricate passage within the Barrier Reef which skirts the eastern shores
of Australia, and of exploring the sea which lies between the northern end of this reef
and New Guinea and the Louisiade Archipelago. A very competent naturalist, Mr. J.
Macgillivray, was appointed to the vessel, but Captain Stanley, justly appreciating the
largeness of the field likely to be opened to students of Natural History by the cruise,
desired to increase the strength of his expedition in this department of science. To
this end, he applied to Sir John Richardson, at that time the Medical Inspector of Haslar
Hospital, to recommend him an assistant-surgeon who should possess some knowledge
of Natural History, or who, at any rate, had sufficient zeal and love for it to be likely to
convert opportunity into knowledge.

Since the spring of 1846, when I joined the medical service of the Navy, I had
been doing duty at Haslar, under the orders of Sir John Richardson, who, always thoughtful
and kindly in act, though sparing of words, to his subordinates, had, I suppose, noticed
my bent; for, in the summer, I found that, without any solicitation on my part, he had

endeavoured to obtain for me an appointment to the Haslar Museum. Failing in this,

* Extract from a Memorandum by the Lords Commissioners of the Admiralty prefixed to
the first edition of the “Manual of Scientific Inquiry prepared for the use of Her Majesty’s Navy,’
edited by Sir J. F. W. Herschel, and published in 1849.

vill PREFACE.

my chief still kept me in mind, and I shall not easily forget the day when, with Captain
Stanley’s letter in his hand, he came to offer me that share in an exploring expedition which
had been one of my childish ambitions, and which afforded the largest scope for the
faculties, or, at any rate, the tastes, which had grown up with my manhood. I need
not say how gladly I accepted the proffered appomtment.

The “Rattlesnake” sailed in the winter of 1846; and as a full history of her voyage
has been given to the world by Mr. Macgillivray, it is needful only to state that, after
safely effecting her circumnavigatory voyage and successfully surveying the regions she
was sent to explore, the ship returned to England, and was paid off on the 9th of
November of the year 1850.

Although occasionally in circumstances which might give rise to anxiety, no serious
evil befel the “Rattlesnake,” or her crew, until her last return to Sydney, in the spring
of 1850, when her Commander, whose health had been shattered by the trying climate
of New Guinea, and by the still more wearing responsibilities of his office, sank with
lamentable suddenness.

I will not allude to the private circumstances which intensified the sadness of this
sudden blow to all who were witnesses of it, but I can truly say that nothing could
have been more unfortunate for the scientific interests of the expedition or for the personal
welfare of the officers who had performed its duties and shared its fatigues. __

Captain Stanley possessed large means of influence, and as it was an eminent virtue
of his to stand by his officers, there can be no doubt that, had he lived, his lieutenants,
already men of standing and experience, would not have been left for years without
promotion; nor would his and their hydrographic labours have been indistinguishably
merged in those of other persons; nor would the official naturalist of the expedition
have been refused the means of publishing the narrative of the voyage; nor, in all
probability, would the present work be making its appearance so many years after date.

In truth it is to the explanation of the circumstance that all the original illustrations
in my plates were drawn between the years 1847 and 1850 inclusively, and that all my
observations, with the exception of those on Porpita, were made so long ago, that I feel it
necessary to devote these prefatory remarks.

The facts are these. I made a good many observations during our cruise, and sent
home several papers to the Linnean and Royal Societies; but of these doves, or rather
ravens, which left my ark, I had heard absolutely nothing up to the time of my return;
and, save for the always kind and hearty encouragement of the celebrated William MacLeay,
whenever our return to Sydney took me within reach of his hospitality, I know not whether I

should have had the courage to continue labours which might, so far as I knew, be valueless.

PREFACE. ix

On reaching England, however, I found not only that the Royal Society had thought my
Memoir on the Medusze worthy of publication, but helping hands were stretched out
to me on all sides; and among the men of science, 1 met with many generous friends
whose sympathy and appreciation were bestowed ina measure far beyond my deserts. Among
these, the genial and noble-minded Edward Forbes supported me with all that energy which
he was wont to throw into his advocacy of the cause of a young man; and now that I
have succeeded (though, alas! not replaced) him in the professorial chair he then held,
and have some personal experience of an analogous variety of occupations and weight
of responsibilities, I cannot reflect without emotion on the patient attention which he
bestowed upon me, and the self-sacrificing zeal with which he exerted all his “ power, amity,
and authority” in my favour.

On reviewing the materials which I had collected with this experienced friend, it
seemed that some of my work might fitly be sent in the form of papers to the learned
societies; while, on the other hand, the more copious observations upon the MJeduside,
Calycophoride, and Physophoride, might better be published in a separate form; and
Forbes and other friends of weight were of opinion that the work had sufficient value
to justify the Government in aiding its publication.

We were the more readily led to hope for this aid, as, in the year 1849, the Lords
of the Admiralty had caused a Manual of Scientific Inquiry to be published, and had
prefixed to it a Memorandum containing the enlightened and liberal expressions and promises
quoted at the head of these pages. And it will be observed that their Lordships’ distinct
promise of encouragement to naval officers who should do good work was wholly
irrespective and independent of any other aid given by the Government to Science.

Clearly, therefore, supposing my work to have any value—and in favour of this suppo-
sition I had all sorts of high testimony—I had a claim upon the Admiralty for aid in bringing
it out, and a simple person might have imagined that that claim would be strengthened by
being the first that had been made (to my knowledge) since the publication of the Minute.

I do not know that I can accuse their Lordships of refusing my application, for they
took the simpler course of rendering it impossible I should make one. Before my ship was
paid off, I applied in due form, through my commanding officer, for a simple permission to
publish; and, when one thinks of the Memorandum, there is a touch of humour about the
reply which he received: “I have been commanded to convey to you their Lordships’ full
sanction to his (my) so doing (¢.e. publishing). . . . . . . But I have to add that
their Lordships will not allow any charge to be made upon the public towards the expense.”
Nevertheless, the writer hopes that “he will produce a work alike creditable to himself, to

his late Captain by whom he was selected for it, and to Her Majesty’s Service.”

x PREFACE.

,

Even could J have flattered myself that I had succeeded in producing “eminently
useful results,” I fear this letter would have effectually damped any hopes of “pecuniary
reward or promotion,” or other aid, which I might have formed; but my friends were not
content that I should make bricks without straw, and, by dint of considerable exertion, they got
me a nominal appointment, so that my assistant-surgeon’s pay ran on, while they endeavoured
to obtain the £300, required for the publication of my book, from the Government.

It would be wearisome were I to narrate the history of their other efforts at length. In
vain the Presidents of the Royal Society and of the British Association, separately and con-
jointly, officially and unofficially, solicited the Treasury ; in vain did I visit and write to, and I
fear, bore, numerous persons in authority about this unfortunate grant. It must be confessed
the business was troublesome enough while it lasted; but, in looking back, I would fain
only remember with gratitude the zeal of the friends who aided me, and the long-suffering
courtesy of the various Government officials, who listened so attentively to the claims of that
Natural Science about which, unless I am greatly mistaken, they neither knew nor cared
very much.

During the three years the contest lasted, I reckon that the Admiralty was good enough
to give me, in the form of pay, rather more than fifty pounds over the sum required,
although, with steady consistency, their Lordships from the first refused to enable me to
publish the work which they paid me for publishmg. I by no means quarrel with an
arrangement, which, although very annoying at the time, has been of the utmost service
to mie; for when, in 1854, their Lordships, as I suppose, weary of our pertinacity, cut the
knot by calling upon me to serve afloat, new prospects had. presented themselves, and, in
giving up my commission, I obtained the long-sought funds for publication—the adminis-
trators of the Government Grant no longer objecting, that the Admiralty was pledged to
supply its officers with funds for the publication of work done in its service.

I offer my hearty thanks to the Government Grant Committee for this aid, and, in
conclusion, I must apologise to them and to the Ray Society—who two years ago under-
took to publish my book—for the delay which has occurred in bringing it out. I can
only plead the pressure of new and heavy official duties in palliation of my seeming
dilatoriness ; and I may add that, since 1850, so much has been done by the eminent German
observers, whose works are incessantly quoted in the following pages, that the literature of the
subject, slight enough when my observations were made, has attained considerable dimensions ;
and its study has retarded the progress of my book, as much as it has increased my
knowledge of the organization of the Oceanic Hydrozoa. As it is, I have been obliged
to omit the Meduside, which were included in my original plan, and the illustrations of

whose organization are already engraved. I trust that they may yet some day see the light.

THE OCEANIC HYDROZOA

SECT. I. MORPHOLOGY OF THE HYDROZOA.

TuE body of every Hydrozoon is essentially a sac, composed of two membranes, an
external and an internal, which have been conveniently denominated by the terms ectoderm
and endoderm. The cavity of the sac, which will be called the somatic cavity, contains a
fluid, charged with nutritive matter in solution, and sometimes, if not always, with suspended
solid particles, which performs the functions of the blood in animals of higher organization,
and may be termed the somatic fluid. The ectoderm is commonly ciliated, at any rate while
young; the endoderm is also very generally ciliated, though not always, nor in all parts.
The cilia of the endoderm, aided by the contractions of the walls of the body, are the sole
means provided by nature for the circulation of the nutritive fluid in the Hydrozoa; the cilia
of the ectoderm, similarly aided by contractility, constitute the only respiratory mechanism.

Notwithstanding the extreme variety of form exhibited by the Zydrozoa, and the multiplicity
and complexity of the organs which some of them possess, they never lose the traces of this
primitive simplicity of organization; and it is but rarely that it is even disguised to any
considerable extent. I know of no Hydrozoon in which the two primary membranes, but
little altered, cannot be at once detected in the walls of almost every part of the organism.

This important and obvious structural peculiarity could hardly escape notice, and I find
it to have been observed by Trembley, Baker, Laurent, Corda, and Ecker, in Hydra; by
Rathke, in Coryne; by Frey and Leuckart, in Lucernaria; and it is given as a character of
the Hydroid polypes in general (//ydre, Corynide, and Sertulariade) in the second edition
of Cuvier’s ‘Lecons.’ I pointed it out as the general law of structure of the Hydroid
polypes, Diphyde, and Physophoride, in a paper’ sent to the Linnean Society, from Australia,
in 1847, but not read before that body until January, 1849; and I extended the generaliza-
tion to the whole of the 4ydrozoa in a ‘ Memoir on the Anatomy and Affinities of the Meduse,’
read before the Royal Society in June, 1849.

' «Observations upon the Anatomy of the Diphyde, and the Unity of Organization of the
Diphyde and Physophoride.” Aun abstract of this essay was published in the ‘ Proceedings of the
Linnean Society’ for 1849.

f I
/

2 THE OCEANIC HYDROZOA.

Professor Allman, in his valuable ‘Memoir on Cordylophora’ (‘ Philos. Trans.,’ 1853), has
adopted and confirmed this morphological law, introducing the convenient terms “ ectoderm and
endoderm,” to denote the inner and outer membranes; and Gegenbaur (‘ Beitrage zur niheren
Kenntniss der Schwimmpolypen,’ 1854, p. 42), has partially noticed its exemplification in
Apolemia and Rhizophysa ; but it seems, singularly enough, to have failed to attract the attention
of the other excellent German observers, to whose late important investigations I shall so
often have occasion to advert.

The peculiarity in the structure of the body-walls of the Hydrozoa, to which I have just
referred, possesses a singular interest in its bearing upon the truth (for, with due limitation, it
is a great truth) that there is a certain similarity between the adult states of the lower animais
and the embryonic conditions of those of higher organization.

For it is well known that, in a very early state, the germ, even of the highest animals, is a
more or less complete sac, whose thin wall is divisible into two membranes, an inner and an
outer; the latter, turned towards the external world; the former, in relation with the nutritive
liquid—the yelk. The inner layer, as Remak has more particularly shown, undergoes but
little histological change, and, throughout life, remains more particularly devoted to the
function of alimentation, while the outer gives rise, by manifold differentiations of its tissue, to
those complex structures which we know as integument, bones, muscles, nerves, and sensory:
apparatus, and which especially subserve the functions of relation. At the same time the
various organs are produced by a process of budding from one, or other, or both, of these
primary layers of the germ.

Just so in the Hydrozoon: the ectoderm gives rise to the hard tegumentary tissues, to
the more important masses of muscular fibre, and to those organs which we have every reason
to believe are sensory, while the endoderm undergoes but very little modification. And every
organ of a Hydrozoon is produced by budding from one, or other, or both, of these primitive
membranes ; the ordinary case being, that the new part commences its existence as a papillary
process of both membranes, including, of course, a cecal diverticulum of the somatic cavity.

Thus there is a very real and genuine analogy between the adult Hydrozoon and
the embryonic vertebrate animal; but I need hardly say that it by no means justifies the
assumption that the Mydrozoa are in any sense “arrested developments” of higher organisms.
All that can justly be affirmed is, that the Hydrozoon travels for a certain distance along the
same great highway of development as the higher animal, before it turns off to follow the road
which leads to its special destination.

The entire double-walled body of the Hydrozoon, whether it be a minute, simple, oval sac,
as in the embryonic condition, or such a vast and complex mass as a tree of Plumularia, an
Agalna three feet long, or a Rhizostoma of still more massive proportions, will be termed,
in the course of the ensuing pages, a Aydrosoma.

The simplest condition of this hydrosoma is that observable in the common fresh-water
Hydra, one end of whose body is expanded into a kind of disc, whereby the creature
adheres to its support, while the opposite extremity presents a widely-open mouth, opening
into a cavity which extends through the whole length of the animal, and surrounded by a
circle of long tentacular organs. Here, then, the body exhibits only three distinct morpholo-
gical constituents: a disc of attachment—which, with its homologous organs in other Hydrozoa,
may be termed the /ydrorkiza; a sac for the digestion and (as there is, in this case, no dis-

MORPHOLOGY. 3

tinct somatic cavity) for the distribution of nutriment—the po/ypite ; and, lastly, organs for
prehension—the éevfacu/a. Furthermore, at particular seasons, tubercular elevations are
developed, which contain either an ovum or spermatozoa, and are the reproductive organs.

A polypite and reproductive organs are, in fact, the sole essential constituents of any
Hydrozoon, but, so far as I know, no member of this group has yet been discovered of so
simple a composition. Organs of prehension and of fixation, or of flotation at least, are
always superadded, and, in the majority, there is more than one polypite. But when this is
the case it becomes necessary to distinguish between the polypites and the common trunk on
which they are supported. To the latter, Professor Allman’s term of ca@zosare is very use-
fully applicable ; and it will be found convenient, in treating of these more complex forms, to
speak of the hydrosoma as composed of a coenosare and appendages, the latter being those
specially modified parts of the hydrosoma which subserve the functions of support, locomo-
tion, alimentation, and so forth.

I will now proceed to point out the principal modifications which are undergone, first by
the coenosarc, and next by the appendages, throughout the //ydrozoa.

The Cenosare.

The ccenosare of the Corymde and Serfulariade has the form of a branching stem,
resembling that of a plant, and presenting almost as many diversities in form and habit. It
may be slender and creeping, or twining; or it may simulate a tree, with stout, erect trunk,
and multitudinous branches, arranged according to a definite pattern; or its ramifications may
run into a sort of fleshy expansion, as in Hydractinia.

In these orders, especially in those forms which possess an erect and branching stem,
the coenosarc is usually strengthened by the development of a strong cuticular layer upon the
exterior of its ectoderm. This structureless, or at most laminated, cuticular substance, may
remain in close contiguity with the rest of the ectoderm, or, as in the Campanularia, may
become separated by a greater or less interval. In the latter case, it seems at first sight as if
the wall of the ccenosare were composed of three membranes instead of two; but the
examination of young organs will clearly show that the outermost or cuticular layer is nothing
but an excretion from, or metamorphosis of the outermost substance of, the ectoderm.

In the Calycophoride, Physophoride, Lucernariade, and Meduside, no such thick and
hard cuticular layer is developed, and, consequently, the ceenosare remains, throughout life, soft
and flexible. In the two former orders it is never tree-like, and when it gives off branches
they are exceedingly short. In some few of the Lucernariade, on the other hand (Rfizos-
tomid@), the coenosarc is regularly branched, but, nevertheless, is extremely different from that
of the Sertulariade.

The coenosare of the Calycophoride is slender‘and filiform; that of the Physophoride
varies from extreme slenderness and elongation to a spheroidal or discoidal shape.

In both these orders it is excessively contractile, a property which it owes to the
abundant muscular fibres developed in its walls, principally in the ectoderm. So far as I have
observed, these fibres are always disposed longitudinally (except perhaps in Stephanomia) ; but
other investigators describe transverse fibres.

4 THE OCEANIC HYDROZOA.

The transverse section of the filiform and tree-like coenosares is usually nearly circular,
but in some Physophoride (Forskalia, e.g.) it is said to be reniform, from the presence
of a deep longitudinal groove on one side.

The coenosarc, as has been stated, always contains a cavity filled with nutritive fluid—
the somatic cavity. The endoderm lining this cavity is in many, if not in all, Corynide and
Sertulariade, provided with cilia, whose motions are, in many cases, so directed as to give
rise to currents in opposite directions on opposite sides of the cavity. The like phenomenon
has been observed in his “ Agalma rubra” (LHalistemma, mihi) by Vogt (p. 64); and cilia
have been noticed on the endoderm of the ccenosare of many Physophorida. 1 have
observed them in Physalia, Velella,and Rhizophysa. Gegenbaur (p. 42) saw them in Rhizophysa
and Apolemia. Kolliker, however, denies the existence of cilia on the endoderm of the
ceenosare in Morskalia, Agalma, and LTalistemma and is silent with respect to the other species
of Physophoride which he describes, except Velella and Porpita, in which the ciliation is
distinctly mentioned. Leuckart (Z. U., p.4) agrees with Kolliker in denying cilia to the first-
named genera.

Will (‘ Hore,’ p. 78) expressly affirms that the inner wall of the ccenosare of Diphyes
exhibits as lively ciliary motion as that of the polypites; but I could never verify this state-
ment, nor find cilia on the endoderm of the csenosarc (except at its proximal end) in any of the
Calycophoride. On the contrary, I have distinctly observed and noted the fact, that solid
particles, which, so long as they are in the cavity of the polypite, exhibit a lively rotatory
motion, impressed upon them by its cilia, lose that rotation the moment they pass through
the pyloric valve into the somatic cavity, and then either remain stationary, or are forced sud-
denly along it, backwards or forwards, by the contractions of its walls or of the attached organs.
Leuckart and Kélliker have been equally unable to find cilia in any part of the somatic
cavity of the Calycophorid@, except its proximal dilated end; but Gegenbaur affirms that it is
ciliated throughout in his Diphyes gracilis (= D. Sieboldii of Kolliker, in which, however,
that observer distinctly states no cilia exist), and in Praya maxima (‘ Beitriige,’ p. 21).

In the Zucernariade and Meduside, the inner surface of the endoderm is, so far as I have
seen, everywhere ciliated.

The Appendages.

Before attempting to describe the structure and relations of the manifold appendages of
the Hydrozoa, it will be necessary to determine the corresponding ends of the hydrosoma
throughout the series of forms, a task which is not quite so easy as it may at first appear
to be. The Hydra is fixed by one end, which is ordinarily lower than the other, and may
be regarded as the base; and the end which answers to this is inferior, or basal, in
all’ the Corynide and Sertulariade. But the Calycophoride swim with the corresponding end
upwards or forwards. The Physophoride float with it upwards. The Lucernariade have it
sometimes upwards and sometimes downwards. It would be impossible, therefore, to
designate the corresponding ends by the terms “upper and lower” throughout the series
without great risk of confusion. Under these circumstances I think it will be best to discard
such phraseology, and to employ the terms “ proximal” and “ distal” for the same object.

MORPHOLOGY. 5

In fact, the growth of every hydrosoma is, absolutely or relatively, stationary towards one end
of its axis, while it takes place with rapidity towards the other. This growing extremity,
therefore, is, as it were, constantly moving away from the opposite end, and the open mouths
of all the polypites are turned more or less in its direction. Hence, I shall call this the distal
end, while the other, comparatively fixed, extremity may be regarded as proximal. The latter
is the basal or lower end in Hydride, Corynide, and Sertulariade. It is the anterior end, in
the actively swimming Diép/yes, the upper end, in Hippopodius and the Physophoride. It is
the upper side of a Medusa, Cyanea, or Rhizostoma ; the lower, or attached, end of a Lucernaria,
and of a larval Cyanea or Rhizostoma.

The distal extremity of the hydrosoma is always, so far as I am aware, either
cecal, or ends in a polypite, but is never modified into any other appendage. The
proximal end is variously metamorphosed in the Hydride, Corynide, and Sertulariade ; and
becomes a Aydrorhiza, either expanding into a disc, or sending out many radicles, by which it
attaches itself to other bodies.

In the Calycophoride (Pl. V, figs. 3 and 4), the proximal end of the ccenosarc dilates a
little, and becomes ciliated internally, forming a small chamber, which gives off the ducts, by
whose intermediation the systems of canals, which embrace the cavities of the organs of loco-
motion, are brought into communication with the somatic cavity. At its upper end, this
chamber is a little constricted, and so passes, by a more or less narrowed channel, into a
variously shaped sac, whose walls are directly continuous with its own, and which will hence-
forward be termed the somatocyst. The endoderm of this sac is ciliated, and it is generally so
immensely vacuolated as almost to obliterate the internal cavity and give the organ the
appearance of a cellular mass.

The somatocyst very commonly contains large, strongly refracting globules of an
apparently albuminous matter, of precisely the same character as those which may be observed
occasionally to pass through the pyloric valves of the polypites, into the somatic cavity ;
and I do not doubt that the globules result from the accidental accumulation of such products
of digestion. Not unfrequently an air-bubble may be seen in the somatocyst, whither it has
travelled, there can be but little question, by the same channel, being either swallowed with
the prey or accidentally sucked in by a polypite, whose mouth has been raised above the
surface of the water. Such a chance bubble has of course no relation whatever with the air
contained within the float of a Physophorid; and it is somewhat surprising that any one
acquainted with both structures should have imagined the existence of even an analogy, still
less of a homology, between them.

The float or pxeumatophore characteristic and diagnostic of the Physophoride is,
indeed, a most remarkable and well-defined structure (Pls. VI, VIII, &c.)

In these Hydrozoa the proximal end of the ccenosare expands into a variously shaped
enlargement, whose walls consist of both the ectoderm and endoderm, and which encloses a
wide cavity in free communication with that of the ccenosare, and, like it, full of the nutritive
fluid. From the distal end or apex of this cavity depends a sac, variously shaped, but always
with tough, strong, and elastic walls, composed of a substance which is stated to be similar

o
to chitin in composition,’ and more or less completely filled with air.

1 At least in Velella, See Leuckart, Z.N.K., p. 114,

6 THE OCEANIC HYDROZOA.

In the adult, this sac, which I shall term the preumatocyst, is sometimes open at
the apex (Physalia, Rhizophysa), and can communicate with the exterior by a pore which
traverses the ectoderm of the pneumatophore. In other cases its apex is shut, but the pneu-
matocyst has other external openings (Vedella, Porpita); and, lastly, in Physophora, Forskalia,
Agalma, Halistemma, and Athorybia, no external opening at all has hitherto been discovered.

In certain of these genera (Lorskalia, Agalna, Halistemma), the pneumatocyst appears to
be widely open below. I suspect that more careful examination would show that it
only becomes very thin; but however this may be, it is a mistake to suppose that there
is any communication between the interior of the pneumatocyst and that of the coenosarc.’

In fact, as I have particularly noted,’ in Physalia, Velella, Rhizophysa, Physophora, and
Agalma, the endoderm of the pneumatophore is reflected on to the pneumatocyst, where
the latter is in contact with the walls of the pneumatophore, and completely invests it,
forming a loose bag over the apparent inferior aperture, in those genera which possess one.

The pneumatocyst is thus firmly held in its position by the reflected endoderm, and
in some genera (forskalia and Agalma) additional support is afforded by septiform processes,
which pass from the lateral walls of the pneumatocyst to those of the pneumatophore.
These were first described by Milne Edwards in his admirable Memoir on Stephanomia
(Forskalia) contorta,> in which he says, that “the air-vesicle is open below, and retained
in a central position by membranous partitions disposed in a radiating manner, and stretched
between its parietes and those of the great pyriform cavity (of the upper end of the stem),
nearly in the same way as the mesenteries by which the alimentary canal is surrounded
in the Alcyonian polypes.”

I find no notice of these suspensoria in the works of either Vogt, Gegenbaur, or Kojliker,
and Leuckart expressly states (Z. U., p. 6) that he has ‘sought in vain for the suspensoria
described by Milne Edwards in Sfephanomia.” I observed them, however, very distinctly
in Agalma. (See the description of that genus 7”/rd.)

A very peculiar structure is attached to the distal surface of the pneumatocyst of
Rhizophysa (P). VIII). A great number of elongated, and more or less branched, processes, in
fact, project freely from it into the cavity of the pneumatophore. Each process consists of a
cellular axis, invested by the ciliated endoderm. The cells of the axis are clear and very
large, measuring as much as jth of an inch in length, and have an opaque, oval ‘‘nucleus’’
xath of an inch in diameter, with an oval or circular nucleolus of ;4,th of an inch in diameter.
Quoy and Gaimard would seem to have originally observed these appendages, and they
have since been carefully described by Gegenbaur. Nothing of the kind appears to have
been seen in other Physophoride, but in Velella and Porpita, the hepatic organ occupies the
same position, and hence one is led to suspect a relation between the two structures. It does
not appear to me that these ramified processes have any real resemblance to the pueumatic
filaments attached to the distal surface of the pneumatocyst in the last-named genera.

* See Leuckart, Z. U., p. 6, and Z. N. K., p. 67, who states of the pneumatocyst of Apolemia ;
“Die untere Oeffnung derselben die in den Reproductionskanal hineinfuhrt.”’

* Gegenbaur describes the same structure in Rhizophysa, and Leuckart in Apolemia and Forskalia,
so that the rule doubtless holds good for all Physophoride.

3« Aunales des Sciences Naturelles,’ t. xvi, 1841.

MORPHOLOGY. 7

In form, the pneumatocyst (and the pneumatophore which contains it) varies from that
of a spheroid (Riizophysa, Athorybia) or of an elongated oval (Physophora) to that of a
cylinder with a narrow inferior neck (Aga/ma), or an irregular, pear-shape (Piysalia), or the
figure of a flattened disc (Velella, Porpita). No less does one genus differ from another
in the proportional size of this apparatus. In Péysalia, Velella, and Porpita, it occupies
almost the whole of the hydrosoma, and constitutes the largest and most conspicuous
part of the body, while in such genera as 4yal/ma and Forskalia it attains to only such
insignificant proportions as to be readily overlooked..

Its functional importance, of course, depends very nearly on its relative size. When
it is large it must necessarily play a very considerable part in determining the habits of
its possessor, though we have at present no information as to whether it acts as a permanent
buoy or whether it can be voluntarily emptied and filled again.’ When the pneumatic
apparatus is so small in proportion to the rest of the organism, as in many Physophoride,
on the other hand, its office in the economy ceases to be easily comprehensible, and can
hardly be very important. :

The last modification of the proximal end of the hydrosoma of which I have to speak
is an organ—I mean the disc or wmbrella of Lucernariade—which is commonly confounded
under one head, with others—the nectocalyces—of quite distinct structure; apparently,
because, like the latter, it acts as an organ of propulsion. Nevertheless, in development,
in structure, and in even in mode of action, the umbrella is altogether distinct from any other
organ possessed by the Hydrozoa.

’ Iam inclined to think that the rudiments of this structure are visible in the Hydride,
and in some Corynide and Sertulariade. Yn examining a Hydra carefully, it is seen that
the tentacles do not immediately surround the oral aperture; but they arise in a circle
at some distance below it, so that the oral aperture is supported on a kind of conical
crater, from whose base they take their origin.

In some Campanulariad@, this separation of the polypite into a distal and a proximal
portion is still more marked. The latter ends distally in a truncated disc, from whose edges
the tentacula spring, while the oral division of the polypite is longer than the basal, and
arise from the truncated face of the latter by a narrow neck. The like structure is
observable in the “Hydra tuba,” the larval form of the Lucernarian Medusa, and from
hence to the structure of Zwcernaria, there is but a step. In the latter genus, in fact, the
discoid, proximal portion of the polypite, or proximal end of the hydrosoma, is greatly
enlarged, and produced into eight obtuse lobes, each of which gives rise to a few short
tentacles; while the short distal division of the animal lies in the middle of the disc.

In Zubularia, the buds which give rise to the reproductive organs are developed from
the surface of the polypite within the margins of the proximal disc and internal to its

1 There is very little good evidence to be met with on this much disputed question. I have seen
part of the contained air to all appearance voluntarily expelled from the pneumatocyst of Rhizophysa,
and Forskal appears to have witnessed something of the same kind, for he says of his Physophora
filiformis : “ Vivee, vesica aere plena, tamen subsidere possunt ; dum corpus arctando, se reddunt specifice
graviores.” Eschscholz found, on irritating a young Physalia five lines long, that it “ suddenly expelled
all the air from the bladder, and sank to the bottom of the glass” (p. 159).

8 THE OCEANIC HYDROZOA.

series of tentacula. In Zucernaria, the reproductive organs are developed in the same
way from the surface of the body on the distal side and within the margins of the disc,
into which the proximal division of the body is expanded.

The mode of origin of the Lucernarian Meduse from their larve is such, that their
umbrella is clearly, like that of Zucernaria itself, nothing more than a lobed expansion of the
body-walls, and as in Zucernaria, the reproductive organs are developed on the distal surface
of this expansion. The peculiar lithocysts of these Zucernariade are developed in the deep
notches which mark the ends of the lobes of their umbrella.

It will be seen presently, that in the mode of development just described, in its
lobed margins, and in the absence of any muscular membranous valve attached to its circum-
ference, the umbrella is wholly different from all other organs of natation.

Such is the general structure of the axis of the hydrosoma, and of its anterior and
posterior terminations, throughout the Mydrozoa. The different orders exhibit many
remarkable variations in their number, kind, and mode of attachment. To begin with the
last-mentioned point, it is to be observed that no regularity is traceable in the arrangement of
the appendages in many Corynide; while in others, and in the Sertulariade in general, the
branchings of the ccenosare and the disposition of the appendages upon it, follow a very
definite law, to which the regular and symmetrical forms of the organisms are due. The
appendages may be developed on one or on both sides of the coenosarc.

In such Physophoride as possess a filiform coenosarc (e.7., Agalma, Morskalia, Stephanomta),
the appendages appear to be always fixed to only one side of it. And even the strikingly
radiate disposition of some of the appendages in Forskalia and Stephanomia, for example,
does not result from their forming exceptions to this rule—for their appendages are all really
attached to one side of the stem only—but is due partly, perhaps, to a spiral twisting
of the ccenosarc, but in a much more important degree, to the manner in which these
appendages are forced by their peculiar form, to adapt themselves to one another.

Whether the appendages in Physophora, Velella, Porpita, and Physalia follow the same
law is not certainly made out.

In the Calycophoride 1 am inclined to believe that all the appendages are primarily
attached to one side of the ccenosarc, and that the subsequent opposition of the necto-
calyces to one another, and of the hydrophyllia to the polypites is a secondary modification.

The following distinct kinds of appendages exist in the Mydrozoa, and will now
be successively described. 1. Polypites; 2. Tentacula; 3. Hydrocysts; 4. Hydrothece ;
5. Hydrophyllia; 6. Nectocalyces; 7. Reproductive organs, consisting of Gonodlastidea and
Gonophores; 8. Lithocysts.

1. Polypites.

By this term I understand the principal organ of alimentation of a Hydrozoon. It
would be wrong to call it merely a stomach, for it is much more; and the word “ polype ”
has been, and is, used in so many senses that it is better avoided.

Every polypite is essentially a sac, open at one end, which serves as a mouth, while at
the other it communicates with the somatic cavity. The oral end is usually produced
into a thin, flexible, and very disténsible lip, whose edges are either quite simple (Hydride,

MORPHOLOGY. 9

Corynide, Sertulariade [all?], Calycophoride, Physophoride), or are produced into longer
or shorter folds or processes (many J/eduside, Lucernarian and other).

I have above referred to the division of the polypite into a distal and a proximal portion
in many L/ydrozoa. In the Calycophoride and Physophoride, the polypite presents, in many
cases, a further division into a proximal (“basal-stiick”’ of the Germans), a median, and
a distal division. There is no distinct line of demarcation between the two latter, but
in many cases the median and basal divisions are very sharply separated, not only by their
texture, but by a distinct valve (Pl. V).

This, in the Calycophoride, is a very well marked structure, though I do not find it noted
by any of those writers whose works I have consulted. It isa strong, circular fold of the
endoderm, whose lips, when the valve is shut, project into the cavity of the gastric, or median,
division of the polypite. As the oily or albuminous globules which result from the digestive
process are formed, they usually accumulate close to the valve, and are kept constantly
rotating by the cilia which line the gastric chamber. After remaining for a while in
this position, the fundus of the gastric chamber contracts, and forces the globule through the
valve, which appears to dilate at the same moment. The position and functions of this
apparatus, therefore, fully justify the appellation of a pyloric valve.

The proximal division of the polypite usually takes on the form of a peduncle, which is
sometimes very long, and gives origin to various appendages. Its walls are ordinarily thin and
muscular, but the endoderm is sometimes much vacuolated, and partially obliterates its cavity.

The median division of the polypite is the widest of the three, and has the thickest walls.
In it the process of digestion goes on, and hence it may with propriety be termed the gastric
division. The inner surface of the endoderm is richly ciliated, and not only is its general
thickness considerable, but in many Calycophoride and Physophoride it is developed into
larger or smaller, slender, conical, villous elevations. These villi are larger and more
distinct in Physalia (where they attain the length of ;th of an inch or more) than in
any other Hydrozoon I have examined (see the description of that genus, zz/rd) ; but they are
very well developed in Athoryéia. In both genera they exhibit in their interior one or more
clear spaces or vacuoles, sometimes obscured by a quantity of dark pigment, and they
contain numerous thread-cells (Pls. IX, X).

In Lhizophysa, Physophora, Diphyes, Abyla, and other Calycophoride, the villi are
represented by shorter processes of the endoderm, which are sometimes obsolete, all that
remains of the villus being the characteristic clear vacuoles, imbedded in the endoderm,
and the production of the endoderm into ragged-looking filaments over them.

Kolliker and Vogt describe similar organs in Hippopodius and Praya, so that such
short villi would seem to obtain universally among the Calycophoride.

In the Agalmopsis of Sars, in Agalma, Forskalia, and Apolemia, the villi take on the form
of longitudinal ridges, which usually contain much pigment, and thus give a very marked
character to the gastric division of the polypite. They have been confounded with reproduc-
tive organs.

Gegenbaur (p. 28) describes the vacuoles in the villi of Praya maxima as cells. “The
contents of these cells differed greatly, sometimes appearing perfectly clear, at others
yellowish or brownish, and, in this case, frequently consisting of minute particles. The

colour then shines through the walls of the polypite. I corisider these to be hepatic cells,
2

10 THE OCEANIC HYDROZOA.

analogous to those cellular elements which coat the gastric cavity and the intestines of

ce 7

so many of the lower animals.” In his Diphyes gracilis, Gegenbaur states that the “cells
possess a distinct double contour and a nucleus, neither of which have I ever observed.
“There can be no question,” he adds, “that these elements must be regarded as glandular
cells; and their ultimate fate is in favour of this view. For we find between them empty
depressions, with sharply defined edges, which correspond exactly in form and size with
those in which the cells in question are imbedded. These depressions can be hardly
anything else than the spaces formerly occupied by such glandular cells, which have burst
and emptied their secretion.”

Kolliker (p. 26) gives a still more definite account and figure of the structure of the villi
and vacuoles in 4thorybia. He considers them to be “glandular sacs of the simplest kind,”
also that they are “open sacs, lined with cells,” and inclines to the opinion that their function
is hepatic.

Leuckart takes the same view of the cavities in the villi of the ph iuscg as myself,
regarding them not as cells, but as vacuoles (Z. N. K., p. 68).

Vogt (p. 102) appears also to consider the cavities of the villi in Praya to be vacuoles
and not cells; and he gives an account of an experiment, which would be well worth repeating.
“Having mixed indigo in the water of a vessel containing a lively Praya, I saw after
some time that the digestive cavities were streaked with blue, the colouring matter being
detained by the villi (Zowrrelets); and I convinced myself by microscopic examination that the
colouring granules existed only in the cellceform spaces, which are nothing else than shallow
depressions or widely open glandular sacs.”

Without by any means denying the posstbility that the vacuoles (for such mere
excavations full of fluid I must confess they always appeared to me to be) contain a special
secretion, I am inclined to think that the villous emmences in which they are imbedded have
other functions. I once observed a half-digested mass in the stomach of an Athorybia, all the
villi in the neighbourhood of which were much elongated, and thrust into it. The ends
of these villi contained fewer thread-cells than usual, while many thread-cells were scattered
through the mass of food. Is it not possible that when the living prey is introduced into the
gastric cavity, its struggles may be restrained and cut short, not only by the mechanical
application of the elongated villi, but by the shooting out of the threads of the numerous
thread-cells with which they are provided :

Allman’ has described structures corresponding very closely with these vacuolated villi
in Cordylophora, and it is therefore probable that something of the same kind will be found in
other Hydrozoa. But, so far as I know at present, the only structures in the Lucernariade
to be compared with the villi are those solid tentacular filaments, with vacuolated axes,
which project from the endoderm into the stomach or into the somatic cavity. They are,
like the villi, covered at their extremities with abundant thread-cells.

The villi and vacuoles are confined to the gastric division of the polypite. The walls of
the distal or buccal division are thin and smooth, but richly ciliated internally.

1 «Anat. et Phys. of Cordylophora,” ‘ Phil. Trans.,’ 1853,

MORPHOLOGY. 11

2. Tentacula.

All the //ydrozoa possess more or less numerous filiform appendages, abundantly provided
with thread-cells, and subserving purposes of offence or defence, which receive the general
name of tentacula, though they differ very widely in structure and place of attachment.

In the Hydride, Corynide, and Sertulariade, numerous tentacula are always attached
to the body of the polypite itself, and they usually form a circle not very far below the mouth,
though sometimes they are scattered irregularly (Coryne, Cordylophora), and sometimes form
a double circlet (Zwbularia). The Calycophoride and most Physophoride have single
tentacula springing from the base of the gastric division of the polypite, or from the peduncle
just on the proximal side of that base. But in the Pdysaliade and Velellide, the tentacles
are wholly distinct from the polypite, arismg by themselves from the coenosarc. In the
Lucernariade and Meduside, finally, the tentacles are developed from the margins of the
umbrella and from its under surface.

The tentacles of the Hydride, Corynide, Sertulariade, Lucernariade, and Meduside, are
always extremely simple in structure, consisting, at any rate primarily, of tubular processes of
the endoderm and ectoderm, enclosing a diverticulum of the somatic cavity, and sometimes
clavate at the ends, or presenting little papillary elevations, but hardly ever branched. The
internal cavity is sometimes persistent, but it very commonly becomes almost obliterated
by the vacuolar thickening of the endoderm; and when this has occurred, the tentacles
usually appear as if they had a solid cellular axis.

Besides these ordinary tentacles, certain Sertulariade possess organs which must be
ranged in the same category, though they differ greatly from them in position and in external
appearance. These xemafophores, as they have been termed by Mr. Busk, are cecal processes
of the ccenosarc, invested by a continuation of its hard cuticular layer, so as to be quite
firm and inflexible. The cuticular investment, however, is open at the end, and in the
soft substance beneath the opening le a number of large thread-cells. These bodies are
particularly characteristic of the Plumulariade.

The tentacles of the various genera of the Physophoride and Calycophoride differ very
widely in structure, gradually increasing in complexity as we advance along a series, the
lowest term of which is Ve/el/a, and the highest Physophora.

The tentacles of Vele/la, in fact, differ in no essential respect from those of the Hydride
and Sertulariade ; they are simple cecal processes of the wall of the ccenosarc, with a
greatly vacuolated endoderm (Pl. XI). Those of Porpita have the same fundamenta.
structure, but they are branched at the ends.

The tentacles of Apolemia are described by Leuckart and Gegenbaur’ as simple, tapering,
unbranched filaments, which are beset with large thread-cells on one side, and are traversed
by a narrow, ciliated canal. One of these tentacles is said to arise from the base of each polypite.

The tentacles of Physalia (Pl. X) exhibit an advance on this structure. They arise inde-
pendently from the ccenosarc, and each is provided at its base with a large, pyriform, saccular

' ZN. K., p. 69; Gegenbaur, p. 40,

12 THE OCEANIC HYDROZOA.

dilatation—the basal sac. The tentacle itself is a filament which in a large specimen attains
the length of many feet when fully extended. It has no lateral branches, but itis beset along
one side, at regular intervals, with reniform enlargements, full of large thread-cells, which are
disposed transversely to the axis of the tentacle, and look like so many beads threaded upon
it. On the opposite side, the tentacle widens out into a ribbon-like muscular band, which,
attached above to one edge of the basal sac, is the agent of its rapid and extensive contrac-
tions. A canal traverses the whole length of the tentacle, and sends cecal diverticula into the
reniform enlargements, while above, it communicates with the cavity of the basal sac. It has
been supposed that the latter organ, by its contraction, drives the liquid which it contains into
the canal of the tentacle, and thus effects its elongation. Without denying that such may
be its office, I would remark, that the tentacles of other species which are not provided with
basal sacs are just as capable of rapid elongation.

The reniform enlargements to which I have referred may be regarded as rudimentary
latera) branches. If they be supposed to elongate and become filamentary, the result will be
a tentacle very similar to that possessed by Rhizophysa, except that in this genus there is
no basal sac, nor muscular band, and that each tentacle is attached to a polypite. The lateral
branches of the tentacles of RAzzophysa have one wall much thicker than the other, but
it contains only spheroidal thread-cells, and the branches are not divided into distinctly
characterised regions.

In Forskalia, however, while the tentacles have essentially the same structure, each lateral
branch is divided into three distinct portions: a proximal slender part; a median division,
with one wall much thicker than the other, containing numerous elongated thread-cells,
arranged in transverse rows perpendicularly to the wall, and flanked on each side by a longi-
tudinal series of larger oval thread-cells; and, finally, a terminal cylindrical thread, full
of small, rounded thread-cells. I shall term the first of these regions the pedicle, the second the
sacculus, and the third the filament. The muscular band of Physalia is partially represented
by two pairs of muscular cords, which, according to Leuckart (Z. N. K., p. 99), lie in the thin
wall of the sacculus. In the contracted state, the sacculus and the filaments are thrown into
spiral coils.

The structure of the tentacles of Halistemma is essentially the same, except that they
are provided with a more complex muscular apparatus, fora description of which I must refer
to the works of Vogt and Leuckart.

The tentacula of the Calycophoride (Pl. V) resemble those just described, and arise either
from the base of the gastric division of the polypite, close to the pedicle, or from the latter itself.
The larger, oval thread-cells are confined to the distal end of the sacculus, which is usually
bent so as to have a half-moon shape, the thick wall forming the convexity. The filament is
coiled up into a close spire, folded against the straight, thin wall; and, where it joins the
sacculus, the points of five or six oval thread-cells commonly project, like those of the rowel of
a spur. Where the peduncle joins the sacculus it exhibits a small dilatation, which I
conceive to be a rudiment of a part to be described presently as the involucrum.

Leuckart has particularly described a structure in the sacculus of the Calycophoride,
where it has also been noticed by Vogt and Kolliker, which he terms the “ angelband.”
It is “a simple, but strong and sharply defined, muscular cord, which is folded in zigzags,
and lies in the posterior (thin) wall of the canal of the sacculus, partly covers it at the

MORPHOLOGY. 13

sides, and then appears to be coiled almost spirally around it. The upper end passes gradually
into the pedicle, while the lower extends as far as the beginning of the filament. In Praya
and /7ippopodius, this cord has a diameter of about ;’,th of an inch, and differs in no
essential respect from the muscles in other parts of the body, especially in the stem, although
at times a slight transverse striation can be detected in it. In Diphyes, and still more in Abyla,
however, this muscular cord becomes gradually thicker during its course, so as even to attain as
much as ;,th of an inch in 4dy/a, and therewith assumes a very distinct transverse striation,
so that, especially in Aéy/a, it might be compared to the most beautiful transversely striped
muscular fibres. No nuclei can be observed in this cord, nor can its sheath be distinguished
from its contents. It appears as if the transverse striation were caused by a regular jointing,
for the edges of the cord are completely incurved at intervals, corresponding with the
constrictions between the joints. In ddy/a, furthermore, this muscular filament is but
little flattened, and becomes triangular in many places by the mutual pressure of the
superimposed folds. If one of the thin edges be accidentally turned directly towards
the microscope, the transverse striation readily leads one to suppose that two series of
transverse rods are imbedded in the cord, as I indeed previously supposed to be the case.”
(ZN. K., p. 19.)

I confess I entertain great doubts as to the real nature of this structure, which is
particularly worthy of the attention of future observers.

In the tentacles of Stephanomia (Pl. VII), which, in many respects, resemble those of
Halistemma, a new part makes its appearance, in the shape of a sort of hood, which
is developed at the junction of the pedicle with the sacculus, and encloses the latter like a
cup. I term this the czvolucrum. It is a solid, lamellar process of the ectoderm, containing no
internal cavity. The sacculus is very long and spirally coiled, terminates in a single filament,
and has a well-developed muscular band in its thin wall.

In the genera Agalma and Athorybia, the involucrum has become much larger in
proportion to the sacculus, and the latter is terminated, not by a single filament, but by
two filaments, between which the sacculus ends in a clear, thin-walled, median lobe,
devoid of thread-cells, and said to be contractile by Leuckart and Kolliker, though in the
species I observed it exhibited no such faculty (Pls. VI, IX).

Finally, the tentacular branches attain their utmost complexity in Physophora, the
spheroidal involucrum here completely investing the sacculus, which lies coiled up within it,
and having undergone other changes, which will be particularly described under the head of
that genus (PI. VIII).

3. Hydrocysts.

I apply this name to certain singular organs which are found more particularly in the
Physophoride, and which resemble nothing so much as the imperfectly developed polypites of
the species to which they belong. As such, indeed, I always considered them, until
the perusal of the works of Philippi, Leuckart, and Kélliker, led me to modify my
opinion.

These investigators term the organs in question “fihler” and “taster,” and are

14 THE OCEANIC HYDROZOA.

inclined, not without a great show of reason,’ to regard them as organs of prehension and
touch, to which may perhaps be added excretory and respiratory functions.

The hydrocysts are always pyriform sacs, composed of the ectoderm and endoderm,
shut at their apical or distal ends, where they are commonly provided with large thread-cells,
but, at their proximal ends, in free communication with the somatic cavity. Like the polypites,
they usually give origin to a tentacular appendage. But this is always simple and filiform,
and the hydrocysts further differ from the mere polypites in their closed apices and in the
general absence of villi. The latter, however, exist in a rudimentary state in the hydrocysts
of Apolemia, according to Leuckart (Z. N. K., 70), and I have seen them in the closed
sacs, which appear to be hydrocysts, of Physalia (Pl. X). These bodies are not found in
Velella or Porpita, and 1 must confess I am very doubtful, whether the structures to which Ihave
just referred in Péysalia are other than young polypites. In the Stephanomiade they
are attached to the ccenosarc, between the polypites, and are usually in more or less close
relation with the reproductive organs. In Physophora (Pl. VIII) a circlet of large hydrocysts
is interposed between the nectocalyces and the polypites, and in <Apolemia, according
to Kélliker, Gegenbaur, and Leuckart, a group of them surrounds the base of each polypite,
while solitary ones are, in addition, interspersed between the nectocalyces.

4. Hydrothece, and 5. Hydrophylha.

Many Hydrozoa possess appendages whose only function would appear to be to serve as
a protection to other parts of the organism, more especially the polypites. Of these organs
there are two very distinct kinds, the one peculiarly characteristic of the Sertwlariade,
the other found only among the Diphyde and Physophoride. ‘Yhe former, or /ydrothece,
are what are commonly termed “polype cells.” They are cup-like receptacles, entirely
composed of the cuticular layer of the ectoderm, and consequently contain no diverticulum
of the somatic cavity at any period of their existence. They are primarily developed
from the whole outer coat of a budding polypite.

For the second class of protective organs, or Aydrophyllia, there is no generally accepted
English name. I have termed them “bracts” in my earlier memoirs, and the Germans
call them “ deck-stiicke.”

The hydrophyilia differ entirely from the hydrothece, for they always contain a diverticulum
of the somatic cavity, or phyllocyst, and are consequently composed of both ectoderm and
endoderm, though their principal mass is furnished by the ectoderm. Again, they always
commence their existence by budding from a limited segment of the coenosare, or of the pedicle
of a polypite. In the Calycophoride, they are attached to the pedicles of the polypites,
which they eventually surround more or less completely (Pl. V). They are either thin and
foliaceous, as in Diphyes, or thick and facetted externally, as in 4dy/a. Their form and
arrangement vary greatly in the Physophoride. n Forskalia, numbers of them are developed
from the pedicles of the polypites, as wellas from the coenosarc, while in Agalma (Pl. V1) they

1 See Kolliker, p. 21; Leuckart, Z.U.,p.17; and, again, Z. N. K., p. 69, where the strong
resemblance of these bodies to polypites is fully admitted.

MORPHOLOGY. Hi)

are confined to the latter. Where nectocalyces are present, the series of hydrophyllia always
ends on their distal side, these organs being never found intermixed with the nectocalyces,
or on their proximal side. The Velellide, Physaliade, Rhizophyside, and Physophoriade
have no hydrophyllia.

6. Nectocalyces.

Nothing is more peculiarly characteristic of those orders of the Hydrozoa to which the
following pages are particularly devoted, than the structure of their locomotive organs,
which, whatever the modifications they undergo, may be reduced to a very simple type
—that of a cup lined by a muscular membrane, by whose contractions the water is expelled,
the animal being consequently urged, by its reaction, in the opposite direction. Nor
is the essential internal structure of this cup less uniform. Its summit contains a cavity,
which is connected with the somatic cavity on the one hand, and, on the other, sends
off at least four canals, which traverse the walls of the cup in a radiating manner, and
eventually open into a circular canal, which surrounds its mouth. These xectocalycine canals
are lined by a continuation of the endoderm. The cavity of the cup, which, with its muscular
wall, may be termed the mecfosac, does not communicate with this system of canals,
but is freely open externally, and, opposite the circular canal, its wall is always produced into
a contractile, valve-like membrane, which, when at rest, projects inwards all round the
aperture, and narrows it to a greater or less extent (Pls. I, VI, VIII).

The presence of the valvular membrane at once distinguishes a nectocalyx from an umbrella.

These organs of propulsion’ are only known to exist among the Calycophoride, where
they are universal, and the P/ysophoride, which only partially possess them.

In the typical Calycophoride (Diphyes, Abyla, Galeolaria), there are normally only
two Nectocalyces, which are attached upon opposite sides of the proximal end of the
hydrosoma, but in Vogtia they are numerous, and in Hippopodius, according to Leuckart,
amount to as many as twelve or more. In the Physophoride they are often extremely
numerous, and (though all originally developed on one side of the coenosarc) are, in the
adult state, either arranged in two opposite series, or in several series, which are disposed in a
radiating manner around the proximal end of the ccenosare.

In the Piysophoride the inner faces of the nectocalyces are usually excavated
to enclose the ccenosarc, and those of opposite sides simply alternate with, and fit in between,
one another. Each nectocalyx is connected with the coenosare by a pedicle, traversed by
a tubular duct, which opens on the one hand into its apical cavity, and on the other into
the somatic canal, and is, in truth, nothing but the original base of the organ elongated.

In all the Physophoride, and in most of the Calycophoride, that portion of the ccenosare
which supports the nectocalyces lies in the same straight line with the rest; but in
Hippopodius, Leuckart has pointed out that it is bent down, and, as it were, folded upon
the remainder of the coenosarc, whence the distal end of the series of nectocalyces may be

readily mistaken for the proximal, and vice versd.

1 Tt is necessary to distinguish carefully between these simply locomotive organs and those similar

structures which are subordinated to reproduction, and contain a central gastric or genital sac.

16 THE OCEANIC HYDROZOA.

In Praya, the two nectocalyces are attached, nearly on the same level, to the proximal
end of the coenosarc. The inner surface of each is marked by a deep groove, and the smaller
is received within the bounding folds of the groove of the larger, so that the two form,
by their application, a sort of chamber or /ydracium, into which the coenosare can be
retracted, as into a house.

In the typical Calycophoride—Diphyes, and Abyla—a still further change has taken
place (Pl. V). As in Praya, there are but two perfect nectocalyces, but of these, one is
altogether on the distal side of the other, its pointedapex being received into a peculiar
cavity of the proximal nectocalyx, which takes the place of the internal groove in the
Physophoride and Praya. It is to the summit of this cavity that the proximal, slightly
dilated end of the coenosarc, described above, is attached, its continuation, the somatocyst,
being imbedded in the substance of the nectocalyx. Proximally, therefore, the
hydreecium in these Calycophoride is entirely formed by the conical chamber of this
nectocalyx; distally, it is continued by a groove, converted more or less completely into a
covered way, on the inner surface of the distal nectocalyx. The ccenosare, with its
appendages, travels up and down the complex hydreecium thus formed, and can sometimes
be completely retracted into it.

The ducts which connect the nectocalycine canals with the somatic cavity come off,
as I have stated above, from the somewhat dilated, ciliated chamber in which the coenosarc
ends. The duct which goes to the canals of the proximal nectocalyx in Diphyes, runs
along that side of the hydreecium which is nearest to the nectosac, and divides into the
longitudinal nectocalycine canals close to the mouth of the latter. In Adyla, on the
contrary, it is extremely short, and divides directly opposite the middle of the nectosac.

The duct which supplies the canals of the distal nectocalyx almost immediately enters
that beak-like process of the latter, which fits into the hydreecium of the proximal necto-
calyx, and traversing it, divides into its four longitudinal nectocalycine canals, either
opposite the apex of the nectosac or close to it.

I have described above the general arrangement of the four canals in the necto-
calyces, which radiate from the end of the communicating duct. Whatever the form
of the nectocalyx, that pair of these canals which occupies the median plane remains
straight, or as nearly so as the form of the organ will permit; but the two lateral ones not
uncommonly undergo a curious flexure, of which a marked example may be seen in Physophora'
(see the description of that genus, %mfrd). Leuckart describes. in the nectocalyces of
Hippopodius, Praya, Halistenma, and Agalma (and I can testify to their existence in the last-
named genus), what he terms “mantelgefasse.” These are two slender, curved, cecal
diverticula, which are given off on opposite sides of the duct, shortly after it enters the
substance of the calyx, and lie in the same plane as the median nectocalycine canals.

7. Reproductive Organs.

These consist, throughout the Hydrozoa, of spermaria and ovaria, portions of the tissue of
the wall of the hydrosoma, and, I believe, more particularly of the ectoderm, metamorphosed

1 Also, according to Leuckart, in Agalma, Halistemma, and Apolemia.

MORPHOLOGY. 17

into spermatozoa or ova. As an invariable rule, while the reproductive organ is in contact,
on the one side, with the nutritive somatic fluid, it is, on the other, directly exposed to
the water in which the animal lives. The genital apparatus is therefore always external,
and is never contained within the cavity of the body, a character which at once distinguishes
the Hydrozoa from the Actinozoa.

These organs are found in their simplest state in Mydra, in Lucernaria, and in many
Medusa, where they are distinguishable from other parts of the body only by their contents ;
but more usually they assume the form of pouches or sacs, termed yonophores by Allman,"
containing a diverticulum of the somatic cavity. Such is their form in Hydractinia, in
Coryne pusilla, and in some Plumularia.

The reproductive organs of both sexes in Cordylophora, the female organs of Stephanomia,
Agalma, Athorybia, Forskalia, Apolemia, and the male organs of Physalia, exhibit an
interesting series of transitions from the simplest to the most complex condition of the genital
apparatus in the Hydrozoa. For we find in all these cases that the reproductive organ is
no longer simple, but that its investment has acquired more or less the structure of a
nectocalyx, from whose walls those of the actual genital sac, the manubrium of Allman, tend
to become independent. To this investing part I apply the name of gonocalyz.

In Zubularia indivisa both the male and the female reproductive organs are included
within a gonocalyx, in which the four longitudinal canals are distinguishable while the organ
is yet young, though they appear to become obliterated with age. But the organs remain
attached, and their calyces, so far as I have observed, never exhibit contractions. A similar
condition of the reproductive organs has been observed in some Campanularie.

Hippopodius and Vogtia have, according to Kélliker, long manubria (sperm-sacs and
ovisacs), surrounded at their bases by short and shallow hemispherical gonocalyces, which
have the four longitudinal canals united by a circular canal, characteristic of a complete
nectocalyx.

The reproductive organs of both sexes of other Calycophoride, and the male organs, of
Athorybia, Agalma, Apolemia, Stephanomia, Halistemma, and Forskalia, have a still more com-
plete, contractile gonocalyx, provided with its membranous valve, and the manubrium or
reproductive sac suspended from its roof either projects but very little beyond its aperture, or
is much shorter. In many of the Calycophoride these organs become detached, and swim
about for a long time, propelled by the contractions of their calyces.

From these conditions of the reproductive apparatus, the transition is easy to that presented
by those of both sexes in Zaomedea geniculata, many Corynide, and Velella, and by the
female organs of Physalia; in which the reproductive organ is detached before the develop-
ment of the generative products within the manubrial sac suspended from the roof of the
gonocalyx, and swims away as a medusiform zéoid. Not unfrequently, this “ Medusa” has
to undergo great changes in size and form before attaining its sexually perfect state ; and it
is, consequently, necessary that it should feed and support itself for a considerable period.
To this end, the central sac or manubrium no longer remains a mere short receptacle, but

1 « On the Structure of the Reproductive Organs in certain Hydroid Polypes,” “Proceedings of the
Royal Society of Edinburgh,’ Session 1857-8; and “ Additional Observations on the Morphology of the

Reproductive Organs in the Hydroid Polypes,” Idid. December 6th, 1858.
3

18 THE OCEANIC HYDROZOA.

opens at its extremity, acquires the structure, and takes on all the functions of a polypite. At
the same time, tentacular organs and lithocysts may be developed from the margins of the
calyx. Eventually, the ova or spermatozoa appear either in the walls of the polypite, or in
special spermaria or ovaria developed in the parietes of the canals of the calyx cf this inde-
pendent medusiform zéoid, which is occasionally a more complex organism than that from
which it sprang.

This is still more remarkably the case in most of the Lucernariade, whose reproductive
zooids surpass the “ Hydree tube,” whence they proceed, many thousandfold in size and mass,
and no less remarkably, in complexity of organization. Known under the same general title
of “ Medusz ” as the foregoing, and in many ways analogous to them, they are, nevertheless,
widely different in structure and origin. Their swimming apparatus is an umbrella, and neither
in structure nor in mode of development a gonocalyx, and they are developed, not by budding
from a limited area of one side of a ccenosarc or its polypite, but by the transverse constric-
tion of a polypite into superimposed segments, each of which becomes lobed at its margins,
and assumes the form of an umbrella with a central polypite.

The arrangement of the tentacles, again, is quite different from that which obtains in the
other Meduse, and the lithocysts differ in being covered by a hood, and in not having each
mineral particles which they contain, spheroidal and enveloped in a distinct vesicle.

In Cyanea, the polypite is simple and single, and the spermaria or ovaria are saccu
lated portions of the distal (or under) wall of the umbrella, inclosed in a peculiar chamber
formed by the outgrowth, into a thick ridge, of the ectoderm round each reproductive organ,
so as to leave only a small, circular aperture, leading into a chamber, from whose roof
the plaited membrane containing the reproductive elements depends.

In this genus, moreover, the edges of the produced angles of the lips of the polypite, which
hang down like four great arms, are so folded as to form little cups, into which the fertilised
ova are received, there to undergo their first changes.

In the Ldizostonide, a complex, tree-like mass, whose branches, the stomatodendra, end
in, and are covered with, minute polypites interspersed with clavate tentacula, is suspended
from the middle of the umbrella in a very singular way. The main trunks of the dependent
polypiferous tree, in fact, unite above into a thick, flat, quadrate disc, the syndendrium, which
is suspended by four stout pillars, the dendrostyles, one springing from each angle, to four
corresponding points on the under surface of the umbrella, equidistant from its centre. Under
the middle of the umbrella, therefore, there is a chamber whose floor is formed by the
quadrate disc, while its roof is constituted by the under wall of the central cavity of the
umbrella, and its sides are open. The reproductive elements are developed within radiating,
folded diverticula of the roof of this genital cavity.’

In passing from Hydra to Rhizostoma, we thus see the reproductive organs acquiring a
greater and greater relative mass, when compared with the organism from which they spring,
and, as it were, grouping round themselves and subordinating to their own perfection
a greater and greater number of morphological elements. First, they are parts of the
body wall, indistinguishable in form from the rest; then they are distinct sacs; then they are

T have described at length and figured, the structure here indicated, in my ‘ Memoir on the
Anatomy of the Meduse,” ‘ Phil. Trans.,’ 1849,

MORPHOLOGY.» 19

sacs with a gonocalyx ; then that gonocalyx becomes a well-developed contractile organ ; next
the reproductive apparatus is detached, and swims about independently by means of its gonocalyx
or umbrella; and, finally, it acquires total independence, feeding and nourishing itself, and attain-
ing the most complex organization exhibited by the class to which its originator belongs.

Even this, however, is not the whole extent’ of specialization attained by the reproductive
system in the Hydrozoa, for, in certain members of the class, the reproductive organs,
or zooids, themselves are developed from, or in, organs especially devoted to that object.

Thus, among the Corynide, the gonophores of /Hydractinia echinata are developed neither
from the ordinary polypites nor directly from their connecting ccenosare ; but the latter gives
rise here and there to cylindrical processes (4/astostyles), each of which is about as long as a
polypite, but is terminated distally, not by a mouth, but by a pyriform enlargement, which
gradually diminishes to a point, and is covered with small, irregular lobes or tubercles.
The circumference of the widest part of the enlargement is produced into ten or twelve
such particularly well-marked conical enlargements, which are thickly beset with thread-
cells. From the stem of the process below these, the gonophores are developed, and from
its base there arises a short, cylindrical body, dilated at its extremity into a globular head,
full of minute, dark reddish orange granules. At first sight this body looks very like the
lithocyst of a Cyanea, but I have been unable to find evidence of its possessing a similar
structure.

Some Sertulariade exhibit organs similar in principle to these, though differing in the
details of their structure. Thus, pedunculated, urn-shaped bodies rise from the ccenosare
of Laomedea gelatinosa. The peduncle has the same structure as the ccenosarc, containing a
central canal, and having the cuticular layer of its ectoderm more or less distinct from
the deeper substance. Where the peduncle enlarges into its urn-like dilatation, the walls of
the latter may be clearly seen to be continuous with this cuticular layer of the ectoderm only ;
at the distal end of the urn its walls turn abruptly inwards on all sides, thus forming
a concave face, the middle of which is produced into a short, spout-like, open mouth.
The continuation of the rest of the substance of the peduncle (consisting of the inner moiety
of the ectoderm, with the whole endoderm, and their contained cavity) traverses the axis of
the urn nearly to its end as a sort of columella—the blastostyle of Allman—but its termination
is hidden by the mass of buds in advanced stages of development which are clustered round it.
These buds may be traced down the sides of the blastostyle, until, at the base of the urn, they
are to be detected in their primary condition of czecal processes of its wall. Each eventually
acquires the form of a ‘Medusa,’ with marginal tentacles and lithocysts, and leaves the
theca by its superior aperture to swim about by itself, and eventually develope the repro-
ductive elements.

The relation of the reproductive organs of many P/ysophoride to the peduncles of those
peculiar czecal polypiform bodies described above under the name of hydrocysts, or, as in
Velella, to true polypites, is very similar to that between the reproductive organs of
Hydractinia’ and their support, or between the budding medusiform zéoids of Laomedea and

' Vogt (p. 138) and Leuckart (Z. U., p. 16, and Z. N. K., p. 71) have already drawn attention to
this resemblance, and have included, under the common term of “ proliferous individuals,” not only these

bodies, but the small polypites of the Veledlide.

20 THE OCEANIC HYDROZOA.

the columella of the urn; and it will be so convenient to have a special name for these
organs set apart for the production of generative buds, that I will term them gozodlastidia.

Among the Meduside, Willsia (itself possibly only a zéoid) possesses stem-like gonoblastidia
terminated, as in Hydractinia, by an enlarged end, abounding in thread-cells.

I have already spoken of the detachment of the medusiform gonophores of many
Hydrozoa. But in many Calycophoride, and, perhaps, in some Physophoride, the organism
undergoes a still further subdivision in the natural course of its development.

This process will be found described in detail below, as it occurs in Adyla. It will be
seen that each segment of the ccenosarc, provided with a polypite, its tentacle, reproductive
organ, and hydrophyllium, as it acquires a certain size, becomes detached, and leads an
independent life—the calyx of its reproductive organ serving it as a propulsive apparatus.
In this condition it may acquire two or three times the dimensions it had when detached, and
some of its parts may become wonderfully altered in form.

It is by no means improbable that some of the P/ysophoride may undergo a similar
disruption, whole groups of organs, as in Apolemia, or at any rate the gonoblastidia,
becoming detached.’ Eschscholz (p. 159), who observed the ready separation of the latter
in Physalia, imagined that they developed directly into young Physalie.

The preceding section was written before Professor Allman’s valuable papers, cited at p. 17, came into
my hands. I have adopted his terminology as far as possible; but I have been compelled to make
certain modifications in it, for reasons which will be stated in a note at the end of this general intro-
duction to the study of the Hydrozoa.

SECT. II. SPECIAL MORPHOLOGY OF THE HYDROZOA.

I have now briefly described all the organs* of the Hydrozoa and their leading modifica-
tions, or, in other words, I have given some account of the general morphology of the class.
The various modes in which these organs are grouped together is the subject matter of
the present section on the special morphology of the group. Of these modes there are,
at most, six or seven, and each, when clearly expressed, defines the plan of structure
common to the members of one of the several groups or orders into which the zoologist
divides the class—for a scientific classification is, after all, nothing more than a conve-
nient mode of expressing the facts and laws established by the morphologist.

The first type, or plan of structure, is that exhibited by the order of the Hydride,
containing the single genus Hydra, whose hydrosoma consists of only a single polypite,
surrounded towards the distal end by a circlet of simple filiform tentacula, and expanded
at the proximal end into a discoid hydrorhiza. The reproductive organs are simple
spermaria or ovaria developed in the walls of the polypite, and the ectoderm developes
no hard cuticular layer.

1 See Leuckart, Z. U., pp. 69, 70.
* T have reserved the consideration of the Lithocysts for a future occasion, when the Meduside

observed during my voyage will be described.

SPECIAL MORPHOLOGY. 21

The second order, or the Corynide, have the hydrosoma developed into a ccenosare of
very various forms, supporting many polypites without theca, upon which are arranged
in circlets, or irregularly, many filiform tentacula. The ectoderm of the ccenosare (always P)
developes a strong cuticle. The reproductive organs are produced by budding from the
bases of the polypites, or from the coenosarc, or from special gonoblastidia, but they present
every variety in structure. The hydrosoma is fixed by a hydrorhiza.

The Sertulariade, which form the third order, have a coenosarc, with a strong, chitinous
cuticular layer, which is usually branched, and supports polypites enveloped in theca.
Each polypite supports a subapical circlet of filiform tentacula. The reproductive organs
vary in structure, but are always developed either from the ccenosare or from gonoblastidia.
The hydrosoma is fixed by a hydrorhiza.

The fourth order, or the Calycophoride, have an unbranched ccenosarc, which is flexible
and contractile, and has no hard, chitinous cuticular layer. The hydrosoma is free, and is
propelled by nectocalyces attached to its proximal end. The polypites have only one
tentacle, developed near their basal or proximal ends, and provided with lateral branches
ending in saccular enlargements. They have no thece, but are sometimes enveloped in
hydrophyllia. The reproductive organs are always medusiform gonophores, produced by
budding from the peduncles of the polypites.

The fifth order contains the Physophoride, which have an unbranched or very slightly
branched, flexible, and contractile ccenosare, which has no hard, chitinous outer layer. The
hydrosoma is free, and its proximal end is modified into a pneumatocyst. It may be
propelled by nectocalyces or not. The polypites have either a single basal tentacle, or the
tentacles, whose structure varies greatly, arise directly from the ccenosarc. ‘There are no
thecee. Hydrophyllia are commonly, but not always, developed from the peduncles of the
polypites, or from the coenosarc. The reproductive organs vary, but are never simple
sacs. They are developed upon gonoblastidia.

The Zucernariade, which constitute the sixth order, have the base of their hydrosoma
developed either primitively, or in the medusiform zéoids to which they may, by a process of
fission, give rise, into an umbrella. The ccenosarc has no chitinous cuticle. The tentacles are
simple, and are developed from the umbrella or its representative. There are no nectocalyces
or hydrophyllia. The reproductive organs are developed in the wall of the umbrella of
the primitive polypite, or in that of the medusiform zéoids produced from it.

Finally, I have retained the term Meduside as, at any rate, a temporary denomination
for all those members of the group termed Cryptocarpe by Eschscholz, Gymnophthalmata
by Forbes, Craspedota by Gegenbaur, with whose origin we are unacquainted. It is not
yet proved’ that any of them are developed directly from the eggs of similar organisms,
and until this is the case the order can be regarded only as a provisional one.

Using the term in this sense, the Meduside have a hydrosoma consisting of a single

' In using this expression, I by no means wish to question the great probability of the suppo-
sition that those ciliated embryos, which were observed by Miiller (¢ Miiller’s Archiv,’ 1851) and
Gegenbaur (‘Zur Lehre yom Generations-wechsel,’ 1854) to pass directly in the ‘“ Medusve ” Cginopsis
mediterranea and Trachynema ciliatum, proceeded from the eggs of similar Meduse. But, I repeat,

there is no proof of the fact.

22 THE OCEANIC HYDROZOA.

polypite, with or without tentacles, suspended from the roof of a bell-shaped body, having the
structure of a nectocalyx, and like it, provided with a marginal valve, and with radiating
and circular nectocalycine canals. The reproductive organs are usually simple spermaria
or ovaria, developed in the walls of the canals or in those of the polypite.

SECT. Ill. GENERATION AND DEVELOPMENT OF THE HYDROZOA.

The substance of the spermarium in all the Hydrozoa in which I have traced its develop-
ment (and I have studied to this end members of each of the orders) becomes differentiated
into minute, clear, spherical vesicles, of about ;4;th to z'sth part of an inch in diameter, in which
I have been able to observe no further structure. The vesicle gradually becomes pointed at
one end, while at the other it developes a slender cilium, and, eventually, the vesicle assumes
the characters of the head of the spermatozoon, while the cilium becomes its tail. The
head of the spermatozoon is always broader at that end to which the cilium is attached.

The ovarium, in like manner, breaks up into a number of spheroidal, or originally,
polygonal masses, each of which contains a clear space, with or without a distinct wall—the
germinal vesicle, and in its centre a smaller, thick-walled vesicle—the germinal spot. I have
never observed any vitelline membrane around these ova. Allman describes none in
Cordylophora, and Gegenbaur (p. 49) distinctly denies the existence of any in the ripe ova of
those Calycophoride and Physophoride which he examined.

Gegenbaur (p. 49) has observed the direct contact of the spermatozoa with the ova
in the Calycophoride and Physophoride. The spermatozoa never entered the ovisacs,
but as soon as the ova were detached they were surrounded by the spermatozoa, which fixed
themselves by their heads to the yelk, their vibrating tails radiating in all directions.
Gegenbaur, however, does not appear to have witnessed the direct penetration of any
of them into the substance of the yelk. Complete yelk-division takes place in the ordinary
way in all Hydrozoa, and Gegenbaur has made the important observation that, in certain
Corynide,' Calycophoride and Physophoride, the germinal vesicle does not disappear, but that
its division immediately precedes that of the yelk—so that its progeny must eventually
become the “embryo cells” of the division masses.

Towards the end of the process of yelk-division, cilia appear upon the surface of the
embryo.

Thus far all the Hydrozoa appear to follow a like course, but from what is at present
known it would seem that the different orders diverge somewhat in their further progress.

In Tubularia and Cordylophora, in Campanularia (?) in Cyanea (and hence, probably, in
all Corynide, Sertulariade, and Lucernariade), a cavity rapidly makes its appearance in the
centre of the germ-mass, which thus becomes a blastodermic vesicle, and then the walls
of this vesicle are differentiated into ectoderm and endoderm.

From Gegenbaur’s observations it would appear that in the Calycophoride and Physo-

‘See the account of the development of the ova of the Corynidan medusiform zéoid, Oceania

armata, in Gegenbaur’s instructive Essay, ‘ Zur Lehre vom Generations-wechsel,’ 1854.

GENERATION AND DEVELOPMENT. 23

phoride, on the other hand, the blastoderm, whence the organs are developed, is, at first,
confined to a comparatively small portion of the divided yelk, where it appears as a
thickened elevation. What becomes of the rest of the germinal mass, and what organ is
first formed in the Physophoride, is not known, but in the Ca/ycophoride, at least in Diphyes,
Gegenbaur states that the first-formed protuberant blastoderm gives rise to a nectocalyx.
The short and broad peduncle of this body is traversed by a canal, which expands within the
remainder of the germinal mass into a ciliated cavity, which would appear to correspond
with the internal cavity of the germ in the other orders. The nectocalyx attains a consider-
able relative size, and carries about the germinal mass with its contained cavity as an
appendage at its proximal end. The first rudiments of the polypites bud out between this
appendage and the nectocalyx (See Pl. V).

Thus far, Gegenbaur’s direct observations go. He concludes from them that the
nectocalyx which is first formed is the distal one of the two possessed by the adult,
and that the mass of the yelk with its central cavity is converted into the somatocyst,
which lies within the substance of the proximal nectocalyx of the adult.

The first conclusion is grounded on the fact that the nectocalycine duct divides into its
canals close to the apex of the nectosac, which, as Gegenbaur justly points out, is an arrange-
ment characteristic of the distal nectocalyx. But I confess this argument loses much of its
apparent weight in my mind, when I reflect on the immense changes the minute nectosac
must undergo before it can attain its adult form, and on the readiness with which, during this
metamorphosis, the relations of the point of division of the duct to the sac might be changed.

Again, without by any means denying the possible or probable validity of the second
conclusion, I must remark that mere similarity of tissue is hardly a sufficient ground
for assuming the identity of an embryonic with an adult structure.

Of the mode of formation of the body of the embryo and of the pneumatocyst in the
Physophoride nothing is at present known. The youngest forms hitherto observed are those
described by Gegenbaur (in Physophora, Forskalia, and Agalma) and myself (in Physalia, infra),
and they were already provided with a pneumatocyst of large proportional size, and
were terminated by a single large polypite, with a rudimentary tentacle.

In the Corynide, Sertulariade, and Lucernariade the ciliated embryo, after swimming
about for awhile, fixes itself by one end, which is modified into the hydrorhiza, and then,
elongating, becomes the coenosarc, whence all the appendages are developed.

A wonderful uniformity pervades the first stages of the development of all the
appendages of the Hydrozoa, notwithstanding their striking ultimate differences ; but before
passing to this part of the subject I must direct attention to the laws which govern the
development of appendages in general upon the coenosarc.

In the Corynide, Sertulariade, and Lucernariade new polypites are very generally,
if not always, developed at, or near, the distal end or surface of the coenosarc, so that
the distal polypites are the youngest. Whether this law also holds good with regard to the
reproductive organs of the first-named groups is perhaps doubtful.

Among the P/ysophoride, on the other hand, a precisely opposite general law obtains ;
not only the new polypites, but the new nectocalyces and reproductive organs, and
even the branches of the tentacles, being developed on the proximal side of the old ones;

24 THE OCEANIC HYDROZOA.

so that the distal appendages are the oldest." Where the organs are arranged in distinct
groups, as in Apolemia, the same rule holds good, and the young groups are formed on
the proximal side of the old ones.

The like is true for the distinct groups of organs (polypites, tentacles, hydrophyllia, &c.)
which are arranged upon the coenosare of the Calycophoride, the young buds being
continually developed on the proximal side of the old ones; so that close to the proximal end
of the coenosare there is a point which may be regarded as a sort of centre of growth,
whence new parts are continually being thrust out towards the distal end.

The nectocalyces of the Calycophoride, however, do not appear to follow the same
law as the other appendages. For if the deflexed axis which supports the nectocalyces
of Hippopodius is simply the bent-down proximal end of the ccenosarc, as the study of
its development (Leuckart, Z. N. K., p. 68) would seem to prove, the young nectocalyces
certainly originate upon the distal side of the old ones, or, in other words, they diverge
from the same centre of growth as the other appendages of the ccenosarc, and, cen-
sequently, they must be successively pushed out in the opposite direction.

If, again, Gegenbaur is right in considering the first-formed nectocalyx of Diphyes to
be the distal one, then the locomotive appendages of this genus would follow yet another law,
the second being developed on the proximal side of the first formed, while the third bud,
destined to replace the first or distal nectocalyx, is certainly produced from the hydrosoma
on the distal side of the first, and the successor of the third again arises on its distal side, and
so onwards. If, on the other hand, the first-formed nectocalyx of Diphyes is really the
proximal one of the adult, this genus follows the same law as /Hippopodius, the young
nectocalyces being developed on the distal side of the old ones, while all the other appendages
are developed on the proximal side of the old ones, and the two sets thus diverge from a
common centre of growth.

However this may be, the knowledge of the fact that new appendages and parts of
appendages are constantly being formed in definite localities is of great importance to
the student of the Hydrozoa; for he is thereby enabled to study their development as easily
on adults as on embryos, and to find in almost every specimen a complete series of stages
of every organ. The result is that this part of our knowledge of the Hydrozoa is tolerably
complete.

Every appendage (except the hydrothec# and lithocysts) commences its existence as a
cecal process of the ectoderm and endoderm, the latter bemg very generally, if not always,
ciliated upon its surface, and containing, of course, a diverticulum of the somatic cavity.

If the appendage be a polypite, it gradually elongates, assumes its characteristic form,
and eventually opens at its extremity. The villi, where they exist, are developed as
outgrowths of the endoderm. The polypite is at first a perfectly simple sac, and the ultimate
division into peduncular, gastric, and buccal portions is the result of the different form
assumed, and histological differentiation undergone, by its corresponding regions.

The pyloric valve of the Calycophoride is formed by the inward growth of the walls
of the polypite at the junction between its peduncular and gastric divisions.

1 Whether the hydrophyllia invariably follow this law may be doubted, and it is not certain that
it obtains even for the other organs in such Physophoride as Velella and Porpita.

GENERATION AND DEVELOPMENT. 25

The tentacula originate in cecal diverticula either of the wall of the polypite or of
some part of the coenosarc, and, in their simpler forms, merely elongate and acquire
thread-cells in their ectoderm, while their endoderm may or may not undergo that process of
vacuolation which, in some cases, obliterates the central canal.

In all the Calycophoride and Physophoride which I have examined, except Velella and
Porpita, the canal of the tentacle remains freely pervious, its endoderm being thin and
not vacuolated. When the tentacle is branched, its branches appear successively as close-set
buds on one side of the proximal end of the tentacle, the younger buds being always, as I have
stated above, developed on the proximal side of the older ones. The buds are, at first, simple,
short czeca, and such, with a slight change in form, they remain in Physalia; while in
Rhizophysa they elongate and become filiform.

The complex tentacular appendages of the Calycophoride (Pl. V), of Athorybia (Pl. TX),
or Physophora (Pl. VIII), arise in the same way, but, as the bud elongates, it becomes divided
into three portions: a proximal, slender division (the peduncle), with walls of the same thick-
ness all round; a median, thickened, cylindrical sac, whose walls are much thicker on one
side than on the other; and a distal, more slender, cylindrical portion. Shortly after this
has taken place, the rudiments of the large, oval, and small palisade-like, thread-cells appear
in the thick wall of the median division, which gradually acquires the characters of the sac-
culus. In the Calycophoride as in Halistemma and Stephanomia, the distal division elongates
very greatly, and becomes the single filament. In Athorybia and those other P/ysophoride
which have a median lobe and two filaments, it divides at a very early period into three
lobes ; the lateral ones elongate into the filaments, and acquire numerous thread-cells in their
thick walls, while the walls of the median process or lobe, which undergoes far less elonga-
tion, remain thin and free from thread-cells.

The involucrum is formed as a process of the ectoderm of the distal end of the
peduncle. In Physophora, as will be more particularly described below, the distal end of
the peduncle itself undergoes a singular dilatation, and helps to form the envelope for
the sacculus.

The rudiments of the hydrocysts are at first indistinguishable from those of the polypites ;
and, indeed, as I have already remarked, they are in their perfect condition like nothing
so much as polypites which, having reached a certain stage of development, have subse-
quently merely increased in size.

A hydrotheca is simply the cuticular investment of its polypite, and, in the young bud,
is represented merely by the outer layer of the ectoderm. But, as growth proceeds,
this becomes more and more widely separated from the body of the polypite, assumes its
characteristic shape, and eventually opens at its distal end so as to allow of the protrusion
of the distal moiety of the polypite. This process of development may be very easily traced
in the Campanularia.

Hydrophyllia, on the contrary, are always developed as buds precisely similar in
composition to those which give rise to polypites or tentacles, and are therefore composed of
both ectoderm and endoderm. As they grow they assume their characteristic form, the
ectoderm enlarging out of all proportion to the endoderm and its contained diverticulum of
the somatic cavity. Consequently, when they have attained their full growth, they appear

4

26 THE OCEANIC HYDROZOA.

like masses of a solid and glassy substance, traversed by an apparently insignificant and
narrow cavity, which assumes very various shapes, and in whose wall the endoderm is often
hardly distinguishable.

All nectocalyces, whatever their ultimate form, are developed in the same way. The
bud from one side of the ccenosarc, in which they originate, becomes somewhat rounded and
enlarged distally, so that its proximal end takes on the form of a narrow peduncle, and the
diverticulum of the somatic cavity becomes pyriform. The distal wall of this cavity now
thickens, so as to form a prominent, hemispherical boss, whose convexity is turned inwards—
and which thus alters the form of the cavity from that of a pear to that of a cup. The nec-
tocalyx gradually enlarges, and at the same time the cup-shaped cavity is so modified as to
appear prolonged at its circumference into four equidistant cornua or czeca, which embrace
the hemispherical thickening, running along its sides parallel with the axis of the organ
(Pls. Vil; VIII, xX).

I have elsewhere described the production of these canals to be the result of the union
of the sides of the boss with the walls of the cup-shaped cavity, but I by no means deny the
possibility of its being partly due also, to the extension of ceecal processes of the cavity
into the solid substance itself. Whatever be the minuter steps of the process, however,
the final result is that the primitive central cavity becomes proportionally very small,
and that it is continued by four longitudinal canals to near the distal end of the young organ.
The czecal ends of these nectocalycine canals now send out a lateral process on each side,
and the adjacent lateral processes eventually unite so as to form a circular canal, so that the
cavities of the four longitudinal canals are now in free communication distally as well
as proximally, while the canal of the peduncle (the nectocalycine duct) connects the whole
system with the general somatic cavity.

During these changes the central prominence or boss, which has become very large,
does not remain solid. An irregular cavity appears within it, and gradually pushes aside its
substance until the latter is reduced to a thin coat, which acquires a muscular structure.
The cavity is at first closed, but it does not remain so; for an aperture appears at the
distal end of the organ, and puts the nectosac in free communication with the surrounding
medium. The membranous valve is a process developed from the muscular wall round this
aperture.

The multifarious accessory ridges and ornaments, all the peculiarities of form, all
the singular contortions of the lateral canals, and the curious processes of the nectocalycine
duct, which Leuckart terms ‘‘ mantel-gefiisse,” are superinduced in the course of growth
upon the primitive model whose formation has just been traced.

The gonophores are always, when they are distinct from the rest of the walls of the
hydrosoma, originally developed as ceecal processes of the ectoderm and endoderm, between
which, and as I believe, by a histological modification of the deeper layer of the ectoderm,
the generative elements make their appearance. They differ from one another throughout
the series of the Hydrozoa, simply according to the progress which they make towards the
development of a gonocalyx around this sac; or, in other words, towards acquiring a
medusiform structure. In Hydra and in Plumularia 1 can find no trace of such a structure
at any period of development. In Cordylophora, in the androphores of Physalia, in the
gynophores of Athorybia and other Physophoride, the system of canals is more or less irre-

THE CALYCOPHORID&. 27

gularly developed; but it is only in the Calycophoride, in the Velellide, and in some
Corynide and Sertulariade, that the development of the medusiform generative zéoids is to
be observed in full perfection.

The first stages in the development of these bodies precisely resemble those exhibited
by the ordinary nectocalyces ; but when the longitudinal canals are making their appearance,
the central boss does not remain convex at its apex. On the contrary, it becomes excavated
at this point, and the excavation gradually extending distally, eventually takes the form of a
saccular diverticulum of the cavity of the organ, occupying the axis of the boss, and extending
for a greater or less distance towards the distal end of the organ. It, therefore, lies in the
middle, between the four longitudinal canals, and nearly takes the place of, though it by no
means represents, the central cavity of the nectocalyx. The latter is, in fact, in consequence
of the appropriation of the greater part of the substance of the boss to form the walls of this
central cavity, reduced to a mere fissure, which gradually separates that part of the organ
which contains the system of canals (as the calyx) from a free central sac (the manubrium),
connected with the calyx only opposite the peduncle. An opening is formed at the distal end
of the calyx, and becomes surrounded by a valvular membrane in the same way as has been
described above for the nectocalyx: finally, as the fissure widens, that wall of it which
belongs te the calyx becomes more and more widely separated from the central sac, and
acquires a muscular structure.

The calyx is thus exactly comparable to a bell, and the manubrial sac to its clapper,
while the peduncle is its handle. The future fate of the body thus formed varies. In some
cases it remains attached by its peduncle, and the ova or spermatozoa making their appear-
ance at a very early period in the walls of the sac, which never opens at its apex, are
detached, apparently by the dehiscence of its outer wall; or the manubrium, having the
same structure and being nothing but a genital sac, does not get rid of its products until the
whole body has been detached from its peduncle, and swims about independently; or the
reproductive elements may not have appeared in the walls of the manubrium when the organ
is detached, and then the sac opens at its end and becomes functionally and structurally a
polypite. The reproductive elements are ultimately developed either in its walls or in those
of its system of canals; or it may be that such bodies are in some cases (Willsia, Lizzia ?)
mere gonoblastidia, whence the true reproductive apparatus is eventually developed.

"Sars (‘ Fauna,’ p. 38) has however shown that the androphores are detached as free medusiform
bodies in his ‘ Agalmopsis elegans.’
The note upon the nomenclature of the parts of the reproductive organs of the Hydrozoa
which should have been inserted here (see p. 20) will, I find, take so much more space than I antici-
pated, that it must be deferred to the end of the work.

28 THE OCEANIC HYDROZOA.

SECT. IV—THE GENERA AND SPECIES OF THE CALYCOPHORIDZE.

The generic term Diphyes was proposed by Cuvier to designate the singular animal
originally described by Bory de St. Vincent under the name of “ Biphore biparti.’ Cuvier
imagined that the nectocalyces—the two most obvious organs of the Diphyes—were distinct
animals, temporarily united, and possibly in copulation. It was not until 1827 that Quoy
and Gaimard established the family of “ Diphides,” or Diphyde, to include this genus, together
with their Calpe, Abyla, Nacella, Cuboides, and Hnaneagonum.

Two years later, Eschscholz formed the group of Stphonophore, to include these and
other ‘“Acalephe,” with the definition, ‘No central digestive cavity, but separate suckers.
Swimming organs either special cavities or vesicles filled with air, or both combined.”
The assemblage is very natural, though its definition requires some rectification. Its author
divides it into three families: “1. The Diphyide, having a soft body coalescent at one end
with a cartilagimous part, and possessing a second swimming organ with a swimming cavity.
2. The Physophoride, having a soft body provided at one end with a vesicle filled with air.
3. The Velellide, whose body contains a cartilaginous or calcareous shell, in whose many
cells air is contained.”

Eschscholz’s family of Diphyide contains the genera Hudowia (including the Pyramis of
Otto), Lrswa, Aglaisma, Abyla (including Calpe, Abyla, and Rosacea of Quoy and Gaimard),
Cymba (including Hnneagonum and Cuboides, Q. and G.), and Diphyes. Hippopodius is
arranged with the Physophoride.

Quoy and Gaimard subsequently added the genera Cucubalus and Cucullus; and De
Blainville, Amphiroa, Praia, Galeolaria (Sulculeolaria), and Tetrayonum. Unfortunately, the
author of the ‘Actinologie’ winds up the list with Woctiluca! and Doliolum!!

Kolliker made a substantial addition to our knowledge of this group by his account of
the structure and determination of the affinities of [ippopodius, and by his description of a
new genus, Vogtia; while his researches, conjoined with those of Vogt, Leuckart, and
Gegenbaur, added largely to our acquaintance with the general organization of the order ;—
and finally Leuckart, in his excellent general account of what has been done of late years
towards the elucidation of the relations of these animals, proposed the term Calycophoride
instead of Diphyde, which is inapplicable to such genera as /ippopodius and Vogtia.

These Calycophoride he divides into two families: 1. Diphyide, including Abyla,
Diphyes, Galeolaria, and Praya. 2. Hippopodiide, containing Hippopodius and Vogtia.

The “ monogastric” Calycophoride, such as Eudoxia and Ersea, which were’ erected by
Lesson into a distinct group, are considered to be only derivative forms.

I agree with Leuckart in adopting the ordinal name Calycophoride, instead of Diphyde ;
and with Sars, Vogt, and Leuckart in regarding the monogastric genera¢as derivative forms,
properly classed under the genera whose zéoids they are.

Nevertheless, in the present imperfect state of our knowledge, it will be more convenient

THE CALYCOPHORID/A. 29

to retain generic and specific names for the monogastric Calycophoride (which I shall term
Diphyozéoids) as the temporary titles of spurious genera and species, which will be suppressed
in the course of further investigation.

The following arrangement will furnish a general view of the known modifications of the
Calycophoridan type. The names of those species only which appear to me to be adequately
described are cited.

I. Fam. DIPHYDA.

Calycophorid with not more than two nectocalyces, which are polygonal. The hydroecium of the
proximal one is complete, or closed on its posterior side. The hydrophyllia well developed.

a. The proximal nectocalyx mitrate, as large as, or larger than, the distal one. The hydrophyllia
spathiform and smooth externally.
Genus I, Diruyss.

1. D. dispar.
2. D. appendiculata.
3. D. Chamissonis.
4, D. mitra?
5. D. Kochii ?

Sub-genus GALEOLARIA.
1. G. filiformis.
2. G. biloba)
6. The proximal nectocalyx polyhedral and smaller than the distal one. The hydrophyllia facetted.
Genus II. Asya,

1. A. pentagona.
2. A. Bassensis.
3. A. trigona.
4. A. Vogtii.
5. A. Leuckartii.

II. Fam. SPH HRONECTID A.

Calycophoride with probably not more than two nectocalyces ; the proximal one being spheroidal,
with a complete hydrecium. No hydrophyllia?

Genus III. SpH#ronectes.
1. S. Kollikeri.
III. Fam. PRAYIDA.

Calycophoride with only two nectocalyces, whose hydreecia are both incomplete or grooye-like.

The polypites are provided with hydrophyliia.

' Sars, ‘ Fauna littoralis Norvegie,’ Erstes Heft, p. 45, tab. vii, fig. 16, describes a “ Diphyes
biloba,” as distinct from “ Diphyes truncata,” which is a Galeolaria filiformis, The extreme shallow-
ness of the hydrecium (judging by the figure) leads me to think that “D. biloba” is also a

Galeolaria.

30 THE OCEANIC HYDROZOA.

Genus IV. Praya.
P. cymbiformis.
(P. diphyes, P. maxima.)

IV. Fam, HIPPOPODIID A.
Calycophoridz with many nectocalyces, whose hydreecia are incomplete. No hydrophyllia.

a. The nectocalyces horseshoe-shaped, smooth.

Genus V. H1proropius.
1. H. gleba,

5. The nectocalyces produced into several pointed processes.

Genus VI. Voatta.
1. V. pentacantha.

Fam. DIPHYD.

Genus DIPHYES (Cuvier).

Both the nectocalyces have a mitrate form, with five more or less convex faces, and
hence, a pentagonal transverse section. The proximal nectocalyx is as large as, or larger
than, the distal one. The hydrophyllia are folded round like the spathe of an arum, and have
their outer surfaces not facetted or irregular, but smoothly convex.

1. DipHyes DIsPAR (Chamisso and Hysenhardt, 1821). Pl. I, fig. 1.

Biphore biparti, Bory de St. Vincent, 1804.
Diphyes, Cuvier, 1817.
Diphyes dispar, Chamisso and Eysenhardt, 1821.
— Bory, Quoy and Gaimard, 1824.
Diphyes, Quoy and Gaimard, 1827.
— angustata, Eschscholz, 1829,
— dispar, Idem.
— campanulifera, Idem.
— Boryi, De Blainville, 1830.
— dispar, Lesson, 1830.
— Bory, Quoy and Gaimard, 1833.
— vregularis, Meyen, 1834.
— turgida? Gegenbaur, 1854.

The proximal and distal nectocalyces are of the same length; but the proximal is the
thicker.

DIPHYES. 31

The latter (fig. 1) is compressed from side to side, and ends superiorly’ in a point whose
obtuseness or acuteness varies much in different individuals. The anterior contour presents a
slight, the posterior, a very marked, convexity. The ridges of the anterior face are more
or less strongly serrated, and end below in three inflexed points. The hydreecial aperture is
quadrilateral, and very large in full-grown individuals; occupying more than two thirds
of the antero-posterior diameter of the lower end. Its lateral edges are cut off very obliquely
from behind downwards and forwards, and its anterior boundary (1 @) is formed by two very
broad, quadrate, overlapping septal plates.

The hydreecium extends for about half the length of the organ.” The somatocyst is narrow
and subcylindrical, and does not extend so far as the upper cecal extremity of the nectosac.

The latter is long and narrow; it widens superiorly, and then narrows suddenly into
a tubular cecum. The superior third of the distal nectocalyx (fig. 1 4, 1 ¢) is elongated into a
pointed process, which is completely received within the hydreecium of the proximal nec-
tocalyx. The anterior margin of this process forms an obtuse angle with the anterior margin
of the main body of the organ; but the posterior margin passes at a right angle into the
truncated inferior half of the anterior end, so as to form a sort of step. The anterior and
posterior sides of the inferior two thirds of the organ are nearly parallel; the anterior contour
being nearly straight, while the posterior is a little convex.

The nectosac is subcylindrical and rounded superiorly, the nectocalycine duct abutting
anteriorly a little below its apex. Of the five pointed processes which surround its mouth, three
are the ends of the serrated ridges which extend along the posterior face of the organ.
The other two correspond with the terminations of the antero-lateral iongitudinal ridges, and
when viewed from above are seen also to form the margins of a broad, transverse septal
plate, which continues backwards the floor of the hydreecial canal (fig. 1 ¢).

The latter is bounded laterally and in front by two thin plates, which commence about the
middle of the superior process, opposite whose lower part their anterior margins bend inwards,
and finally overlap for a short distance.* The inflexed edges next separate for a_ brief
space, so as to leave an oval interval, and then the left again overlaps the right, by a broad
inflected process. Inferiorly the latter ends suddenly, and the two plates run parallel for a
short distance, to terminate by abruptly truncated inferior edges.

The hydrophyllium (fig. 1 e) is a delicate and transparent glassy plate—its edges over-
lapping both above and below its point of attachment. Its truncated distal margin is pro-
duced and acuminated at its angles.

Length of the superior nectocalyx ¢ 7) Le neh.
Depth PEA 2 i »
Thickness ” » 3 ”

‘In this and all the following descriptions of the species of Calycophoride it is supposed that
the apex of the proximal nectocalyx is directed upwards, and that the side on which the hydrecium
lies is posterior.

* Eschscholz indeed says, ‘“ Cavitate ductus nutritorii ante medium corporis desinente,” which is
in strictness true; but the difference between the length of the hydroecium and half the length of the
organ is very slight indeed.

° This overlapping is well figured by Chamisso and Eysenhardt, loc. cit., tab. xxxii H.

32 THE OCEANIC HYDROZOA.

Length of the inferior nectocalyx : ep einchy
Depth » » Lait
Thickness 55 a , ; S45

Ccenosarc, between two and three inches long.
Polypite, | inch.
Sacculus of the tentacle, ;4. inch.

The three specimens on which this description is founded were taken on the 15th of
July, 1847, in the Southern Pacific Ocean, in lat. 36° 31’ south, and about fourteen miles from
the eastern coast of Australia. One was quite colourless, except the sacculi of the tentacles,
which had the usual reddish hue. In the other two the nectosacs were irregularly stained
with red.

I took several specimens of what I believe to be this species once again, on the eastern
coast of Australia (December 21st, 1848). At least, the only difference I could detect was
the absence of the median division of the transverse plate which forms the anterior boundary
of the hydreecium in the superior nectocalyx. In the largest of these specimens the
superior nectocalyx was even thicker and blunter than in the figured specimen; while the
smaller forms showed insensible gradations to such a figure as that represented in 1 d.
It will be observed, however, that this young specimen has the characteristic somatocyst and
tubular caecum of the nectosac. The detached superior nectocalyces of young specimens of
the same species were taken in the South Atlantic and in the Indian Ocean.

MM. Quoy and Gaimard, took D. disyar in the Straits of Gibraltar, but it is not
noted by the German investigators of the Mediterranean Calycophoride, unless, as I am
inclined to think, it is identical with Gegenbaur’s D. turgida.

I have taken some pains to unravel the history of this, the first discovered of the
Calycophoride, and to trace its identity through the manifold changes of name which it has
undergone. It is described in the following words by its original observer, Bory de
St. Vincent :

“Salpa (bipartita) /anceolata bipartita.—Le Biphore biparti est si transparent qu’on ne le
distingue pas dans l’eau ; lorsqu’on I’en tire, il a l’air d’une lame de cristal, et ne présente
aucune sorte d’organisation. Son corps est oblong, et comprimé latéralement. La _ partie
antérieure présente cing angles saillans; le cédté dorsal étant caréné, linférieur est
au contraire sillonné. On reconnait dans cette partie deux cavités longitudinales; de
la supérieure sort un filet, souvent trés-long et rarement entier; il supporte, quand il
existe, une foule de petits corps qui ressemblent aux glandes pedicellées de plusieurs
végétaux. La partie postcrieure est amincie a son extrémité, et a quelques rapports avec la
forme du fer d’une lancette; le coté supérieur, qui est plat, est épais d’une ligne et demie.
Dans cette épaisseur se trouve une cavité interne, dans laquelle entre une sorte d’éperon
qui termine la partie antérieure; et c’est li toute l'union apparente de deux parties de
Yanimal. Il y a aussi une autre cavité inférieure dans la moitié de derricre, dont l’ouverture

échancrée au point de jonction des deux parties, forme un cran 8 leur union.’

' Bory de St. Vincent, ‘ Voyage dans les quatre principales Iles des Mers d’Afrique,’ 1804, t. i,
pp. 134, 135, note ; and ‘ Collection de Planches,’ pl. vi, fig. 3 a, B, c.

DIPHYES. 33

It is clear from this description that Bory de St. Vincent regarded the proximal
nectocalyx as the posterior, the distal as the anterior, which is not their relation in
swimming. In his figure a, the coenosare is represented as issuing from the nectosac
of the distal nectocalyx, and the minuter details of structure are not given. Nevertheless,
its identity with the species under description is unmistakeable. It was taken about midway
between Tristan da Cunha and the Cape of Good Hope.

Cuvier (as has been stated above) gave the generic name of Diphyes to this species
in the ‘ Regne Animal’ (1817), but conferred no specific denomination upon it.

Chamisso and Eysenhardt (De Animalibus quibusdam, ‘Nova Acta, x, 1821, p. 365)
describe and figure a D. dispar, which they justly consider to be the same as Bory’s
species, and which is certainly identical with mine. They took it in the equinoctial
Pacific.

In the (‘ Zoology of Freycinet’s Voyage,’ 1849), Quoy and Gaimard say (under the head
of “ Diphyes Bory”)—“ Only a single species of this genus is known: it was discovered by
M. Bory de St. Vincent, who has figured it under the name of Biphore biparti. It is the same
as that which we reproduce here, and which we dedicate to this naturalist. After him,
MM. Tilesius and Chamisso have also given figures of it” (p. 577). Iam inclined to judge
from the figure (tab. Ixxxvi, fig. 2), which, however, is not so good as that of Bory or that
of Chamisso and Eysenhardt, that the “Diphyes Bory” is really the same as Bory’s species.
At any rate, Quoy and Gaimard admit that it is so, and therefore one does not see how they
were justified in suppressing the name D. dispar, given by Chamisso and Eysenhardt three
years before, and creating a new one.

The French voyagers and naturalists give the Atlantic, the Indian Ocean, the coasts of
New Holland and of Timor, as the geographical range of their Diphyes.

Curiously enough, Quoy and Gaimard confer no specific name at all on the Diphyes
which they figure and describe in the ‘Annales des Sciences, 1827, and leave one in doubt
whether they consider it identical with that previously described by Bory de St. Vincent and
themselves or not.

Eschscholz (1829), not recognising its identity with Chamisso and Eysenhardt’s D. dispar,
and with Quoy and Gaimard’s D. Boryi, with both of which he was acquainted only by
figures and descriptions, calls it D. campanulifera; while he establishes a new species
(D. angustata) wpon a form which he observed himself. I can find in his figures and
descriptions, however, nothing to justify the separation of angustata from dispar, campanulifera,
or Bory?.

Lesson did good service in his ‘Centurie Zoologique’ (1830), by rehabilitating the name
Diphyes dispar, and by reproducing side by side with his own drawing the various figures of
this Diphyes already given, so as to show the identity of the forms described by Bory,
Chamisso, and Quoy and Gaimard. He took the species frequently in the Indian Ocean
and about the Moluccas.

In the ‘Zoology of the Voyage of the Astrolabe’ (t. iv, 1833, p. 83) Quoy and Gaimard,
still ignoring the name given by Chamisso and Eysenhardt altogether, and not even referring
to these able naturalists, describe and figure anew the Diphyes Boryi. Jn its general outline
the new figure perfectly resembles that given by Lesson and others, but while in all the

preceding figures, in which the hydreecium and the nectosac of the proximal nectocalyx
5

34 THE OCEANIC HYDROZOA.

are depicted, the hydroecium hardly extends beyond the middle of the nectosac, it is made
to be as long, or even longer, in this. Nevertheless, in the ‘ Annales’ (1827), a figure of a
species, which is supposed by the French voyagers themselves to be the same, is given
with the two parts in the ordinary proportion.’ I make no question that there is some
error here, and that D. dispar was the only Diphyes seen by Quoy and Gaimard.

The ‘ Beitriige zur Zoologie’ of Meyen (‘ Nov. Acta,’ t. xvi, Supp. 1, 1834) contains a long
and in some respects valuable description of a species of Diphyes, which he considers to be
new, and calls D. regularis ; without, however, entering into any comparison with the species
already described by Eschscholz and Quoy and Gaimard, or in any proper way defining the
characters of the supposed new species. In fact, D. regularis is nothing but D. dispar, as an
attentive examination of the figures and description at once shows. Meyen has fallen into
an error in describing four cavities in the proximal nectocalyx instead of three, and he regards
the canals of the gonocalyx as muscles “ intended by their contraction to facilitate the birth of
the ova.” However, his account is the best that had been .up to that time given of the
anatomy of a Diphyes. As I have hinted above, I have a strong suspicion that the Diphyes
turgida of Gegenbaur (‘Beitriige,’ p. 62) is no other than this species; but it is not figured,
and the description is very brief.

DipHy&s APPENDICULATA. PI. I, fig. 2.

Diphyes appendiculata, Eschscholz, 1829.
— elongata, Hyndman, 1841.
— Sieboldii, Kolliker, 1853.
— gracilis, Gegenbaur, 1854.
— acuminata, Leuckart, 1854,
— Kochii (?), Will, 1844,
— — (?), Busch, 1851.
Muggica pyramidalis (?), idem.
Eudoxia Lessonii (Diphyozooid).

The proximal nectocalyx is larger than the distal ; and is acuminated at its apex.
The posterior contour is slightly and evenly curved ; the anterior is a little concave below the
apex, then strongly convex, and finally retreats a little at its inferior boundary. The anterior
face presents three serrated ridges, which end below in but very slight points around the
aperture of the nectosac, which therefore appears to be placed in the midst of a truncated
inferior facet, occupying the anterior half of the inferior extremity of the organ. The pos-
terior half is produced for some distance beyond this, so that the hydreecium opens far below
the nectosac. The aperture of the hydrcecium is as usual quadrate, and its anterior boundary
is formed by two overlapping quadrate plates. The inferior edges of its lateral walls are cut

1 De Blainville improves on this (‘Manuel d’Actinologie,’ tab. v) by making the hydreecium and
nectosac of equal size in the unmagnified figure of the “ Diphye de Bory,’ and very unequal (or in

their true relations, as I conceive) in the magnified one !

DIPHYES. 35

off obliquely from behind downwards and forwards. The hydreecium is very short, not
equalling more than a fourth the length of the whole organ, and its apex is curved backwards,
so that its contour resembles that of a Phrygian cap. The somatocyst is connected with
its anterior face, below its apex, by a narrow duct, and then expands into a wide subcylindrical
sac, which terminates superiorly at about the junction of the inferior and middle fourths of
the length of the organ. The nectosac gradually widens from its mouth, and having attained
about two thirds of its entire length, gradually narrows again to end in a rounded apex,
close to the superior extremity of the organ.

The distal nectocalyx is hardly more than two thirds as large as the proximal. The
nectosac occupies not more than two thirds of its length. It is elongated, wider in the
middle than at the ends, and is rounded superiorly. The inferior aperture is situated in the
middle of a truncated facet, with almost obsolete ridge points, as in the proximal necto-
calyx.

The distal portion of the organ, which contains the nectosac, is obliquely truncated
proximally, passing into a slender beak-like process of the proximal moiety, which is inserted
into the hydroecium of the proximal nectocalyx, and carries the nectocalycine duct.

Two perpendicular plates rise from this process, and run down parallel with one another,
to end below in free, pointed and serrated processes, which extend far beyond the aperture
of the nectosac, and whose posterior edges are united for about half their length by a trans-
verse plate (fig-26). This continues downwards the floor of the hydreecial canal, which is
bounded at the sides by the two perpendicular plates, and in part by a transverse lamina,
which stretches from one plate to the other for the greater part of their length.

The hydrophyllia are like those of Diphyes dispar, but are without the inferior truncated
edge and acuminated angles.’

Length of the proximal nectocalyx = inch.
: Depth 3 io = nee

Length of the distal nectocalyx op

Depth 95 s Just

This species was particularly abundant in Bass’s Straits, in February, 1848. Several
specimens were also taken in the southern part of the Indian Ocean. It has been taken? on
the Irish Coast, and it seems to abound in the Mediterranean. Leuckart says he has seen it
with a hydrosoma many inches long when fully extended, and with as many as fifty
completely-developed polypites.

I entertain no doubt that the species which I have here described is identical not only
with the D. appendiculata of Eschscholz, but with the D. Steboldii of Kolliker. Leuckart
admits that the D. gracilis of Gegenbaur is the same as Kolliker’s D. Siedoldii. And in this I
fully agree with him, though Gegenbaur says that the hydreecium of the distal nectocalyx

‘ Compare Leuckart’s account of their structure, Z. U., p. 67.
> See a ‘Note on the occurrence of the genus Diphya on the Coast of Ireland, by Mr. G. C,
Hyndman (‘ Annals of Natural History,’ vii, 1841.) Mr. Hyndman names the species D. elongata,

but from his sketches I entertain no doubt that it is D. appendiculata.

36 THE OCEANIC HYDROZOA.

in his species is open. I believe, however, he has overlooked the anterior wall of its
hydreecial canal. If he has not, then D. acuminata is not D. Sieboldii, in which Kolliker has
not only described (p. 38), but has figured (tab. xi, fig. 8), the wall in question. Leuckart
also admits the great similarity of his D. acuminata with D. Sieboldi, but says that his
species is ‘distinguished by the presence of a closed canal for the passage of the coenosare
(Durchlass-kanal) in the posterior swimming organ.” However, this is contrary, as we see,
to Kélliker’s express statement.

Under the head of Eudowvia Lessonii (infra) will be found my reasons for believing that
form to be the free Diphyozooid of D. appendiculata.

Dipuyes CHAMISSONIs (n. sp.) Pl. I, fig. 3

The proximal nectocalyx is very wide above, where it ends in an obtuse point. The
anterior and posterior contours are nearly parallel; the posterior is straight; the anterior, a
little convex. The anterior ridges are serrated, and produced below into three strong incurved
points. The nectosac is wider superiorly than at the mouth, and ends above, close to the
apex of the organ, in an obtusely pointed dome.

The hydreecium has about half the length of the nectocalyx, and is straight, conical, and
widely open below ; its anterior wall is a single plate, deeply emarginate below, and somewhat
convex forwards; the posterior wall is also emarginate ; and the inferior edges of the lateral
walls are but very slightly oblique.

The somatocyst is subcylindrical, and not so long as the hydroecium.

The hydrophyllia resemble those of Diphyes dispar.

I took the detached proximal nectocalyces, with the coenosare attached, of this species,
repeatedly, on the east coast of Australia and in the Louisiade Archipelago, during the years
1847, 1848, and 1849; but I was never able to obtain the inferior nectocalyx.

The specimen figured was about half an inch long ; the fully formed hydrophylha measured
one twelfth of an inch.

I am unable to identify this with any of the described species of Dip/yes.

DIPHYEs MITRA, n. sp. (? APPENDICULATA). PI. I, fig. 4

The proximal nectocalyx is considerably wider above than below; and is obtusely
pointed at its apex. The ridges of the anterior face are serrated, and end below in almost
obsolete points. The hydreecium attains hardly more than one fourth of the length of the
organ; it is obtusely conical, with a slightly recurved apex, beneath and in front of which
arises the narrow neck of the somatocyst; the latter hardly extends beyond the middle
of the length of the nectosac, and its walls are not vacuolated. The anterior wall of the
hydroecium is formed below by two triangular plates, which extend inwards from its lateral
walls, and overlap one another throughout their superior halves.

Hydrophyllium a very thin plate, whose edges are bent upon themselves, but do not
completely overlap ; the middle part of the inferior margin produced into a lobe.

DIPHYES. 37

Nectocalyx not more than one quarter of an inch long.
This obviously young and imperfect Diphyes was taken only once in the Indian Ocean,
to the south-east of Mauritius, May 27th, 1847.

Dieuyes Kocati.

Diphyes Kochii, Will, 1844.
Muggieca pyramidalis, Busch, 1851.

Will, in his ‘ Horee Tergestinze,’ 1844, describes and figures a Diphyes Kochii, which he
found in the Adriatic, and which appears to resemble the form just described in many respects.
However, it differs from D. mitra in the gradual widening of the nectosac towards its lower
end, so that the nectocalyx does not exhibit the peculiar inferior narrowing characteristic of
D. mitra. All the specimens of D. Kochi: examined by Will were devoid of a distal nectocalyx,
and the proximal nectocalyx was not more than a line and a half long.

I am inclined to suspect that both D. Kochii and my D. mitra may be the young of
D. appendiculata." The Muggiea pyramidalis of Busch (‘ Beobachtungen ueber Anatomie und
Entwickelung einiger wirbelloser Seethiere,’ 1851) is identical with Diphyes Kochit.

* Gegenbaur’s account of the development of D. appendiculata, to which I have referred at p. 23
(see also Pl. V), shows that the nectocalyces of this species must undergo a very considerable change
of form in the course of their development.

38 THE OCEANIC HYDROZOA.

Sub-genus GALEOLARIA (Lesueur and De Blainville).

Distinguished from the other species of the genus Diphyes by the obsoleteness of the
hydroecium of the proximal nectocalyx.

GALEOLARIA FILIFORMIs. PI. III, fig. 5; and Pl. XII.

Physophora filiformis, Delle Chiaje.
Galeolaria, Lesueur, De Bainville, 1830.
Sulculeolaria quadrivalvis, idem.
Beroides, Quoy and Gaimard. MSS. cited in Lesson’s ‘ Acalephes,’ 1843.
Epibulia aurantiaca, Vogt, 1851.
— _filiformis, Leuckart, 1853.
Galeolaria aurantiaca, Vogt, 1854.
— filiformis, Leuckart, 1854.
Diphyes quadrivalvis, Gegenbaur, 1854.

I have only obtained a single detached distal nectocalyx of this species, and I must
therefore borrow a brief description of its most important characters from other observers,
more especially Leuckart and Gegenbaur.

Originally described by Delle Chiaje (‘ Descriz. et Notom.,’ t. v, p. 135), and confounded
by him with Forskal’s Physophora filiformis, subsequently termed by Quoy and Gaimard
“ Beroides,” one or more species of this genus formed the basis of two genera, Sudculeolaria
and Galeolaria, founded by Lesueur, and adopted and published by De Blainville in his
manual. De Blainville arranges both genera among his Diphyde, and rightly suspects that
Sulculeolaria may be only an organ; but with respect to the Galeolaria, he hazards the unac-
countable suggestion that “ these animals differ essentially from the Diphyde, approaching the
Beroide.”

The Diphyes, truncata, so well described by Sars (‘Fauna litt. Norv.,’ 1846), is clearly
a Galeolaria, but the Norwegian naturalist appears to have been unacquainted with
De Blainville’s manual, and hence has not referred to that genus or to Sw/culeolaria.

Vogt observed the Galeolaria and Sulculeolaria forming one animal, which he described,
at first, under the title of Zpibulia, the name which Eschscholz had mistakenly conferred upon
the Lamarckian R/izophysa (Physophora filiformis, Forskil). Subsequently, Vogt adopted the
name of Galeolaria, and Leuckart follows his example. Gegenbaur is inclined to suppress
both Galeolaria and Sulculeolaria, and calls his Galeolaria a Diphyes. But the difference
appears to me amply sufficient to warrant at least a subgeneric distinction.

GALEOLARIA. 39

For although the proximal and distal nectocalyces closely resemble those of Diphyes
in their general structure and arrangement, the hydreecium of the former is reduced to
a mere depressicn, and, according to Leuckart, the anterior rounded ends of the lateral walls
of the hydroecial canal of the distal nectocalyx embrace the adjacent end of the
proximal one.

The septal wall, which bounds the hydreecial canal of the distal nectocalyx anteriorly,
again, is divided by a median emargination into two wide lobes, with rounded margins
(Pl. III, fig. 5 @); and the pointed processes which surround the lower apertures of the
nectocalyces are large and strongly incurved. One, or even two, additional nectocalyces,
smaller than the others, and apparently destined to replace them, are commonly attached to
the proximal end of the coenosarc.

The somatocyst is short, and of even diameter throughout. The course of the necto-
calycine canals in the proximal nectocalyx is the same as in Diphyes; but, according to
Leuckart, the lateral and posterior median canals are united by peculiar anastomosing trunks,
The arrangement of the canals in the distal nectocalyx differs in the great loop, directed
forwards, which the lateral canals make.

The polypites have the same structure as in Diphyes. The hydrophyllia are spathe-like,
expanded, and deeply excavated on one edge.

The distal nectocalyx figured in PI. III, fig. 5, was about five eighths of an inch long, and
was obtained in the Indian Ocean, near Timor. The species has been taken in other parts of
the Indian Ocean, and it appears to abound in the Mediterranean. Leuckart (Z. N. K., p. 33)
records the same, or a very similar species, from Greenland ; and Sars took it and another
species of the same genus (G. d7/oba) on the coast of Norway.

40 THE OCEANIC HYDROZOA.

Genus ABYLA (Quoy and Gaimard).

The proximal nectocalyx differs widely from the distal one in form and size, the latter
having an elongated, more or less mitrate shape, while the former presents many straight-
sided, plane, tetragonal, pentagonal, or hexagonal facets.

The ‘proximal nectocalyx is shorter than the distal. The hydrophyllia are thick
and facetted, or produced into points on their outer surface, and their edges do not overlap
spathe-wise.

ABYLA PENTAGONA. PI. II, fig. 2.

Pyramis tetragona (?) Otto, 1828.
Calpe pentagona, Quoy and Gaimard, 1827.
Abyla pentagona, Eschscholz, 1829.
_ — Delle Chiaje.
— — Kolliker, 1853.
—_— — Leuckart, 1853.
— trigona, Vogt, 1854.
Aglaisma Baeri, Eschscholz, 1825.
— — Busch, 1851.
— pentagonum (?), Leuckart, 1853.
Eudoxia cuboides, idem, 1853.
Aglaismoides (?), Huxley (infra).

The proximal nectocalyx has not more than one third the length of the distal.
If it be placed with the axis of its nectosac vertical and its hydrcecium backwards,
it will be found to be four-sided and to have two parallel, flat, pentagonal faces, one of which
is anterior and the other posterior. Each pentagon has one of its angles superior, and
one side consequently inferior. The lateral faces are divided into two portions, one supero-
lateral, four-sided, looks upwards and outwards, and meets its fellow of the opposite side
in a ridge which joins the superior angles of the anterior and posterior faces. The other,
infero-lateral, also four-sided, looks downwards and outwards, and is produced posteriorly
and inferiorly into the lateral wall of the hydreecium, as the posterior face is produced into its
posterior wall. Anteriorly, the lateral plates are united by an undivided, anterior septum,
which, like the posterior, has its lower edge deeply emarginated and serrated.

The nectosac opens upon the truncated anterior moiety of the inferior face of the organ.
It is subcylindrical, and ends above before reaching the superior edge of the nectocalyx. The

ABYLA. 41

hydreecium is a conical, wide-mouthed cavity, extending through less than half of the length
of the organ. The somatocyst arises from it by a short and narrow duct, which opens into a
large oval sac as wide as the nectosac, opposite whose extremity it suddenly contracts, and
sends off from its anterior wall a narrow, cecal process, which extends very nearly to
the upper edge of the nectocalyx. The walls of this organ are greatly vacuolated.

The nectocalycine duct passes off at right angles from the base of the coenosarc, and
divides opposite the middle of the nectosac. The distal nectocalyx is wider distally than
proximally. Its distal extremity is pentagonal, and of the five angles the four lateral are
prolonged into strong, serrated points (of which the left hand and anterior is the largest) ;
while the posterior is not at all produced, and as its ridge is correspondingly obsolete, while
those from the others are tolerably well marked, the organ appears to be only quadrilateral.
The posterior part of the proximal extremity is truncated, but anteriorly it sends off a short trian-
gular process, through which the nectocalycine duct passes. From the edges of the anterior
face two plates arise, which inclose the hydreecial canal (fig. 2 a). The wider, right-hand, plate
bends over, to end by a free, convex, serrated edge posteriorly, while its left, free edge is
smooth. It overlaps the other, which is narrower, rises perpendicularly from the anterior
surface of the organ, and ends in a number of very strong, distant, curved teeth.

The hydrophyllia are subeubical, sending off a process from behind the middle of their
inferior surface. The phyllocyst has the form of a cross, with short, lateral arms.

Length of the proximal nectocalyx ‘ ; 3, inch.

oi

ale

>» distal ” ”

»”

a-

oh hydrophyllium

The history of my acquaintance with this species is somewhat instructive. On the
15th of April, 1847, while in the South Atlantic (lat. 38° 9’ south, long. 52° 31’ east), I first
made acquaintance with a detached proximal nectocalyx, with which only a single polypite
was connected, and which exhibited no trace of a distal nectocalyx. In the Indian Ocean,
off Timor, and in the South Pacific,’ I found other specimens in precisely the same condition ;
and it was not until the 22d of December, 1848, that I met with one having the small
inferior nectocalyx represented in fig. 2c. The lower part of this organ had the form of
a pentagonal prism, with three strong posterior longitudinal crests, and two anterior ones.
All these terminated in strong incurved points below, and a strong plate projected backwards
internal to the right anterior ridge. The proximal moiety was an irregular pyramid, traversed
by the nectocalycine duct, which connected its extremity with the ccenosare. This organ
was so small as not to project beyond the hydreecium.

This specimen, like the others, had only one polypite, and on subsequent occasions I
repeatedly took the animal in the same state, and without any sign of a distal nectocalyx ;
until, at Jast, on the 21st of July, 1849, while off the south-east coast of New Guinea, I
obtained the entire individual represented in fig. 2. Even in this, however, the coenosare

‘ Leuckart (Z. U., p. 41) remarks on the wide geographical distribution of Abyla pentagona.
Quoy and Gaimard took it at Gibraltar, and “in different seas” (‘ Zool. de Astrolabe,’ iv, p. 90) ;
Delle Chiaje and Costa at Naples; Kdélliker at Messina.

6

42 THE OCEANIC HYDROZOA.

was incomplete, both the hydrophyllium and the reproductive organ of the terminal zéoid being
imperfectly developed.

The two genera Abyla and Calpe were established by Messrs. Quoy and Gaimard’ to
include certain species of Calycophoride which they met with in the Straits of Gibraltar,
in Bass’s Straits, and elsewhere, and which were distinguished from Dipiyes by the irregularly
cuboidal form and small proportional size of the anterior nectocalyx.

Eschscholz, justly perceiving that no difference of generic importance existed between
Abyla and Calpe, suppressed the latter name in the ‘System der Akalephen,’ and united
all the species into the one genus 4éyla, which he characterised thus: ‘ Ductus nutritorius
tubulis pluribus obsitus; pars corporis natatoria cavitate parva natatoria interna, extrorsum
se aperiente, instructa.”

The genus thus defined contained the two species 4. pentagona and A. trigona (neither of
which was observed by Eschscholz himself); and Eschscholz adds, as a doubtful appendix
to the genus, the two Mosacee of Quoy and Gaimard, which, however, are nothing but
detached nectocalyces of Praya.

In a previous work (‘Isis,’ 1825) Eschscholz had established another genus of Diphyde,
Aglaisma, having the following diagnosis : “'Tubulus suctorius unicus ; pars corporis nutritoria
cavitate parva natatoria interna instructa.” And this genus he retains in the ‘ System.’

Leuckart, during his investigations at Nice, met with some individuals of the Aglaisma of
Eschscholz, which he considers distinct from the 4. Baer of the latter writer, and terms
A. pentagonum. is description of the nectocalyx agrees down to the smallest particulars with
the structure of the corresponding part in the species of 4dy/a at present under description ;
and he shows that this so-called Aglaisma is essentially nothing but the detached anterior
nectocalyx of Abyla pentagona, the precise result to which I was led in a different way in 1849.

Leuckart considers that this “Aglaisma-form” of Adyla is not a regular stage in the
development of the animal; but that an adult 4dy/a has undergone mutilation—that it has
lost its ccenosare and inferior nectocalyx—and while in the act of reproducing them,
necessarily takes on the Aglaisma form. In one “dglaisma” he found a second polypite, with
its tentacle, budding from the neck of the ordinary one ; and he states that a more or less
developed rudiment of the inferior nectocalyx is always present.

This was certainly not the case in my specimens, and I suspect Leuckart has fallen
into error in describing the rudimentary inferior nectocalyx as a “four-sided” pyramid.
(Z. U., p.52.) Gegenbaur appears to have made similar observations to those of Leuckart and
myself. He says (‘ Beitriige,’ p. 14): “I have often seen young, entire Abyle pentagone with
only one polypite; and young Dépiyes in the same condition.” So that he regards these
“Aglaisma-forms” as young, and not as mutilated adults.

Leuckart’s careful investigations enable me to add some other particulars of interest
to my account of the structure of Ady/a pentagona. I did not trace the four canals which are
given off from the short nectocalycine duct of the superior nectocalyx. But, according to

1 « Annales des Sciences Naturelles,’ t. x, 1827
2 Eschscholz retained Abyla because he considered the form so named to have been more completely
observed. But there is another reason for suppressing the name Ca/pe, in the fact that it had been

applied by Treitschke to a genus of Lepidoptera in 1825 (Agassiz, Index, p. 176).

ABYLA. 43

Leuckart, one descends directly from the end of the nectocalycine duct along the posterior
wall of the nectosac to its mouth. Another passes upwards to the blind end of the
nectosac, over which it runs, and down the middle line of its anterior face. The two
other canals form an arch, convex upwards, on the lateral faces of the sac, and descend to its
aperture, round which all four canals are united by a circular canal.

The inferior nectocalyx of the Mediterranean Abyla pentagona occasionally attains an inch
in length. It has four longitudinal canals (conf. Kélliker, p. 46). To the upper end of the
coenosare a rudimentary accessory nectocalyx is almost always attached, and the basal part
of the coenosare, as far as the first rudiments of polypites, is ciliated internally.

Those groups of organs or zéoids which are provided with perfectly formed tentacles (and
which occur at about the junction of the upper with the middle third of the coenosare), exhibit
two other appendages; one, the rudiment of the hydrophyllium, and the other that of the
reproductive organ. The latter becomes larger and larger, and finally its calyx assumes the
form of a four-sided pyramid, while the generative elements are developed within its manu-
brium.

The rudiment of the hydrophyllium assumes at first the form of a trefoil leaf, with
a median and two lateral lobes; and the phyllocyst acquires an elongated figure, with two
lateral processes corresponding with these lobes. The inner surface of the hydrophyllium now,
according to Leuckart, approaches the coenosarc, and the two lateral lobes bend towards one
another as if they were about to embrace it. From this time the lateral lobes grow rapidly,
the emargination which separates them from the middle lobe is gradually effaced, and thus,
when the organ has attained one fifth of a line in length, it lies like a saddle on the ccenosare.
The phyllocyst sends a long and slender canal downwards, parallel with the ccenosarc, a
short process upwards, and two lateral processes at right angles to them.

“In the stage just described the hydrophyllium (Deckstiick) has, as I have said, the
form of a saddle: its lateral ale are as yet isolated, so that the genital capsule—which now
measures about one eighth of a line—projects outwards like a campanulate appendage
between them. But these lateral ala soon begin to coalesce, and the hydrophyllium enters
into a new stage of development.

“The first traces of coalescence appear at the upper end of the alee, which had previously
approached one another in the middle line in front of the coenosarc, and finally meet.
The confluence proceeds rapidly downwards and inwards, so that the previously foliaceous
hydrophyllium is very soon changed into a solid body, whose axis is traversed by the
coenosarc. In the lower half the coalescence is confined to the outer edges of the
hydrophyllium. There remains here an internal dome-shaped cavity, whose bottom is
pierced by the continuation of the hydrosoma, and, besides the polypite and tentacle, incloses
the stem of the reproductive organ. At the point of insertion of the last, which now
measures about a quarter of a line, and already has the form of a quadrangular pyramid with
a contractile nectosac, the rudiment of a second reproductive capsule has appeared as
a simple vesicular body, 3th of a line long.

““At the commencement of the process just described, the outer surface of the hydro-
phyllium was tolerably evenly curved; but gradually four longitudinal ridges appear upon it,
which run down at regular intervals, become bounded by plane surfaces, and gradually

44 THE OCEANIC HYDROZOA.

change the body into a cuboid, whose posterior surface is continued downwards into a
hood-like (Schirmartig) appendage.’’'

Gegenbaur (‘Zeitschrift fiir Wiss. Zool.,’ v, p. 451) has objected that the hydrophyllium
in Adyla does not grow round and embrace the ccenosarc, as Leuckart describes, but remains
entirely on the side on which it was first formed ; and with this conclusion, notwithstanding
the latter observer's reiteration of his own view (Z. N. K., p. 23), my own observations lead
me to agree. ‘The hydrophyllium figured in Pl. III, fig. 2d, in fact, measured about sth of
an inch in length, and yet, in opposition to Leuckart’s statements quoted above, it had already
assumed the cuboidal form; and assuredly its lateral lobes had not coalesced round the
coenosarc. On the other hand, it seems very possible that the hydrophyllium may eventually
infold the pedicle of the polypite.

The hydrophyllium attains, according to Leuckart, the length of half a line, while the
ceenosare is still entire. The polypite, tentacles, and gonophores associated with each
hydrophyllium have also increased in size, and now the smallest shock, or even a sudden
contraction of the ccenosare, is sufficient to detach the group of organs as an independent
Diphyozéoid, or “ monogastric Diphyes.”

Leuckart states (Z. U., p. 60) that he has repeatedly watched this process of
separation under the microscope, which results from the ccenosare breaking through
midway between every two groups of organs, and that the young Diphyozéoid exhibits the
projecting stump of the ceenosarc for a long time. Eventually the hydrophyllium enlarges to
a length of 1} line, and has the form of a tall four-sided pyramid, with a square superior face
and obliquely truncated inferior extremity, whose posterior wall is prolonged into a triangular
blade, pointed below, and having a produced point on each margin. The inferior face
presents the opening of a wide cavity, which extends upwards to the middle of the organ.
The phyllocyst sends off a superior and an inferior narrow cecum, and two very wide
lateral branches, which embrace the dome-like roof of the cavity. The calyx of the
reproductive organ attains a leneth of two lines and more, and is marked by four strong
longitudinal crests, ending below in strong points. At the upper end is a conical process,
whereby the organ adheres to the coenosarc. Not infrequently, beside a gonophore of
two lines long, another two thirds of a line in length, and already exhibiting contractions,
may be seen. Leuckart terms the fully formed Diphyozioid of Abyla pentagona, Eudowia
cuboides. But this organism differs widely from the Hudovia of Eschscholz, and I believe it
to be identical with one of the species of Aglaismoides (infra), which in that case will form an
additional synonym of Abyla pentagona.

Gegenbaur’s account of the zéoids of Adyla pentagona agrees in all essential respects
with that just cited from Leuckart, but he describes the phyllocyst as sometimes having four,
instead of two, lateral lobes; and states that the inferior extremity is emarginate and not
pointed. :

Vogt’s figures and descriptions of the zéoids before their detachment are not very
clear, but coincide, so far as they go, with those of Leuckart and Gegenbaur. In Kolliker’s
specimens the hydrophyllia appear to have been not yet developed, but he gives the first, and
a remarkably clear, description of the nectocalyces and the mode of their adjustment to one
another.

} Leuckart, Z. U., p. 59.

ABYLA. 45

AByLa Bassensis. PI. II, fig. 1.

Diphyes Bassensis, Quoy and Gaimard, 1833.
Abyla quadrilatera, De Blainville, 1830.
Bassia quadrilatera, Quoy and Gaimard, 1824.
Calpe bassensis, Lesson, 1843.

Sphenoides (?), Huxley (infra).

The proximal nectocalyx has the same general form as in the foregoing species; but,
whereas in the latter, the base of the quadrate process, which contains the lower portion of the
hydreecium, does not occupy half the inferior face of the organ, it here takes up a much
greater space, and is proportionally shorter and wider, its lower edges being somewhat
everted. The apex of the large and bell-shaped hydroecium extends nearly to the middle of
the organ, and is fully as long as the nectosac, which is wider and shorter, and has a pro-
portionally narrower mouth. The nectocalycine duct passes to its posterior face rather above
its middle. The somatocyst is ovoid, aud has no narrow cecal prolongation.

The distal nectocalyx is prismatic, and widest in the middle of its length. The prism
appears at first to have only four sides and as many longitudinal ridges, but in reality there are
five ; that in the middle of the convex posterior face being almost obsolete, and marked only
by a very small inferior pointed prolongation.

The postero-lateral crests are somewhat stronger, and their inferior points more prominent,
and the antero-lateral ones are still more marked; the right antero-lateral pomt is the
strongest. The anterior surface presents a deep longitudinal groove, bounded on each side
by the antero-lateral ridges. From the right-hand ridge a plate stretches transversely to the
left-hand one, and is fixed to the latter for the greater part of its length, but it ends inferiorly
in a free, convex, deeply serrated edge (fig. 1 a). Beneath this plate may be seen indications
of the existence of another, so that the anterior half of the hydreecial canal is probably formed
by the union of two plates, one from each antero-lateral ridge.

The inferior surface is truncated, and shelves downwards anteriorly into the septal ridge,
which connects the terminations of the two antero-lateral crests.

The rounded aperture of the nectosac lies in the middle of the broad, flat inferior
surface, of which it occupies only a small portion. It is, as usual, surrounded by a membranous
valve, and leads into a large nectosac, which gradually widens, attaining its greatest dimen-
sions about the middle of the organ, and then narrowing again till it ends at about the junc-
tion of the upper with the next fifth of the long diameter of the organ. The posterior and
superior surface of the upper end shelves forwards and downwards, and where it joins the
anterior surface, is the point of division of the nectoduct, which ascends thence obliquely to
the floor of the hydreecial canal.

The upper end of the distal nectocalyx is deeply notched, so as to present a sort of
articular surface, into which the solid angle of the proximal nectocalyx, formed by the junction
of the inferior and the posterior faces, is received. The posterior facet of the notch looks
forwards and upwards, and is formed by the upper face of that portion of the distal necto-
calyx which contains the nectosac. The lower half of the posterior face of the superior
nectocalyx rests upon it.

46 THE OCEANIC HYDROZOA.

The anterior facet is prolonged into an obtuse process, which carries the nectocalycine
duct, and by which therefore it is attached to the ccenosarc. This facet is excavated anteriorly
by the superior aperture of the hydrcecial canal, over which the lower aperture of the hydre-
cium of the proximal nectocalyx fits.

The hydrophyllium has a very remarkable form (fig. 1c). It consists of a central thick median
part, produced into two thin and flat lateral ala. The former has a slightly concave anterior
margin (towards the ccenosarc), and a flattened upper surface, which forms nearly a right
angle with its posterior contour. From the middle of the upper surface a small recurved pro-
cess arises. Another is developed at the junction of the upper and posterior contours, and a
third arises about the middle of the posterior edge. About the middle of the upper edge of
each ala a similar process arises. The lower edges of the ale are truncated, and produced
into a sharp recurved point where they meet the anterior edges.

The phyllocyst presents a globose superior division, passing below into a long czecum.

Length of the proximal nectocalyx ; inch.
Breadth mH uf Nts
Length of the distal nectocalyx Deans
Breadth - 33 As
Hydrophyllium an

The specimens on which this description is founded were taken in great abundance on the
23d of June, 1847, during a beautifully calm and moonlight night, while the ship was passing
within a few miles of the southern shore of Tasmania. The edges of the larger specimens
were all coloured deep blue. I obtained 4éby/a Bassensis again in Bass’s Straits in 1848;
and I took a young specimen in the South Pacific, May 15th, 1850.

Quoy and Gaimard (‘ Astrolabe,’ t. iv, p. 91) give an account of a Diphyes Bassensis
which they discovered at the entrance of Bass’s Straits, and which, I have no doubt, notwith-
standing the brevity and insufficiency of their description, and the imperfection of their figures
(‘ Zoophytes,’ pl. iv, figs. 18, 19, 20), is that here described. Their specimens measured from
eight to ten lines in length. The species has not to my knowledge been described since. I
suspect that my Sphenoides australis is the Diphyozooid of this species.

ABYLA VoeoTit (n. sp). Pl. II, fig. 3.

The proximal nectocalyx has five faces arranged around its vertical axis. Of these the
posterior is pentagonal, and larger than any of the others, its superior angle forming the apex
of the organ. Two supero-lateral four-sided faces shelve away from a median superior
ridge, as in the foregoing species; and there are likewise two infero-lateral four-sided faces.
But the anterior pentagonal face is replaced by two quadrilateral faces, which unite to
form a median longitudinal ridge. The inferior face is consequently pentagonal.

The posterior face is deeply emarginate below, where it forms the wall of the hydreecium,
and the edges of the emargination are strongly serrated. The inferior margin of the infero-
lateral face is also serrated, and appears deeply notched, from the fact that its posterior part is
produced to form the lateral wall of the hydreecium. From the outer surface of this wall, on

ABYLA. 47

the right-hand side, a small serrated ridge projects perpendicularly and ends below in the
pointed lower extremity.

The hydreecium is bell-shaped, and its axis is directed obliquely upwards and forwards.
Its summit does not quite reach the middle of the organ. A narrow tubular somatocyst,
without processes, continues its direction towards the solid angle formed by the junction
of the two supero-lateral faces.

The nectosac oblong, narrow, rounded above, is inclined parallel with the somatocyst,
and extends about as far ; its lower opening is circular. The distal nectocalyx was not obtained.

The hydrophyllium in the most perfect specimens had the form of a quadrilateral pyramid,
with its apex truncated, to form a flat, quadrangular area, whose angles are prolonged (Pl. II,
fig. 3 0).

The peduncle of the polypite traverses the centre of the upper face of the hydrophyllium,
and dilates within it into rather a wide phyllocyst, before becoming continuous with the
polypite. The anterior face of the pyramid is excavated by a deep arched notch, extending
nearly to its upper boundary. The edge of the notch is serrated. The produced angles
of the upper face are separated on this side by a much deeper depression than on any
of the other three. |

Length of the proximal nectocalyx : : *; inch,
Length of the hydrophyllium ; 2 Sear

The end of the ccenosarc of the single specimen of this species (taken off the south-east
coast of New Guinea) which I observed, was torn off. The largest or terminal zéoid is figured
in 34; it possessed a gonophore (3 c) of 4th of an inch in length.

In my notes I find it stated as above, that the hydrophyllium is developed round the
pedicle of the polypite. This is so unusual a circumstance, however, that I suspect I have
confounded the pedicle of the polypite with the ccenosarc. j

ABYLA TRIGONA. PI. III, fig. 1.

Abyla trigona, Quoy and Gaimard, 1827.
— — Eschscholz, 1829.
Diphyes abyla, Quoy and Gaimard, 1833.
Amphiroa alata (?), Lesueur, 1830.

The proximal nectocalyx has six faces arranged around the inferior moiety of its longi-
tudinal axis. Of these the anterior (fig. 1 @) is an oblong parallelogram with its inferior
angles produced and a little divergent; the posterior (fig. 1 4) is triangular, with its
apex downwards; the antero-lateral faces are pentagonal; the postero-lateral, rhomboidal.

The superior extremity of the proximal nectocalyx is cut into four facets—an anterior,
hexagonal with concave lateral edges; a posterior, quadrilateral, meeting the last in a transverse
ridge; and two lateral rhombic faces, which fit in between the anterior and posterior facets
of the superior surface above, and between the antero-lateral and postero-lateral facets below.

The inferior face presents an anterior quadrilateral facet, in the midst of which is the

48 THE OCEANIC HYDROZOA.

small aperture of the nectosac, surrounded by a five-toothed raised margin. Posteriorly,
the inferior face slopes very obliquely upwards and backwards, and presents a triangular
space entirely occupied by the aperture of the hydreecium.

The hydreecium is bell-shaped and very large, occupying nearly the whole length of the
organ. Its inferior aperture is narrow from side to side, but measured antero-posteriorly it
equals about half the diameter of the nectocalyx in this direction.

The somatocyst—a large oval sac, occupying nearly all the space left between the
posterior wall of the hydrcecium and the posterior contour of the organ—gives off no ceca.
Its walls are highly vacuolated, and the duct by which it communicates with the coenosare
opens into its apex.

The nectosac, finally, is subcylindrical, and tapers above to a point, against which
the nectocalycine duct applies itself.

The distal nectocalyx (PI. III, fig. 1 ¢) is pyramidal. Five more or less prominent
longitudinal crests mark the outer surface of the organ, and are produced at its lower
extremity into as many serrated points. The posterior of these is directed downwards and
backwards; the lateral ones are curved sharply inwards; and the anterior pair pass down-
wards and forwards. Of these last that on the right-hand side is produced forwards into a
crest which almost immediately becomes confounded with a large serrated perpendicular
plate, which traverses the anterior face of the organ, and gives off from its inner surface a
transverse plate, deeply toothed on its free edge.

The left-hand anterior point is prolonged upwards as a slight ridge for some distance,
and then rises into a very strongly serrated vertical plate, which is overlapped by the
transverse plate of the other side—the two enclosing the hydroecial canal. Arrived at the
conical, superior portion of the organ, the two crests diminish in height and become obsolete.

The nectosac is hardly more than two thirds as long as the whole organ, sub-
cylindrical, and rounded above. The nectocalycine duct abuts against its apex.

The hydrophyllium is cuboidal, but narrower behind than in front, where it is somewhat
excavated. The phyllocyst is oval, and two curved, narrow ceca proceed from it above.

Length of the proximal nectocalyx inch.

ale

Length of the distal nectocalyx » nearly.

wh

A single specimen of this species was taken in Torres Straits, off the south-east coast of
New Guinea, on the 24th of July, 1849; but I observed it neither before nor afterwards.

The ccenosarc was imperfect and its appendages incompletely developed. The hydro-
phyllium described and figured was detached, but I have no doubt it belonged to this
specimen. Its resemblance to the Amphiroa alata described below is so striking, that I
imagine the latter must be the Diphyozooid of Adyla trigona.

It is with some little hesitation that I identify this species with the <Adyla trigona of
Quoy and Gaimard, described and figured in the ‘Annales des Sciences’ for 1827, and
subsequently in the ‘Zoology of the Voyage of the Astrolabe,’ t. iv, p. 87, pl. 3, iv, figs.
12—17; but there are so many points of similarity that I prefer to run the risk of making
a species too few rather than one too many.

ABYLA. 49

ABYLA Leuckarti. PI. III, fig. 2.

The superior nectocalyx is long and broad, but thin and compressed from side to side,
with six faces disposed around its longitudinal axis, and a single superior face which cuts the
anterior and posterior ones nearly at right angles.

The anterior face (24) is flat and elongated, with four parallel sides. Its inferior angles
are produced into two curved, serrated points.

The posterior face (2a) is very narrow above, wider in the middle, and below ends in a
point at the margin of the hydreecium.

The antero-lateral faces are by far the largest; they are somewhat concave above,
convex in the middle, and concave again below. They are separated by well-marked serrated
ridges, concave forwards, from the narrow postero-lateral faces.

The superior face is hexagonal and somewhat concave; its posterior side is very
small.

The inferior surface exhibits an anterior quadrate division facing downwards, and
having in front the serrated points in which the lower angles of the anterior face end,
while behind are two larger serrated points developed from the inferior edges of the lateral
faces. Between these processes is the aperture of the nectosac itself, surrounded by four
short, incurved points.

The posterior division is a triangular space which looks downwards and backwards and
is wholly occupied by the aperture of the hydreecium. The latter extends through nearly
the whole length of the middle of the organ. Behind it is a large oval somatocyst with
vacuolated walls, which occupies nearly all the space between it and the posterior face, and is
connected by a duct from its apex with the coenosarc.

The nectosac, cylindrical, narrow, and rounded above, extends as far upwards as the
hydreecium, and receives the short nectocalycine duct close behind its apex.

Length of the proximal nectocalyx . - 4 inch,

A single specimen of this species, consisting of a detached superior nectocalyx, was
taken in the South Pacific, not far from the east coast of Australia, in January, 1850.

Only one large polypite was attached to the remains of the coenosarc, but several young
ones were in the course of development.

I find no previous notice of this species, and I venture, therefore, to name it after the
able Professor of Zoology at Giessen, to whom and to his colleague Frey, we are indebted for
the most important improvement in the classification of the Animal Kingdom that has been
made since the time of Cuvier—the establishment of the Celenterata as a sub-kingdom.

50 THE OCEANIC HYDROZOA.

Fam. SPHMRONECTIDZE.
Genus SPHASRONECTES (nov. gen.)

The proximal nectocalyx (which alone has been hitherto observed) is spheroidal, and
of a gelatinous texture. The hydrcecium is completely closed behind. Hydrophyllia (?)

SPHERONECTES KOLLIKERI. PI. III, fig. 4.

Rosacea, Huxley, 1851.

Proximal nectocalyx a spheroid of gelatinous consistency, presenting anteriorly’ a wide
circular aperture, surrounded by a valvular membrane. This leads into a spacious, irregularly
hemispherical sac, which reaches as far as the centre of the spheroid. Four canals run along
the sides of the nectosac, and terminate above in a short, narrow duct, which passes to the
closed extremity of a tubular hydrcecium, whose other end terminates in an aperture in front
of the median plane of the spheroid. The hydrcecium traverses more than two thirds of
the diameter of the spheroid, and contains the slender coenosare which is affixed to its closed
end, and there receives the nectocalycine duct. It gives off a narrow cecal somatocyst,
which bends a little downwards, and ends near the proximal surface of the organ.

There was no second nectocalyx nor any indication of the attachment of one, nor were
there any hydrophyllia developed.

Length of the superior nectocalyx 5 ; ¢ inch.

I obtained altogether three specimens of a single species of this genus—in the Indian
Ocean in 1847, on the east coast of Australia in 1848, and in Torres Straits, off the south
coast of New Guinea, in 1849.

In all the ceenosare was incomplete, though some of the polypites were fully developed
and provided with tentacles of the ordinary characters. In one I observed rudimentary
reproductive organs attached to the pedicle of a polypite.

The wall of the somatocyst was vacuolated in some specimens, in others not. There was
a very distinct pyloric valve, and the tentacle arose immediately beneath it, from the wall of
the polypite, into whose cavity its canal directly opened.

In one specimen the lip of the nectosac appeared to be obscurely four-lobed, and in all,
the sacculi of the tentacles had a deep-red colour.

1 Supposing the nectocalyx to have its hydreecium vertical, which is the position in which Diphyes

and Adyla have been described,

SPHARONECTES. 51

The absence of the hydrophyllia may have been accidental, but it does not seem easy to
comprehend how any second nectocalyx could exist in this genus, at least, in its ordinary
place attached to the proximal end of the ccenosarc.

From the description contained in Lesson’s ‘Acaléphes’ I formerly imagined that this was
the Rosacea’ of Quoy and Gaimard, but it is clear that this species, though presenting many
points of resemblance with Praya (= Rosacea, Q. and G.), cannot be included in that genus ;
the tubular form of the hydrcecium and the very large proportional size of the nectosac
at once excluding it. I have, therefore, established a new genus for it, and I propose to call
this species S. Kdllikeri, after the distinguished author of ‘ Die Schwimmpolypen von Messina,’
and of so many other valuable contributions of science.

' See ‘ Reports of the British Association,’ 1851, p. 79.

52 THE OCEANIC HYDROZOA.

Fam. PRAYIDAE.
Genus PRAYA (De Blainville).

The proximal and distal nectocalyces are similar, and of a gelatinous texture. The
hydroecia of both the nectocalyces are open, groove-like, and applied together so as to form
a tube in which the upper part of the ccenosarc lies. The hydrophylha are thick, reniform,
and gelatinous.

PRAYA DIPHYES? PI. III, fig. 3; and Pl. XII, fig. 5.

The body figured in PI. III, fig. 3, 3 @, is thus described in my notes :

An ovate, reniform, gelatinous mass, presenting a small, hemispherical cavity at its lesser
extremity, and widely excavated below. From the centre of this inferior excavation depended
a twisted tubular cord (coenosarc), surrounded in its upper part by a mass of nascent organs.
The cord expanded into a broad muscular band at its point of attachment, and through
the centre of this ran a narrow canal, which appeared to connect the cavity of the cord with
a narrow, thick-walled cavity (somatocyst), extending along the roof of the inferior excavation
from one extremity to the other. From this cavity a small canal was given off to the
hemispherical natatorial cavity (nectosac), and on reaching it divided into four canals.

Length . 5 5 : + inch.

This specimen was taken on the 23d June, 1847, in the Indian Ocean. I obtained
another in Torres Straits in 1849, and one more in the South Atlantic in 1850; neither of
these had any polypites.

In the last the hemispherical cavity was in active contraction, and the adjacent extremity
of the organ was more distinctly bilobed.

I have little doubt that this fragmentary organism is the same as that described by Quoy
and Gaimard under the name of Rosacea plicata. As Leuckart (Z. N. K., 39) has well
pointed out, however, the two species of Rosacea described by the French naturalists are only
isolated parts of what in its entire state is known as the genus Praya ; and on comparing my
figures with the excellent and well-illustrated accounts of this genus which have been
recently published by Vogt, Kélliker, Leuckart, and Gegenbaur, I entertain no doubt that my
specimens were merely the detached proximal nectocalyces of a Praya.

Leuckart states, indeed, that he has frequently found Praye which have lost one

PRAYA.

Y
Qo

nectocalyx, with the other floating horizontally, and allowing the coenosarc to depend perpen-
dicularly from its under surface.

In order to enable those into whose hands this work may fall to comprehend the
relations of this detached organism to Praya, I add the following short account of that genus,
founded upon the statements of the authors just mentioned.

The nectocalyces are two, placed at the upper extremity of a ccenosarc, which some-
times (in P. maaima on Gegenbaur’s authority), attains a length of three feet, and are nearly on
the same level. One is usually smaller than, and more or less embraced by the edges of, the
other. They are both of a gelatinous consistence, reniform, and more or less bilobed below.
A shallow nectosac occupies the lower and outer endof each. Both nectocalyces are attached
by a muscular expansion to the proximal end of the coenosarc; and the somatocyst is a long,
curved canal, which sends a blind process downwards between the nectosac and the
hydreecium. The latter is a widely open groove, or shallow pit, in both nectocalyces,
and the chamber into which the coenosarc is retracted is formed by the application of the
edges of these grooves to one another.

In the natural position, therefore, the long axis of this composite hydreecium is vertical,
and, consequently, the long axes of the nectocalyces are also vertical; and the mouths
of their nectosacs are directed outwards and downwards.

Each hydrophyllium is a thick, gelatinous, and reniform body, bent upon itself, rounded
and solid at one extremity, and divided at the other into a median thick and two lateral
lamellar lobes. The phyllocyst is prolonged into four cecal processes.

The calyx of the reproductive organ is provided with two strong, longitudinal crests,
and the same hydrosoma presents gonophores of both sexes.

Two species of doubtful distinctness’ (P. diphyes. and P. maxima) are described. They
have been taken in the Mediterranean and in the tropical parts of the Atlantic and Pacific
Oceans.

* See Leuckart, Z. N. K., p. 40.

54 THE OCEANIC HYDROZOA.

Fam. HIPPOPODIID.
Genus HIPPOPODIUS (Quoy and Gaimard).

The nectocalyces are numerous, horse-shoe shaped, smoothly convex on their inner faces,
and form a double series along the (deflexed) proximal end of the ccenosare.

H. GueBa. PI. XII, fig. 6.

Gleba hippopus, Forskal, 1775.
— excisa,! Otto, 1823.

Hippopodius luteus, Quoy and Gaimard, 1827.
Hippopus excisus, Delle Chiaje.
Protomedea lutea, De Blainville, 1830.
Stephanomia hippopoda, Quoy and Gaimard, 1833.
Elephantopes Neapolitanus, Lesson, 1843.
Hippopodius Neapolitanus, Kolliker, 1853.

— luteus, Vogt, 1854.

— gleba, Leuckart, 1854.

I have not observed this or any other member of the genus Hzppopodius, but, in order
to give completeness to my account of the Calycophoride, 1 subjoin an abstract of Leuckart’s
description of it, referring to his and the other works just cited for further details.

Each nectocalyx is shaped like a horse’s hoof, with its lower face turned upwards.
It is a wedge-shaped bevelled segment of a cone, in which may be distinguished an upper
surface which forms an angle with the longitudinal axis of the cone formed by the union
of all the nectocalyces, a less inclined lower surface, and a horseshoe-shaped lateral surface.
The upper face is turned towards the axis, and is provided with a deep longitudinal groove
with strongly produced edges, which end below and internally in two obtuse processes,
between which is a rounded projection. The lower, smaller, face of the nectocalyx exhibits a
large circular aperture which leads into a flat nectosac. At the anterior edge (or that turned
away from the axis) of the nectosac the ridge between the lower and lateral or anterior face of
the nectocalyx is raised up into four well-marked prominences, which are separated by
concavities, and die away superiorly upon the anterior surface of the organ. The upper edge
of this surface also shows one median and two lateral prominences, which correspond with the
intervals between the prominences of the lower edge.

1 Though Forskal only figures the animal he discovered, Otto, who adopts his genus, gives not only

a figure but a description of the nectocalyces. Ought not the genus therefore to be termed Gleba ?

HIPPOPODIUS.

cr
on

The substance of the nectocalyces has a semicartilaginous consistence. Their number
amounts to as many as twelve. The lowest are the largest, and new ones are constantly
budding forth at the upper end of the series.

The nectocalyces of opposite sides are wedged together by their inner faces—each one
receiving the end of its opposite neighbour and the lower surface of that which lies above it,
between the ridges of that face. The apertures of the nectosacs of the two lowermost
nectocalyces alone are uncovered, the rest being hidden by the upper surface of the nectocalyx
below them. As usual, there is a muscular valve, which is narrower below than above.

The axis of the cone formed by the nectocalyces appears to be, as it were, a branch
from the upper end of the coenosarc; but, from Leuckart’s observations, it results that this
apparent branch is, in reality, merely the upper end of the coenosare bent down upon itself :
for, in the youngest individuals, which may not have more than two nectocalyces, the
arrangement of the parts is just as in Diphyes or Adbyla. The coenosare gives off a series of
nectocalycine ducts, one below the other, along the same lateral line ; though, in consequence
of the position of the nectocalyces, they appear to be right and left.

The extended coenosarc sometimes measures full six inches, and bears twenty or thirty
well-developed groups of organs, besides the numerous buds which remain inclosed within
the cone formed by the nectocalyces.

The tentacles are like those of the other Calycophoride. HHippopodius is moncecious,
and the gonophores are remarkable for the brevity of their gonocalyces and the length
of their manubria. The genus has not hitherto been taken beyond the limits of the
Mediterranean.

56 THE OCEANIC HYDROZOA.

Genus VOGTIA (Kéliiher).

The nectocalyces are numerous, similar, and convex on their inner sides ; externally they
are concave, and produced into five points, of which the three upper are much longer and
stronger than the two lower. The nectosac is small and rounded.

VoGTIA PENTACANTHA. PI. XII, fig. 7.
Vogtia pentacantha, Kolhiker, 1853.

Only a single species of the genus Vogtia has been discovered, and we are indebted to
Professor Kolliker for all we know of its structure. Unfortunately, only two specimens passed
into that observer’s hands, and hence his account is necessarily brief and incomplete.

The nectocalyces, whose form has been described in defining the genus, are four or five
lines wide. Their arrangement upon the ccenosarc has not been accurately determined,
but they formed a double opposed series with their convex faces turned towards one another.

The ccenosare of Vogtia is like that of Hippopodius, but it is shorter and somewhat
thickened at the lower end, where it bears the polypites and the other organs. The polypites
were only two and large; each had a tentacle similar to those of Hippopodius. Gonophores
of both sexes were borne by both specimens, and were attached by short peduncles to the
coenosarc close to each polypite—a single long gynophore being accompanied by two still
larger androphores. The gonocalyces are like those of Hippopodius, but shorter in proportion
to the long manubria.

Vogtia has hitherto been observed only in the Mediterranean.

é EUDOXIA. 57

one

THE DIPHYOZOOIDS.

Eschscholz, in defining his genera of Diphydan Stphonophore, lays great stress upon
the character of the digestive apparatus. Certain genera have a “tubulus suctorius unicus, ’
others have the “ ductus nutritorius (coenosarc) tubulis pluribus obsitus.”

As I have already had occasion to mention, Lesson seizing upon the distinction thus
indicated, divided his “Diphyde” into two very natural groups—the Polygastrice (tubulis
pluribus) and the Jonogastrice (tubulo unico).

In the first division he arranged the genera Diphyes, Calpe, and Adyla; in the second,
Cymba, Enneagonum, Cuboides, Cucubalus, Cucullus, Eudowia, Amphiroa, Ersea, and Aglaisma.

Neither by Will, Busch, nor myself was the true nature of the “monogastric Diphyde”’
suspected, but Leuckart’s investigations cited above, appear to me to leave no doubt that
these apparently independent forms are nothing more than detached zéoid forms, or
independent portions of the “ polygastric Diphyde.’ I have grouped them together, there-
fore, under the head of “ Diphyozooids ;” and though it is necessary, so long as the genesis
of each Diphyozéoid has not been worked out, to retain their distinct generic and specific
titles, and even to invent new ones for undescribed forms, yet these names can only be
regarded as provisional.

Genus KUDOXIA (Pschscholz).
Diphyozdoids derived from DiPHyYEs (°).

The hydrophyllium is lamellar and evenly convex externally. The walls of the cavity
which contains the polypite are incomplete on one side. The phyllocyst is simple and
conical.

Eupoxia Lessoni (Lschscholz). PI. III, fig. 6.

Ersea Quoyi, Eschscholz, 1829.

— Gaimardi, Idem.
Diphyes cucullus, Quoy and Gaimard, 1833.
Ersea pyramidalis, ? Will, 1844.
Eudoxia Eschscholzii, ? Busch, 1851.

— campanula, Leuckart, 1853.
Galeolarie pullus, Vogt, 1854.

This Diphyozdoid attains the length of four tenths of an inch. The hydrophyllium occupies

from less than a half, to two thirds of the whole dimensions. It is thick and irregularly
8

58 THE OCEANIC HYDROZOA.

conical, terminating in a rounded apex. One face—the anterior—is excavated in its superior
half by a wide and shallow groove, which is continued into the wide cavity which occupies
the inferior half, and which has somewhat the form of a pyramid with a truncated apex.
The posterior wall of the cavity is tolerably thick, and terminates below in a short,
transversely truncated margin. The lateral walls of the cavity become very thin at their
anterior margins, which are cut away from above downwards and backwards. The right-
hand margin slopes gradually into the inferior edge of the posterior face. The left-hand
margin, on the other hand, cuts the edge at an obtuse angle, which is produced into a point.

The phyllocyst is about two thirds as long as the interval between the apex of the
hydrophyllium and that of its inferior cavity. It is usually broad at the base, tapers
to an obtuse point above, and has no lateral processes. Its walls are largely vacuolated.

The calyx varies a good deal in size and proportion. It is a little convex anteriorly and
posteriorly, and its edges are sharply serrated; superiorly and posteriorly it is prolonged
into a short process traversed by a canal which divides into the ordinary four radiating
canals a little below the apex of the nectosac.

Length of the hydrophyllium, one fifth of aninch. Length of the calyx, one fifth of an inch.

Ludoxia Lessonit 1s one of the commonest of Diphyozéoids. I took it in all the seas
which the ‘‘ Rattlesnake” traversed in her circumnavigatory voyage, and it appears to be
common in the Mediterranean and Adriatic,

Eschscholz characterised the genus Hudowia (‘ Isis,’ 1825, and ‘ System’) thus: “ Tubulus
suctorius unicus. Pars corporis organa nutritoria fovens simplex (cavitate natatoria haud
instructa);” a definition which will apply equally well to any Diphyozéoid whatever.

In the ‘System’ another genus, Lrswa, immediately follows. It is defined, “'Tubulus
suctorius unicus. Pars corporis nutritoria cavitate parva natatoria, tubuli instar prominent,
instructa.” And Eschscholz states that the only distinction between Hrsea and Ludowxia
is the presence in the former, of a very small swimming cavity, which projects like a short
tube from the nutritive piece, or hydrophyllium. Of this genus he makes two species,
BL. Quoyi and F. Gamardi.

On comparing Eschscholz’s figures and descriptions with the various states in which T
have observed Hudowxia Lessonii, I entertain no doubt that his two Lrsee are merely Eudoxia
Lessonii with differently developed calyces.

I have found it to be the rule in this species that there should be a single large calyx,
devoid of any manubrium, and, consequently, presenting the appearance of an ordinary
nectocalyx, while one or two much smaller gonocalyces containing manubria, with ova
or spermatozoa, are attached to the pedicle of the polypite. Sometimes there were two such
large empty calyces.

It was the constancy with which these empty calyces occurred which, more than
anything else, led me to believe the monogastric Diphyde to be independent existences ;
and it may still be a question whether these empty calyces are all gonocalyces whose manubria
have discharged their contents and disappeared, or whether they are sometimes or always

true nectocalyces."

1 Consult on this point Leuckart’s excellent history of Hudoxia campanula (= Lessonii), in Z. U.,

p. 43.

EUDOXOIDES. 59

Leuckart (Z. U., p. 69), has conclusively shown that his Hudowia campanula is the
Diphyozéoid of his Diphyes acuminata. But I can find no distinction between /. campanula
and #. Lessonii, and I have already endeavoured to prove that Dipihyes acuminata is identical
with D. appendiculata. Consequently, Hudoaia Lessonii must be regarded as one of the
synonyms of the latter species, and the other Ludozia are probably nothing but Diphyozéoids
of other species of Diphyes.

Eupoxia Bosant (fschscholz). PI. Ill, fig. 7.

This species, of which I only obtained one specimen, is distinguished by the more
flattened form and symmetrically oval outline of the hydrophyllium, whose margin is
produced into a point on each side, at about the junction of the lower with the two upper
thirds. The inferior edge is convex and rounded, not truncated.

In the sole specimen obtained the phyllocyst, broad at the base, was very short and
irregular. I suspect that it had undergone some abnormal alteration.

The calyx is greatly elongated, and its edges smooth.

The gonocalyces, which in the present instance were male, were subcylindrical, smooth,
and rounded, without those strong crests so characteristic of the preceding species.

Length of the hydrophyllium, three sixteenths of an inch.

This species agrees so well with the figure and description of #. Bojani given by
Eschscholz, that, notwithstanding one or two minor differences, such as the lateral points
of the hydrophyllium, which Eschscholz does not notice, I feel justified in supposing it to
be the same. It was taken on the southern coast of New Guinea.

Genus KUDOXOIDES (nov. gen.)

oe

Diphyozooids derived from

The hydrophyllium is elongated and pointed, flattened on one face, evenly convex on the
other, and is provided with a deep conical cavity for the polypite. Phyllocyst simply conical.

Eupoxo1pgs sagittata. PI. IV, fig. 1.

I have seen only four specimens of the hydrophyllium of this species, and each of these
was devoid of a calyx, and had only imperfectly developed tentacles and reproductive organs.
Nevertheless, the form of the hydrophyllium is so peculiar that I am compelled to distinguish
it generically from other Diphyozéoids. It is narrow and pointed superiorly, and its anterior
face is flat, the inferior angles of this face being pointed and well defined. The conical cavity
in which the polypite is lodged occupies rather less than half the length of the piece, and

60 THE OCEANIC HYDROZOA.

is closed in for the greater part of its extent anteriorly. The phyllocyst, slightly pyramidal,
has no processes, and ends near the apex of the organ.
Length of the hydrophyllium, one fifth of an inch.

This species was also obtained on the southern coast of New Guinea.

Genus AGLAISMOIDES (nov. gen.)
Diphyozooids derived from ABYUa (2).

The hydrophyllium is a frustrum of a pentagonal prism, with one face partially
bevelled off. The phyllocyst has four cecal processes. The cavity for the polypite is deep
and conical, or hemispherical.

AGLAISMOIDES Escuscnouzit. PI. IV, fig. 2.

The hydrophyllium is a pentahedral prism with truncated, more or less concave,
terminal planes, whose edges are usually strongly serrated. In the position in which
all these Calycophoride ave described the posterior face of the hydrophyllium is formed
by one of the terminal planes, and is consequently pentagonal. One of the five angles
is directed downwards, and, therefore, one of the five faces of the prism is directed upwards.
The lower half of the anterior face is, as it were, bevelled off below, and the bevelled area
corresponds with the extent of the lower opening of the cavity which shelters the polypite,
which is almost hemispherical, and has its axis directed upwards and backwards.

It extends through not quite half the diameter of the organ, and the edge of its inferior
boundary is produced on each side into a tooth.

The phyllocyst is short and wide, and gives off four cecal processes. Of these two are
slender and vertical, the longer passing towards the inferior angle of the hydrophyllium,
the other approaching the middle of its superior face, and being inclined backwards towards
the posterior face. The two lateral processes are shorter and wider than the vertical ones,
and pass forwards and outwards towards the antero-lateral angles, exhibiting a slight
curvature or bend about the middle of their length.

The calyx is four-sided, with serrated edges, and a broad, truncated, and slightly
excavated inferior surface, in the middle of which is the small aperture of the nectosac.
This cavity has an elongated bell shape, and the duct connecting the radiating canals
with the ccenosarc terminates close to its apex. The calyx exhibited no manubrium.

Length of the hydrophyllium, one eighth of an inch.

This Diphyozooid occurred in all the seas which I traversed. Its curious resemblance
to the Aylaisma form of Adyla, with which it used constantly to be taken, is indicated by its
name ; and its general characters, more particularly the form of its phyllocyst, leave no doubt
on my mind that it is the Diphyozéoid of an Abyla.

AGLAISMOIDES. 61

AGLAISMOIDES ELONGATA. PI. IV, fig. 3.

The hydrophyllium is a vertically elongated prism; its superior face is square and flat,
and a little narrower than the inferior; the lateral and anterior faces are more or less convex,
and cut off square below; the posterior face is pentagonal, being prolonged inferiorly into a
short, triangular process, which bends a little forwards.

The inferior face is excavated by a large conical cavity whose apex reaches the centre of
the organ in a vertical direction. In width the cavity occupies about the posterior half of the
hydrophyllium. The posterior wall of this cavity is prolonged by the triangular process
above mentioned, and on the face of this are seen two short longitudinal crests, which run
down from the side walls of the cavity and terminate abruptly below. These inferior angles
are pointed and curved. The phyllocyst gives off, or rather is divided into, four processes :
Two of these, wide, lateral, pass at first horizontally outwards, and then are deflected at right
angles forwards to terminate in ceca which are bent downwards. Another process is very
slender, and passes vertically upwards, nearly reaching the superior face of the hydrophyllium.
The last, equally slender, runs downwards and backwards parallel with and close to, the
posterior wall of the chamber, and ends nearly opposite the commencement of the triangular
process.

The gonocalyces are four-sided prisms tapering to a conical end above, and having their
edges produced into crests which end below in four inflected points.

Length of the hydrophyllium, one tenth of an inch.

Aglaismoides elongata occurred on the east coast of Australia, in Torres Straits, and about
the Louisiade Archipelago.

It is impossible to compare these species of Aylaismoides with the figures and descriptions
which Leuckart and Gegenbaur give of the detached zioids of 4byla pentagona, without
being struck by the many points of resemblance which they offer.

Indeed, if I had only Leuckart’s figures before me, I should be inclined to identify my
A. Eschscholzii with those zéoids, but I cannot reconcile Gegenbaur’s account of the same
objects with my own figures and notes. I prefer therefore to leave the question open.

In the first specimens which I obtained the gonocalyx was very small, not extending
beyond the cavity of the hydrophyllium. Subsequently, I found not only large single
gynophores, but on one occasion two large androphores and a small gynophore attached
to one polypozéoid (PI. IV, fig. 3 4).

Genus SPHENOIDES (nov. gen.)

Diphyozéoids derived from ABYLA (°?).

The hydrophyllium is wedge-shaped below, but is bevelled into four quadrate faces on
its upper surface. ‘The cavity for the polypite is open in front. The phyllocyst is like that of
Eudowia, but has a long, slender, cecal diverticulum from its posterior and inferior part.

62 THE OCEANIC HYDROZOA.

SPHENOoIDEs AusTRALIs. PI. IV, fig. 4.

Sphenia (mihi), 1851.

Hydrophyllium wedge-shaped, presenting a sharp inferior edge and a wide superior
surface, produced into a point in the middle, round which four quadrate facets are
symmetrically arranged; the two posterior are slightly excavated and look backwards
and outwards. The two anterior look forwards and outwards, and are separated anteriorly
by a deep triangular notch. The posterior wall is broad and convex above, but ends below
in a point. It has four sides, the two upper of which are much shorter than the two lower.
The lateral faces are broad and irregularly pentagonal, their anterior edges being somewhat
concave. Close to the lower termination they give off a sharp downwardly directed point.

The anterior face is occupied by the wide triangular aperture of the cavity of the organ,
which extends much further back in its lower than in its upper moiety, so that its posterior
contour is very convex forwards above, and equally concave below.

Where this contour cuts the centre of the hydrophyllium the phyllocyst arises, and
is continued for more than half the distance to the apex of the organ, where it ends
in a point. Just above its origin it sends down in the middle line posteriorly a slender
czcum, which runs parallel with and close to, the posterior contour of the cavity, and ends
near its lower boundary.

The gonocalyx has the form of a four-sided prism whose edges are produced into four
more or less marked crests, ending below in as many points, which surround a flat space in
which the circular aperture of the nectosac lies. The summit of the organ is roof-shaped ;
two of the faces of the prism being, as it were, bevelled off, and the calyeime duct leads to one
gable end of this roof, by which, therefore, the gonocalyx is attached to the coenosare.

The elongated nectosac extends through the length of the prismatic portion of the organ,
and four longitudinal canals, ending in a circular canal around its mouth, run down its walls
from the calycine duct. It contains a large manubrium with well-developed ova or
spermatozoa. When the gonocalyx has attained its full size it becomes detached, and swims
about as an independent organism.

Length of the hydrophyllium, half an inch. Length of the detached gonocalyx, one
fourth of an-inch.

I first found this species on the 8th of February, 1848, in Bass’s Straits, where
it abounded. Subsequently, I took it off Timor, and in different parts of the Pacific and
Indian Oceans.

The individuals figured in Pl. IV, fig. 4*, are probably only the young of the others.
I am strongly inclined to suspect that this is the Diphyozéoid of Ayla Bassensis. In fact, the
hydrophyllium of the latter requires very little modification to convert it into that of
Sphenoides. The abundance of the species in Bass’s Straits is in accordance with this view.’

1 I have briefly referred to the characters of this Diphyozéoid, and more particularly to those
of its reproductive organs, in my note, ‘ Ueber die Sexual organe der Diphyden und Physophoriden,’
published in Miiller’s ‘Archiv,’ for 1851.

AMPHIROA. 63

Genus CUBOIDES (Quoy and Gaimard).
Diphyozooids derived from ABYLA (?)

Hydrophyllium with six flattened and four-sided faces. The chamber for the polypite is
deep and conical, and its inferior aperture occupies one face of the cube. The phyllocyst
is pyriform or narrow above, and very broad and slightly bilobed below.

CUBOIDES VITREUS, Quoy and Gaimard. PI. IV, fig. 5.

Hydrophyllium with six equal, slightly concave faces, and prominent angles. The
inferior face is occupied by a wide infundibuliform cavity, whose apex reaches to the centre
of the cube. The phyllocyst is pyriform, with a broad base, which is much dilated and
bilobed anteriorly, while above it narrows into a short caecum which approaches the upper
face of the organ within about a fifth of the whole diameter.

Gonocalyx a quadrilateral prism with its edges produced into four strong, serrated
crests, ending below in sharp points. The upper portion of the organ is prolonged into
a large pyramidal process, which is traversed by the calycine duct, and forms a strong,
longitudinal crest posteriorly.

Length of the hydrophyllium, one fourth of an inch. Length of the gonocalyx, one
eighth of an inch.

This species was taken twice, once on the east coast of Australia, and once on the south
coast of New Guinea.

The genus Cuboides was established by Quoy and Gaimard (‘Annales des Sc. Nat.,’
x, 1827, p. 19), on apparently the very same species as that which I have described and
figured. Leuckart (Z. N. K., p. 20) imagines that Cuboides vitrevs may be the Diphyozéoid
of Abyla pentagona, but this idea seems to me to be negatived by the form of the
hydrophyllium—more especially of its phyllocyst. On the other hand it might very well be
derived from the species which I have named 4. Vogtii, as a comparison of PI. IL, fig. 3 4,
with Cuboides will readily show.

Genus AMPHIROA (Lesueur).
Diphyozéoids derived from ABYLA (°).

Hydrophyllium with a trapezoidal section, the longer side being convex ; inferior face
obliquely truncated in front (where the wide and deep chamber for the polypite opens);
four-sided behind. The phyllocyst is large and oval, with two slender, curved, lateral and
superior ceca.

64 THE OCEANIC HYDROZOA.

AMPHIROA ALATA (Lesueur). PI. V, fig. 1.

The hydrophyllium is depressed and broad, with six unequal faces; the superior face is
a little convex and four-sided; its anterior side is convex and very much longer than
the posterior, which is straight; the lateral edges of this face are nearly straight. The
anterior face is a parallelogram, elongated, and convex from side to side. The lateral faces
are slightly concave and four-sided ; their inferior edges are concave and prolonged downwards
posteriorly. The posterior face is a vertically elongated parallelogram. The inferior face is
flat and four-sided posteriorly, but anteriorly it is obliquely truncated and occupied by a wide
aperture leading into a conical cavity, which extends beyond the centre of the organ vertically.
A very short, narrow duct runs from the apex of this cavity to just below the summit of a
great oval phyllocyst with vacuolated walls, which occupies almost all the space between the
posterior wall of the cavity and the posterior face of the hydrophyllium. From the upper
end of this there is given off on each side a long, slender ceecum, slightly concave upwards,
which ends close to the junction of the anterior and lateral faces.

The gonocalyx is a prism of apparently four sides, but its lower face is surrounded
by five points like those in the nectocalyces of ddyla and Diphyes, with which the organ
presented a further resemblance in the junction of two of the points by a strong, transverse
plate, and their continuity upwards with strong crests, between which a deep groove is
included. The upper end of the organ is pyramidal.

Length of the hydrophyllium, one fourth of an inch. Length of the gonocalyx, one
fourth of an inch.

Amphiroa alata occurred only in Torres Straits.

In one specimen of this Diphyozéoid (PI. V, fig. 14) obtained July 26th, 1849, I found
three reproductive calyces; one was very small, the other two large and fully developed.
Of these, the larger contained an ovisac, in which were four or five large ova. The smaller
contained a sperm-sac, with incompletely developed spermatozoa.

This is one of the species drawn and named by Lesueur, but first published by
De Blainville, in his ‘Actinologie.’ The description given in the latter work is very
imperfect, but the figure leaves no doubt in my mind as to the identity of Lesueur’s species
with mine.

I suspect this Amphiroa to be the Diphyozooid of Abyla trigona; at least, the hydro-
phyllium of the latter (PI. III, fig. 1 e) has the same general form and large oval phyllocyst,
with slender lateral processes.

AMPHIROA ANGULATA. PI. V, fig. 2.

The hydrophyllium has the same general form as in the foregoing species, but its
anterior face is concave instead of convex, and its upper edge is angulated in the centre,
while its lower edge is notched opposite the same point. Furthermore, the posterior wall of

ENNEAGONOIDES. 65

the cavity is prolonged beyond the general level of the truncated flat part of the lower face,
and the lateral walls send down thin processes, deeply notched on their free edges, to meet
this prolongation.

Length of the hydrophyllium : . 4th inch
Breadth _,, % A a F en
Length of the gonocalyx . p : 5 2

Amphiroa angulata was only taken once, in Torres Straits, and I am strongly inclined to
believe that it may be nothing but a younger form of the last species.

Genus ENNEAGONOIDES (nov. gen.)

Diphyozooids derived from - (?).

Hydrophyllium having the general form of a cube with produced angles, but a pointed
process is developed from one solid edge, so that this and the adjacent face are pentagonal,
and the body exhibits altogether nine points. The chamber for the polypite is conical.
The phyllocyst is simple and oval.

ENNEAGONOIDES Quoyr. PI. IV, fig. 6.

Hydrophyllium six-sided, with four quadrangular and two pentagonal faces, and much
produced solid angles. One of the faces is inferior, and presents towards one angle the
rounded aperture of the cavity for the polypite, which is conical, and is connected above with
a small elongated phyllocyst. Supposing the side on which this opening lies to be anterior,
then the axis of the cavity and that of the phyllocyst are directed obliquely backwards, upwards,
and to the right side.

The superior, anterior, left lateral, and posterior, faces are square, but the right lateral
face is pentagonal.

The organ would be a simple cuboid, were it not for the development of the triangular
process (3, fig. 6 a, 4, c), which converts the two adjacent sides into pentagons.

I regret to find that I have omitted to note the size of this Diphyozéoid, which was

taken on the east coast of Australia in 1849.
I obtained but one specimen of Hnneagonoides Quoyi and it had neither polypite

nor reproductive organs.

I was at first inclined to regard this species as the Huneagonum hyalinum of Quoy and

Gaimard (‘Annales de Sc. Nat.,’ 1827), but they state that the hydrophyllium of their genus
9

66 THE OCEANIC HYDROZOA.

contains two oblong, lateral cavities, besides that which lodges the swimming organ, and
one would be led by their figure to imagine that one of these cavities is a natatorial cavity,
and that their Zxxcagonum is nothing but an ill understood “Aglaisma-form” of ddyla. 1 can
testify that it is no easy matter to obtain a clear conception of the complex figure of the
singularly shaped hydrophyllium of this and other Calycophorida.

The genera Cymba and Cucubalus of Quoy and Gaimard appear to be Diphyozooids ; but
the figures and descriptions given by these naturalists are not such as to enable me to
arrive at any very confident opinion on this point.

Gegenbaur (Beitriige, p. 9) carefully describes and figures a species, Diplophysa inermis,
which appears to be very distinct from any of those enumerated above. However, it has
some resemblance to the Cucubalus of the French voyagers.

67

SECT. V. THE GENERA AND SPECIES OF THE PHYSOPHORID/:.

The systematic arrangement of the Physophorid@ is beset with difficulties which are
partly due to the nature of the animals themselves, but still more to the confusion introduced
by the mistakes of systematists.

The term “ Physsophora” was first used by Forskal, and under it may be found, in his
‘Descriptiones Animalium,’ good and clear descriptions of three very distinct though allied
animals—Physsophora hydrostatica, Physsophora rosacea, and Physsophora filiformis, all of
which were discovered in the Mediterranean.

Peron and Lesueur (‘ Voyage aux Terres Australes’) represent a Physophora muzonema,
obviously closely allied to the P. Aydrostatica of Forskal, and they give the name of Rhizophysa
planestoma to another Physophorid, which is as nearly related to P. filiformis. At the same
time they figure two new forms, Stephanomia amphitridis, and S. uvaria, without expressing
any opinion as to their zoological relations." Subsequently, on the return of these naturalists
from a visit to the French coasts of the Mediterranean, Lesueur communicated to the Société
Philomatique (from whose bulletin it was copied into the ‘Journal de Physique’ for 1813,
t. Ixxvi, p. 119, under the title of ‘Mémoire sur queique nouvelles espéces d’Animaux
Mollusques et Radiaires recueillies dans la Méditerranée pres de Nice’) a notice of some of the
animals which they had observed, and which are stated to “belong to the genera Salpa,
Stephania (sic), Physsophora, Pyrosoma and Hyalea.”

The word “ Stephania” appears to be a misprint for Stephanomia, but I can find no
evidence respecting the relation of this S¢ephanomia to that already figured.

Eschscholz (‘System’ &c.) separated the Stephanomia uvaria (the original drawing of
which, in the Banksian Museum, he had carefully examined) with great justice from the
Stephanomia amphitridis, and erected the first into a new genus, 4polemia, while he restricted
Stephanomia to the latter.

Forskal’s Physsophora hydrostatica, Peron’s Physophora, and a species described by Quoy
and Gaimard (P. Forskdlii), are combined by Eschscholz into the restricted genus Physophora,
while the Ph. rosacea of Forskal is very properly united with the Rdizophysa heliantha of
Quoy and Gaimard into a new genus, Athorybia.

Eschscholz is less happy in his treatment of the third Forskalian ‘“ Physsophora,”
P. filiformis. Tmagining that this was simply a mutilated individual of some Physophorid

1 Peron does indeed give a highly coloured and poetical popular sketch of Steph, amphitridis

(t. i, p. 45), but it has no scientific value.

68 THE OCEANIC HYDROZOA.

normally provided with nectocalyces, he makes a new genus (Hpibulia) of it, with the
definition, ‘Tentacula sacculis simplicibus obsita. Partes cartilagineze adhuc incognita ;”
and separates it from Peron’s Rhizophysa, whose swimming organs he also imagines he
has seen. Lamarck had taken a more just view in uniting both Peron’s and Forskal’s species
under the common head of RAizophysa, but, on the other hand, he had erred, as Eschscholz
points out, in including Forskal’s P. rosacea in the same category.

The genus Agama is a new and most important addition to the list of Physophoride.
Eschscholz includes under it, and I suspect rightly, the Stephanomia Amphitrite of Chamisso
(Cuneolaria incisa of Eysenhardt), which is certainly not the species of that name figured by
Peron, and the Pontocardia cruciata of Lesson.

To these genera of Physophoride, Eschscholz adds the Hippopodius and Discolabe of
Quoy and Gaimard, and the long known Physalia of Lamarck, while Velella, Rataria,
and Porpita are united into a family by themselves, the Velellide.

Leaving out Zpibulia, Eschscholz’s family of Physophoride contains nine genera:
three, Physophora, Athorybia, and Rhizophysa, founded on Forskal’s three Physsophore; one,
Stephanomia, on Peron and Lesueur’s St. Amphitridis; another, Apolemia, on their St. uvaria;
two being Quoy and Gaimard’s Hippopodius and Discolabe ; one Lamarck’s Physalia ; and one
altogether new, Agalma. Excluded from this group, and forming the distinct family of
the Velellide are the Velella and Porpita of Lamarck, and a new genus, Rataria. Hippopodius
has since been shown to be one of the Calycophoride, and Rataria is only the young of

‘elella. Discolabe is doubtful. The rest are all, I believe, good, sound genera.

Although Eschscholz separated the Velellide as a distinct family from the Physophoride,
he very clearly apprehended the close relations of the two; he refers particularly (p. 166) to
the resemblance of the “shell” of Rataria and Velella to the air-vesicle of Physalia, and
it seems to have been only the fancied resemblance between Porpita and Fungia, which
led him to separate the two families.

I find the two groups first united, though in an indirect sort of way, by Deshayes,
in the second edition of Lamarck (1840), tome iii, p. 94, and Index, p. 764.

Neither Quoy and Gaimard, nor Lesson, nor De Blainville, in their various publications,
have, so far as I have been able to discern, added a single real genus to those established by
Eschscholz, while in many respects their works are far behind his; and those of Quoy
and Gaimard, and of Lesson, have introduced the most lamentable confusion, especially in
regard to the limits of the genus Sfephanomia.

The masterly investigations of Milne Edwards published in the ‘Annales des Sciences
Naturelles’ for 1841, inaugurated a new epoch in the study of these singular animals,
and they were well seconded by the author of the ‘Fauna Litteralis Norvegiee’ (1846), which
contains a remarkable memoir on some Physophoride and Calycophoride taken on the shores
of the Island of Florée, in the high northern latitude of 613°. Sars considers (though at
p. 55 he seems to have some misgivings on the subject) that they all belonged to the same
species of a new genus, 4yalmopsis, which he thus defines :

‘‘Partes cartilaginese superiores seu natatorie ut in agalmate; inferiores numerosee
solide, triangulares, sparsee, non tubum componentes, sed modo una earum extremitate
canali reproductorio affixee ceterumque libere, pro emissione tubulorum suctoriorum ac
tentaculorum ubicunque fissuras preebentes. Canalis reproductorius longissimus, tubulos

THE PHYSOPHORIDA. 69

suctorios, vesiculos varie forme et tentacula offerens. Tentacula sacculis clavalis (clava
varie forme) obsita.”

Sars remarks (p. 40) upon the features which this genus has in common with both the
Agalma of Eschscholz and the Stephanomia (Amphitridis) of Peron, and, indeed, it would
be difficult to separate it by any character, save that assigned to the tentacula, from
Agalma. If the tentacula really have a “clava varie forme,’ Ayalmopsis differs not only from
Agalma, but from all Physophoride at present known, except Rhizophysa, whose tentacles,
however, are so differently and so much more simply constructed, that they can hardly serve
as terms of comparison.

After a careful study of Sars’ memoir, I must confess that I am strongly inclined to
believe that his Agalmopsis elegans embraces in reality species of two distinct genera, one
a true dyalma, with the tentacles characteristic of that genus, the other a form closely allied
to what I have described below as Stephanomia, believing it to be the S. Amphitridis
of Peron.

Sars’ “ first kind of tentacles” (1. c., tab. v, figs. 5 and 6, p. 35) are, in fact, exactly like
those which I have figured in Stephanomia, and unlike those of any other genus of Physophoride
with which I am acquainted. The “4yalmopses,” enumerated by Sars at p. 36, under (c),
which were the largest, and were taken from the end of November until March, possessed
only this one kind of tentacle; and the most of those taken in autumn, and enumerated
under (a) p. 35, had only these, and those peculiar organs whose structure is not very
clearly represented in Tab. VI, fig. 10. I suspect these were all Stephanomie.

The individuals enumerated under (4), taken contemporaneously with the last named,
had tentacles exactly like those of Ayalma, and tentacles of the first kind, in addition,
according to Sars. But with every respect for the accuracy of observation of the justly-
esteemed Norwegian naturalist, I think it very possible that he may have overlooked the real
mode of termination of the younger tentacles of this form, and have been led to imagine
their ends single when they were really trifid. In this case they would all belong to
Agalma.

Whatever be the real nature of the species described by Sars under the head of
Agalmopsis, however, there is, as Vogt has justly pointed out, an insuperable objection
to the genus, as defined by its author; the sole distinctive character given by him being
accidental, and depending on the extended state of his specimens, whose solid appendages
would, had they been contracted, have formed a tube, as in Agalma.

Leuckart (Z. N. K., p. 73), admitting the justice of Vogt’s elimination of the genus Ayal-
mopsis, as defined by Sars, appears, nevertheless, to wish to retain the term, as the designation
of that section of Agalma which has naked tentacular sacs, and a long, straight coenosarc.
Unfortunately, in this case, the sub-genus 4galmopsis would not contain either of Sars’
species, so that there would be considerable risk of confusion. Under these circumstances,
I propose to retain the name 4yaima for those species whose tentacular sacs are provided
with two filaments and a median lobe, and to give to the Agalma rubrum of Vogt, and its
allies, the generic appellation of Halistemma.

The following systematic arrangement of the P/ysophoride appears to me to be in
accordance with our present knowledge. Under each genus I have given the names of those
species which have been carefully examined and may be considered to be established.

70 THE OCEANIC HYDROZOA.

I. Fam, APOLEMIADA,

Physophoride provided with nectocalyces and hydrophyllia; the latter united with the other
organs into groups, which are arranged at considerable intervals along the coenosarc. Tentacula without
lateral branches, Pneumatocyst small. Ccenosare filiform.

Genus I. Avotemia (Eschscholz).

A. uvaria.

II. Fam. STEPHANOMIAD A.

Physophoride provided with nectocalyces and hydrophyllia; the latter arranged with the other
organs in a continuous series. Tentacula with lateral branches terminated by sacculi. Pneumatocyst
small. Coenosare filiform.

A. Sacculi without involucra, and ending in a single filament.

a. With biserial nectocalyces.

Genus II. HatistemMa (mihi).
H. rubrum (elegans ?)

6. With multiserial nectocalyces.

Genus III. Forsxauia (Kélliker).

F.. contorta.
F, Edwardsii.
F, Ophiura.
F. prolifera.

B. Sacculi provided with inyolucra, Filament single.

Genus IV. Sreruanomia (Peron and Lesueur).

S. Amphitridis.

GC. Sacculi with involucra, Filaments two, with a median lobe.

Genus V. Acatma (Eschscholz).
A, Sarsii.

A. clavatum.
A, Okenii.

III. Fam. PHYSOPHORIAD&.
Physophoride with nectocalyces, but without hydrophyllia. The distal end of the filiform

coenosare dilated. Tentacular branches with involucrate sacculi. Pneumatocyst small.

Genus VI. Puysornora (Forskdl).
P. hydrostatica.

P. musonema ?

SYSTEMATIC ARRANGEMENT. 71

IV. Fam. ATHORYBIAD A.

Physophoridee without nectocalyces, and with hydrophyllia. Tentacular branches with involucrate
sacculi, with two filaments and a median lobe. Pneumatocyst occupying almost the whole of the

globular ccenosare.

Genus VII. Arnorysia (Eschscholz).

A. rosacea.
A. melo.

V. Fam. RHIZOPHYSIAD &.
Physophoridx with a small pneumatocyst and a filiform coenosarc ; without either nectocalyces or

hydrophyllia, and having lateral branches without sacculi, to their tentacula.

Genus VIII. Rutzopnysa (Peron and Lesueur).

R. filiformis.

VI. Fam. PHYSALIADA.

Physophoridz with a pneumatocyst which occupies almost the whole of the thick and irregularly
fusiform ccenosare ; without either nectocalyces or hydrophyllia; and without lateral branches to the

tentacula, which are provided with basal sacs.

Genus IX. Paysatia (Lamarck).

P. caravella.

VII. Fam. VELELLID A.

Physophoridee without nectocalyces or hydrophyllia; with short, clavate, simple, or branched,
submarginal tentacles. A single central, principal polypite. Pneumatocyst flattened, divided into

chambers by numerous concentric partitions, and occupying almost the whole of the discoidal ccenosarc.
a. Pneumatocyst cristate. Tentacula simple.

Genus X. Vewnxia (Lamarck).

6. Pneumatocyst without a crest. Tentacula branched.

Genus XI. Porprra (Lamarch).

72 THE OCEANIC HYDROZOA.

Fam. STEPHANOMIADZE.'

Genus STEPHANOMIA (Peron and Lesueur).

Nectocalyces and pneumatophore unknown. Tentacula with lateral branches terminated
by an involucrate sacculus with a single filament.

STEPHANOMIA AMPHITRIDIs. PI. VII.

On the 4th of May, 1848, during a calm which had already lasted some days, I saw
from the deck, floating at the surface of the sea near the ship, the beautiful organism a
part of which is figured of the natural size in Plate II, fig. 1. I threw the towing net over it,
and succeeded in capturing it and transferring it, apparently uninjured, to a basin of sea-water,
where it remained entire all that day ; by the next morning, however, all the hydrophyllia had
become detached though the polypites remained perfectly lively. I did not perceive
any motion in the individual hydrophyllia, but when the living mass was irritated, either by
the least touch or even by pinching one of the depending tentacles, the ccenosarc
shortened, and thus brought them into close contact. The ordinary position was horizontal,
with the tentacles hanging down in the water, and when I first saw the animal it was
curved to one side.

The hydrophyllia were perfectly colourless, and so transparent as to be almost
imperceptible when in the water; the coenosarc had a whitish hue; the enlarged portion
of the polypites was pink or scarlet. The reproductive organs were colourless, as were
the stems and lateral branches and filaments of the tentacles, while the tentacular sacs
were scarlet.

The body bore handling very well, and a piece could be cut off from one end without
causing any detachment of other parts.

The ccenosarc had a diameter of about one twentieth of an inch, with thick, muscular
walls. It was obviously broken at each end.

The hydrophyllia surrounded the ccenosarc in whorls of four (figs. 2, 3); they were
attached by triangular, striated processes of the ectoderm, whose base was inserted upon a
triangular ridge, which traversed the middle of the internal face of each hydrophyllium
(fig. 4). From this a clear linear canal is continued almost to the extremity of the organ.
The hydrophyllia were in general leaf-shaped, and it appeared to me that a distinction might
be drawn between those which were situated at the sides and those which were’superior and
inferior ; the latter (fig. 5, 5 a) being distinguished by a sort of shoulder on one side of their
upper extremity, which is absent in the lateral hydrophyllia (figs. 6, 6 a, 6 4).

' IT commence with this genus because I have had no opportunity of examining any Apolemia,
Halistemma, or Forskdlia.

STEPHANOMIA. 73

The polypites are of two kinds, large and small; the former are attached at regular
intervals along one side of the coenosare by a short peduncle, measure six tenths of an inch
to one inch in length, and resemble those of other Physophoride.

Their ventricose enlarged portion is marked by numerous red strie, which are simply
elevations of the endoderm, containing thread-cells and colouring granules, and are
homologous with the villi of the Calycophoride (PI. VI, fig. 11).

The small polypites (hydrocysts?) are attached to the coenosarc between the large ones.
They possess no coloured villi, and Iam not sure that their extremities are open; they are
frequently as long as the large stomachs, but never nearly so wide; and their apices are
commonly provided with large thread-cells. A rudimentary tentacle is developed from
one side of their base (fig. 12).

The tentacles are attached to the pedicle of the large polypites, or to their base close to
the pedicle, but which I could not determine. They consist of a stem, which gives rise on
one side to a series of branches, which carry at their extremity deep-red sacculi. In structure
most of these parts resemble those of the Calycophoride, but the sac requires especial notice.
It is very long, and is usually coiled up and partly inclosed within a cup-shaped, hood-like
process of the ectoderm of the peduncle. On each side of the ordinary imbricated series of
thread-cells there is a single row of larger thread-cells; the latter are oval, and ,th of an
inch in length, while the others are elongated, somewhat curved, and not more than
wath of an inch in length. The thin wall of the sac contains a quadruple series of peculiar
bodies, represented in Pl. V, fig. 9. They are elongated, slightly curved, and pointed at
one end, while the other end is obtuse, and presents three or four elevations. They are
disposed transversely to the axis of the sacculus.

The filament, which terminates the sacculus, is long and cylindrical, and contains a vast
number of thread-cells of ,j;th of an inch in length.

The mode of development of these organs exactly resembles that of their homologues in
the Calycophoride. They commence as cecal processes, which elongate and become constricted
into three portions—pedicle, sac, and filament. The thread-cells are at this time all alike,
and not more than ;4,th of an inch in diameter. The involucrum commences as a circular
projection of the ectoderm of the peduncle.

The gonophores (fig. 12) are attached by branched footstalks to the ccenosarc, between
the larger polypites. There was no very great regularity in their occurrence, but I think
that, as a general rule, there were both ovarian and seminal organs between every two
polypites. The same stem that supported them frequently gave rise to a small polypite,
or hydrocyst.

The androphores are distinguishable from the gynophores by their greater absolute
length, their more elongated form, and by their greater clearness and transparency. The
largest androphore I measured was one tenth of an inch long; the largest gynophore three
fortieths of an inch.

The calyx of the androphore is terminated by four obtuse elevations, containing large
thread-cells. In essential structure it exactly resembles the corresponding part in Dipiyes.
The thickened outer wall of the spermsac contains round vesicles, about ;th of an inch in
diameter, none of which were developed into perfect spermatozoa in the specimen I observed.
The androphores were developed from simple papilla, in the same way as those of Diphyes.

10

74 THE OCEANIC HYDROZOA.

The gynophores, or female organs, in their essential structure and mode of development,
resembled those of Diphyes; but they offered two points of difference: 1. They never
contained more than a single ovum; 2. The surface of this ovum appeared as if wrinkled or
covered with anastomosing canals; but this appearance arises, as in Athorybia, merely from
the irregular separation of the ovisac from the walls of the calyx.

AGALMA. To

Genus AGALMA (B2schscholz).

Nectocalyces biserial. Branches of the tentacula terminated by involucrate sacculi, with
two filaments and a median lobe.

AGALMA BREVE. PI. VII.

A. Okenii {?) Eschscholz, 1829.
A. intermedia (?) Quoy and Gaimard, 1833.

To the naked eye this animal appeared like a prismatic mass of crystal, traversed by a
delicate filament, terminated at one extremity by a pink spot, and at the other by an irregular
pink mass. The pink spot is the pneumatophore, or float; the filament, the coenosarc; the
pink mass, the polypites and their appendages; while the crystalline prism is formed above
by the nectocalyces, below by the solid hydrophyllia, which are appended to and embrace the
coenosarc.

The pneumatophore is oval, and measures about 4th of an inch in length. In the first
specimen I examined the walls of the organ were of so deep a colour that all the details of its
structure could not be made out; but in the second, which was somewhat smaller, the
pneumatophore was colourless, and the arrangement of its internal parts was easily deter-
mined.

The pneumatocyst appears to be open below; but however this may be, its cavity does
not communicate with that of the coenosarc, for the endoderm, which is reflected upon and
closely adheres to its outer surface, does not stop short of the lips of the aperture, but extends
completely over it. It is not tightly stretched over the aperture, but forms a sort of loose,
bag-like end, extending far below it. The contents of the pneumatocyst could be readily
forced into this sac, and they returned again when the pressure was removed.

The sac thus formed by the reflected endoderm does not hang loosely and freely in the
cavity of the pneumatophore, but is connected with the walls of the latter by a number of
vertical, mesentery-like, partitions. These terminate in free arcuated edges opposite the lower
end of the reflected endodermal sac. Immediately below the pneumatophore a number of
budding nectocalyces, in all stages of development, make their appearance, the youngest being
the highest. Of the fully-formed and functionally active nectocalyces, however, there were only
five, three on one side, and two on the other. Viewed laterally, these organs appear fusiform,
while from above they have a horseshoe shape, in consequence of the deep concavity of their
internal edges. Above and below they present a deep and wide groove, whose edges are
bounded by well-defined ridges, or rather crests.

76 THE OCEANIC HYDROZOA.

The nectosac has a comparatively narrow, oval, or rounded aperture, fringed by the
usual valvular membrane. Its cavity near the aperture is subcylindrical, gradually widening
internally. It then suddenly dilates, and forms a very wide, blind, sac, more or less divided
into two lobes by a median constriction. The cavity is much wider than it is deep.

Below the nectocalyces four thick and solid hydrophyllia are attached, so as to lie nearly
in the same plane. They have the form of pyramidal wedges, with square bases." The
latter are turned outwards, while the apices are connected with the coenosarc by a duct which
extends, as a cecal phyllocyst, through the axis of the hydrophyllium, terminating at some
distance from its base. In some specimens there was a second set of such appendages, but
in others, these hydrophyllia were succeeded by four different ones, much larger and more
foliaceous, though still very thick. Internally, they are concave; superiorly, convex. Exter-
nally, they present two or more somewhat excavated facets, separated by thick ridges. Their
lateral edges are sharp, and coarsely serrate, and they taper more or less to a point below.
Like the preceding, these organs contain long and narrow cecal phyllocysts, which traverse
their axes, and nearly reach their apices.

The polypites lie amongst and between the hydrophyllia. Bunches of what appear to
be young polypites (hydrocysts), accompanied by rudimentary hydrophyllia and tentacles, are
attached to the coenosarc, between the fully formed ones, and are either on the same pedicle
with, or close to, the reproductive organs.

The sacculi of the tentacula are nearly a sixth of an inch long, and possess a long
median prolongation or lobe, flanked on each side by a filament of about double its length.
The involucrum is very large, and apparently capable of containing the whole sacculus.

The reproductive organs are developed more particularly towards the lower end of the
coenosarc, the male and female organs being placed close to one another. The gynophores
are very numerous, and about half as large as the androphores, which are fewer in number.
The gynophores are borne upon special stems, or gonoblastidia, each of which is simply a
process of the coenosarc, and contains, of course, a diverticulum of the somatic cavity. On all
sides the gonoblastidium gives off short bud-like processes, whose development is always the
more advanced the nearer they are to the free end. It would appear, therefore, that new
ones are continually developed at the base of the gonoblastidium. The smallest of these pro-
cesses is a mere cecal process of the endoderm and ectoderm, and is rather less than 5th of
an inch in length. It next becomes pyriform, and the endoderm acquires so great a thickness
at the apex and at the neck of the organ, that the included cavity assumes a more spherical
form, with a narrower neck. The thickened apical endoderm now presents a clear space of
about ;4,th of an inch in diameter, containing a spheroidal, pale, solid body. These are the
rudiments of the germinal vesicle and spot.

The largest gynophores are oval bodies, attached by a short pedicle, and about ;i:th of

1 In describing his Agalma Okenii Eschscholz states that, of the solid, cartilaginous pieces,
“some are similar to a very depressed pyramid, whose base presents two longer and two shorter sides.
‘'he broader lateral faces meet at the apex of the pyramid earlier than those which ascend from
the narrower sides. Other pieces are very irregular; they present a broad base, then a large, convex
surface, and many small excayated ones, which cause one side of the piece to be notched (zackig).”

Figs. 1 e, 1 f of pl. xii, which represent these solid pieces, have a close resemblance to mine.

AGALMA. 77

an inch long. The contained ovum is nearly as large as the gynophore, and has a pale,
granular yelk, a clear, spherical germinal vesicle of th of an inch, and a thick-walled,
vesicular, germinal spot of ;4;th of an inch.

The gonocalyx remains in a very rudimentary state, closely embracing the ovum. It
exhibited no terminal aperture in any specimen I examined, and its canals were narrow,
straight, and unconnected by any circular canal at their extremities. The inner wall of the
calyx was only separated from the wall of the ovisac or manubrium over irregular spaces, thus
giving rise to a system of canals like those in the same position in the gynophore of Athorybia,
only less complete.

The androphores are oval bodies, seated on very short peduncles, and ,th of an inch or
more in length. They commence their development as processes of the endoderm and ecto-
derm, in which the four canals are developed in the ordinary way ; but some of these would
appear to become obliterated with age, as those which were fully formed rarely possessed
more than from two to three canals, and exhibited only indications of the circular canal.

The manubrial spermsac was not distinctly separated from the calyx in the largest
specimen I examined, nor did any exhibit fully-developed spermatozoa.

I obtained on one occasion the young 4gal/ma (possibly of this species), about two lines
long, which is represented in Pl. VII, fig. 12. The unilateral attachment of all the
appendages was very obvious in this young individual. There were—Istly, immediately below
the pneumatophore, a series of young nectocalyces, the largest of which measured about
roth of inch in length, and had an apical opening to its rudimentary nectosac, with four canals
not yet united by a circular canal; 2dly, a series of ceca, the rudiments of the hydrophyllia ;
3dly, polypites in various stages of development. The only perfect one was terminal, and half
as large as the rest of the animal. It was suspended by a pedicle, and presented a pyloric
valve at its junction therewith. The upper third of the polypite had a globular form, and
was of a dark reddish colour. Its endoderm was raised up into longitudinal ridges, in which
a great number of round fatty-looking particles were imbedded. Besides these, other smaller
villous processes, similar to those in the polypites of Diphyes, were scattered about. A coiled
filament, probably the rudiment of a tentacle, arose from the neck of the polypite, and gave
off lateral buds, the most fully developed of which were cylindrical processes, terminated by
rounded heads contaiming many thread-cells.

I am unable to identify the Aga/ma which has just been described with any published
species. It presents some points of resemblance with the 4. Ofeni2 of Eschscholz, others with
the Stephanomia intermedia of Quoy and Gaimard; but there are well-marked differences in
each case. I therefore give it the specific name of 4. dreve.

78 THE OCEANIC HYDROZOA.

Fam. PHYSOPHORIAD.

Genus PHYSOPHORA (E&schscholz).

Ceenosare dilated at its distal end; provided below the pneumatophore with necto-
calyces. No hydrophyllia. | Hydrocysts forming a series on the proximal side of the
polypites. A spheroidal involucrum completely inclosing the sacculus.

Puysopuora (sp. °) Pl. VIII, figs. 1—12.

The anatomy of certain species of Piysophora has been studied of late years by Philippi,"
Vogt, and Kolliker. Only three specimens presented themselves during my voyage, and
neither of these was an adult, while two were exceedingly small. Nevertheless, as I paid
particular attention to points which happen to have been comparatively neglected by Vogt
and Kolliker, I trust that the additions which I am able to make to their account of this
interesting genus will not be without value.

The hydrosoma in the largest of Vogt’s specimens measured rather more than two
inches in length, and Kolliker’s appear to have had about the same dimensions (23 inches).
In my largest specimen the hydrosoma was only three quarters of an inch long.”

The ccenosare enlarges above into the pyriform pneumatophore ; below this it is slender
and cord-like, and gives attachment to a variable number of reine abkry: Its inferior
extremity is enlarged, and supports the remaining appendages.

With regard to the structure of the pneumatophore, Kolliker states only that it has
an apical pigment-spot ; and Vogt appears to me not to have comprehended its organization
at all, inasmuch as, in his description of the adult Physophora, he mistakes the walls of
the pneumatophore for those of the pneumatocyst, having apparently overlooked the latter.
This is the more singular as, in his account of the young Physophora, at p. 58, he states that

* Ueber den Bau der Physophoren und eine neue art derselben, Physophora tetrasticha. ‘Miiller’s
Archiv.,’ 1843, pp. 59—67, tab. v.

* Peron and Lesueur’s P. muzonema is stated to have been four inches long ; the P. Forskélii
of Quoy and Gaimard two inches; while Forskal’s original P. hydrostatica had a length of only
an inch and a half. The axis of Philippi’s P. tetrasticha had apparently a length of about
two inches.

° La bulle d’air est entourée immédiatement comme je viens de le dire, d’une expansion muscu-
laire faisant suite aux couches musculaires du tronc” (p. 43). Philippi denies the existence of any air

in the pneumatophore (p. 63).

PHYSOPHORA. 79

the air-vesicle “has the form of a pear, and is enclosed in a fibro-flocculent tissue, which lines
the interior of the pneumatophore (partie pyriforme), surrounds the pigment-spot, and
descends, forming four organs, in the midst of which passes down a rounded mass of
the same substance, and closely invests the air-bubble, which thus perfectly resembles the
clapper suspended in the middle of a bell.”’

I found the structure of the pneumatophore to be quite similar to that which obtains in
Agalma; that is to say, the air is contained in a distinct pneumatocyst, enclosed within
a reflection of the endoderm of the pneumatophore. No cellular processes, like those of
Rhizophysa, were observed to proceed from the under part of the reflected endoderm,
and I could discover no pore communicating with the exterior, even in the youngest
specimens.

The ccenosarc, immediately below the pneumatophore, gives attachment to a number of
budding nectocalyces, which are succeeded by a double series of perfect ones. Of these
there may be as many as five on each side."

Kélliker says that these are so like those of his Agalmopsis, that it is unnecessary
to describe their form particularly; they alternate and embrace the ccenosare by their
excavated sides. This was also the case in Vogt’s specimens and in mine. According to
the last-named investigator, the nectocalyces are horseshoe shaped, with an orbicular aperture
on the truncated summit of their external curve, surrounded by a membranous valve,
and leading into a nectosac, which is “very small, and does not extend through more than
half the thickness of the organ.”

M. Vogt’s figures, however, represent the nectosac with much larger relative proportions,
and Kdlliker represents the nectosac with much the same proportions as in my specimen,
in which the nectocalyces were about a quarter of an inch long, and had an irregularly
cordate form. Each was attached by a broad, thick pedicle to the ccenosarc, and its
axis was inclined (as Vogt notices) at an angle of 45° to the latter. The inferior and
external face of the organ was broad and convex above, but suddenly narrowed towards
its free end. The superior and internal face was deeply excavated in the middle line, so
as to present a convex lobe on each side, between which arose the pedicle of attachment.
The excavation was continued into the internal and inferior face, while the inferior and
external face was truncated and almost wholly occupied by the rounded, or somewhat
four-sided, aperture of the nectosac, provided with its narrow membranous valve.

The edge of junction of the inferior internal, and inferior external, faces is prolonged
into a thin, broad, subquadrate plate, whose lower free edge is emarginate in the middle line,
and pointed at the angles.

The sides of this plate are continued into two ridges, which run along the margins
of the inferior internal face to about its middle, and then become constricted, but are
continued upwards, on each side, on to the superior internal face, and thence downwards
on the superior external face, as far as its truncated inferior edge, where they end in
obtuse points. Superiorly and internally these ridges form broad plates, which overlap

1 In P. tetrasticha Philippi found four series of nectocalyces, each series containing four
of these organs (p. 59). It would appear, nevertheless, that they were all really attached to one side

of the coenosare. '

80 THE OCEANIC HYDROZOA.

those of the opposite nectocalyces, and thus enclose the csenosare in a sort of hydreecial
canal. The nectosac occupies nearly the whole cavity of the organ. Near its mouth it is
subcylindrical, but almost immediately it becomes exceedingly wide and is divided into two
lobes, by an inflexion corresponding with the groove in the internal face of the nectocalyx.

On very careful examination, the ends of four delicate canals could be traced opening
into a circular canal which surrounded the aperture of the nectosac, and the origins of
these canals could be seen radiating from the end of the canal of the pedicle, but in the fully
formed organs I could not satisfactorily trace their intermediate portions.

M. Vogt says of these canals, which he calls “ batonnets” (though he recognises them
to be canals): “ Une entoure liris orbiculaire de l’ouverture en servant de support a ce
rideau musculaire,! tandis qu’un autre se porte d’arri¢re en avant. Deux courbes latérales
embrassent le bouton median’ par laquelle la cloche est fixGe en arricre, se réunissent au
batonnet qui court dans le ligne médiane, tandis que deux autres se dessinent sur les ailes
postérieures proéminentes de la cloche”’ (p. 42).

I do not quite understand this description, and I will, therefore, without further
comment, state the results of a careful study of the development of these organs in my
Physophora. The smallest buds measured ;sth of an inch in length, and were simple cecal
processes of the ectoderm and endoderm of the ccenosarc, immediately beneath the
pneumatophore, with an internal cavity, whose walls were ciliated, and which was in free
communication with the somatic cavity (Pl. VIII, fig. 10 a).

In a bud of ,th of an inch in diameter, the apical region of the wall had developed
a large, internal, spheroidal prominence, and its cavity had thus become cup-shaped (fig. 10 4).
In buds of one eightieth of an inch in length, the cavity had become divided into four canals,
and a central hollow began to appear in the spheroidal prominence (fig. 10 c).

In a bud of one sixtieth of an inch long, the central hollow, or rudimentary nectosac, had
acquired a very large size, but was not open at its apex; and the four canals were united by
a circular canal. The whole organ had become broader in proportion to its length, principally
by the widening of its base. This was still more marked in young nectocalyces of one
fiftieth of an inch in length, in which the nectosac had a larger cavity, and proportionally
thinner walls. In these the lateral canals had acquired a slight double curvature, so as
to have the outline of an italic s (fig. 10d).

A nectocalyx, one twenty-fourth of an inch long, consisted of an extremely broad and
expanded base attached to a short pedicle, and not yet excavated internally, surmounted by a
short, truncated, and subcylindrical apical portion. There was no aperture in this, nor were
any of the characteristic crests of the adult organ developed. While the median longitudinal
canals remained unchanged, the lateral ones had become singularly contorted (fig. 10).
Supposing the pedicle of the organ to be posterior, and its natural superior and external face
superior, the lateral canals pass at first outwards, and then turn sharply inwards on the
under face, where each forms a loop by bending as suddenly outwards again. On reaching
the lateral face, it curves upwards and backwards to attain the superior face, on which it
passes inwards, and afterwards bends at right angles forwards; it then passes inwards again
and finally forwards to the circular canal. I traced the development of the organ no further

' The membranous valve. > The pedicle.

PHYSOPHORA. 8]

than this, but it is probable that in the adult organ the convolutions of the lateral canals
are even more complicated.

K6lliker considers the lower enlarged end of the stem of Physophora to be a simple
dilatation of the ccenosarec; but Vogt affirms that it is not only enlarged, but bent upon itself,
so that its discoid form is more apparent than real; and he adds, that the more developed
appendages are always situated at the distal end of this coiled enlargement of the coenosarc ;
the young buds of the hydrocysts and polypites being situated at its proximal end, or
where it joins the narrow stem-like portion of the coenosarc.

Philippi states that, in P. ¢etrasticha, a membranous fold runs down one side of the axis, to
which, without doubt, the nectocalyces were attached. Below, the axis becomes somewhat
wider, makes a turn, and ends in a tolerably large vesicle, which, in the living animal, is
hidden by the appendages. It is placed not in the middle, but on one side; upon the other
side it exhibits a wide aperture surrounded by a plaited membranous fold like a collar.
Philippi supposes that this aperture is the mouth, and the cavity into which it leads
the stomach, wherein he is undoubtedly mistaken. Externally, he found no trace of the
attachment of the hydrocysts, but, from the middle line below, the racemose reproductive
organs depended.

In my notes upon the small Physophore, I find it stated that the bulbous expansion
of the stem ends below in a sort of solid papilla, in which there were no indications of any
inferior aperture ; and on this last point Vogt and Kdlliker are also agreed against Philippi.

According to Philippi and Kélliker, the hydrocysts are arranged, most externally, in
a-complete circle around the inferior enlargement, and internal to these are the polypites,
with a tentacle attached to each. The male and female reproductive organs are attached in
pairs beside (neben) the polypites. | According to Vogt, the hydrocysts are external,
the polypites internal, and the groups of reproductive organs lie between them.

In my specimens the hydrocysts were external, the polypites internal; but I can say
nothing about the reproductive organs, as they were not developed.

The hydrocysts are closed sacs with pointed apices and broad bases, by which
they are fixed into the ccenosarc. Vogt says, “L’extrémité par laquelle le tentacule est
fixé au disque, est taillée comme un bec de plume de maniere a s’adapter a la surface arrondie
du disque. Cette extrémité n’est point fermée; elle s’adapte facilement sur une ouverture
conduisant du disque dans la cavité du tentacule. En arrachant ce dernier on remarque
qu'un fil de matiére Clastique entre dans le disque méme et sert ainsi d’attache ” (p. 45).

In my specimens this was by no means the way in which the hydrocysts were connected
with the ccenosarc, but the broad bases of the former coalesced completely with the wall of
the latter (fig. 3), which extended inwards so as to leave only a narrow, oblique, valvular
aperture of communication between the cavity of the hydrocyst and that of the ccenosarc.’

' Philippi (p. 61) describes the hydrocysts as having “ flattened articular surfaces.” “In thie
middle of the articular surface is a small knob (Hicker), probably the point of attachment, and down
from its upper part depends a fine thread, which, when extended, is longer than the hydrocyst,
but is often spirally coiled up, and then is hardly a line long” (p. 61). According to this observer,
the hydrocysts are true “ prehensile arms” (Fangarme), for, on placing the end of a forceps in their

neighbourhood, “ they all seized it, held it fast, and loosened their hold only in a couple of seconds.”

11

82 THE OCEANIC HYDROZOA.

The endoderm and ectoderm were exceedingly distinct in the walls of the young
hydrocysts. There were no villi; but the apices of the organs contained several large
thread-cells.

Vogt does not mention or figure any small tentacular appendages to the hydrocysts,
and Kélliker distinctly states (p. 22) that he “could find no trace of the fine tentacles
observed by Philippi attached to these organs in P. fetrasticha.” In the young forms
of my specimens, on the other hand, they were very obvious as delicate, filiform pro-
cesses of the ectoderm and endoderm of the hydrocyst, which arose from the upper wall,
close to its attachment to the ceenosarc. The internal cavity of each tentacle communicated
freely with that of the hydrocyst, and towards its apex delicate thread-cells were imbedded in
its outer wall. I did not particularly examine the polypites, but Vogt and Kolliker agree in
stating that they are affixed by slender pedicles, and that the tentacles are attached at
the junction of these pedicles with the enlarged digestive division of the organ. Villi are
developed from the endoderm of the latter.

The stem of the tentacle’ is as usual single, and carries a number of lateral branches, to
whose extremities very singular saccular bodies are attached, of whose structure I can speak
with some confidence, as I worked out the principal steps of their development.

In their early stages these lateral branches exactly resemble those of the tentacles
of the Calycophoride, that is to say, they commence as simple, cecal, double-walled
processes of the stem, which gradually elongate, and become divided into three portions.
The onewall of the middle division becomes greatly thickened, and rows of elongated thread-
cells disposed vertically to its plane are developed in it. The distal division remains short,
and acquires only small thread-cells.

The proximal division elongates and becomes a slender pedicle, devoid of -thread-cells
in the greater part of its length. When it joins the middle division, however, it becomes
dilated, and not only small thread-cells, but several very large oval ones, appear in it.
Similar thread-cells make their appearance in the proximal end of the middle division. Thus
far the development of the organ (which has now a length of about one fortieth of an inch)
exactly resembles that of the corresponding part in Dip/yes, and it is obvious that the basal
division answers to the pedicle and involucrum, the middle division to the sacculus, and
the distal to the filament (fig. 5).

The next change that takes place is, that the sacculus lengthens and becomes coiled upon
itself, while the involucrum at the same time dilates and acquires a quadrate outline (fig. 6).

In the most perfect organ (which measured about one sixtieth of an inch in diameter.
but was still colourless, and, consequently, incomplete) the involucrum has become
immensely dilated, so as to form a sort of hemispherical cup closed on all sides, except
at one point, where a small aperture is left, through which the extremity of the filament,
divided into three lobes, protrudes’ (figs. 7, 8).

} Attaining as much as a foot in length in P. ¢etrasticha, Philippi, 1. c., p. 61.
> Compare Philippi’s fig. 9. The lobes are described as ‘a few short cirri.” But this observer
has not comprehended the structure of those singular organs, and while he definitely asserts they are

not prehensile organs, supposes they may be branchiz.

PHYSOPHORA. 83

The structure of this involucrum differs from that of the corresponding part in
Stephanomia and Agalma, inasmuch as it is only partially formed by a lamellar expansion
of the ectocyst; the greater part of its wall being constituted by a cup-shaped dilatation of
the pedicle itself, from whose edges the proper involucrum rises. As a consequence
of this enlargement of the pedicle, the attached end of the sacculus, distinguishable at
once by its large oval thread-cells, has been carried to the distal extremity of the involucrum,
instead of remaining at its proximal end. And at the same time, as the sacculus has
increased in length far beyond the diameter of the sac, it has become coiled into a
helix with several close turns. By pressure it can be extruded from the involucrum.

As I have stated, none of those organs which I examined were fully formed. Indeed,
they must have been far removed from their adult condition, for they were not more than
one sixtieth of an inch in diameter; while, according to Philippi and Kélliker, the fully
formed sacs measure a line or a line anda half. Both Kolliker and Vogt, however, figure
and describe intermediate states, which completely bridge over the interval between the adult
forms which they more particularly investigated, and the very young ones of which I
have just given an account, but of which they say nothing.

Vogt’s pl. v, fig. 10, and Kolliker’s pl. v, fig. 3, represent successive intermediate stages ;
Vogt’s being but little more advanced than my oldest. The sacculus here makes seven
or eight turns, and its proximal end is represented as lying free, while the distal end,
containing many large oval thread-cells, is attached to the wall of the involucrum.

Kolliker’s figure represents a rather more advanced stage, the distal extremity of
the involucrum being produced into a point; otherwise, so far as the figure is concerned,
I should have said there was no essential change. His description, however, presents
difficulties.

“These (organs) consist of an oval vesicle produced into a point at its free end, and whose
thick granular walls, composed apparently of small cells, inclose an oval cavity (provided with
a small aperture near the pedicle), which contains a clear fluid and a large spirally coiled
filament. This commences at the upper (distal) end of the investing capsule, with a rounded
free end, runs down, making six or seven turns, towards the pedicle, and ends near the
aperture in a narrow cord, which ascends through the cavity of the spiral, and is inserted
into the upper (distal) end of the capsule. On examination the spiral thread turns out to be
a cylinder full of thread-cells, and the cord to be a muscle, which, divided into two bands, is
continued on to the spiral thread, and runs along its concave side to its free end” (p. 22).

From this description it will be observed that the general structure of the organ appears
to be the sameas that which I have described, except that the spiral is turned round, its free
end being distal instead of proximal, and the proximal end remaining connected with the
original place of attachment by a muscular band.

Such a Jouleversement is almost inconceivable, and Kolliker’s fig. 2, and Vogt’s figs. 9
and 11, convince me that it really has not taken place, for in all these I find the end of the
spiral cord, which contains the large oval thread-cells, distal, ¢. ¢., in the same position as in
my figures.

I conceive, in fact, that the real state of the case is this. As the sacculus enlarges, a
strong band of muscular fibres is developed within its thin wall, and it is the edge of this
band which, in the retracted state, appears like a distinct muscular cord traversing the axis of

84 THE OCEANIC HYDROZOA.

the spire. I am strengthened in this view by Vogt’s statement (p. 51) that when the sacculus
(his “ banderola’”’) is protruded, ‘ it is accompanied through its whole length by a very delicate
muscular filament, which is attached to its inner edge, so that it is hidden completely by the
folds of the ‘ banderola’ when the latter is retracted into the capsule.” It is true that Vogt
figures (fig. 9) the end of the sacculus, which contains the large thread-cells, free; but as the
lining membrane of the involucrum is forced out of the aperture, this may well be the result
of pressure.

In other respects the structure of the adult organ, as described by Kolliker and Vogt, is
readily reducible to the type of the young form. ‘The cavity of the base of the involucrum
appears to become filled up by vacuolated tissue, so that Kolliker describes the wall of the
capsule as consisting of three layers. ‘‘Close round the inner cavity (whose outline is
never twice alike, and which always presents a sinus towards its apex), in fact, is a granular
substance, having the same appearance as the wall of younger capsules, only that a reddish
colouring matter is deposited on one side of it. This is succeeded (but on one side only,
toward the apex) by a conical appendage formed of coarsely vesicular tissue ; and, lastly, the
whole is invested by a partly striated, partly vesicular-looking substance, whose extreme point
is also coloured reddish” (p. 23). This innermost layer is, I doubt not, the inner wall of the
dilated base of the involucrum. If the latter (as it appears to do) gradually grew out into.a
cone beyond the sacculus, and the endoderm of its cavity became vacuolated, as is so constantly
the case in the appendages of these animals, we should have the “ conical appendage with
coarsely vesicular tissue.” And if the ectoderm thickened and altered, it would give rise to
the striated or vesicular outer substance.

The reproductive organs are stated by Kolliker to resemble exactly those of his Agal-
mopsis (Agalma). Vogt’s account of their structure and development is not very clear, and, if
I may judge by the analogy of other Physophoride, is in many respects not quite correct ;
but it tends towards the same conclusion.

The largest of my Pdysophore was taken in the southern part of the Indian Ocean on the
17th of June, 1847. With it I obtained a very small one, not more than one third of an inch
long, and one of the same size had been obtained on a previous day (June 7th, 1847).

In these small specimens (fig. 2) the ectoderm of the stem-like portion of the ccenosare
appeared to consist of longitudinal fibres ;i.th of an inch in diameter. The lower end of the
coenosare expanded into a somewhat obliquely set, spheroidal, or elliptical, bulbous enlarge-
ment, round which were arranged three hydrocysts, two large and one small, which I took
at the time for young polypites. These, which have been described above, were arranged
nearly in one plane. Below them were numerous small buds, which appeared to be nascent
polypites. Arising from one side of the ccenosarc, just below the pneumatophore, were
several rudimentary nectocalyces; the largest and lowest, one sixtieth of an inch in length,
had four straight longitudinal canals united by a circular canal, but its nectosac was not
yet open. This is, I believe, the youngest state in which Physophora has yet been observed,
as the young form described by Vogt has four large hydrocysts and a polypite with open
mouth and long tentacle; however, I confess I can understand neither the figures nor the
description of the young tentacular sacs of this specimen.

I will not attempt to refer my Pdysophore to any of the species yet described, for I doubt
whether it is possible, with our present information, to separate one species of Physophora

PHYSOPHORA. 85

satisfactorily from another. I will only remark that if the hydrocysts of P. Philippi (KOll.)
and P. hydrostatica (Vogt), which I am strongly inclined to think with Leuckart are one and
the same, are really devoid of tentacles, mine is a totally distinct species.

In conclusion, I will remind the reader that fig. 1, Pl. VIII, is no better than a diagram,
the origimal sketch having been very hasty and imperfect. 1 imagined I should meet with
plenty more Physophore, and could finish my drawing at any time; but in three years and a
half I never took another. The other drawings were made very carefully, and are, I believe,
accurate.

86 THE OCEANIC HYDROZOA.

Fam. ATHORYBIAD/E.

Genus ATHORYBIA (2schscholz).

The hydrosoma, which is not distinctly separated into pneumatophore and ceenosarc, Is
spheroidal, the pneumatocyst nearly filling its cavity. There are no nectocalyces. The
hydrophylha are arranged around the coenosarc, on the proximal side of the other appendages.

ATHORYBIA ROSACEA (2). PI. IX.

Athorybia rosacea, Eschscholz. 1829.
= — Kolliker. 1853.

That portion of the hydrosoma which represents the pneumatophore is spheroidal, but
somewhat pointed above, and has its apex surrounded by a series of radiating dark-brown
stria, each made up of dots. Its general colour is pink. The polypites have a lightish-red
hue, shading off into pink at their apices. The tentacles are yellowish or colourless, with
dark-brown sacculi.

The hydrophyllia are lanceolate, tolerably straight in their upper halves, but greatly
incurved below. The upper and’ outer surface of each is convex, the lower and inner concave.
Six longitudinal rows of small thread-cells mark the outer surface. The basal or upper portion
of the hydrophyllium is broad, and presents, on each side, a notch or shoulder, where it joins
with the narrower neck, by which it is connected with the broad lanceolate blade. These
shoulders overlap, or are overlapped by, those of neighbouring hydrophyllia. The axis of
each hydrophyllium is traversed by a narrow, tubular, czecal phyllocyst, which runs nearer the
inferior surface, and is wider and ciliated internally in the younger organs. The larger and
more conspicuous hydrophyllia are arranged in a circle around the globular hydrosoma above
its equator, but a few smaller ones are to be observed among the appendages of the under
surface. These were polypites, tentacles, and reproductive organs.

Of the polypites, the distal ones,’ or those attached more towards the centre of the under
surface, especially one or two, were the larger and more active. Their inner wall was provided
with villi for about its basal two thirds; some of these villi were short and thick, others thinner
and elongated. The latter were often jagged or produced into small points at their
extremity. The vill contained large clear spaces (vacuole) in their interior.

Young undeveloped polypites were scattered among the perfect forms.

The tentacles sometimes appeared to come off from the wall of the ccenosarc itself,

} The upper and outer series were so similar to the others that they are not specially distinguished
in my notes, so that I cannot say whether they have the structure described by Kolliker in the “ Filer,”

or hydrocysts of his Athorybia, or not.

ATHORYBIA. 87

but in other cases I could distinctly observe that they arose from the pedicle of the polypites.
In their perfect state they were similar to those of the Calycophoride, except in the structure
of the sacculus, which nearly resembles that of Agalma.

This organ is very long, and is coiled up within a large involucrum. The distal end of
the sacculus is prolonged into a conical median lobe, devoid of thread-cells, and two lateral,
comparatively short, filaments (fig. 9/).

The involucrum and the filaments contain many small thread-cells (;i:th inch); those
which lie in the thickened wall of the sac are, as usual, elongated, about ,J,th of an
inch long, and of a deep-brown colour. Thread-cells of a different kind are disposed obliquely
to these and imbedded in the proximal half of the walls of the sacculus; they are from
two to eight in number on each side, measure ;ith of an inch in length, and have
an elongated oval form.

The lateral branches of the tentacula are first visible as papillae about {th of an inch
in length, composed of the endoderm and ectoderm, and containing a diverticulum of
the somatic cavity. These papilla elongate and become tricuspid at their extremities, small
thread-cells at the same time making their appearance in their walls towards the proximal
end; next a constriction appears in the middle of the papilla, and one wall becomes
much thicker than the other. The lateral processes of the tricuspid distal end now
elongate rapidly, the central one remaining comparatively stationary, and the thread-cells
acquire their characteristic forms (figs, 92—9e).

The involucrum finally makes its appearance as a process of the ectoderm of the basal
division of the organ. Up to this time the sacculus has remained straight; its coiling up only
takes place subsequently.

The reproductive organs are scattered irregularly upon independent branched stalks
or gonoblastidia (fig. 12), one pedicle (always?) carrying both androphores and gynophores.
Small hydrocysts are also developed on the same pedicle. The androphores are longer and
rather thicker than the gynophores. The calyx of the former presents a terminal circular
aperture, and the characteristic longitudinal canals, though I am not sure there were four of
them.

In the most advanced androphores the manubrium had a reddish colour, and its apex
projected beyond the mouth of the calyx; it contained nearly fully developed spermatozoa.
Each gynophore contained but a single ovum. The calyx exhibited its four regular canals,
united, in the fully developed organ, by a circular canal surrounding a terminal aperture.

The gynophore appeared to contain many large superficial anastomosing canals, an
appearance produced, as in the case of Ayala, by the irregular and incomplete separation
of the outer surface of the manubrium from the inner surface of the calyx.

I took this Athorybia in the Indian Ocean in April, 1847; on the East Coast of Australia
in May, 1848.

The only complete anatomical description of any Athorybia extant is that given by
Killiker of 4. rosacea, which occurred abundantly at Messina, and presented many varieties.
The hydrophyllia were attached below the air-vesicle to the short, depressed, conical
ceenosarcal portion of the hydrosoma, in two or three circles, of which the uppermost was

88 THE OCEANIC HYDROZOA.

sometimes the shortest, so as to suggest that here, as in other Physophoride, the youngest
organs are nearest the proximal end of the body. Some of these organs were lanceolate and
compressed at the base, others obliquely truncated at the base, and rounded at the apex. Five
or six lines of thread-cells ornamented their outer surface, and each contained a narrow canal
which nearly reached its apex, and was not ciliated. The hydrophyllia could be raised
and depressed, and the animal could propel itself by alternating these movements.

There were twenty to forty hydrophyllia, but never more than eight polypites.

Irregular cavities occur in the walls of the latter organs, which Kélliker considers to be
elandular. .

A series of hydrocysts (Fiihler), some fourteen or twenty in number, and always more
numerous than the polypites, are attached by short peduncles around the margins of the
hydrosoma. ‘Their walls are muscular and ciliated internally and externally, and their apices
are surrounded by thread-cells.

There is a-tentacle for each polype, whose general structure appears to be very similar
to what I have described above, but there is a good deal of apparent difference in the sacculi
and their appendages.

K6lliker says—“The saccculi (nesselknopfe) consist of many parts. Their pedicle
divides into two organs; a stalked, elongated capsule, and a crescentic, proper urticating cord
(nesselstrang); to whose extremity two prehensile filaments (fang-fiden) and a pyriform vesicle
are attached. All of these parts are hollow, and contain the same clear liquid as the stem.
The urticating cord shows the ordinary structure. First, large thread-cells, one of which
is shown with its thread projected in fig. 10; then smaller thread-cells, inserted in rows
in the thickened wall, between which a reddish-brown pigment is deposited. Finally,
there are two muscular cords in the concave part of the organ. The prehensile filaments
are thread-like, short, and very contractile, so that they can shrink into a very small
space, and are beset throughout with small thread-cells; while on the other hand,
the pyriform sac appears clear, and although quite similar in form and_ position
to the analogous organ in 4Agalmopsis Sarsii, exhibits no movements and is not ciliated.
The stalked elongated capsule, lastly, exhibits im its walls, and, as it would seem, also,
in its interior, a beautiful reticulation, in whose spaces, besides a clear fluid, yellowish
or reddish fatty-looking globules are often contained. Besides, the yellow middle part
is (though not always) surrounded by transverse fibres, and, at one point, covered with
dark organs like urticating organs, and drops of fat. The signification of this vesicle, which,
like the other parts, was observed colourless, especially among the uppermost of these organs,
is altogether doubtful. In the younger and youngest sacculi, such as were frequently to be
observed at the root of the tentacles, it was not unusually closely applied to the concave side
of the urticating cord, but I could find no close ‘relation between the two parts, nor did
it seem that the capsule took any share in the movements of the urticating cord, for, indeed,
it exhibited none itself.” :

On comparing this description and its accompanying figures with my own notes
and drawings, I entertain no doubt that the structure of the tentacles in the two
Anthorybie is essentially the same; that, in fact, the “ stalked elongated capsule” is the
involucrum; the “ urticating cord” the sacculus; the pyriform sac the dilated median lobe ;
and the “small tentacles” the filaments.

ATHORYBIA. 89

The reproductive organs of both sexes are found combined in 4. rosacea, as in my
species, but, apparently, they are somewhat differently arranged. A solitary androphore
and a bunch of female organs are attached to the under surface of the hydrosoma, beside
each polypite. The male organs have essentially the same structure as that I have described.
As to the female organs, Kdélliker confirms the account which I had given in ‘ Miiller’s
Archiv’ of the peculiar canals which lie between the ovisac and the calyx, and shows that
their structure is in all essential respects the same as in my species. The entire calyx is,
he says, covered with cilia, and when detached swims about by their agency.

Forskal gives the length of the hydrophyllia in 4. rosacea as half an inch, whence
the diameter of the whole animal must be about an inch. Lesson says, “ Largeur | pouce,”
but Kélliker’s specimens would not seem to have attained half that size, which is about
the dimensions of those I observed. Notwithstanding the apparent discrepancies, I am
inclined to regard the forms I have described as of the same species as that described
by Kélliker. It will be better at any rate to do so provisionally, for M. Kolliker’s observations
show that the Avhorybie are liable to great variation, and it is anything but clear that
the A. heliantha of Eschscholz and the 4. melo of Quoy and Gaimard are distinct from
A. rosacea.

12

90 THE OCEANIC HYDROZOA.

Fam. RHIZOPHYSIDA.

Genus RHIZOPHYSA (Peron and Lesueur).

Hydrosoma filiform; neither nectocalyces nor hydrophyllia. Tentacula branched, the
branches peculiarly modified at their extremities, but without true sacculi.

RHIZOPHYSA FILIFORMIS? PI. VIII, fig. 13.

Physsophora filiformis, Forskal, 1775.
Epibulia filiformis, Eschscholz, 1829.
Rhizophysa filiformis, Gegenbaur, 1853.

The pneumatophore is pyriform, and, like the rest of the body, of a pale, pinkish hue,
but with a very deep red patch surrounding the aperture of the pneumatocyst.'

The coenosarc, long and cylindrical, is hardly at all twisted upon itself, and gives
attachment at intervals to slender polypites, about half an inch long.

The endoderm of these organs exhibits short villi, containing clear spaces. A tentacle
arises at the junction of the polypite with its pedicle, and the stem of this organ when
contracted is not more than an inch long, thick at its base, thin at its extremity, and
usually coiled upon itself. It is capable of great extension (to as much as two or three
inches in length), and then appears like a slender thread. Floating at the top of the water
in a glass vessel two or three inches deep, the animal could fix itself by means of the
lowermost branches of its tentacles to the bottom, and holding on by them, raise and lower
itself at will.

The stem of the tentacle gives off a series of lateral branches in which no distinction
into pedicle, sacculus, and filament is discernible. These branches arise from only one side
of the stem, and those which are nearest the attached end are mere small, oval buds, while
the distal ones are delicate threads as much as half an inch in length. Each thread is
a double-walled, caecal tube, with one wall thicker than the other in the proximal portion,
and alone containing the thread-cells, which are scattered over the whole surface in the lower
part of the organ. The thread-cells are spheroidal, and measure about ~,th of an inch
in diameter.

The polypites appear to be attached indifferently to either side of the ccenosare (?), and
at the base of the pneumatophore several small and partially developed ones could be
observed. There was no indication of any nectocalyces.

* Whose structure is particularly described above.

RHIZOPHYSA. 91

I sought carefully for reproductive organs, but the only trace of them to be discovered
was two bunches of small, oval bodies, situated at the base of and between, two of the
polypites. These were oval, double-walled diverticula of the ccenosare, with a ciliated

internal cavity, and about jth of an inch in length. Their ectoderm contained a few
thread-cells.

I took a single specimen of this species in the Indian Ocean on the 2d of January, 1849.

It was in this animal, as I have stated above, that I saw the air spontaneously expelled
from the orifice of the pneumatocyst.

The only account of any species of Rizophysa which suffices the wants of the modern
naturalist is the excellent description of 2. fiiformis by Gegenbaur, in the ‘ Beitriige,’
to which I have so often had occasion to refer. This animal attains a foot and a quarter
in length, but has about the same sized pneumatophore and general thickness as that which
I have just described.

In Gegenbaur’s description of the structure of the pneumatophore, the facts agree pretty
closely with those which I have observed, but his interpretation of them is different. He seems,
in fact, to regard the layer of endoderm, reflected over the outer wall of the pneumatocyst,
as the whole wall of the pneumatophore, and hence he imagines that the apex of the
pneumatocyst is, at the superior pore, naked. The ciliated, cellular processes are described
just as I saw them. I did not observe the layers of small, round, yellowish cells which
he describes between the endoderm and the air-sac, whose presence, however, is extremely
interesting when we consider the position of the liver in Vededla.

Gegenbaur having supposed the reflected endoderm to be the ectoderm, naturally denies
that the apical pore can have any communication with the interior of the air-sac; but
its real connexions are in favour of such a communication, and, as I have stated, I
distinctly saw air-bubbles escape from it.

The young polypite is at first, according to Gegenbaur, a simple process of the wall
of the ccenosarc, whose cavity at first freely communicates with that of the latter.
The rudiment of the tentacle buds forth from the base of the polypite as this budded from
the coenosarc; and at the same time a process arises just in front of this rudiment, and
gradually reduces the communicating passage to a narrow canal, which can be completely
shut by muscular contraction. Eventually, the polypite opens at its apex, becomes
functionally active, and villi appear on its inner surface.

Thus far Gegenbaur’s observations agree so closely with my own that I can see no
ground for supposing that we examined different species, but, in his account of the tentacles,
he describes structures which, I think, I could hardly have overlooked, though, as I examined
but one specimen, I may have done so; or, perhaps, they were not developed in the tentacles
I examined. They consist, he states, of a stem with secondary branches, whose buds
are closely aggregated at the root of the stem. At the end of each branch are small,
mostly greenish capitula, which, when microscopically examined, appear as prehensile organs
(fang-organe) of a structure quite different from what obtains in other ‘“ polypi nechalei.”
Their forms are various, but they may be reduced to the different states of three typical forms.

In the first, the end of the branch is somewhat dilated and divided into three lobes, each
of which is beset with large thread-cells.

92 THE OCEANIC HYDROZOA.

In the second, the expanded end of the branch divides dichotomously (so as to resemble
one of the oral tentacles of a Bougainvillia), and a single large thread-cell is contained in the
end of each of the terminal branches.

Of the third kind there is not one for ten of the others. The end of the branch enlarges
into a globular knot, containing many thread-cells, and sending out, at a right angle to the
axis of the branch, a firm, conical process. “It is composed of a few cells, and carries
on one side a coat of clear, muscular fibres, by means of which it is frequently set in motion,
being usually bent towards that side on which the muscular fibres lie, and then starting back
into its first position.

“At the root of this beak-like process a great number of fusiform bodies are attached,
one end being buried in the organ, the other free. No motion was observed in them.
Their contents are finely granular and condensed towards the middle, where the outlines
of a nucleus are sometimes discoverable. ‘The points are almost always clear and glassy.”

Fully developed sexual organs were found by Gegenbaur in none of the numerous
specimens which he examined. But he describes vesicular bodies seated on the coenosarc
between the polypites, which he believes were their rudiments. He found, however, no trace
of either generative products or medusiform buds, so that the real nature of these organs
is still open to doubt.

Lesson (‘ Acaléphes,’ p. 490) gives a figure of Riizophysa filiformis, to whose coenosare,
spheroidal, granular masses are attached between the polypites. These he calls in the text
“ paquets d’ovaries,” and in his definition of the species these “ paquets d’ovaries” are said to
be ‘“‘jaunes.” Nevertheless, in the figure they are of a lively pink hue! By way of making
the confusion perfect, Lesson praises Delle Chiaje’s “ description de cet Acaléphe,” which he
subjoins, and which obviously relates to a totally different animal.

PHYSALIA. 93

Fam. PHYSALIADA.

Genus PHYSALIA (Lamarch).

Hydrosoma irregularly pyriform. Pneumatocyst occupying almost the whole of the
somatic cavity, and produced into czca, arranged in a longitudinal series. Tentacula without
branches, but provided with reniform enlargements along one edge, and with basal sacs.

The term Pfysalia was employed by Lamarck to denote certain remarkable animals
which, before his time, had passed under the various scientific names of Urtica marina,
Arethusa, Holothuria, Salacia, &c., and were popularly known by the titles of “ Portuguese man-

“Caravella,” “ By-the-wind Zeyler,” &c. appellations,

of-war,” “ Kriegschiff,” ‘ Galere,”
which the Physalie enjoy in common with the Ve/e//a, and all expressive of the fact,
that the great bladder which constitutes the most conspicuous part of the animal is impelled
hither and thither by the winds, as it floats on the surface of the sea.

Cuvier discerned the true position of P/ysa/ia in the animal kingdom, and Eysenhardt’s
essay, ‘Zur Anatomie und Naturgeschichte der Quallen,’*

upon its structure. He points out the analogy of the polypites of Piysalia with the mouths

contains some valuable notes

of the Rhizostomide, and of the central cavities and canals of each, but he considers the long
tentacles to be external genital organs, and seems not to understand the true nature of
the pneumatocyst.

Eichwald again (‘Observationis nonnull circa fabricam Physaliaw, auctore Dr. E.
Eichwald*) describes the manner in which the tentacles are, as it were, gathered up,
along the line of their attachment to their basal sacs, but he considers them to be repro-
ductive organs (*funiculos proliferos, cognominare liceat,” 1. c., p. 456). The attachment of
the pneumatocyst to one point only of the hydrosoma, and the free communication of
the cavity interposed between the pneumatocyst and the wall of the hydrosoma, with
the cavities of all the appendages, are noted. The crest, which Eichwald calls a branchia,
and certain vessels distributed upon it, are fully described, but the reproductive organs
are supposed to be polypites in various stages of development (p. 461), while the author's
conception of the reproductive process is expressed in the following paragraph (p. 468) :

“Hisce forsitan concludendum, physalias veluti infusoria seu polypos hydriformes
prolem gemmee instar, in funiculis proliferis progignere, qui tunc per fimbrias ei nutrimentum
adducerent donec tubuli suctorii ab initio filis tenuibus comparandi ei evolverentur.”

A somewhat more satisfactory account of the anatomy of Physalia is due to Eschscholz
(‘System,’ p. 147, et seq.)

1 Nova Acta, t. xvii, pt. 2, 1821.
2 «Mém. de l’Acad. Imp. des Sc. de St. Petersbourg,’ t. ix, p. 453, 1824, (The essay was read in
1822.)

.

94 THE OCEANIC HYDROZOA.

“The very large swimming bladder is distended with air,” says this excellent observer,
‘in such a manner that its longest diameter is horizontal. In all the species there may be
remarked at one end of the bladder a prolongation, also full of air, which is not provided with
suckers or prehensile filaments, and at whose surface, near the end, a distinct pit is perceptible,
which, as soon as the bladder is compressed, opens and allows of the exit of the contained air.
In some species this part equals half the entire length of the bladder, while in others
it is very short. The opposite end of the bladder, on the other hand, is, in all species,
covered on one side with suckers, which, in young individuals, are much more imperfect
than the others which lie in the middle of the vesicle. This end also possesses a pit on
its upper surface, which appears to be an opening of the bladder,’ and is in some species
provided with a particular solid process. Along the upper surface of the bladder there runs
a plaited ridge, which, in its common condition, is also filled with air ; but the animal can, at
will, press the air out of it, when the ridge collapses into a membranous fold, but the bladder
becomes thicker and more distended.

“The bladder consists of a double membrane, an outer solid, and an inner much more
delicate one; when the animal has lain in spirit, the inner membrane can be separated
from the outer as a separate bladder.

“On the lower side of the bladder are the organs of nutrition, which consist of suckers
and prehensile filaments. The former arise either singly” from the bladder, or many
spring together from a common stem. The prehensile filaments are so far to be called
simple that they neither branch nor give off lateral offsets. They consist of a rounded
filament, which is covered throughout its whole length on one side with a series of reniform
acetabula (saug-warzen), and on the other side is supported by a narrow membrane, which
accompanies it from the root to the point.

« At the root of each prehensile filament, of which there are many of different sizes on a
single animal, is a long, pointed receptacle of fluid (fliissigkeits-behilter), attached throughout
almost its whole length to the filament, and only free at its apex. The acetabula of the
prehensile filaments appear to be the organs which secrete the mucus which produces the
well-known urtication of the human skin, and by which animals which are seized are at
once paralysed.”

Eschscholz compares the receptacles at the bases of the tentacles to the ambulacral
vesicles of the Hchinodermata.

After denying the justice of the statement of a Dr. Blume respecting the existence
of a nervous system in Physalia, Eschscholz goes on to say :

“That the Physalie have the power of emptying the air out of their bladder, and
so diving down, as the older writers relate (though their statements are only suppositions),
rests upon no direct observation. Piysalie which are touched, pricked, cut, irritated with
acrid substances, or placed in brandy, would certainly not fail to dive if they could, in order
to escape such treatment. But it is only by actual compression of the bladder that I have

1 This observation has not been verified by subsequent observers.
> T entertain great doubts whether this is ever the case in adult Physalie. Leuckart denies it
in Ph. utriculus (‘Ueber den Bau der Physalien, p. 195), which is in perfect accordance with

my own observations.

PHYSALIA 95

succeeded in pressing the air out of the aperture. However, a young P/ysalia of five lines in
length possessed this power: after I had touched the animal a few times, it suddenly expelled
all the air from its bladder, and sank to the bottom of the glass.

“ Besides the prehensile filaments and suckers, we find between these organs, on the
under surface of the bladder, one or many bundles of short threads, which may be regarded
as young. Different parts are distinguishable in these bundles, namely, a long thread closed
at its apex; then a tubular or funnel-shaped body; anda small globe at the root of the two
preceding. These parts of the bundle were easily detached when the animal was touched,
which clearly showed them to be young, as this phenomenon is observed in all the lower
animals. Of the three organs here mentioned I hold the first, the long thread, to be the
receptacle of fluid of an undeveloped prehensile thread, the funnel-shaped body for a sucker,
and the small globe for the vesicular body, which is not yet filled with air. So that these
three parts form together the most essential parts of the body of the PAysalia.”

De Blainville (‘Manuel,’ p. 113) has fallen into the most marvellous errors respecting
Physalia. Not content with discovering its mouth and an anus, he attributes to the Physalie
a stomach (the air-bladder), a foot (the crest), branchiee (the tentacles), and generative
apertures, besides a hepatic plate, vessels, and a central circulatory organ! No wonder, then,
that he places Piysalia among the Molluscs.

Von Olfers, the author of the next essay of importance on the structure of the P/ysalia,'
has not added very much to what was made known by Eschscholz.

He describes the chambered structure of the crest, and mentions the sphincter of
the aperture, but he affirms that a probe introduced into the latter leads into the space
between the outer bladder and the inner, in which last he was unable to detect any opening.
Like Eichwald he ascribes to the inner, air-containing, bladder cecal processes, which lie
in the compartments of the crest.

Von Olfers further mentions the villi of the polypites (or large and small suckers),
and the communication of the cavity of the latter with the cavity which lies between
the external vesicle, or body-wall, and the internal proper air-bladder. Less accurate in some
respects than his predecessor, he affirms the tentacular receptacles to have an aperture at
their apex, though he admits he has never seen the supposed mouths in a state of diia-
tation.

The thread-cells, while iz stu in the villi and tentacles, are described as small, round
glands; while, when partially detached and adherent by their thread only, to the latter,
Von Olfers has mistaken them for Vorticelle.

The supposed budding young, or “brut,” of Eschscholz, are called “keimbiindel”—
germ-bundles, and are carefully analysed by Von Olfers into bodies of three kinds: 1. Small
suckers, similar to the large ones, but smaller; 2. Club-shaped bodies; 3. Ovate or
pyriform bodies.

Von Olfers states particularly that he does not consider these to be reproductive organs
(l. c., p. 163), nor does he agree in Eschscholz’s view of their nature; but he imagines that
the ovate or pyriform bodies are “the germs of young P/ysalie, which, probably, become

1 «Ueber die grosse Seeblase (Physalia arethusa) und die Gattung der Seeblasen im Allgemeinen.’

Abhandlungen der Kén. Akademie d. Wissenschaften zu Berlin, 1831.

96 THE OCEANIC HYDROZOA..

developed into the clavate bodies and then fall off, either alone or m connexion with
one of the small suckers as young buds.”

In 1847, during the voyage of H.MLS. ‘Rattlesnake,’ I’ sent from Australia to the
Linnean Society, a memoir on the structure of Piysalia, which was read before that learned
body on the 21st of November and the 5th of December, 1848, but the following abstract
from the ‘ Proceedings of the Linnean Society,’ vol. ii, p. 3, 1855, is all that has appeared of
its contents.

“The specimens of Physalia on which Mr. Huxley’s observations were made, were
collected on board the ‘Rattlesnake, between the 25th of February and the 3d of March,
between lat. 25° and 37° south, and long. 5° and 7° west. They varied in size from one sixth
of an inch to two inches, in the long diameter of the float. The author first describes the
general appearance of the specimens, of which he doubts whether the largest were adult,
and then proceeds to a minute examination of their details, dividing them for this purpose
into the float, air-bladder, and the appendages, of greater or less length, which depend from
it, when the animal is in its natural position at the surface of the water. The smaller
specimens he states to be the best adapted for examination.

“The float is described as consisting of an outer coat, an inner coat, and an air-sac,
contained within them, attached only to one spot of their parietes, and there communicating
with the exterior by a small constricted aperture, which was always found on the upper surface.
The disposition of the appendages is very irregular, but the larger tentacles are generally
placed more externally, the smaller and nascent organs more towards the centre. These
appendages are of three kinds, and consist of stomachal sacs, tentacles, and cyathiform bodies.

“Of each of these the author gives a detailed description, in their more perfect form,
as well as in their undeveloped state, as nascent organs; and then proceeds to inquire,
first, what is the physiological importance of the organs described, and, secondly, what
zoological place should be occupied by an animal provided with such organs, so disposed.

“Each of these questions the author treats at considerable length. Of the function of the
stomachal sacs in receiving the prey there can be little question; but it may be doubted
whether the digested nutritive matter circulates in the ciliated water-carrying canals,
or is absorbed into totally different channels.

“Tn the latter case, the purpose of the stomachal villi would plainly seem to be to absorb
nutritive matter, and convey it, through their central canal, to the wide interspace existing
between the outer and the inner membrane; but the author states that he has never
seen in this interspace any corpuscles analogous to those described by Will’ as_blood-
corpuscles. He suggests that the villosities noticed by Dr. Milne Edwards in the stomachal
sacs of Apolemia are the same organs, and not ovaries, as Dr. Milne Edwards considers them ;
and observes that similar organs exist in a Dipiya (Ludowia), hereafter to be more fully
described.

“The functions of the tentacles, both as prehensile and defensive organs, admit of little
doubt, and on this subject the author notices an erroneous view of M. Lesson, who describes

1 By accident, the author of the communication is called ‘William’ Huxley in the ‘ Proceedings.’
2 T should state that Will’s ‘ Hore Tergestine’ was one of my few books, and that for a long

time I made great efforts to see the vascular system which he has imagined to exist in many Hydrozoa.

PHYSALIA. 97

them merely as ducts for conveying a (hypothetical) acrid fluid from a (hypothetical)
poison-gland. He also controverts M. Lesson’s opinion that certain of the colourless
tentacles are to be regarded as branchie; being quite sure that there is no difference
between these and the ordinary tentacles, except in the absence of colour. As regards
the function of the cyathiform bodies, he has no other than analogical evidence to offer.
The only organs in the Acalephe with which he conceives them to have any resemblance are
the natatorial organs of the P/ysophore. But their little adaptation to a similar purpose, and
the entire absence even of their rudiments in young Physalie, discourage this comparison ;
while, on the other hand, they bear a singular resemblance to the female generative organs of
a Diphya, and this resemblance extends even to the younger stages of both.

“Mr. Huxley concludes by referring Physalia to the position assigned to it by Eschscholz
among Physophore, and near Discolabe or Angela. In fact, he regards Physalia as, in all its
essential elements, nothing but a Physophora, whose terminal dilatation has increased at
the expense of the rest of the stem, and hence carries all its organs at the base of this
dilatation.

“The paper was illustrated by pencil drawings of the structures described.”

In 1851, Professor Leuckart published a very valuable essay, ‘On the Structure of the
Physalie and of the Siphonophore in general, which contains some important additions
to our knowledge of the structure of these animals. On one or two points, respecting which
Leuckart’s observations are more complete than Von Olfers’ and my own, I will quote him at
length.

“The crest is an integral part of the outer wall of the body, and may be regarded as a
duplication of it. Between its two lamelle there remains a cavity, which is not continuous
through its whole length, being divided by a number of transverse septa, which are visible
even from without, into a corresponding number of czecal chambers or compartments.

“These septa are especially formed by the inferior muscular layer of the wall of the
body, whose elements here consist of large, trabecular, transverse fibres, composed of
fine fbrille.

“The length of the septa is very different, and they are alternately larger and smaller.

“According to these differences, four groups of septa may be distinguished. The first
group contains the longest, which traverse the whole height of the crest, from its ridge to its
base. Of such septa I count six in the present species, a number which, notwithstanding the
difference in size (I examined individuals of two to three and a half inches in length) of
the body, appeared to be constant.!_ These six septa divide the cavity of the crest into seven
successive chambers. Each such chamber is divided in the middle by a septum of the
second order, which extends from the ridge half-way down. In this way fourteen chambers
are formed in the upper part of the crest, and this number is increased by further repetition
of the dichotomous division to twenty-eight and fifty-six. The last septa are the shortest,
being little more than constrictions of the outermost ridge of the crest.

“The position of the crest is commonly stated to be on the upper surface of the bladder.

1 Leuckart adds in a note: “This character would, therefore, seem to be not without importance
in distinguishing species. In Ph. arethusa I counted twelve such septa, with which the statements and

figures of Von Olfers agree.”

13

98 THE OCEANIC HYDROZOA.

This, however, is not exactly the case. At least, if the place of attachment of the appendages
is decisive, and the opposite surface is to be regarded as the upper, the crest lies horizontally
and on one side, hardly more remote from the under than from the upper surface.”

With respect to the ‘‘ keim-biindeln,” Leuckart’s account of their structure contains no
addition to that of Von Olfers, which is not to be wondered at, as only spirit specimens were
at his disposal. He states (p. 209):

“So much is certain, that these vesicles (the ovate and club-shaped bodies of Physalia—
medusoid bodies of other Stphonophoride) are not sexual organs, but are more or less
developed individuals of a second generation, which arise asexually by budding and are capable
of sexual propagation, whilst the Siphonophoride themselves remain constantly without sexes.”

As we’ shall see, Leuckart nearly, though not quite, divines the true nature of these
bodies in this passage.

In the same year 1851, the lamented Johannes Miller, the great anatomist of Berlin,
published in his ‘Archiv’ a translation of a brief communication which I had addressed to him,!
and which comprised among other matters a description and figures of the reproductive
organs of P/ysalia, based upon the observations and drawings which will be given in
the present work.

The last memoir upon Physalia to which it is necessary I should advert, is that published
in the ‘Annales des Sciences Naturelles, for 1853, by M. De Quatrefages.

The author justly corrects the statement of Leuckart which I have quoted above,
so far as to point out that, in the natural position of the animal, the crest is vertical,
while the appendages are lateral; and he describes extremely well and faithfully the singular
somersaults which the Pdysalia describes during life by raising up the pointed end of
its body, changing its centre of gravity and tumbling over; but I must confess that I cannot
regard the fact of the performance of these manceuvres as any proof that the Physalie
are in the habit of tacking (virent de bord), or that it “ décéle une volonté bien déterminée
et comme réflechie” (p. 114). On the contrary, as I have over and over again watched
the operation going on with great vigour in a dead calm, I cannot but think that if the
Physalie really imagine they are tacking or wearing ship, they exhibit a remarkable absence
of reflective capacity.

In discussing the functions of the pore, M. De Quatrefages contributes the following
remarkable observation :

“One of my Physali@, which had already served for many investigations, suddenly began
to collapse. The loss of the gaseous matter evidently took place through the pore in
question. The bladder collapsed completely and floated on the surface of the vessel, without
giving any other signs of life than obvious contractions when I irritated it.

“T believed the animal was dead, and setting it down for such in my notes, resumed the
occupation with which I was then engaged. A quarter of an hour afterwards, at the most,
having chanced to cast a glance at my Piysalia, I was greatly surprised to see it distended
anew, although a little smaller than it was at first, and erecting its crest in the air at the surface
of the water, just as when it was first brought to me.”

» «Ueber die Sexual-organe der Diphyde und Physophoridx.’ Miiller’s ‘Archiv,’ 1851, p. 380,
taf. xvii.

PHYSALIA. 99

M. De Quatrefages goes on to say, that, to account for this reappearance of the air,
we must suppose either that it is secreted, or that it enters by the pore, and he adopts
the latter hypothesis, conceiving that the walls of the air-vesicle are so strong and elastic
as to act like a perforated, hollow, caoutchouc ball, which yields its air when compressed
by the hand, and, expanding when the compression is removed, sucks it in again. The
muscular fibres of the body-wall act, in M. De Quatrefages’ opinion, as the compressing
force, and when they are relaxed the vesicle resumes its primitive form. I must say
that this hypothesis appears to me to be quite inadmissible. The wall of the pneumatocyst
is, so far as my recollection serves, kept tense by the air which it contains, so that when
it is pricked or cut, the organ at once collapses, by its own elasticity, quite independently
of the muscular body-wall. And the extreme thinness of the membrane of the pneumatocyst
is, to my mind, incompatible with the notion that its elasticity is sufficient to overcome
the weight of the wall of the hydrosoma.

As little can Tagree in M. De Quatrefages’ view of the functions of the air-vesicle,
which he supposes to be respiratory. A chemist, I think, would not lay much weight on the
two eudiometrical experiments whose results are given; and the dense and tough, though
thin, elastic membrane of the pneumatocyst, covered as it is by the endoderm (the mucous
layer of M. De Quatrefages), is hardly the structure in which a physiologist can recognise a
special respiratory apparatus.

On injecting a Physalia by one of the polypites, M. De Quatrefages made an observation
which seems to point to the existence of a system of canals in this Physophorid similar
to those which cover the upper surface of the pneumatophore in Ve/el/a and Porpita.

“T saw that the coloured liquid had passed into a perfectly free cavity on almost
the whole of the lower face and a part of the sides of the body, but, beyond a certain limit,
this cavity seemed to me to become changed into an extremely close network’ of canaliculi,
ascending even on to the upper face. Muscular (?) bands passing from the parietes of
the body to the air-vesicle, properly so called, kept the two concentric bladders in connexion
and traversed the cavity. I regret not having ascertained if this last, or the
network, which is a prolongation of it, passed into the crest, but I am led to believe
it did so.”

I do not propose to discuss the histological details into which the author enters,
but I must remark that he has unfortunately confounded the tentacular sacs with polypites,
while he supposes the latter to be hepatic organs. The account of the development of
the polypites and of the reproductive organs is also, in many respects, imperfect.

The perusal of the works of the various writers who have occupied themselves with the
establishment of specific distinctions among the Pdysaliz, simply makes one long for the
advent of a Caliph Omar in this department of zoological literature. A sort of unpleasant
vertigo is the only result I can report of my study of the systematic labours of Von Olfers,
Lesson, and Lamarck, but those who are inclined to take up the inquiry will find a very
elaborate discussion of the synonymy and characters of the various so-called species in the
already-cited memoir of the first named of these naturalists.

Eschscholz, on the other hand, is, ‘as usual, clear and intelligible, and, as his species are

1 This network seems to have been first noticed by Tilesius. See Eysenhardt, I. c., p. 414.

100 THE OCEANIC HYDROZOA.
founded upon personal observation of the animals in their living state, they might very
advisably (if their distinctive characters are really well founded) be taken as the starting
point by future observers. I therefore subjoin his definitions.

1]. PHYSALIA CARAVELLA.

The first species is Physalia caravella, especially distinguished, according to
Eschscholz, by the fact that the polypites arise froma common stem. There is a considerable
number of tentacles of equal size, and the polypites and tentacles are all crowded together at
one end of the hydrosoma, which is obtuse, in such a manner that the rest of the organ,
which is naked, and produced into a proboscidiform process, is almost twice as long as
the part provided with polypites.

When the hydrosoma is so placed that its naked end is anterior, the left side is
nearly straight; the right side, on the other hand, widens towards the middle, and behind
this is strongly concave. The series of polypites begins here, and continues to the
posterior very blunt end; at a corresponding point the crest is emarginate, and it terminates
at both ends of the hydrosoma, in specimens where this measures cight inches, at one inch from
the end. The colour is bright purplish red, with dark extremities, and blue lines in the
folds of the crest. The polypites are violet, with whitish points ; the larger tentacles are red
with dark purple acetabula; the smaller tentacles blue; the bundles of buds are reddish.

This species attains a length of eight inches, and a width of two and a half inches
and inhabits the Atlantic from the Azores to the Brazilian coast.

2. PHYSALIA PELAGICA.

The polypites arise singly, and one of the tentacula is often much larger than the others.
That end of the body which bears the polypites, &c., has no fleshy process, but is filled
to its point with air. The hinder end of the body is bent towards the right side, while
the appendages are attached to the left side on a median prominence. The naked end
of the hydrosoma is not separated by any marked excavation from the rest; and
is a third or a fourth part as long as the whole. The crest begins as a slight elevation
in the neighbourhood of the anterior naked end, gradually increases in height posteriorly,
and ceases suddenly, opposite the root of that posterior process which bears the appendages.

In young individuals, the hydrosoma is only pale blue. In the adult, both ends are
green, and the crest appears to be purple in its highest part. The tentacles are blue,
with dark acetabula. In the older individuals, besides the large tentacle, there is a second,
smaller one, distinguishable from the rest. The polypites are dark blue, with yellow points.
Between the posterior process and the middle part, of the hydrosoma, there is a naked
spot, where the series of appendages is interrupted. This species attains a length of two and
a half inches, and inhabits the Atlantic, especially about the Cape of Good Hope.

PHYSALIA. 101

3. PHYSALIA UTRICULUS.

A long, fleshy, proboscidiform process at that end of the hydrosoma, which is
provided with appendages, distinguishes this species at first sight from the two preceding ;
the polypites arising singly from the hydrosoma, and the inequality of the tentacles,
among which one is very much larger than the other, are characters which it has in
common with P. pelagica.

In the largest specimens, the naked anterior end of the hydrosoma elongates into a
narrow, conical process, which is almost longer than the other part, provided with appendages.
At its extreme end, a small, fleshy, conical process of half a line in length is observable,
and is to be regarded as a part of the hydrosoma containing no air. In like manner, the
posterior end, beset on one side with short polypites, is not filled with air, but appears
as a narrow, fleshy, and thence moveable process, six lines long, which is continued by
two crura into the sides of the hydrosoma. Behind the middle there is observable
on the lower surface, a great, round, bluish spot, which arises from a flat, fleshy part
connected with the other thicker, fleshy streak. The long and not high crest commences six
lines behind the anterior end, and does not extend to the root of the posterior process.
Physalia utriculus attains a length of three and a half inches. As to its colour, the crest and
the middle part of the pneumatophore are greenish, the two extremities blue. It inhabits the
tropical regions of the Pacific.

I must confess I feel anything but satisfied that these are really distinct species, but this
is a point which can only be settled by those who study these singular animals with modern
appliances and a knowledge of what has already been done.

To such observers I would suggest that the number of primary and secondary septa in
the crest, and the exact mode of grouping of the appendages are more likely to yield
good distinctions than any other characters.

PHYSALIA UTRICULUS ? (Zsch.) Pl. X; Pl. XII, fig. ¥.

The Plysalie which I examined appeared to me to belong to one species, whose
description I subjoin as I find it in my notes.

The hydrosoma tapers to a point at one end, whence it gradually enlarges, and
becomes subcylindrical about the middle; the other end is somewhat suddenly widened,
so that, viewed from above, the organ has, roughly speaking, a T shape. Along the middle
of the upper surface of the subcylindrical and pointed part of the hydrosoma, its
wall is raised up into a transversely plaited crest, which dies away on the enlarged end.
The rest of the upper surface and the greater part of the under surface of the pointed
and subcylindrical portions are quite smooth, and distended by the air contained in the
pneumatocyst. The under surface of the transversely enlarged end, on the other hand, is
very irregular, and gives attachment to‘the numerous appendages.

102 THE OCEANIC HYDROZOA.

The general colour of the hydrosoma is a pale, delicate green, passing gradually into
a dark, indigo blue on the under surface.

The ridge of the crest is tipped with lake, and the pointed end is stained deep bluish
green about the aperture of the pneumatocyst.

The bases of the large tentacles are deep blue ; the polypites deep blue at their bases,
and frequently bright yellow at their apices; the velvetty masses of reproductive organs
and buds on the under surface are light green.

This description applies to individuals whose hydrosoma is four inches long and
upwards. In young specimens, the form and colour are apt to vary much.

The structure of the hydrosoma is best made out in very young individuals, small
enough to be placed bodily under the microscope, as, in consequence of the extreme
contractility of the walls of this organ in the larger ones, it commonly collapses and contracts
into a formless mass the moment an incision is made into it.

In such young Physalia, of one fifth to three eighths of an inch in length (Pl. X, figs. 1, 2),
the wall of the hydrosoma, which presents a minute, closely shut aperture or pore
at one extremity, is very obviously composed of two distinct membranes,! the outer, or
ectoderm, being denser and more transparent, while the inner, or endoderm, is softer, more
opaque, and ciliated on its inner free surface.

In the region of the pore (fig. 3), the endoderm is reflected on to the outer surface of an
ovate sac, which occupies only a comparatively small portion of the cavity of the hydrosoma.
The walls of this pneumatocyst are dense, thick, and elastic, and its inner surface is covered
with a thin layer of granular matter. The end of the pneumatocyst on to which the
endoderm is reflected, is in close contact with the ectoderm, and when viewed from within
presents a depression, surrounded by circular fibres, and towards which short, radiating folds
of the membrane converge, which corresponds with the pore visible upon the outer surface,
and is the inner end of the pneumatic aperture. By pressure, the air contained in the
pneumatocyst could be made to pass out throngh this aperture.

In adult specimens, the outer crescentic aperture of the pneumatocyst is visible on
the upper surface of the pointed apex of the hydrosoma, about half an inch from the
end.

The ectoderm is composed of parallel, elongated, caellzform masses, arranged perpendi-
cularly to the plane of its surface; the endoderm of an outer layer of delicate (muscular ?)
fibres about ;4th of an inch in diameter, invested by a granular, internal substance, which is
richly ciliated.

The appendages are large tentacles, small tentacles, polypites, gonoblastidia, and repro-
ductive organs.

The tentacular appendages arise from the outermost of the numerous protuberances
developed from the under surface of the hydrosoma, or what is in reality the representative
of the stem-like ccenosarc of other Physophoride.

The larger tentacles of large specimens sometimes attain a length of many feet, and are

1 [ do not find this structure distinctly alluded to by my predecessors. They often speak of an
inner membrane, indeed, but always mean thereby the wall of the pneumatocyst. M. De Quatrefages

refers to it in other parts of the organism, as the mucous layer.

PHYSALIA. 103

remarkable for their deep blue colour, which is not merely confined to the reniform elevations,
containing thread-cclls, with which they are beset, but extends in a less degree on to
the stem of the organ.

Each tentacle consists of two very distinct parts—the short, broad, basal sac (c*), and
the filiform tentacle proper (c’) (fig. 11).

The former is a sort of conical bag, attached by a broad base, and terminating in a free,
pointed apex, which is nothing more than a cecal process of the hydrosoma, containing a
diverticulum of the somatic cavity. Its inner surface is ciliated. The outer presents many
thread-cells, whose number increases towards its base.

On one side, the edge of the sac is rounded and thick; on the other, it thins and
expands into a broad, muscular membrane, which is continued down into the muscular band
of the tentacle.

The tentacle proper has a common base with the sac, but is otherwise separate from it.
It is essentially a czecal tube, whose canal communicates with the cavity of the sac and with
that of the hydrosoma, by means of the common base. On the side turned away from the
basa] sac, the ectoderm of the tube is much thicker than on the opposite side, where it
expands into a thin muscular band, continuous above with the muscular expansion of
the basal sac. The endoderm of the tentacle is ciliated.

The proximal end of the tentacle is gathered up into folds, which are attached to
the edge of the tentacular sac by the muscular membranous expansion, just as the
folds of an intestine are held by a mesentery. The rest of the tentacle hangs down
straight.

At the attached end of the convoluted part of the tentacle, the thickened wall was quite
even and smooth, but a little lower down, the wall had become sacculated (fig. 11 a),
each sacculation involving both the ectoderm and the endoderm, and containing a short, broad
cecum of the tentacular cavity. At first, the sacculations involve only a small segment of the
tentacular wall, but by degrees they extend transversely until, at length, they leave between
their extremities only sufficient space for the longitudinal band of muscular fibres, into which
the ectoderm of the opposite wall of the tentacle is metamorphosed, and for the tentacular
canal, which, within the cavity of each fold, expands into a cecum, corresponding with the
fold in shape. Throughout the convoluted part of the tentacle the folds are very close
together; but, in the beginning of the straight portion, they separate from one another,
and throughout the rest of its length they are, in the uncontracted state of the organ,
a long way apart.

When fully formed and extended, consequently, the tentacle looks like a delicate, narrow
ribbon, having transverse, reniform thickenings at regular intervals (fig. 12). The substance
of each thickening has a dark blue colour, and imbedded within it are myriads of close-set,
colourless, spherical thread-cells, varying from ;ith of an inch in diameter, to a sixth of
that size. Under a low power a number of fine filaments, the extruded threads of many of
these cells, might be seen radiating from the surface of each reniform thickening.

The small tentacles arise indiscriminately among the appendages of the rest of the
hydrosoma ; they precisely resemble the larger ones in structure, but they are, in addition
to their small size, distinguished by their pale colour, the blue coloration being either absent
or entirely confined to the reniform bodies.

104. THE OCEANIC HYDROZOA.

The polypites (fig. 4) attain a length of three quarters of an inch or more. Like the
tentacles they exhibit great irritability and contractility, and are in constant motion, at
the same time assuming every variety of form from that of a cylinder to that of a pear,
or that of a wide-mouthed and narrow-necked flask. The latter is the shape most commonly
assumed.

The lip is covered with many thread-cells, and is usually colourless or yellowish white ;
the dilated body of the organ is blue, and appears, even to the naked eye, to be dotted
over with numerous dark spots (4'). The ectoderm of the polypite is thin and dense. The
endoderm, thicker, occasionally exhibits an appearance as if it were longitudinally fibrillated,
and its inner surface is covered with long and beautiful cilia, which are especially obvious
about the oral aperture. The ectoderm and endoderm are in close contact about the mouth
of the polypite, but elsewhere a narrow, clear space exists between them; and in the
enlarged, globular portion of the polypite this space was increased at intervals by the
endoderm becoming folded (fig. 6). The folds were not at regular intervals, nor, so far
as I could observe, did they ever extend completely round the circumference of the sac.
They were certainly not produced by any mere contraction of the endoderm, inasmuch
as the state of extension of the polypite had no influence whatever upon their existence.
The space left between the endoderm and ectoderm appeared (as might be expected)
greenish, and as if filled by a liquid, but no trace of globules or granules could be detected
in it.

The dark spots which have been referred to above, arose from the presence of delicat«
villous processes of the endoderm (figs. 5—7). These villi were found wherever the folds
existed; and, whenever I succeeded in tracing a villus completely to its attachment, I found
its base to pass into one of these folds. The villi were especially numerous towards the
proximal end of the polypite. They were cylindrical, or slightly conical, but became almost
dendritic when compressed, in consequence of the thrusting asunder of the large globular
masses, of which their free ends appeared to be composed. These masses were but
indistinctly visible towards the base of the villus, but, towards its apex, they projected,
so as to give the extremity of the organ a slightly lobed appearance, and were filled
with densely packed, coloured granules.

The surface of the villi was richly ciliated, and thread-cells were imbedded here and
there in their substance. The axis of the villus was traversed by a clear canal, continuous at
its basal end with the interspace between the endoderm and ectoderm. The apical
termination of this canal was obscured by the pigment of the villi. It would seem as
if the endoderm had a distinct basement membrane, for, in two instances, where the soft
investment of the villi had been torn asunder and compressed, they remained connected with
the endoderm of the polypite by a delicate membranous tube of the same diameter as
the central canal.

The greater part of the surface of the ccenosare is covered by a bluish-green, granular
mass, mixed with small polypites and numerous tentacles of the smaller sort (figs. 13, 13a, 134).
On microscopic examination, it is seen that the granular mass is formed by an infinity
of small appendages, attached to the extremity of ramified processes of the hydrosoma, whose
internal cavities are, of course, continuous with that of the hydrosoma.

Most of these appendages are imperfect, and are to be found in every stage of

PHYSALIA. 105

development. They are—l, polypites; 2, tentacles; 3, hydrocysts; 4, androphores ;
5, medusiform bodies, probably gynophores.'

1. In their youngest condition (fig. 9) the polypites are pyriform processes of the
ectoderm and endoderm (the latter ciliated internally) about ,,th of an inch in diameter.
From these rudiments a regular series of gradations could be traced up to elongated sacs
about one fortieth of an inch long, and ,ith of an inch wide, with rather pointed apices,
about which more especially, thread-cells were accumulated (fig. 10). The endoderm, ciliated
on its inner surface, sometimes presented an appearance of longitudinal fibrillation, and
a distinct interval, or, at any rate, a clear space, was visible between the endoderm and
ectoderm. Towards the proximal end, the transverse folds and villi of the latter had begun
to make their appearance. The villi are at first papillary, obtuse eminences, without any trace
of an internal cavity. As they become larger, clear vacuolar spaces appear in their interior,
so that they resemble the villi of the Calycophoride, and dark pigment granules are deposited
in their apices. In polypites of larger dimensions, the folds and villi were fully developed,
but their distal extremities were still closed. Eventually, these open and the polypites are
complete.

2. The rudiments of the tentacles (fig. 11 4) are, in the youngest state, hardly distin-
guishable from those of the polypites. The first formed part is the sac, and from the side of
the base of this, the tentacle buds out as a cylindrical process, one of whose walls eventually
thickens and becomes sacculated in the way described above.

Many of the protuberances of the hydrosoma are covered with nothing but these
polypites and tentacles, but others, distinguishable even by the naked eye by their rounded,
velvetty appearance, exhibit only one or two small polypites arising from the midst of a
mass of branched, tubular processes of the hydrosoma (gonoblastidia), to which the three
remaining kinds of appendages are attached.

3. The hydrocysts (fig. 13) I find described in my notes as “young stomachs,” or
polypites, which, indeed, they resembled in every respect, save being open at the extremity.
They were usually placed at the extremities of the branches of the gonoblastidia.

4. The gynophores (fig. 15) seemed also to be very generally situated towards the end of
a branch, though this appearance may possibly be due to their long pedicles. They first
appeared as rounded papilla, about ,:th of an inch in length, and having the endoderm
at their apex much thickened, so that it projects into the cavity of the papilla (fig. 15 @).

The series of changes from this condition to that of the fully formed medusiform bud
(fig. 15 e), is quite similar to that passed through by the nectocalyces of other Physophoride,
and, therefore, need not be detailed. In the largest of these (fig. 15 e) the ectoderm of the calyx
was thick and yellowish. Four wide, straight nectocalycine canals opened into a circular
canal, which surrounded the prominent mouth of the organ. Broad transverse fibres were
distinguishable in the thick inner wall. The peduncle of attachment, thick and subcylindrical,
contained a canal which widened at the base of the medusiform body into a cavity whence
the four longitudinal canals arose.

1 Called “cyathiform bodies” in my memoir, communicated to the Linnean Society (supra)
in 1847. At that time I had not discovered the androphores, which, however, were briefly described

and figured in my paper in ‘ Miiller’s Archiv’ for 1851, above referred to.

14

106 THE OCEANIC HYDROZOA.

These bodies attain a diameter of one sixteenth of an inch or more.

5. The androphores (fig. 14) commence as papillary elevations of the lateral walls of the
branches of the gonoblastidia, whose endoderm rapidly thickens, and consists of pale, circular
granules ;;th of an inch in diameter. A small, ceecal prolongation of the central cavity next
extends up one (or probably both) sides of the organ (fig. 14 5).

When these rudimentary androphores have attained a length of about ;;th of an inch
(fig. 14 ¢), they are more completely pedunculated. The canal of the peduncle sends a short,
broad process, in the middle line, into the thickened apical mass, which now consists of two
portions—an inner membranous wall, resembling in all respects, and, apparently, continuous
with, the endoderm of the peduncle, and an outer thick mass lying between this and the
ectoderm, which consists of an aggregation of pale, spherical granules ={;th of an inch in
diameter. Two narrow, opposite canals pass from the upper end of the cavity of the peduncle,
and run close to its outer wall to terminate in blind ends.

Finally, the organ increases in size, and its contents assume more and more the form of
spermatozoa (fig. 14 d).

I can produce no direct evidence of the fact, but I entertain little doubt that the gyno-
phores are detached as free swimming medusiform zoids, as in Ve/el/a, and only develope their
ova subsequently. The androphores, on the other hand, must dehisce while still attached.
It might seem at first as if this arrangement were not very well calculated to ensure the
impregnation of the ova, but when we consider in what enormous troops the Physalie
are ordinarily found, the difficulty disappears.

All my Pfysaliz were taken either in the South Atlantic or in the Indian Ocean.

VELELLA. 107

Fam. VELELLIDZ.

Genus VELELLA (Lamarck).

The flattened pneumatocyst, which has more or less the form of a parallelogram, is
provided with a diagonal vertical crest. The tentacles are simple.

Under the head of Holothuria spirans, Forskal (‘ Descriptiones, p. 104, and ‘Icones,’
tab. xxvi, fig. /, and /' to 2°) gave the first, and, at the same time, the best, concise scientific
account of a Ve/ella extant. He was acquainted with the concentric chambers of the
pneumatophore, and with the fact that air is contained in them (“Nucleus annulis ovatis
concentricis.” . . . “In ejusdem nuclei annulis bullz aeris residui se manifestant vario
motu.” . . . “Haud raro fundum vasculi petiverunt emisso prius aere”); being, in these
respects, far in advance of some of his successors. The descriptions of Bosc (‘ Histoire
Naturelle des Vers,’ 1802); of Leach (‘Tuckey’s Voyage,’ 1818, p. 419); of Chamisso
and Eysenhardt (‘De Animalibus quibusdum, Nova Acta, 1821); of Quoy and Gaimard
(‘ Voy. de Uranie, Zoologie,’ 1824, p. 586); of Eschscholz (‘System,’ 1829); and of Lesson
(‘Voyage de la Coquille, Zoologie,’ 1830, t. i, part 2, p. 48); contain no additions of
importance to the facts which Forskal had made known.

De Blainville (‘Actinologie, 1834, p. 306) threw out the suggestion that the “ Rataria”
of Eschscholz would turn out to be the young of Veled/a, a surmise which has proved correct.
Delle Chiaje (‘ Descrizione e Notomia,’ tomo 4°, 1841, pp. 105-8) figures and describes at
some length the vascular system. He points out the communication of the chambers of
the pneumatocyst with one another, and notices the gonocalyces, which he suspects to
be the genital organs.

M. Hollard (Sur l’Organisation des Velelles,” ‘Annales des Sciences Naturelles,’ 1845)
endeavoured to prove that the small polypites or gonoblastidia of Vedel/a are “ aquiferous
tubes,” and not digestive organs. On the other hand, he is the first to have pointed out
very distinctly the nature of the liver (though Delle Chiaje called it “fegato”), and,
apparently unacquainted with Delle Chiaje’s work, he draws particular attention to the
gonocalyces, which he considers to be ovaria. The thread-cells which they contain he
takes for embryonic Ve/el/z, while the yellow corpuscles are ova less advanced. P

Krohn (“ Notiz iiber die anwesenheit eigenthiimlicher Luftkaniile bei Ve/ed/a und Porpita,”
Wiegmann’s ‘Archiy.,’ 1848) confirmed Delle Chiaje’s statements as to the communication
between the chambers of the pneumatocyst, and at the same time communicated the
important discovery of the existence of tubular prolongations of the under wall of the
pneumatocyst containing air. Krohn adds a fuller account of the liver than had been given
by either Delle Chiaje or Hollard.

108 THE OCEANIC HYDROZOA.

Forskal pointed out the peculiarity in the direction of the crest, in his species of Velella:
“Ab uno ejus margine ad alterum oblique transiens ita ut quamcunque extremitatem advertas,
pone ad dextram ante ad sinistram spectet ;” and Chamisso and Eysenhardt, supposing that
direction to be constant, made it the basis of their subsections of the genus, distinguishing
right-handed and left-handed Velelle.

“Falli autem videtur, si unicam Velellarum speciem existere, contendit, tres enim forme
facile distinguuntur, si ad directionem vele respexeris et longitudinem corporis cum ejus
latitudine comparaveris. Vela semper obliqua est, quod auctores omnes annotavere; sed
duplici modo. Cum enim laterum corporis longiorum alteruam dextrum, alterum sinistrum et
breviorum alterum anticum, alterum posticum nominaveris, tunc vela vel ab antica parte
ad posticam simulque a sinistra ad dextrum tendit (Vela sinistra, Velella sinistra), vel ab
antica parte ad posticum simulque a dextra ad sinistram cursum dirigit (Vela dextra, Velella
oblonga et lata). Directio hec cum ratione diametri sequentes constituit formas.”

Chamisso then goes on to describe his three species, but a statement of Eschscholz, who
was his companion in Kotzebue’s voyage, renders it difficult to understand what value is to be
attached to them.

“During the voyage of the Rurick with Captain Kotzebue, we observed that the direction
of the shell in the membrane which surrounds it is not alwaysthesame. In some, the nearer end
of the shell was to the right hand, in others, to the left. Chamisso described these Velella,
in 1823 [s7c], in the tenth volume of the ‘ Nova Acta Akad. Leop.-Carol.,’ t. xxxii, and drew
attention to the variation in the position of the shell; but unfortunately both descriptions and
figures reverse everything (geben alles verkehrt an). But, inasmuch as I drew and described
the animals at the same time, I can rearrange them. No great weight, however, can be placed
upon the position of the shell, for having taken three specimens out of a fleet of Velel/e in the
Atlantic, I observed that in two the shell lay from right to left, while, in the other, it lay from
left to right.””

I cannot understand this statement of Eschscholz, for it will be observed that Chamisso
does not speak of the position of the shell, but of that of the crest; and to complete the
confusion, on comparing Eschscholz’s figures with this description I find that they also are
all reversed. Thus it is said to be the character of the first division that “ If one of the longer
sides of the animal is turned towards the observer the shell lies from the anterior angle of the
left side to the posterior angle of the right side.” On turning to the figure of V. septentrionalis
(fig. 1, t. xv), I find that in placing it in the position indicated the shell lies from the anterior
angle of the right side to the posterior angle of the left side.

In addition, Eschscholz establishes the genus Rataria, but I have no doubt whatsoever
that, as De Blainville supposed, these are nothing but young Ve/el/z, such as I have figured
in Pl. XI, fig. 9. Eschscholz describes, altogether, ten species, but I doubt the sufficiency
of his definitions.

My own observations on this genus are but fragmentary, relating chiefly to the develop-
ment of the gonocalyces, which eventually become free medusiform zoids ; the distribution of
the vascular canals in the soft parts; and the structure of the young forms. In the following

1 «System der Akalephen,’ p. 169.

3 «De Animalibus quibusdam,’ Nova Acta, tom. x, pars secunda, 1821.

VELELLA. 109

pages I shall therefore borrow freely from the excellent memoirs of Kolliker and Vogt,
contained in the works which I have so often had occasion to quote.

The body of Velella (fig. 1) presents a firm, horizontal, rhomboidal disc, convex upwards
or proximally, in the middle and fringed by a delicate membranous limb. The crest passes
diagonally from one angle of the rhomboid to the other, and, like the horizontal disc, presents
a firm central portion, and a soft marginal limb. It is triangular in shape (at any rate in most
species), the apex of the triangle being immediately over the centre of the horizontal disc.

In its natural position the Vé/e//a floats on the surface of the sea, with its crest vertical
and exposed to the air,’ so as to act as a veritable sail; and, consequently, very little is to be
seen of its appendages, which are all situated on the under surface of the horizontal disc ; and
indeed of its firm part, as none are attached to the membranous limb.

The large central polypite presents a projecting, conical, often plaited, oral extremity,
which is susceptible of considerable dilatation. Its base is more particularly prominent in
the line of the crest, so that the polypite is not altogether symmetrical. On its base, and on
the under surface of all firm portions of the horizontal disc, are scattered innumerable minute
polypites, which perform the office of gonoblastidia, and slender simple tentacles, with slightly
enlarged extremities, spring in a single series from the line of junction between the firm part
and the membranous limb.

The difference between the “firm part” and the membranous “limb” in the horizontal
and vertical portions of the body of the Vélel/a, is caused by the presence in the former of a
hard resisting body, which is commonly known as the shell of the Ve/el/a, and consists of a
horizontal and a vertical plate, which are perfectly continuous with one another. The notion
entertained by Lesson that this shell, the pneumatocyst, is composed of several pieces, has arisen,
as Kdlliker has well pointed out, from a mistaken view of the nature of certain markings on
its surface. Thus there isa linear groove-like depression (fig. 1 1, w), which traverses the upper
face of the horizontal portion of the pneumatocyst, nearly at right angles to the crest, and
rises upon this in the middle line until it reaches its apex. A slight ridge on the under sur-
face of the pneumatocyst answers to this groove, while a longitudinal depression, increasing in
depth from the margins to the centre, corresponds with the attachment of the crest.

The horizontal division of the pneumatocyst (accidently inverted in fig. 2) consists of two
thin lamina, passing into one another at their free edges, and united by a number of con-
centric? vertical septa, between which are corresponding chambers filled with air. All these
chambers communicate together by means of apertures in the septa. Of these each septum
presents two, placed at opposite points of its circumference, and all nearly in the middle line
of the pneumatocyst. Kélliker made the interesting discovery that many of the chambers
have an additional opening, by which they communicate directly with the exterior. These
apertures are situated in the proximal or upper wall of the chambers, along a line about
midway between that of the openings just described and that of the vertical plate of the
pneumatocyst. Of the thirteen apertures observed by KOlliker, six ‘lay on one side of the

’ At least, I always observed it in this position. Delle Chiaje and KOlliker, on the other hand, _
seem to have seen it with the crest turned downwards.
* Vogt and others have affirmed the horizontal portion of the pneumatocyst to be formed of a

spirally coiled tube; but this is an error.

110 THE OCEANIC HYDROZOA.

vertical plate and seven on the other; one aperture lies in the wall of the central chamber,
the other six at tolerably even intervals between this and the margin. Consequently, as
there are more than six concentric chambers, some of them must communicate with these
stigmata, only indirectly.

The vertical plate of the pneumatocyst is a thin solid plate, wholly without chambers.
Its concentric markings are due to mere ridges on its surface.

A character which distinguishes the pneumatocysts of Ve/ella and Porpita from those of
other Physophoride is the attachment to their distal or under surfaces of the long filamentous
processes containing air, discovered by Krohn, whose cavity is subdivided at intervals by im-
perfect partitions, whence they appear jointed. In Ve/el/a there are, according to Kolliker, ten
or fifteen of these ‘‘ pneumatic filaments” attached to the five or six innermost chambers ; and
this observer agrees with their discoverer in asserting that they traverse the liver without
branching, and end cecally in the walls of the central polypite.

Vogt, on the other hand, states that these filaments are inserted into the centre of the
disc (‘“‘ par quatre canaux dispos¢s en croix’’), and ramify in the liver. Leuckart (‘Z. N. K.,’
p- 115) is uncertain as to the mode of origin of the canals, but confirms their rami-
fication.

The substance of the pneumatocyst is structureless, and consists of a nitrogenous sub-
stance, similar to, if not identical with, chitin. (See Leuckart, ‘Z. N. Ke; pe tla)

This singularly modified pneumatocyst is inclosed on all sides by the soft parts
of the body or hydrosoma which extend beyond its free edges, and constitute the limb to
which I have already referred:

If the crest is cut off and examined under the microscope (fig. 3), the soft parts over the
vertical plate of the pneumatocyst will be found to be traversed on each side by a series of
canals, about j;th of an inch in diameter, which radiate from the central third of the base
of the crest towards its free margin. The innermost of these canals are nearly straight; the
outermost are more or less curved, in correspondence with the free margin of the crest.
From these canals branches are given off, some of which terminate in czeca, while others
anastomose together, so that an irregular network of vessels spreads over the whole surface
of the crest. Near the termination of the vertical plate of the pneumatocyst, the canals,
becoming wider and less marked, take on a decided blue colour, and pass into a canal, which
runs along the line of junction of the vertical plate of the pneumatocyst with the limb (p’).
From the outer side of this canal, again, branches are given off into the limb, which run
obliquely towards its margin, and in their course give off wide cecal processes on each

. side, which but rarely anastomose.' Arrived near the edge of the limb, the ends of all the
branches are connected by a blue marginal canal (p”), which gives off numerous short ceca
on its outer side. The walls of all these canals are ciliated.”

' Vogt (p. 19) says, in describing these canals, “Ils s’anastomosent nulle part entre eux ;” but
so far as the species I observed is concerned, this is certainly not the case.

* They were certainly so in my specimens. M., Vogt, however, states, “ Le dois remarquer que je
n’ai jamais pu observer aucun mouvement dans toutes ces ramificationes vasculaires, méme en observant
des individus pleins de vie qui tout en étant placés librement sous le microscope, nagaient dans eau.”

p.21.) Kolliker, on the other hand, finds a ciliated epithelium in all the larger canals. (p. 51.)

VELELLA. 111

The proximal wall of the hydrosoma, or that which covers the upper surface of the disc,
presents a system of similarly ciliated, wide, anastomosing canals, which communicate with a
longitudinal canal, situated, over the diagonal groove which marks the upper surface of the
horizontal portion of the pneumatocyst.!. On the limb of the disc these canals divide
dichotomously into straight sacculated branches, which end in a marginal canal.

All these ciliated sinuses are continuous with a series of similar sinuses in the distal
layer of the hydrosoma investing the under surface of the animal, which eventually commu-
nicate with the cavities of the great central polypite and of the small or gonoblastidial polypites.

The roof of the wide digestive cavity of the central polypite is, in fact, formed by a dark-
brown mass, which fills the inferior hollow of the pneumatocyst, and may probably be regarded
as a liver, while, according to Kdlliker, it is traversed in all directions by thin-walled canals or
sinuses, zith — jth of an inch in diameter, containing cells with brownish granular contents.
The canals commence in a series of cleft-like apertures, discovered by Krohn, which are readily
visible with the naked eye, in the roof of the cavity of the polypite, and from these, forming
many anastomoses, the sinuses pass towards the convex surface of the organ. Here they
communicate with a whitish network of superficial canals, containing cells, like those of the
liver, but colourless.

At the margins of the liver these whitish vessels become continuous with the system of
dorsal canals above described, and, according to Kdlliker, in the following way. Very
numerous and close-set offshoots of the whitish vessels pass from the margin of the liver on
all sides, into that portion of the soft substance of the body which covers the under surface
of the horizontal division of the pneumatocyst, and, frequently dividing and anastomosing,
pass beyond the points of attachment of the gonoblastidial polypites and tentacles to the margin
of the pneumatocyst. Here they divide into two sets. Some pass into the limb, and, keeping
near its under surface, subdivide and anastomose, becoming continually smaller, until they
nearly reach the free edge ; the others, on the other hand, bend upwards, round the edge of the
pneumatocyst, and enter those which ramify on its upper surface.

The sinus which I have described as running along the edge of the vertical plate of the
pneumatocyst appears to be what Killiker describes as the “ marginal vessels of the crest”
(Randgefisse der senkrechten platte), which he states arise from the ends of the liver,
and passing along the line of insertion of the membranous limb, meet and open into one
another.

Besides these ramifications the sinuses of the distal layer of the hydrosoma all commu-
nicate with the cavities of the small polypites, the canals of whose peduncles open into them.

The tentacles (fig. 3) of those Velelle which 1 examined, appeared to be nearly solid, the
endoderm being, as in many Sertulariad@, so extensively vacuolated, that the sides mect and
nearly obliterate the central cavity. Thread-cells’ abound more particularly towards their
slightly enlarged ends. Kélliker’s account of the structure of the tentacles, however, differs
widely from this, since he states that, in /. spirans, they possess an external longitudinally

1 According to Kdlliker this canal is continued on to the crest in the prolongation of the groove
upon the vertical plate noticed above.
® These are, according to Kdlliker, arranged in two rows in Velella spirans. In a species from the

Pacific Leuckart observed four rows.

112 THE OCEANIC HYDROZOA.

fibrous and an internal circularly fibrous muscular coat; and he says that a “ cavity occupies
their whole length,’ and opens at its base into a vessel.

Vogt (p. 22), in describing the same species, says, “ these tentacles are hollow cylinders,
completely closed at their extremities ..... One might regard them as very strong, thick,
and muscular tubes, whose interior is filled by a transparent liquid, which doubtless plays a
certain part in the process of elongation of these organs. The principal muscular fibres
which form this tube seem to be disposed in ogives, so as to present the summit of the
ogive towards the interior of the tube. Other bundies are distinguishable which run
circularly, while still others are disposed longitudinally. Some of these bundles traverse the
interior tube in such a way as to simulate a cellular structure, which does not appear to me
really to exist.’”’ From this description I am inclined to believe that the structure of the
tentacles in the Mediterranean species is in reality similar to what I have observed.

Vogt (p. 22) describes a peculiar glandular structure in the margin of the limb of the
horizontal disc of Veledla spirans in the following terms: “I have already frequently mentioned
the clear blue line which runs along the edge of the limb. On microscopic examination it is
seen that this is composed of a simple series of great glandular sacs arranged side by
side, whose circular apertures are all situated on the dorsal face of the limb, and which
secrete a transparent and viscous mucus. The structure of these glandular sacs is very
curious, for they are wholly made up of great cylindrical cells, disposed in a radiating manner,
so that the free and rounded extremities of the cells are turned towards the centre of the
sac, whilst the other extremity is fitted into its wall. On the edge of the latter the rounded
contour of the cells intended to replace the cylindrical cells (of which there are apparently
many series) is visible..... The inferior face of the limb is covered only by the ordinary
integument, and exhibits the bottom of those glandular sacs, which here have no aperture.”
KGlliker (p. 60) would seem to have observed these glands, and similar ones in Porpita, and
Leuckart (‘Z. N. K.,’ p. 115) confirms Vogt’s discovery in Vedella.

The smaller polypites, or gonoblastidia (fig. 4), as they might, perhaps, be more correctly
termed, are supported upon longish, slender, peduncles. Their apical ends are very variable
in shape, sometimes wide and trumpet-shaped, with everted lips, sometimes contracted and
plaited. The wall of the polypite’ is, as usual, composed of a distinct ectoderm and
endoderm; the former being provided with villous elevations on its inner surface.” The
outer surface of the ectoderm also presents many rounded elevations, each of which contains
a great number of spherical thread-cells.*

The wall of the peduncle has the same composition as that of the polypite itself.

} Kélliker (p. 53), however, states that he has seen no hepatic strie or glandular cavities in the
walls of the small polypes.

* Vogt has observed the two membranes of which the wall of these buds is composed, but
imagines that the endoderm is a structure peculiar to them (p. 27), whereas it is continuous with the
endoderm of the polypite. As I have stated above, these buds were discovered by Delle Chiaje, and
have since been observed by Hollard (who entirely mistook their nature), Leuckart, K6lliker, and all
the later investigators.

° Half digested Crustacea, &c., have been observed in these organs by Lesson, Vogt, and Kdlliker,

so that there is no doubt as to their function.

VELELLA. 113

Attached to its outer surface are numerous groups of little oval bodies (figs. 5, 6, 7), the
smallest of which are simple cecal diverticula of the ectoderm and endoderm of the peduncle.
These, as they increase in size, undergo all those changes which have been described as
characterising the development of a medusiform zéoid, so that in such bodies of less
than ,jth of an inch in length, one is able to distinguish a circular aperture in the free
truncated end of the calyx, and four wide, longitudinal canals in its walls. Along the
sides of these canals many peculiar, reddish-yellow vesicles, of about ,j,th of an inch
in diameter, containing two or three small granules, are discernible, and round thread-cells
appear in the outer wall of the calyx.’

On the 23d of August, 1850, while traversing the South Atlantic Ocean, I found in the
vessel in which a full-grown Vede//a that I was examining was contained, a number of free
bodies more or less like these medusiform buds (fig. 8). The smallest were not more than
noth of an inch in length, and were motionless. They very closely resembled the still
attached zéoids, but there was an indication of four lobes around the mouth of the calyx,
and a small ovate sac containing much dark pigment, depended from the summit of the
roof of the calyx into its cavity. From these every intermediate stage could be found to
little ‘“M/eduse,” one sixtieth of an inch in length, or thereabouts, which propelled
themselves by the vigorous contractions of their bell-shaped calyces (fig. 8 a).

The circular mouth of the deep nectosac is surrounded by a narrow membranous
valve, and externally to this by four rounded lobes, whose surface, like that of the calyx, is
here and there beset with thread-cells. Four longitudinal canals traverse the calyx close to
its inner surface, and end below in cecal extremities, lines of the brownish-yellow vesicles
marking their course. An oval sac of ;j;th of an inch in length depends from the summit of
the cavity. It is closed at its extremity, and near its base is coloured by much dark pigment.
No trace of their original attachment is visible in these free gonophores.

The substance of the above observations, accompanied by illustrative figures, was
published in Miller's ‘ Archiv.’® for 1851. They were, in the main, confirmed by
M. Vogt, in his ‘ Recherches,’ published in 1854, but in the meanwhile Professor Gegenbaur
had indicated the ultimate fate of the Vele//a-zéoids. This naturalist observed free-swimming
medusiform bodies perfectly similar to those just described ; and in the larger ones he found
two tentacles, with greatly enlarged ends, developed from opposite sides of the rim, at points
corresponding with the ends of two of the longitudinal canals. In these the yellow vesicles
formed eight series, and indications of the development of four more longitudinal canals, in the
intervals of the old ones, from the common cavity were discoverable.

In fact, Gegenbaur adduces reasons for believing that, eventually, sixteen such canals
are developed, and that the zooid becomes the “ Medusa,” which he has termed Chrysomitra
striata (Pl. XII, fig. 13).

The calyx of Chrysomitra is about three lines high, and has the same diameter at the
mouth. In the centre it is about a line thick, becoming gradually thinner towards the edge.

1 Vogt gives a view of one of the medusiform buds, like my figure, but he has missed the four
canals. He states, further, that “the yellow cells lie in the interior of the canals ;” but this, as
Gegenbaur has pointed out, is certainly incorrect.

2 «Ueber die sexual-organe der Diphyiden und Physophoriden.’
15

114 THE OCEANIC HYDROZOA.

The inner surface has a deep-yellow colour, and the conical manubrium or polypite presents
four reddish-brown lateral enlargements, which are the sexual organs.

From the common cavity at the base of the manubrium sixteen radial canals arise,
and pass to a circular vessel surrounding the edge of the calyx.

On the surface of the calyx there are sixteen radiating rows of thread-cells, one
corresponding with each longitudinal canal. Where the latter opens into the circular canal
there is a semicircular dilatation, containing externally a whitish body, formed of vesicles like
drops of oil, the larger of which are coloured brown, red, or violet. The valvular membrane
lies within these enlargements, and is seen with difficulty. There are neither lithocysts nor
ocelli.

Only one of Gegenbaur’s specimens possessed a tentacle, and that was single. It
arose by a broad base from one of the enlargements just described, and attained a length of
half a line, ending below in a knob with a depressed centre.

The yellow colour of the inner surface of the calyx is produced by yellow cells, which
form close networks between the longitudinal canals. Two of the specimens observed were
male, and one female, but the sexual products were in none fully developed.

Of the earliest stages of the development of Vele//a nothing is known, but a very
young form (younger than almost any hitherto observed’) is depicted in Pl. XI, fig. 9. The
animal was not one tenth of an inch long, and the horizontal disc of the adult was
represented by a bell-shaped, membranous expansion, continued above into a broad crest, half
as high as the whole depth of the animal. It was symmetrically disposed, and, its
superior edge, far from being pointed, was rather concave, and in the centre presented a
curious thickening.

The central polypite was already open at its distal extremity, and around its base
were a few short, czecal processes, the rudiments of the gonoblastidia or of the tentacles.
The margin of the disc was occupied by a single series of large, oval vesicles.

The somatic cavity was divided by a series of vertical septa, which passed con-
tinuously over the pneumatocyst into the crest, near whose free edge they terminated
abruptly. At their other ends, or near the margin of the disc, they also ended abruptly,
and between them other very short septa were interposed. The somatic cavity and its
continuation into the crest were thus broken up into a series of parallel canals, united at
their ends by two marginal canals (p,y) at right angles with one another, one in the
disc, the other in the crest. The canals were ciliated, and I could observe granules
passing from one into the other.

The pneumatocyst shone through the disc, and did not extend into the crest at
all. By carefully tearing away the fleshy substance with needles, I succeeded in isolating it
(fig. 10). It then appeared as an almost hemispherical body, convex above, and flat below.

1 Bschscholz’s Ratarie, rightly suspected by De Blainville to be young Veledle, are said to
have been a line long. Those observed by M. Vogt’s friend (Vogt, ].c., p.83) were five millimetres
(one fifth of an inch) long, and were evidently much further advanced. The smallest Velelle observed
by Killiker were three or four lines long, and differed in no essential respect from the adult except
that they were more “Rataria-like,” possessed no medusoid buds, and exhibited a smaller number of
“small polypes,” tentacles, and chambers in the pneumatocyst.

VELELLA. 115

On two of its sides, in a plane perpendicular to that of the crest, there was a double
crescentic mark, caused by a depression. The air did not completely distend the pneumatocyst,
but appeared to be divided into seven or eight lobes below, so that, at first sight, the organ
itself appeared to be lobed, but this was not really the case. It was, in fact, in the
smallest specimens a simple vesicle, about one twentieth of an inch in diameter, with strong
and thin walls, which, when it was burst and the air expelled, fell into sharp folds.

In the individual figured the commencement of the first chamber is indicated by a line,
concentric with the outline of the pneumatocyst, which corresponds with the base of
the first septum.

A Felella about twice the size of the foregoing (fig. 14) had a flatly campanulate
dise, with a nearly circular outline. The crest was oblong, raised to an obtuse point in the
middle, and somewhat narrowed at its attachment. The edges of the disc and crest had
a blue fringe, internal to which were short, radial lines, of a yellowish-red colour, arising
from an aggregation of corpuscles similar to those contained in the medusiform zéoids.
The rest of the animal was colourless.

The pyramidal pneumatocyst inclosed a considerable quantity of air, and presented
several concentric lines on its surface. The central polypite appeared somewhat four-
lobed at its base, and contained the half-digested remains of small crustaceans. A
single series of tentacles surrounded the polypite. Like those of the adult they were
slightly enlarged at their extremities, which were covered with thread-cells, and presented
a vacuolated axis marked by a dark line in its centre. They were attached in the re-entering
angle between the limb and the central polypite. Apparently proceeding from the outermost
margins of the latter were several small, ovate sacs, with two or more minute papille,
containing thread-cells at their apex. The ectoderm and the ciliated endoderm could be plainly
distinguished in the thin walls of the central polypite, and these small sacs, containing
a cavity whose walls were likewise ciliated, and had precisely the same structure as those of
the central polypite, were, I believe, simple processes of the latter.

The crest (fig. 12) was thin and membranous throughout, and divided by vertical septa
into a number of canals. The innermost of these passed vertically upwards, but the outer
were more or less curved in correspondence with the outline of the lateral margins of
the crest. At the margin of the crest it was easy to see that the outer wall of the marginal
canal was double, its ectoderm being lined by a distinct endoderm.

A small quantity of brown, hepatic matter had accumulated under the pneumatocyst in
this individual. The disc was thicker and denser than the crest, but I did not work out its
structure in this specimen.

In another of nearly the same size, ora little larger, the structure of the disc (fig. 11) was
essentially the same as in the first-described specimen, but a number of reddish granules had
accumulated in the marginal canal and in the peripheral extremities of the radial canals.

In a still larger individual the radial canals had become much narrower in proportion
to their length (fig. 13). At their central ends their walls were simple and straight,
and separated only by a thick, clear, transparent substance, but, externally, they were
sharply folded, or drawn out into short, lateral sacculi; and, from the ends of these,
ramified processes, as a sort of connective tissue, extended into the clear substance which
separates the canals. At their peripheral ends, the canals, instead of being simply divided

116 THE OCEANIC HYDROZOA.

by a short septum, separated into two widely sacculated branches, full of brown granules.
The canals were much nearer the inferior than the superior surface of the disc.

Velelle one third of an inch long had the disc flat and oblong, and the crest oblique.

Incomplete as these observations are, they seem to me to prove sufficiently that,
in its early stages, Vele/la departs in no respect from the typical structure of the Physo-
phoride, and that the membranous limbs of both the crest and disc must be regarded
as vertical and horizontal processes of that part of the hydrosoma which represents the pneuma-
‘tophore of other Physophoride, and whose cavity has become subdivided by vertical partitions.
The existence of these last, however, is nothing new, for if the pneumatophore of a Physophora
or of an Agalma were flattened into a disc or pulled out into a crest, the septa which retain
the pneumatocyst in its place would assume nearly the same position and relations as the
partitions between the canals in the young Velelle.

The hepatic mass does not at first exist, but appears to be developed in contiguity with
the under surface of the pneumatocyst, like the cellular processes attached to this organ
in Rhizophysa; and 1 imagine that the canals which eventually traverse it are mere sub-
divisions of the somatic cavity,’ produced by the encroachment of the lobes of this organ, and
their mutual anastomosis.

The pneumatic filaments remain as something peculiar to the Velellidan family.

The Velellz on which my observations were made were taken on various occasions in
the Pacific and Atlantic Oceans. Lesson’s ‘Acaléphes’ was the only work which I could
consult, and those who are acquainted with that singular though useful miscellany, will under-
stand why I soon gave up the attempt to identify the Velelle 1 met with, with any of the
sixteen species which are enumerated therein. I can only say that some were right-handed,
and some left-handed, and that Eschscholz’s Rafarie are obviously nothing more than
such young Velell@ as I have described above.

M. Vogt describes a great peculiarity in the mode of growth of the Vele/le. He affirms
that at a particular point of the circumference of the disc, the series of tentacles and of
gonoblastidia always appear to be interrupted, overlapping one another as it were, like the
ends of a brokenring. At this point, the young tentacles and the gonoblastidia are developed,
and an aggregation of their buds is consequently always found here. The buds of the
tentacles are, at first, ceeca placed on the circular vessel of the limb, and “ composed of
two layers—an epidermal, blue layer, and another more transparent layer, in which the
internal cavity of the bud is excavated.” M. Vogt has obviously seen the ectoderm and
endoderm. In the course of development the former becomes covered with thread-cells ;
the latter, increasing in thickness, ‘‘shows itself composed of muscular fibres.” That is,
as I believe, it becomes vacuolated, the intermediate substance between the vacuoles being
the so-called muscular fibres.

Kolliker says nothing about the existence of a special “centre of gemmation,” such
as that described by Vogt; and Leuckart denies its existence, affirming that the deve-
lopment of the new organs takes place among the old ones in all parts indifferently.
CZANT KG? p:elil8:)

1 The homology of these canals with the somatic cavity of other Hydrozoa is well pointed out by
Vogt, p. 35.

PORPITA. 117

Genus PORPITA.

Pneumatocyst circular, produced into radiating lamelle inferiorly, but unprovided with a
crest. Tentacula branched.’

Forskal, whose powers of observation must have been of a very high order, not only
left behind the very recognisable figures of Ve/el/a (which he calls Holothuria denudata), given
in the ‘Icones,’ tab. xxvi, L, but his description (‘ Descriptiones,’ p. 103) is admirable, and
contains everything that was known about the genus up to within a very few years.

“Diameter latitudine digiti. Paginz superioris discus vel nucleus, diameter dimidio
minor ipsius animalis; paulum elatus ut umbo clypei depressi; albidus, striis concentricis et
e centro radiantibus decussatus, rigidus, margine obscure ceruleus. Limbus disco dimidio
angustior, czeruleus, immaculatus, pellucidus, flexilis, planus; linea circumcirca ad oram striis
nigris centrum spectantibus composita. Pagine inferioris tentacula omnia, disco jam descripto,
nucleo scilicet, non limbo affixa. In medio apparet ventriculus limbi la itudine, basi globosus,
superne terminatus ore cylindrico aperto; quod demitti et dilatari potest patulum ejusdem
ambitus cum basi ventriculi; et conspectum dat intestinorum; que exserebantur gelatinosa,
albida, longa. Reliquee parti albz nuclei adnectuntur tentacula parvula, leevia, basi ovata,
adunata in stylum filiformem apice incrassatum; interiora tentacula alba; exteriora magis
magisque czerulescentia et pellucida, exeuntia ad limbi oram, non vero excedentia. Nuclei
margini ceruleo adnectuntur tentacula multorum ordinum, limbo longiora, inzqualia;
exteriora seu superiora plerumque breviora; quorum longissima diametrum corporis parum
superant; brevissima vix limbum excedunt; omnia filiformia, hyalino czrulescentia, linea media
obscuriore ; dimidia parte interlore nuda, dimidia exteriore triplici serie longitudinali ciliato
glandulosa, ad utrumque enim latus et subtus in medio exeunt pedicelli hyalino cérulescentes,
subverticillati, terminati glandula seu capitulo obscure ceruleo. Pedicelli atque capitula
versus apicem gradatim increscunt; series infera terminat tentaculum capitulo medio fere
sessili, inter glandulas laterales pedicellatas. Verticilli pedicellorum sape 5 vel 6 in quovis
tentaculo. Varie incurvat animal et extendit tentacula glandifera ; at nuda ista contrahit seepe
in cylindros crassiores, apice patulos. Limbus interdum margine leviter reflectitur. In
fundo vasculi ubi Holothurias servavi vivas per horam unam alteramve, vidi ova an excre-
menta? dejecta; sudcubica, hyalina, circello fusco in medio baseos, et linea fusca subsinuata
interiore, erecta ad singulos 4 angulos laterales. Oculo tantum armato apparuere.”

From the last passage it is clear that Forskal observed, without knowing it, the
detachment of the medusiform zéoids of Porpita.

Eschscholz (‘System,’ p. 176) calls the substance of the pneumatocyst of Porpita
“kalkartig”’ (calcareous), while that of Velella is termed “knorplig” (p. 168), without

1 At any rate, provided with more or less stalked knobs or capitula.

118 THE OCEANIC HYDROZOA.

any particular reason being assigned for the distinction; the fact being that both have a
cartilaginous consistence, while neither is calcareous. He describes the radiating perpen-
dicular lamelle of the distal surface, and says that the flatness or more rounded character
of the body depends on the depth of these.

The capitula of the tentacles are called suckers (saugnipfen), and are said to be
provided with longer or shorter pedicles. Does the following passage refer to the pneumatic
filaments or to the misunderstood peduncles of the gonoblastidia, or to both?

“ Between the suckers one sees a great number of peculiar organs, which are probably
receptacles for germs (keim-behilter). Mucous globules beset with white points are disposed
in alternating series along a fine median tube” (1. c., p. 176).

I find no addition of any importance to our knowledge of the organization of Porpita,
until the publication of Krohn’s observations upon the pneumatic filaments cited above.

Subsequently to the promulgation of his discoveries, KGlliker’s important investigations,
which were first published in 1853 (‘Bericht iiber einige im Herbste, 1852, angestellte
Untersuchungen, Siebold and Kélhiker’s ‘ Zeitschrift,’ 1853), have brought us much nearer to
an exhaustive knowledge of the organization of this animal. I partly translate and partly
condense his excellent memoir on Porpita, contained in ‘ Die Schwimmpolypen.’

“As in Velella, the pneumatocyst (knorpel-platte) of Porpifa is thickest at the edge,
thinnest in the middle, though, on the other; hand, each of the the air chambers, of which
the outermost are the widest, is, on account of the circular form of the organ, of even
diameter throughout. If they are filled with air, or in a fresh pneumatocyst (schale) still
containing its air, it is observable that the inner contour of each air chamber is a simple
circle, while the outer appears to be undulating; whence it follows that the septa have
tolerably regular, small projections on their inner surfaces. From these septa, also, the
elegant concentric striations, visible even with the naked eye, on the upper slightly convex
face of the pneumatocyst (knorpel-scheibe), arise, whilst the radial striation visible on both
faces, but especially on the under, results from linear elevations of the cartilaginous walls
which, upon the under side, are so pronounced that they may rightly be termed ridges or
lamellee (Leisten oder Blatter).

“The relations of the air chambers of the pneumatocyst to the external parts are much
more interesting in Porpita than in Velella. The pneumatic filaments (luftcanale) discovered
by Krohn, and already mentioned in describing Velel/a, are present, and in much greater
numbers, as are also the pneumatic foramina, which have not as yet been noticed by any one.
The pneumatic filaments arise from the inferior concave surface of the pneumatocyst,
in as many series as there are radiating lamellz, and in fact in such a manner that every air
chamber, with the exception of the outermost, gives off one or many pneumatic filaments.
Since the number of chambers in a middle-sized Porpita (four or five limes in diameter)
is twenty-two or twenty-three, and that of the lamelle of their concave surface (of which,
however, many are very short, and only visible on the outer parts of the shell) is eighty or
ninety, it results that the number of the pneumatic filaments is out of all proportion to
that of the same organs in Ve/el/a. In other respects they are similar, only that they
have thinner walls and are narrower (0:005—0-015”), and are not only provided with
internal projections, but also exhibit external prominences. After arising from the
pheumatocyst, they at once enter the liver, and pass in a wavy but tolerably direct course

PORPITA. 119

through it, to end in the walls of the large central polypite, and of all the little ones which are
disposed around it.

“ Before this, however, they form a plexus on the lower face of the liver, in the membrane
which carries the polypes, but, to all appearance, without divisions or anastomoses, of which
no definite evidence was seen in Porpita, except that here and there a few canals arose
from a common, very short stalk. In the central polypite these pneumatic filaments end,
partly on the attached wall, partly on the lateral walls, in which latter position a strong
magnifying glass even will show them as white wavy lines. In the smaller polypites
four to six, also undulating canals, pass in their walls (?z den Winden derselben) as far
as buds are attached, and then, so far as I could make out, end blindly.

“Though there may be no opening of the aeriferous apparatus of Porpita in this situation,
it is very evident elsewhere, if a proper method of preparation be pursued. If all the soft
parts be removed from the under surface of a Porpita, in fact, and if the convex surface be
then observed with a low magnifying power, there is observable in it, besides the rich
vascular network to be described immediately, whose trunks are arranged radially, many
oval openings disposed serially between the vascular trunks, the outermost of which are
Jargest, while the inner are smallest.’

“Tf the soft parts are now entirely removed, it becomes obvious that each of these
apertures corresponds with a hole surrounded by a low, depressed wall on the upper lamella
of the pneumatocyst, so that it leads directly into an air chamber. In a middle-sized Porpita
I can count forty-five series of pneumatic foramina, and in each series nine to thirteen
stigmata, so that, inasmuch as there are twenty-two to twenty-three air chambers, each series
of stigmata cannot communicate with all the chambers.

“The most stigmata are possessed by the outermost chamber (which has sometimes one
in every series) and the next to it. From thence their number rapidly diminishes, and the
innermost chamber of all, which has the form of a round cell, has only a single central—
sometimes smaller, sometimes larger—foramen, which lies, consequently, in the centre of the
pneumatocyst. The size of the outer apertures, is 0-‘04—0°05” ; of the inner smallest ones,
0:015” —0024”. . . . . . . The pneumatocyst of Porpita consists of the same
homogeneous cartilage-like tissue that forms the skeleton of Veled/a, and there is no ground for
the assumption of authors, from Eschscholz downwards, that it is calcareous” (1. c.,
pp. 57—59).

Professor Kolliker then goes on to show that the soft parts consist of a thin superior,
and a thick inferior layer, united in the broad marginal limb. The upper layer is thin,
and perforated by apertures which correspond with the pneumatic foramina. The lower
layer, much thicker, incloses the liver and the renal organ, and gives attachment to the
appendages. The upper surface of the lower layer exhibits radiating folds and grooves,
which correspond with the elevations and interspaces of the lamelle of the inferior surface of
the pneumatocyst. The brown hepatic cells are lodged in this lower layer, and extend
over nearly the whole area covered by the pneumatocyst.

The roof of the cavity of the great or central polypite exhibits, in the middle, eight
radially disposed, cleft-like depressions, each with a round opening at the bottom, and many

1 Tn fig. 8, pl. xii, however, the innermost foramina are the largest.

120 THE OCEANIC HYDROZOA.

circumferential apertures, all of which communicate with the reticulated canals which
traverse the hepatic mass, and then pass into the radiating, branched, and anastomosing
canals, which traverse the inferior layer of the soft parts. Having reached the edge of
the pneumatocyst, they divide into two sets of branches, some of which pass into the
limb, while others ascend and ramify in the dorsal layer of the soft parts. All the larger
canals were ciliated.

Klliker alludes to the existence of peculiar gland-like organs, similar to those described
by Vogt in Velella, in the limb of Porpita. h

The canals of the peduncles of the smaller polypites or gonoblastidia open into the
hepatic canals, and the buds resemble those of Ve/ed/a in their general structure.

The canals of the tentacles are ciliated, at any rate in young specimens, and they
communicate with the vessels which proceed from the hepatic canals.

Between the circumference of the central polypite and that of the liver there lies,
on the under surface of the body, a white lamella, perforated by many foramina, which gives
passage to the peduncles of the gonoblastidia and to the pneumatic filaments. It consists of
a reticulated tissue, sharply defined, and invested by a thin, fibrous membrane inferiorly,
while superiorly it runs into fine processes and trabeculae, which pass between the hepatic
canals.

The white colour of this plate arises from the presence of innumerable molecules, of
which the smallest look like spheroidal dots, and appear quite dark, while the larger
ones have the form of crystalline needles and rhombs, and are transparent, or are pale
and rounded, and have a more fatty and albuminous appearance. The chemical reactions
of the crystalline substance led Professor Kélliker to suppose it to be guanin, whence he
concludes that this organ has a renal function.

With regard to the species of Porpita, Eschscholz admits the following:

1. PorpiraA MEDITERRANEA.

The pneumatocyst is half as broad as the whole body, somewhat convex, and whitish on its
proximal or upper surface. The limb is dark blue, as inall the other species. The central poly-
pite and the inner gonoblastidia are whitish, the others becoming blue the nearer they are to the
edge. Tentacula bluish,and provided in their outer half with three series of tolerably long-stalked
capitula. The last capitulum of the lower series is sessile upon the point of the tentacle.

This is the Holothuria denudata of Forskal, and the Porpita glandifera of Lamarck. It is
found in the Mediterranean, and attains a diameter of eight lines.

.

2. PoRPITA RAMIFERA.

The pneumatocyst is convex at the sides, and flattened in the middle, and as convex
distally (or on its under surface) as proximally (or on its upper surface). The middle part is

PORPITA. 121

dark brown ; the lateral parts are of a clear blue, and clear rays mark the whole. The blue
limb is very narrow. The tentacles are blue, and not clavate, but thicker at their roots; and
they have on their outer side only, usually four, and sometimes five, very long-stalked
capitula.

This species, only half a line in diameter, was taken in the North Pacific. Is it an
adult P

3. PORPITA GLOBOSA.

The hydrosoma is almost globular. The flat upper part of the pneumatocyst is very
small, has only a line in diameter, a dark-blue colour, and about thirty-two radiating strie.
The broad limb is dark blue; the rest of the body, which consists of the proper globular
part, is supported by the distal lamellae of the pneumatocyst, which are greatly elongated and
expanded downwards and laterally. The pneumatocyst, in fact, extends even beyond the
limb, its sides here inclining downwards and outwards. This part is coloured bright blue,
and is marked with a great number of radiating brown lines. The tentacles spring from
the broadest part of the body ; they are filiform, and in their larger outer moiety are covered
with almost sessile capitula. From the line of their attachment the body contracts again,
presenting a convex contour.

In the only specimen which Eschscholz took he observed a tentacle which, very thick and
coloured in the middle, allowed a yellowish juice to be seen through its walls. The stomach
and the gonoblastidia were yellowish—the latter surrounded by a brownish villous ring, out
of which fine white filaments hung. This last Eschscholz regards as a germ-receptacle. This
species attains three lines in diameter, and was taken near the Cape de Verde Islands.

It would be very desirable to re-examine this remarkable species. The “ brownish villous
ring” and the “fine white filaments” seem clearly to be the gonophores and the pneumatic
filaments.

4. PoRPITA UMBELLA.

The proximal surface of the pneumatocyst is flat, colourless, or whitish ; the limb broad
and dark blue. The tentacles have short stalked capitula on their outer thicker half.
The central polypite and the inner gonoblastidia are white, the outer gonoblastidia are bluish.

Eschscholz considers this to be the Porpita gigantea of Peron and Lesueur. It was taken
in the equatorial and the temperate parts of the Atlantic, and attains an inch in diameter.

5. PoRPITA C@:RULEA.
The pneumatocyst is quite flat, and has a blackish-blue colour.’ Forty-two radiating
' For brevity’s sake I have used the term pneumatocyst in these definitions in a broad sense, so

as to include the layer of the hydrosoma which overlies it.
16

122 THE OCEANIC HYDROZOA.

striae were counted, and these had little tubercles (zacken) disposed along their length. The
blue limb is a line and a half broad.

With a strong magnifying power it may be observed that the outer edge of the limb
(equal to one fifth of the whole) is marked by fine excentric darker striae, which may be
regarded as muscular fibres, by whose aid the marginal limb may be bent upwards or
downwards. The rest of the limb appears dotted. The bluish tentacles gradually thicken
towards the ends, where they are trihedral, and their outer half is beset with three series of
almost stalked capitula. The tentacles vary in length, the largest surpassing the diameter
of the body, the shortest not projecting further beyond the limb than its breadth. In a large
and uninjured specimen there were seven series of tentacles of diferent lengths.

The gonoblastidia are whitish and bluish, as in the preceding species. The bodies,
which appear to be germ-receptacles, have the appearance of yellowish-brown (hellhornfar-
bigen) racemes. The largest specimens attain the diameter of an inch.

It chanced that I fell in with Porpita only once during the cruise of the “ Rattlesnake,”
and the observations I made are of no moment. But some very large Porpite, which, though
they have long been preserved in spirit, are in very good preservation, having recently come
under my notice, I am enabled to add some particulars of interest to what is already known.

The proximal or superior face of the disc in this species is quite flat, and measures,
including the limb, about one inch and a half in diameter, but its thickness does not exceed
one thirteenth of an inch. The limb, one eighth of an inch wide, is thin and flexible. The base
of the large central polypite measures not more than one quarter of an inch in diameter, so
that it occupies but a very small portion of the distal or inferior face of the disc. The small,
or gonoblastidial, polypites are exceedingly numerous, occupy an area half an inch wide
on each side, and, when distended with food, as many of them are, they acquire a transverse
diameter of one eighth, and a length of one fourth, of an inch. The space between the zone
occupied by these polypites, or the gonoblastidial zone, and the limb, is occupied by a narrow
area, rather less than one eighth of an inch wide, which presents a reticulated appearance,
and to which the numerous rows of tentacles, almost all of which were detached in my
specimens, were attached.

This tentacular zone, as I will call it, is somewhat convex downwards and outwards,
while the rest of the inferior surface of the disc is flat.

It is impossible to say what may have been the original coloration of this species.
At present, all my specimens have a uniform pale-yellowish hue, which inclines to brown
over the gonoblastidial area. The proximal surface of the disc is covered with numerous
small tubercles, set at irregular intervals, and remarkably large and prominent in the central
region, while they become smaller and less conspicuous peripherally.

The presence of these tubercles leads me to think that this species may be the
Porpita cerulea of Eschscholz, which is, perhaps, identical with the Porpita Atlantica of
Lesson (‘Acalephes,’ p. 590).’

With respect to the general character of the polypites, the gonoblastidia, and their buds,
I have nothing of importance to add to the account given by Kolliker, except that the roof of

' There is no means of ascertaining whence the specimens were obtained.

PORPITA. 123

the central polypite does not present that arrangement of eight radially disposed, cleft-like
depressions described and figured in P. Mediterranea. On the contrary, it exhibits a great
number (some forty) of septiform membranous bands, which radiate between the centre and
the periphery of the roof of the polypite, becoming deeper as they go, until, at the cireum-
ference, their height is equal to the thickness of the distal layer of the hydrosoma.

Similar but very much smaller bands, not exceeding seven or eight in number, appeared
to me to be developed in the roof of the gonoblastidial polypites, but it was not easy to assure
oneself of the fact.

The hydrosoma exhibits, as in Porpita Mediterranea, a thin proximal, and a thick distal,
layer, between which the pneumatocyst is inclosed. As the other parts are all disposed
in relation to the structure of the last-named organ, it will be convenient to consider its
structure first.

The proximal or upper surface exhibits a beautifully regular, radiating striation, each
stria terminating in a little projection at the periphery of the disc, which thus acquires
somewhat the appearance of a toothed wheel. In the circumferential half, or thereabouts,
of the surface of the disc, these striz are the expression of ridges, with intermediate valleys,
which attain a height and depth nearly equal to half the total thickness of the disc;
but, towards the centre, the inequalities gradually diminish, the valleys becoming filled
up and the ridges covered over, until, in the central half of the proximal surface of the disc,
the general surface is even, and would be quite flat and smooth, if it were not for the
numerous conical tubercles scattered over it.

These tubercles attain a diameter of one sixtieth of an inch at their base, and one
fortieth of an inch in height, and they are disposed without reference to the direction
of the radiating ridges. The apparent radiating striation of the central region of the pneuma-
tocyst is due only to the shining through of the deeper structures.

A radial section of the pneumatocyst (woodcut 4, p. 126) shows that its proximal or
upper wall (/), which is not more than ,jth of an inch thick at the periphery, gradually
increases in thickness towards the centre, until, in the middle, it attains a diameter of one
twenty-eighth of an inch, or, in other words, has twelve times the thickness it possesses at
the circumference.

On the other hand, as it passes into the distal or inferior wall, the substance of the
pneumatocyst becomes thinner and thinner, until, over the whole distal surface, it has a
thickness of less than ith of an inch. Where thin, it appears structureless; but where
thick, it is obviously composed of a great number of superimposed lamelle.

I conclude from this structure that the pneumatocyst has originally very thin walls,
to which very little, if any, addition is made on the distal surface ; while, as the animal grows
and becomes older, new layers are constantly added to the whole of the proximal wall of
the pneumatocyst.

The concentric chambers of the pneumatocyst are more than fifty in number, and
the septa which divide them from one another are thicker where they join the proximal
layer of the pneumatocyst.

The central chamber is octagonal, and has a diameter of one thirty-sixth of an inch.
Viewed with a low power, it seems to be divided into eight compartments by as many
ribs, which unite in a circular disc in its centre, and radiate thence to its angles. In the

124 THE OCEANIC HYDROZOA.

centre of this disc there appears to be a small circular foramen, ;i,th of an inch in diameter,
and each compartment appears to have a similar but rather smaller foramen terminating
a conical prolongation of the compartment. These foramina look so clear and plain that
it is difficult to persuade oneself they do not open on the surface. But they do not.
They are, in fact, covered over by a considerable thickness of the laminated pneumatocystic
substance, which is wholly imperforate.

Each of the eight radiating ribs probably indicates the position of one of the valleys
of the pneumatocyst when it was not larger than the central chamber. Filled up in
the adult state by the additions which have been made to the proximal wall, each valley
would become more opaque than the intermediate ridge, which must necessarily have
less deposit over it, and hence must look like a more solid rib.

Of the fifty or sixty concentric chambers which surround the central one, those midway
between the centre and the circumference are, on the whole, larger than the central or
the peripheral ones.

The pneumatic foramina (/*) have an average diameter of about jth of an inch. Each
is situated at the extremity of an urn-shaped process of the pneumatocyst, which has very
delicate walls, and is always seated upon one of the ridges—never in a valley. The urn-shaped
process (woodcut B, p. 126) is about ;4th of an inch high, and has nearly the same width in
the middle, but it narrows to its peduncle below and to its mouth above.’ The mouth is
surrounded by an everted and deflected lip, about .4,th of an inch wide, whose margins are
slightly denticulated.

Every ridge exhibited a pneumatic foramen for each of the ten or twelve outer chambers
of the pneumatocyst, except im the case of those ridges which were too short to extend
over so many chambers; and all their foramina were, so far as I could observe, open.
Beyond this point, however, the foramina occurred far less frequently, and while they were
continued nearly to the centre along some radii, stopped far short of it along others, so that
in no case have I found more than twenty-four foramina in one radial line; and more than
this, so far as I have been able to ascertain, all the foramina on the central side of the
twelfth or thirteenth are closed by the extension of new layers of the substance of the
pneumatocyst over them.

The thin distal wall of the pneumatocyst is somewhat arched and concave under
the central chamber, but is otherwise smooth. Eight groups of pneumatic filaments (of
which there are four or five in each group) arise from this. wall opposite the pneumatic
foramina in the proximal wall.

The pneumatic filaments are simply tubular prolongations of the thin distal wall,
which have an average diameter of ;{;th of an inch, and exhibit annular constrictions at
intervals. They arise close together, but can hardly be said to have a common stem.

In the succeeding chambers the distal wall is produced into lamellar folds, which radiate
like the ridges of the ‘proximal wall, though they are two or three times more numerous, and
become deeper towards the periphery, where they end in margins which are convex
downwards and outwards. At their deepest they make up about half the total vertical
diameter of the pneumatocyst (woodcut C, p. 126).

The concentric septa, which divide the chambers (/'°) of the pneumatocyst from one
another, present, in a vertical radial section, a contour which is convex peripherally, and slopes

PORPITA. 125

away more towards the centre distally than proximally. They send prolongations down into the
lamellar folds or diverticula, which are directed still more obliquely, so that each diverticulum
is divided into a series of primary compartments (/"), whose axes are directed very obliquely
upwards and outwards when the animal is in its natural position. These primary com-
partments are divided and subdivided into secondary and tertiary compartments, which
become smaller and more numerous towards the free edge of the lamella, from which alone
the little groups of pneumatic filaments are given off. These filaments, similar to those
which arise from the central chamber, pass directly into the distal layer of the hydrosoma,
and sometimes at once go straight through it. The majority run obliquely, so as to
traverse the hydrosoma in all directions. Whether they anastomose or not, I cannot say; but
I have never seen any clear instance of such an occurrence. Eventually, they seem to
pass out on to the distal surface of the hydrosoma, and there terminate in free ends,
those which are in the neighbourhood of the gonoblastidia at first twining about among the
gonocalyces.

No pneumatic filament ever ends 7” the walls of the polypites, they are all on their walls,
and, so far as I have seen, no pneumatic filaments enter the tentacular area.

The hydrosoma, or proper body of the Porpita, which incloses the pneumatocyst, is
composed of a thin proximal, or dorsal layer, and a thick distal or ventral layer. In the
former, the anastomosing canals described by Kélliker are very beautifully displayed. It
in addition exhibits a number of comparatively large apertures, through which the tubercles
(f°) of the pneumatocyst pass. The hydrosoma, gradually thinning, rises for a short distance
upon the bases of these tubercles, and then can be traced no further, so that I have no
doubt the tubercles are naturally bare. Besides these apertures, there are a number of very
much smaller ones, not more than ;{th — ;,th of an inch in diameter, which correspond with
those urn-shaped tubercles which bear the pneumatic foramina.

With regard to the inferior or distal layer of the hydrosoma, the only points I find
worth noting are the structure of the “tentacular area” and the distribution of the somatic
canals. As I have said, I find no pneumatic filaments traversing this region, or, at least,
passing out through it; nor does the hepatic mass extend into it. The reticulated appearance
which it presents is due to the meshwork formed by a peculiar tissue, which consists of
hands running perpendicularly from one face of the distal layer of the hydrosoma to the other,
and then spreading out and passing into the superficial substance. They resemble very
much, in fact, those fibrous bands which pass vertically through the substance of the dermis
in many fishes and Reptilia, and tie together the successive layers of bundles of connective
tissue.

The lamellze of the proximal surface of the distal layer correspond with, and accurately
fit into, the intervals between the lamellar diverticula of the pneumatocyst. Their proximal
edges are thin and sinuous, presenting shallow excavations, separated by sharp points.
They rise in height towards the circumference, and close to it, slope rapidly, so as to present
a convexity the opposite of that of the diverticula of the pneumatocyst.

Each lamella has a wide sinus running along its proximal or dorsal edge, and com-
municating, by a series of more or less vertical channels, with the sinuses which lie in
the hepatic substance.

These last appear to be disposed in several tiers, but the opacity of the mass renders it

126 THE OCEANIC HYDROZOA.

difficult to come to any clear conclusion on this head. Apparently from the discoloration of
the specimens, there is no distinguishable white renal organ, but I have met with many
minute crystals and granules similar to those considered to be Guanin by Professor KOlliker.
Not having had any opportunity of examining Porpita mediterranea, I will not enter upon the
consideration of the apparent differences in structure between that species and the one which
I have studied.

It would be very interesting to trace out the development of Porpita, but, from the
observations I have detailed respecting Ve/ed/a, I have little doubt that the early state of the
two forms is very similar; and that, in Porpita as in Velella, the sinuses must be regarded as
the somatic cavity cut up and subdivided by the union of the opposed layers of the
endoderm.

The most remarkable feature in the organization of both genera, however, is the
penetration of the hydrosoma by the pneumatic and other tubercles on the proximal, and by
the pneumatic filaments on the distal side.

Descriprion or THE Wooncurs.—A. Diagram of a radial section of the large Porpita. a, The distal or
inferior layer of the hydrosoma. 0’. The gonoblastidial polypites with their gonophoric buds, among which the
pneumatic filaments, f7, are seen twining. c. The tentacles, d. The limb. e. The proximal or superior layer of
the hydrosoma. 8. The pneumatic apertures. f°. The tubercles of the proximal layer of the pneumatocyst.
f*®. The air-chambers. f1!. Their prolongations into the lamelle. B. One of the urn-shaped pneumatic apertures.
©, Diagram of a transverse, or tangential, vertical section of Porpita.

APOLEMIA. 127

Of the family of Apolemiade, and of the genera Halistemma and Forskalia, 1 have seen no
specimens, but to render this work more useful to future observers I add figures and descrip-
tions of them, borrowed from the works of Gegenbaur, Kdlliker, and Leuckart.

Fam. APOLEMIAD/AE.
Genus APOLEMIA. (Eschscholz.)

The pneumatophore is small; the hydrophyllia lanceolate in outline and curved, so that
their proximal sides are convex.

APOLEMIA UVARIA (Zschscholz). Pl. XII, fig. 8.
Stephanomia uvaria, Peron and Lesueur, 1807.

Gegenbaur, whose figure of this species I have copied, gives so excellent an account of
its structure (Beitrige, p. 37), that I cannot do better than transfer it (with a little condensa-
tion and borrowing here and there from Leuckart) to my own pages.

The nectocalyces are disposed in two lateral series of three or four in each, and
constitute an oval mass, measuring altogether about an inch and a half in length. They
have the form of a truncated cone whose basal edges are rounded off. The aperture of
the nectosac, which resembles a broad-bodied bottle, and occupies the greater part of
the nectocalyx, is situated at the apex of the cone.

The pyriform pneumatophore is wholly devoid of pigment.

Scattered, filiform, and very contractile hydrocysts are attached to the coenosarc between
the nectocalyces, beyond which the ccenosarc is cylindrical, a line to a line and a half in
thickness, and carries at regular intervals (of about two inches) the plume-like groups of
polypites and other appendages; so that the whole colony, approximating towards the type of
the Calycophoride, consists of similar segments. Attached to the ccenosarc, immediately behind
the nectocalyces, is a compact bundle of young undeveloped organs, which, by the subsequent
outgrowth of particular parts of the ccenosare to which they are attached, become separated
into distinct bundles.

Each group of organs is, according to Leuckart (‘Z. N.K.,’ p. 67), attached by a distinct
peduncle, and consists proximally of a number of /ydrophyllia, succeeded by two or three
polypites and a great number of vermiform hydrocysts, at the base of each of which is a
simple tentacle. The hydrophyllia are ovate or clavate bodies, a few lines in length, which
are attached to the coenosarc by short peduncles. The larger ones are slightly curved, so as

128 THE OCEANIC HYDROZOA.

to have a proximal convex and a distal concave surface. They present many dots and specks,
which are for the most part aggregations of thread-cells. The hydrophyllia are capable of
independent motion, and in the undisturbed condition of the animal they are constantly
elevated and depressed. On irritating the creature they are all closely adducted, so as to
protect the organs over which they lie. The hydrocysts are from six to twelve lines long,
and almost cylindrical, and from the base of each springs a long tentacle.

The polypites are six or seven lines long, even in the contracted condition. They have
six villous or hepatic ridges.

The structure of the simple tentacles has been referred to above (p. 11). The gono-
phores have not yet been found attached to an entire hydrosoma, but Leuckart (‘Z. N. K.,’
p- 71) describes them in the detached groups of appendages.

They appear to be developed in bunches among the hydrocysts, and sometimes in con-
nexion with their peduncles, and the gynophores, at any rate, exhibit distinct calyces and
manubria ; the latter containing a single ovum.

Leuckart observed a specimen of <Apolemia a foot long. Its pneumatophore measured
two lines, its nectocalyces from six to eight lines in length.

HALISTEMMA. 129

Fam. STEPHANOMIAD AK.
Genus HALISTEMMA (mhz).

Tentacular sacculi, without involucra, and ending in a single filament. The nectocalyces
in a double parallel series.

HALISTEMMA RUBRUM. PI. XII, fig. 9.

Agalma rubra, Vogt.
Agalmopsis punctata, Kolliker.
Agalma (Agalmopsis) rubrum, Leuckart.

This species was discovered by Vogt, and was subsequently described by Kolliker,
whose figure (the distal end of the hydrosoma being omitted for want of room) I have copied.
As Leuckart (‘Z. N. K.,’ p. 73), however, has, with the works of Vogt and Kédlliker before
him, studied this species, I prefer to give an abstract of his description of it. Professor
Leuckart considers this form (the type of his Agalmopsis) to belong to a sub-genus of Agalma,
so that the general characterisation of this genus, which he gives at page 72, is applicable to
Halistemma.

The pneumatophore is of moderate size, and is without a pigment spot. The
nectocalvces, of which there are sometimes as many as thirty pairs, are cubical, compressed
from above downwards, and provided at their proximal ends with two pairs of processes, an
upper and a lower, the latter having a very considerable size and a wedge-like form, and
projecting much more than the upper ones. These processes serve to embrace the coenosare,
and to interlock with the opposite nectocalyces. Of the nectocalycine canals, there are not only
the ordinary four radiating and the circular, but in addition curved cecal processes (mantelge-
fisse), as in Lippopodius and Praya. The lateral radial canals form a double loop. The
coenosarc is spirally coiled. The polypites are exceedingly large, and the hepatic ridges have
a reddish colour.

The branches of the tentacles are very large, having a length of as much as a line in the
coiled condition, and of four lines when extended.

The hydrocysts are slender, and without projecting villous ridges, but otherwise
resemble young and undeveloped polypites. Their tentacles are slender, and without
sacculi. Four to eight hydrocysts in various stages of development are interposed between

every pair of polypites.
17

130 THE OCEANIC HYDROZOA.

The hydrophyllia are about eight lines long, and three and a half lines broad, thicker in
the middle than at the sides, and somewhat concave on their inner faces. They are frequently
devoid of lateral processes or teeth. They are not fixed by their upper ends to the coenosarc,
but by a proper peduncle, which is inserted a little behind their proximal ends, which are free
and solid.

The reproductive organs are like those of Stephanomia, and they are attached directly to
the coenosarc.

‘Halistemma rubrum attains a length of three feet, and appears to be not uncommon
in the Mediterranean, especially in winter.

FORSKALIA. 131

Genus FORSKALIA. (Kélhiker.)

The tentacular sacculi are without involucra, and end in a single filament. The necto-
calyces are arranged in many series around the ccenosarc.

Forskatia Epwarpsit (Aolliker). Pl. XII, fig. 10.

The chief points of interest in Kolliker’s description of this species (‘Schwimmpolypen,’
p- 2, et seq.) are the following :

The pneumatophore has a brownish-rose colour. The nectocalyces are arranged so as to
form a cylinder containing eight or nine longitudinal, and twelve to sixteen transverse, rows,
and each has a yellow spot upon its mouth.

The polypites have brownish-red hepatic striae. The sacculi are red. The hydrophyllia
are flat, broad, and terminated by three rounded points; the hydrocysts are colourless, but
occasionally of a lively red at their apices. The manubria of the androphores are yellowish-
white, but orange internally. The animal, which was very common at Messina, attained a
length of from six to twelve inches or more.

The ceenosarc is a straight or slightly undulating tube, with relatively thick walls, in which
longitudinal and transverse muscular fibres are clearly visible. It is exceedingly contractile,
and, in shortening, takes on the form of a close spiral.

The pneumatophore is ovate or pyriform, and a third of a line in length, and the air
contained in the pneumatocyst is divided into a large superior, and a smaller, inferior, bubble.

The nectocalyces have the form of a cone, whose apex is turned towards the coenosarc,
its base outwards. To speak more exactly, the basal surface, which has a rounded quadri-
lateral circumference, presents in the centre a mammillary elevation, truncated at the end,
which exhibits a circular aperture leading into the nectosac. According to the greater or
less projection of this mammillary elevation, the nectocalyx, viewed from above, has a
rhomboidal form, or more that ofa simple cone. In a lateral view the nectocalyces appear to
be oval or rectangular, with rounded edges; the oral region not unfrequently projecting a
little. Many of the nectocalyces are also quite irregularly rounded cuboids, so that their form
can hardly be defined in words.

At the upper end of the cylinder formed by the regularly arranged perfect nectocalyces,
numerous undeveloped nectocalyces may always be discovered.

The nectosac has the form of a short gourd, and the four radiating nectocalycine canals
take a straight course to the circular canal.

The hydrocysts are fixed directly to that portion of the coenosare which lies beyond the
nectocalycine cylinder, whilst the polypites, with their hydrophyllia and tentacles, are attached

132 THE OCEANIC HYDROZOA.

to the ends of special stalks. In this region the somatic cavity, which has the form of a wide
canal, is excentric, being situated nearer the convex side of the coenosare. In many places
it gives off cecal diverticula, which pass at right angles into the substance of the thicker wall.

The hepatic striz are eight to twelve or more in number, and are situated in the
proximal third of the digestive cavity of the polypite.

The stems of the tentacles give off, on one side, branches to which the sacculi are
attached.

The hydrophyllia have a triangular form, their prolonged proximal ends or apices being
attached to the peduncles of the polypites, while the basal or distal ends are divided into
three points. They are slightly concave on the inner side, slightly convex externally, where
they present a slight median ridge.

The hydrocysts are of two kinds—simple, attached immediately to the coenosarc; and
double, one large and one small, attached by a long common peduncle to the coenosare. A
single and double hydrocyst are commonly associated in pairs. Each hydrocyst has a slender
simple tentacle.

The androphores and gynophores are usually, if not always, attached to the base of the
double hydrocyst.'

Two to four androphores and a larger number of gynophores are usually associated
together. In structure they appear, from K@lliker’s description, very closely to resemble those
of Slephanomia amphitridis.

* Whence these should probably be regarded as gonoblastidia.

DOUBTFUL GENERA AND SPECIES. 133

REMARKS ON DOUBTFUL GENERA AND SPECIES OF C4LYCOPHORIDH
AND PHYSOPHORID.

Besides those which have been incidentally referred to in the text, the following genera
and species of Calycophoride and Physophoride appear to me not to be sufficiently well
established, or to be otherwise open to criticism.

Genus ACIES.
A. palpebrans, Lesson. ‘ Voyage de la Coquille, Zoologie, ii, 2, 1880.

Lesson himself seems to regard this as a mere sub-genus of Porpita. No figure is given,
and the description is wholly insufficient.

Genus ANGELA.
A. cytherea, Lesson. ‘Acaléphes,’ 1843, p. 496, pl. ix, fig. 1.

All the author of this genus really knows of it is, he says, derived from a drawing,
“communiqué par M. Rang sans nom et sans renseignements.” Under such circumstances,
it is hardly worth while quoting his definition.

Genus ANTHOPHYSA.

A. rosea, Brandt, ‘Prodromus,’ cited in Lesson’s ‘Acaléphes,’ appears to be an Athoryéia.

Genus APOLEMIOPSIS.
A. dubia, Brandt. ‘ Prodromus.’
“Canalis reproductorius filiformis, tenuis, elongatus, cui vesicula parve, biseriatee, porro

tentacula simplicia verrucis parvis quadriseriatis obsessa ; nec non tubuli nutritorii adnexi et ut
. . . ?
e figura apparet, partes cartilaginez fere semilunares adheerent.’

134 THE OCEANIC HYDROZOA.

Genus BRACHYSOMA.
B. erythrophysa, Brandt. ‘ Prodromus.’

Probably a mutilated and contracted Physophorid.

Genus CUNEOLARIA.

C. incisa, Eysenhardt. ‘Nova Acta,’ x, 1821.

A detached nectocalyx of a Physophorid.

Genus CUPULITA.

C. Boodwich, Quoy and Gaimard. ‘ Voy. de l’Uranie,’ Zoologie, p. 580, 1824.

An imperfect description and figure of what appears to be one of the Stephanomiada.
The authors give up the genus in the ‘ Zoology of the Astrolabe,’ t. iv, p. 53.

Genus DIPHYSA.
¢@

D. singularis, De Blainville. ‘Manuel,’ p. 117, 1830.

De Blainville founded this genus upon a specimen collected by Quoy and Gaimard ; but
his account of its structure is brief and insufficient to enable one to form an idea of the
real nature of the object.

Genus DISCOLABE.

Rhizophyza discoidea, Quoy and Gaimard,
Discolabe Mediterranea, Eschscholz.

Thus described by Quoy and Gaimard in the ‘Annales des Sciences Naturelles,’ t. x,
1827: A little flattened disk, from whose circumference depend about a dozen slightly rose-
coloured appendages, consisting of little orbicular bodies connected with one another. In the
middle of the disc below there is a bundle of transversely striated suckers, whose base is
surrounded by little yellowish bodies. The middle of the upper part of the disc gives off a
very long tube, containing an air-bubble at its extremity, which is reddish. Length, an inch
and a half; diameter of the disc, five lines. Taken in the Straits of Gibraltar. Figured in
Pl. V. B. Eschscholz, who based the genus Discolabe on this form, did not observe any

specimen of it himself.

DOUBTFUL GENERA AND SPECIES. 135

Genus PYRAMIS.

Pyramis tetragona, Otto. ‘Nova Acta,’ t. xi, 1823.
Eudowxia pyramis, Eschscholz.

Apparently the distal nectocalyx of an Adyla.

Genus PHYSOPHORA.

P. alba and P. intermedia of Quoy and Gaimard (‘ Voy. de ]’Astrolabe,’ Zoologie, t. iv)
both belong to the genus Agalma.

Genus PLETHOSOMA.
P. cristalloides, Lesson. ‘ Voy. de la Coquille,’ Zool., ii, 2, 64, 1830.

This is a combination of a contracted piece of one of ‘the Stephanomiade with an Abyla,
apparently 4. pentagona!

Genus POLYTOMUs.

P. cruciata, Lesson. ‘ Voy. de la Coquille,’ Zool., ii, 2, p. 45, 1830.

A detached nectocalyx of one of the Physophoride.

Genus RACEMIS.
R. ovalis, Delle Chiaje. ‘Memorie,’ t. iv, p. 4, and p. 30. 1829.

The account of this genus, given at p. 4 of the fourth volume of the ‘ Memorie,’ is but a
dilution of the following ‘Descrizione technica,’ at p. 30: “ Racemis—Vesicule globose
celerrimo mota predite et in formam ovatam disposite. &. ovalis—Corpore ovali vesiculis
globosis.” The figures given, Tab. L, figs. 11 and 12, are quite unintelligible to me.

Genus RATIS.

R. Meduse, Lesson. ‘ Voyage de la Coquille,’ 1830, Zool., ii, 2, p. 60.

Apparently a young Porpita, but there is no figure, and the description is very

imperfect.

136 THE OCEANIC HYDROZOA.

Genus SARCOCONUS.

A genus built up by Lesson (‘ Acaléphes,’ p. 477), out of the odds and ends of fragments
of Stephanomiade, observed by Chamisso and Eysenhardt, and by Quoy and Gaimard.

Genus STEPHANOMIA.

S. levigata, described by Quoy and Gaimard in the ‘Zoology of the Voyage of the ‘Uranie,’ ’
p. 585, and figured in pl. Ixxxvi, fig. 2, looks like a fragment of Sfephanomia amphitridis.
Of the species described as Stephanomie in the ‘ Zoology of the Voyage of the Astrolabe,’
S. helianthus and 8. melo are Athorybiea; S. hippopoda is a Hippopodius; 8. triangularis is
an Agalma; 8S. imbricata, S. heptacantha, and 8. tectum apparently belong to the same genus ;
S. foliacea is very like a true Stephanomia (mihi); S. alveolata is pretty clearly a Vorskalia;
and S. cirrhosa would seem to be part either of a Halistemma or of a Stephanomia.

Genus TETRAGONUM.

T. Belzoni, Quoy and Gaimard. ‘ Voy. de lUranie,’ Zoologie, p. 579.

Probably the detached distal nectocalyx of an Aédyla.

NOTE ON THE TERMINOLOGY OF THE ORGANS OF THE HYDROZOA.

Since I have strongly acknowledged the value of Professor Allman’s contributions to our
knowledge of the Hydrozoa, and since I have gladly made use of certain of the terms which
he has invented (e. g. gonophore), | feel bound to assign my reasons for not adopting the whole
of his terminology.

The transition between what Professor Allman terms sporosacs and his “ Weduse” is
so gradual, that I know not where the line of demarcation is to be drawn between the two.
If the title of sporosac, however, could be applied to such reproductive organs as those of
Hydractinia, which are simple sacs without a trace of medusoid structure, I think the
term would be useful.

“Hctotheca” and “blastostyle” appear to me to be very good terms when applied to
the central axis and the capsule of such “compound gonophores” as those of Laomedea;
but I think that when the stalk of the gonophores of Zuéularia (with which the stem
of similar organs in the Physophoride is homologous) is also called blastostyle, the same
name is applied to different things. The stem of the “gonoblastidium”’ of the Physophorid
contains, I apprehend, the representatives of both the blastostyle and the ectotheca of Laomedea.

Again, the “generative polypes” of Hydractinia are surely homologous with the stalks
of the gonophores in Zudularia; but to these, and, as I think, with justice, Professor Allman
does not apply the term “ blastostyle.”

When we know a medusiform body to be simply the detached reproductive organ
of a hydrozoon, it seems to be better to avoid all chance of confounding it with a truly
independent organism. I have, therefore, abstained from using the word “ AZedusa,” except
in the sense defined at p. 21; and for me Professor Allman’s Weduse are medusiform
gonophores.

Finally, the term manubrium should, I think, be restricted to the central polype-like sac
of a medusiform gonophore, which is surely the homologue of the whole sporosac of
Hydractinia, and not of its central cavity only.

witire

Se ae

. i:
ny th pau ,
Wart aveh kf Panwa § Peo NA ey Oe) To a
rel dae PEAY nat r {aaa el vy) | t ee bad f
; > tata } wwe

cae «

ae

Lobel er ae
ria et -!
rou peel

139

BIBLIOGRAPHY.

The following are the full titles of the books and detached papers cited in the foregoing
work, arranged alphabetically according to the names of the authors :

Auman. ‘On the Anatomy and Physiology of Cordylophora; ‘ Philosophical Transactions;’ 1853
(with plates).
‘On the Structure of the Reproductive Organs of certain Hydroid Polypes ;’ ‘ Proceedings
of the Royal Society of Edinburgh,’ Session 1857-8.
‘Additional Observations on the Morphology of the Reproductive Organs in the Hydroid
Polypes ;? Ibid., 1858.
De Brainvittr. ‘Manuel d’Actinologie ou de Zoophytologie’ (with an Atlas of plates); 1834.
Bory pe St. Vincent. ‘Voyage dans les quatre principales Iles des Mers d’Afrique ;’ 1804 (with a
‘Collection de Planches’).
Bosc. ‘ Histoire naturelle des Vers’ (with figures) ; 1802.
Buscu. ‘Beobachtungen iiber Anatomie und Entwickelung einiger worbelloser Seethiere’ (with
plates); 1851.
Cuamisso and Eysennarpr. ‘De Animalibus quibusdam e Classe Vermium Linneana’ (cum
tabulis xi, enez pictis); ‘ Nova Acta physico-medica Academiz Czsarez Leopoldino
Caroline Naturz Curiosorum,’ tomi decimi, pars secunda; 1821.
Cuias, Dette. ‘ Memorie su la storia et notomia degli animali senza vertebre del Regno di Napoli ;’
1823—1829; 4 vols. (with plates).
‘Descrizione e Notomia degli Animali invertebrati della Sicilia citeriore osservati vivi negli
anni 1822—1830,;’ 1841—1844.
(Six volumes of text, and two of plates.)
Costa. ‘Note sur lappareil vasculaire de la Velelle (Armenistarium Velella); ‘Annales des Sciences
Naturelles,’ serie deuxiéme, Zoologie, tom. xvi; 1841.
Cuvier. ‘Le Regne Animal; 1817.
Epwarps, Mitne. ‘Observations sur la Structure et les Fonctions de quelques Zoophytes, Mol-
lusques, et Crustacés des Cétes de la France’ (Lues a VAcad. des Sciences, Aout
16, 1841); ‘Annales des Sciences Naturelles,’ deuxiéme serie, Zoologie, tome xvi,
1841.
Ercnwatp. ‘Observationes nonnull circa fabricam Physalie ; ‘Mémoires de l’Académie Impériale

des Sciences de St. Petersbourg,’ t. ix; 1824.

140 THE OCEANIC HYDROZOA.

Escuscuorz. ‘System der Acalephen, eine ausfiihrliche Beschreibung aller Medusenartigen Strahl-
thiere,’ mit 16 Kupfertapeln ; 1829.

Kysennarpr. ‘Zur Anatomie und Naturgeschichte der Quallen:’ 1, Von dem Rhizostoma Cuvierii
(Lam.); 2, Ueber die Seeblasen; ‘Nova Acta physico-medica Academiz Czesarez
Leopoldino Caroline Nature Curiosorum,’ tom. x, pars secunda; 1821.

Vorskax. ‘ Descriptiones Animalium, Avium, Amphibiorum, Piscium, Insectorium, Vermium, quie in
itinere orientali, observavit Petrus Forskal,’ post-mortem auctoris edidit Carsten
Niebuhr; 1775.

“Tcones rerum Naturalium ; 1776.

GecunBaur. ‘Zur Lehre vom Generations-wechsel,’ (with a plate); 1854.

‘ Beitrage zur naheren Kenntniss der Schwimmpolypen, (with plates);? Siebold and
Klliker’s ‘Zeitschrift fiir Wiss. Zoologie ; 1854.

Hoxtarp. ‘Recherches sur Organisation des Velelles, (présentées & lAcadémie des Sciences,
le 2e Octobre, 1843); ‘Annales des Sciences Naturelles,’ troisieme serie, Zoologie,
tome i; 1845, (with figures).

Huxiny. ‘On the Anatomy of Physalia; ‘ Proceedings of the Linnean Society ;’ 1848.

‘Observations upon the Anatomy of the Diphyde and the Unity of Organization of the
Diphyde and Physophoride; ‘ Proceedings of the Linnean Society ;? 1849,

“On the Anatomy and Affinities of the Medusze,’ (with plates); ‘ Philosophical Transactions ;’
1849.

‘Report on the Structure of the Acalephe;’ ‘ Reports of the British Association ;’ 1851.

‘Ueber die Sexual Organe der Diphyden und Physophoriden,’ (with figures) ; ‘ Miiller’s
Archiv ; 1851.

Hynpman. ‘Note on the Occurrence of the Genus Diphya on the Coast of Ireland, (with figures) ;

‘Annals of Natural History,’ vi; 1841.

Koéuiiker. ‘Die Schwimmpolypen oder Siphonophoreu von Messina,’ (with plates) ; 1853.

Kroun. ‘Notiz ttber die anwesenheit eigenthtimiichen Luftkanale bei Velella und Porpita;’
Wiegmann’s ‘Archiv fiir Naturgeschichte,’ I4r Jahrgung, B. 1; 1848.

Lamarck. ‘ Histoire Naturelle des Animaux sans Vertébres,’ 2d ed.; 1835—1845.

Leacnu. ‘Tuckey, Narrative of an Expedition to Explore the River Zaire; 1818.

Lesson. ‘Duperrey, Voyage autour du Monde sur la corvette La Coquille executé pendant les
années 1822—1825,’ Zoologie, (with plates).

‘Centurie Zoologique,’ (with plates) ; 1830.
‘Histoire Naturelle des Zoophytes Acaléphes, (with plates); 1843.

Lesururx. ‘Mémoire sur quelques nouvelles especes d@’Animaux Mollusques et Radiaires, recueillies
dans le Méditerranée pres de Nice ;’ ‘ Journal de Physique,’ t. ixxvi; 1813.

Leuckarr. ‘Ueber den Bau der Physalien und Siphonophoren ; Siebold and Kélliker’s ‘ Zeitschrift
fiir Wiss. Zoologie, Bd. iii; 1851.

‘ Zoologische Untersuchungen,’ Ist Heft, (with plates); 1853.
(Quoted under ‘ Leuckart, Z. U.’)
‘Zur niheren Kenntniss der Siphonophoren von Nizza,’ (with plates) ; 1854.
(Quoted under ‘ Leuckart, Z. N. K.’)
Muyen. ‘Beitrage zur Zoologie;) ‘Nova Acta physico-medica Academie Czsaree Leopoldino

Caroline Nature Curiosorum,’ tom. xvi, Supp. 1; 1834.

BIBLIOGRAPHY. 141

Mixxer, Jonannes. ‘Ueber eine eigenthiimliche Meduse des Mittelmeeres und ihren Jugendzu-
stand ;? Miiller’s ‘Archiv ;’? 1851.

Ourers, Von. ‘Ueber die grosse Seeblase (Physalia Arethusa) und die Gattung der Seeblasen im
Allgemeinen ;? ‘Abhandlungen der Kéniglichen Akademie der Wissenschaften zu
Berlin ;’ 1831.

Orto. ‘ Beschreibung einiger neuen Mollusken und Zoophyten, (with plates); ‘ Nova Acta Akad.
Leopol. Ces. Nat. Cur.,’ t. xi; 1823.

Peron. ‘ Voyage de découvertes aux Terres Australes ;? 1807.

Puruirri. ‘ Ueber den Bau der Physophoren und eine neue Art derselben, Physopliora tetrasticha ;’
Miiller’s ‘Archiv fiir Anatomie, Physiologie, &e. ;? 1843.

Quarreraces, Dz. ‘Mémoire sur Porganisation des Physalies;’ ‘Annales des Sciences Naturelles ;’
1853.

Quoy and Garmarp. ‘Freycinet, Voyage autour du Monde sur les corvettes L’Uranie et La
Physicienne, fait pendant les années 1817 a 1820,’ Zoologie, (with plates) ; 1824.

‘Observations Zoologiques faits & bord de L’Astrolabe dans Je Détroit de Gibraltar,’ (with
figures); ‘Annales des Sciences Naturelles ;’ 1827.
“Dumont D’Urville, Voyage de la corvette L’Astrolabe, executé pendant les années

1826—1829, Zoologie, (with plates) ; 1830—1834.

Sars. ‘ Fauna littoralis Norvegi,’ (with plates); 1846.

Voer. ‘Recherches sur quelques Animaux inférieurs de la Méditerranée ; ‘Mémoires de I’ Institut
National Genevois,’ tome ler, (with plates) ; 1853.

Witt. ‘ Hore Tergestine, oder Beschreibung und Anatomie der im Herbste 1848, bei Triest

beobachteten Akalephen,’ (with plates) ; 1844.

‘a Taha

vi wig. fh ontalhs aatintt ttn (i vere we {i 7
pier tr LANE, A VlitOr AY os Mi a,
fut Th 2 A'S) av an Gh idk » ie
dy rite i nee Pda relate a 1 radi Aaa
, : a } ye bythe
PHC Nia Nai daibaphlora Gai yi i wai ain hi
\ ppg Ha Hoa oe AME ae). dened

~

Vt age Ale on dye’? 7
«

‘.

J ty »
§ Waal) Tbe) yes ig dba Hah tines as atp is ; ali F ue

rf . ; .
7 . pie» ¢ if . 1] bid) fy HI vi 7! AT A aH f , @
obit i maraiyce. wah Walarr ae

¢

Cabrel sie bg inhi! AAS et kina

o, ; . La i _ 7

: : in sa f ee rs : a

BR LER MUL wally t809. 4D iy ahah A sal va Aue Plant ae ena ch er
= AGE plidinla H j | a. | Q bum : im iteanh ahd : i] :

‘Mie y Say) HY gy ti

4

ae iiey that

oS AAR PND IRL, oa Ail DAR
ae hitits wat ry if Atihiaks sian i wer ; ft

+
U Hain k't rier

Pern Buide

143

DESCRIPTION OF THE PLATES.

The letters have the same signification in Plates I to XI. The short ruled tines indicate
the natural sizes of the objects.

a.
b.

Cc.

d,

&

Coenosarc (kowoc, common ; oaeé, flesh).

Polypite ; 47, villi; 4°, pyloric valve.

Tentacle; c', pedicle of the sacculus (reniform body in Physalia); ce’, involucrum ;
c, sacculus; c*, filament; c°®, median lobe; c*, basal sac; c’, muscular band.

Somatocyst (cwua, body; «varie, a vesicle).

Pneumatophore (wvevja, air; gogéw, I bear).

Pneumatocyst (avevua, kvorc); f', the endoderm reflected over it; f°, the cellular
processes of this layer of the endoderm; /*, the mesentery-like lamelle connecting

the pneumatocyst with the pneumatophore ; /°, the air contained in the pneuma-
tocyst.

. Nectocalyx (vnxréc, having the power to swim; xadv&, a cup); g', the proximal

nectocalyx of a Calycophoridan; yg’, the distal nectocalyx ; g°, the nectocalycine
duct and canals, both radiating and circular, and in gonocalyces as well as in
nectocalyces; g*, the nectosac; y’, its aperture, surrounded by its membranous
valve; g°, the hydreecium (orcoc, a house) of the proximal and the hydreecial canal
of the distal nectocalyx.

. Hydrocysts.
:. Hydrophyllium ; 4’, phyllocyst (guAXov, a leaf).

Gonocalyx (yovoc, progeny).

. Gonoblastidium (/sAdorn, a bud).

. Androphore ; 7’, its calyx; 2’, its manubrium ; n®, spermatozoa.
. Gynophore ; o, calyx; 0°, manubrium ; o°, ova.

. Endoderm.

. Ectoderm.

PE Ay BOONE y-

i

Ve
La hue
a - 2 ‘ " 4 é , 5 ‘ ayyn
Pe Ae wanitd Dept! jest HHS mY Legare ny) ht i Des ay
s » 2 f to F ,
leet (0 Aenean

PLATE I.

Fig.
1. Diphyes dispar, complete.
la. An anterior view of the distal end of its proximal nectocalyx.
1 6. A lateral view of the distal nectocalyx.
1 c. A posterior view of the same object.
1d. A young proximal nectocalyx.
le. A polypite, with its hydrophyllium and tentacles, much magnified.
2. Diphyes appendiculata, complete.
2a. A front view of the distal end of its proximal nectocalyx.
2%. A posterior view of its distal nectocalyx; the bristle (y°y°) is supposed to
be passed through the hydreecial canal.
2c. The hydrophyllium of the same species.
3. Diphyes Chamissonis? a lateral view of the proximal nectocalyx.
3a. An enlarged view of the distal end of this nectocalyx.
3 6. The hydrophyllium.
4. Diphyes mitra? a lateral view of the proximal nectocalyx.
4a. A front view of the same.
4%. A polypite, with a gonophore, and a possibly imperfect hydrophyllium.

19

——

_ 4 : Se :
: e EWE adnabidel. ot Onn tn. 4. b, DIPHYES. | EBianine se. ‘

. as é eet ; ° * ae ace *.

"

sj ei : a) ie) : : ee Fels : i
. 0 Bb) 2
a er”

®

a

; mi A t val
; as, oie : ah
% : ! “i i aa)

'
rl

yw

PLATE II.

Fig.

1. Adyla bassensis, complete ; a lateral view.

1 a. The same, so placed as to show the posterior side of the distal nectocalyx.
1 4. The proximal nectocalyx from in front and above.

1 c. The polypite, with its hydrophyllium.

2. Abyla pentagona, complete ; a lateral view.

2a. The distal nectocalyx from behind.

2. The proximal nectocalyx enlarged, from one side.

2c. A young distal nectocalyx of the same.

2d. A polypite, with its hydrophy!hum and a gonophore.

2 e. The hydrophyllium detached.

3. Abyla Vogtir, the proximal nectocalyx ; an antero-lateral view.
3a. The same; a postero-lateral view.

3 6. A polypite, with its hydrophyllium and gonocalyx.

3c. A young gonophore magnified:

PLATE. II.

gee Ff

a.” if
°
4
vA ,
(i
.
e
Ss
at
‘i
Y
’
,
«
f
“
7
5
= 7
=
‘ee
J
i
ie
4
:

A

Iniib oaT OO

4 4 LA | Lad
1 TD 4

Konya MT AS

PLATE III.

Fig.
1. Abyla trigona, complete ; a lateral view.
1 a. The proximal nectocalyx from in front.
1 4. The same, viewed laterally.
1 c. The distal nectocalyx from the left side.
1d. The distal end of the same.
1 e. A polypite, with its hydrophyllum.
2. Abyla Leuckartii; a lateral view of the proximal nectocalyx.
2 a. The same, viewed from behind.
2 6. The same, viewed from in front.
3. Praya Diphyes? (Compare PI. XII, fig. 2.) A lateral view of a nectocalyx.
3a. An anterior view of the same.
4. Spheronectes Kollikeri, complete ; a lateral view.
5. Galeolaria filiformis. (Compare Pl. XII, fig. 1.) A lateral view of the distal
nectocalyx,
5a. A posterior view of the same.
6. Hudoxia Lessonii, complete ; a lateral view.
6 a. The distal end of the gonocalyx.
7. Eudoxia Bojani, complete.
7 a. The hydrophyllium detached.

ili pes y
eed
: d
oo a

7

ABYLA_ PRAYA_ SPHG@RONECTES— GALEOLARIA — EUDONA. ; |

, t TOT
: i Wiser te

PLATE IV.

Fig.

1. Eudoxoides sagittata, two views.

2, 2a, 26. Aglaismoides Eschscholzii, in different positions.

3. Aglaismoides elongata, complete ; but with only a small gonophore.
3a. The hydrophyllium alone.

34. A complete specimen, with two large androphores and a small gynophore.
4, 4a. Sphenoides australis.

44. One of its detached free swimming gynophores.

4*. Possibly young specimens of this species.

5, 5a. Cuboides vitreus.

5 6. One of its gynophores.

6, 6a, 64. Different views of Hnneagonoides Quoyi.

PLATE IV,

TRascre Sc.

CUBOIDES_ ENNEAGONOIDES.

SPHENOIDES _

FUDOXOIDES_ AGLATSMOIDES_

-

a

Ja 3
a

i , iS s.! 1 i
i real 2 1 ae iy
‘ he his , On f une,

if)
- ‘ ey : ! a ' Te

i)
oY A eae
' Ths Tae ‘
i oR | Me ge)
; Fis yan
if H hh Ly i
Te ies Wit a
uA
uah : Na ae
\ 2 (ial ; i
t a Pant iW 4
aa) J 1D)
vee titi viet
i if ‘
vf — : ! j
5 ats A Le
‘ q i
i j F
j
i
iit
; Ty
- iy
| iph
} u a {
: |
®
¥ if
i } Poa
|
: - i] : {

-_)

- " a u
7) - i
- ry Ss
7 : iF

Fi i

' |

.
- a} <p i

{ 4 Vel }

i] i ‘

\ ;

j “

i a
+ v :
7 aT 5 _¢
| :
*
» it
; n
7
: 1e ‘ .
yy 1

itt. am hay

i [yeoeitigl by
Winliven i uy
Ve hori i
yop qed Ay

i" Tt eAy hr

+9 ANA ed

ore ay
Ny Ws ety
rn) vee

a ad ip
He af Aq yi to wt.
pany Va. jv iy a>
We), ity cs
TOA veo J
era
DOU 3 Ly Bi, hon
ar
iD AED i 7
7 Hoel am son 7
writs RE ky meet
- sedi a)
ee Tar wri iF jn gl i) hi J

Fig.

PLATE V.

1, 1 a. Amphiroa alata, without gonophores.

14. The same, with two gynophores.
1c. A detached gynophore.

2.

Amphiroa angulata.

3—15 give the details of the organization and development of the Calycophoride; fig. 15

3.

8.
)-
10,

being copied from Gegenbaur.
Part of the proximal nectocalyx of Diphyes dispar, showing the manner in which the
ceenosarc ends proximally, and the course of the nectocalycine duct.

. The proximal end of the ccenosare of Diphyes appendiculata.
. The proximal or basal end of a polypite of Diphyes, exhibiting the pyloric valve and

the relations of the phyllocyst to the hydrophyllium.

. The same parts (without the hydrophyllium) in Diphyes appendiculata.
. A portion of the proximal end of the coenosarc of Diphyes, bearmg young polypites,

with their rudimentary hydrophyllia, &c.
The basal or proximal portion of a polypite of Spheronectes.
The polypite of Ludowia Lessoni, with part of its hydrophylhum.
11. Polypites of Diphyes, more advanced than in fig. 7.

12a—f. The successive stages of growth and metamorphosis of the tentacular sac in Diphyes.

13.

A gynophore of Hudoxia Lessont.

13a. Part of the wall of the manubrium of such a gynophore, with an ovum in place.

14.

A free androphore of Sphenoides.

14a. Its manubrium burst, and discharging its contents.
14 4, 14 c. Spermatozoa.
15a—g. The successive stages of the development of Diphyes, after Gegenbaur. (See p. 23.)

DH Had natal.

AMPHIROA __ ANATOMY OF CALYCOPHORIDA.

uy
-

7,
i vu,
ie

eee’

i

Al ‘i Al
etal a

ai
Ls

Dy) aie

PLATE VI.
Fig.

to}

1. Stephanomia Amphitridis. A fragment taken on the east coast of Australia.
2, 3. A portion of this slightly enlarged, and viewed in different aspects.

4. A single hydrophyllium, with a polypite, hydrocysts, and part of the coenosarc.
5, 5a, 6a, 6 6. Two views of other hydrophyllia.

6. Another hydrophyllium.

7. The end of one of the tentacular branches, magnified.

8. The same more highly magnified.

9. The “ peculiar bodies” mentioned at p. 73.

0. Thread-cells.

12. Valle

12. Part of the ccenosare, with hydrocysts and gonoblastidia.

13, 14. Gynophores with their single ova.

15. An androphore.

oF

a
5

13

Pug3

LL
6

VE

PLATE

ae Re ees
BBO EE
So

VaR ad nat det

: ]
a
/ hy
|
‘
ae
' 7 TEN * if 7
Ai RNEH WM, Debio Nw °E se

* y an

he us
* / ol neh ic

Ail to; life
,. .

PLATE VII.

1. Agalma breve, complete; magnified. Taken on the east coast of Australia.
2. Its pneumatophore.

3. A fully formed and (figs. 4 and 5) young nectocalyces.

6. A young hydrophyllium.

7. The end of a tentacular branch.

8. Part of the ccenosarc, with hydrocysts and gonoblastidia.

9, 10. Young and fully formed gynophores.

11. An androphore.

12. A young dgalma.

Fe AUIbIE, =, Wiis

AHamre Se.

AGALMA .

CTEM wchnat. del.

.
;
‘
.

: 7 ae iy 7 :

oy
MEE L213, a

ci i i A Pa py) Fy eS Vey. Py flag on 2 aa

: : ‘aay
ip ’ i, f i : 7 7."
, +4
= vig) - _
tl Lp gD ’ ; -
y ue mM y wean "
fs we be ae
re . ee
BN ho ae
17 iy 4 Bi s. ao :
j Be ure i 6 ‘ aan,
sly ; oh | = ie ; +
7 . | » f - a f n
. be : oe 7
Ss ' ‘ ; 7 } ef ;
1 pe ey
a ; 5 }
~ 5 ® ce .
, . ) 1D
i 7
' [ 2 f

.
a jn
.)
= as
— r > 2s
7 i | : ®
a : : 4 ‘ po
a hes ne 7 wes r ava oar ‘ hi ite: 7 «vent
: - : 7
i» wen ¢. * y . Pale Wn r yas Any é ty

OE Nevis iS Tes dh) ih dij ha hal 5 Triste } V. Yiblinnds lo eneLe i) va)eist afok
ba | = Ca pny uae i. af

; TOUR Thais ue) lk i
: A Hise wlll i y. ;
SV Mt ME) RO BPA Pe) VTA WOM seo yb JOM) Ree AF
lil See | 4 he MeO Lak) OY v ia
; 4 Noa ite ae I ae
sik hs aba ontiy in ith Bato bean De 2 1) .) ee

cee
a2 f it oy iv ' ha | A wy i,
Sfifonadt livin! st) hw Sonal Boe na. {7 a -

PCO Bap iid ai Win. Rute gos to pee Ganley "it ‘8 uf! is
® 4 ’ ; i

. ie ieee
ue. z PLL AL ter) “Ww
rn | . i 3 : a ‘.
. ar.
. : : 2
,. - a
* ;

* -
é * 7
a 7
i F Tn a’
rid ro
. ? - , a
°. ,
-
st a
. ®
D
! . * ee.
j : 2
y Pe A
4 ‘ ye
° . :
’
‘ fn >
A
‘ s ‘ih Aes "
a 7
- 4 , i ‘ : ‘ | ; |
'
ry, F a -
fF hi ry” - é | o>
iP) ade! cies | ici ® rua 1: :

Fig.

Ih.
2.

PLATE VIII.

’

A sketch of Physophora —?
A young Physophora —?

3 exhibits the mode of connexion of a hydrocyst with the dilated distal end of the coenosarc.

A.

The pneumatophore.

5—8. Successive stages of the tentacular branches.

S);

A thread-cell.

10 a—12. Successive stages and various views of the nectocalyces.

13.
14.
15.
16.
17.
18,
NG
20.

Rhizophysa filiformis.

Its pneumatophore.

One of the cellular processes developed from the reflected layer of the endoderm.
A polypite, with its tentacle.

Part of a tentacle, with its lateral branches.

Part of the proximal end of such a tentacle, with its budding branches.

A villus.

Part of a tentacular branch.

EMATE Vil

er

Se
5 oe
G
S
De
ae
fees
ea
a
5 e
B
P

SC PHYSOPHORA. RHIZ OPHYSA. ee

eee. _- Cae

ree
§
| ‘ei
' i
5
il
as
4 7
( 7
ae
|
& :
A j
i [
i i
\
=!
.
+t
val
a
i
eh |
Lat
A
tea Pay
r. ‘a
\ iy
ON ANY
i
&
”
a
Aa
.
6
oo

as

it

Luh ~~

yeas
i
i
ii
by

Wea}

hy

ey /
Re he
Wy it

;

“

i

a

‘
j
i ae

8 ic*
pide?
‘ +

Peep

PLATE IX.

Fig.
1 . Athorybia rosacea ; a \ateral view.
2. The same magnified, and viewed from above.
3. The same deprived of most of its hydrophyllia, and viewed from above.
4, 5. Hydrophyllia.
6. A polypite.

7, 8: Some of its villi.

9 a—y. The ends of the tentacular branches in various stages of development.

10, 11. Large and small thread-cells.
12. A gonoblastidium, with androphores, gynophores, and hydrocysts.
13. Gynophores attached to their common stem.

14, 15. Similar bodies enlarged, and viewed separately.

PLATES DX.

TE ad rated. ATHORYBTA..

A Bache sc

a
yy

yp

SORTA

+

eptonnist hr wigaton elquis # vino Wity, Some ye sein (Ym A.1 -
; a 2 Or "ay Uber ahie bla Ba le oie a i ‘es
SMATOONE OTNOY G1) Te oe Y yarict ent 45 “yieig®) #*
a esigglog hak Wy
ae ; * tliv je “P ROTEBY 9 ert, an :
Pn aid wave AF fojitye annals Maul %o aoe 2B
) : git hath yet Mi ae,
alae ie Taued ood abteve onan Bh JH
5 : Ri ooh tne eile? a ay gsr pny Pade re Hid pix any a i i a 1
Ce, tl ' - AN
* > Boimrrnyil we Hie vert yay t hl + x) eA ehh ie a torniud dnd rr ys Ww Oleg, hn BL.
: A aay! d Crh Fyn ia erin allo) phoret! aes w&t 7 Ah
; a
tii whirr ; ae ia Bb Cad au ba adj a yt) a
. he
LG tinntebi svi ioe ry me ipiP ar wot Ay soul Ng: BLT A Ae
j 0, Dy ant uy '
; hae Feelings a Maes ei" i 9 oy Tr 2 Oakey Ob: Sogo ovartnoped A
ae yt ’ "Lb Bronie yi eM Ly

+

PLATE X.

Fig.
1. A very young Physalia, with only a single polypite and tentacle.
2. A somewhat older specimen.
3. Opening of the pneumatocyst in a young specimen.
4. A polypite.
5—7. Various views of villi.
8. Some of the elevations which cover the outer surface of the lip of the polypite.
9, 10. Young polypites.
11. A tentacle, with its basal sac.
11 a. Part of the proximal end of that tentacle ; 11 4, a young tentacle.
12. A portion of the distal moiety of this tentacle, exhibiting the reniform enlargements ;
12a, 124, the thread-cells contained in them.
13. One of the “ velvety processes” magnified.
13a. A similar process of the natural size; and 134, somewhat magnified.
14 a—d. Successive stages of the androphore ; 14 e, young spermatic elements.
15 a—e. Successive stages of the gynophores.

PLATE, X.

pal

VT Basire, Sc

SALIA.

PHY

THE. ad.nak. del.

f

’ ‘ ! 7 , sf
i" ai NAG "Ugh SU a
H ble TH ” ae zy
7 } elt “ #4 » At:
TMA De Baek wom , ey
¥ UMD Airy lid w
Nee) : ve - ¥ ; os
i

vote sent baseiy SW 6. perth if ih
h Bagh 2s yg aay oe TA poiiay Fit ae Hite
: ; ‘chonay ik Re
Wee HOSE of wily i rh A we
OY ie irl this ahTits we (laws A
syencdepngt sj ite Avy wp waly Levit dy Aa
acti oily Ww WHY barton al By)
er vy
vriattyouced tonifpaiod a4 Me
hme) Sa

S13

rn!
i
Ss
ii wt 2. aot

seve | ww tlie a
ry ania asl wo he iby ait Ve wht

Lda ge Seep | au
: a

PLATE XI.
Fig.
1. Disc of a Velella viewed from above, and having the crest cut off.
2. A vertical section of the pneumatocyst. It has been accidently inverted.
3. A tentacle.
3a. A part of the crest, much magnified.
4, A small or gonoblastidial polypite.
5. A lateral view of a young gonophore.
6. A terminal view of the same.
7. A more advanced gonophore.
8, 8a. Detached gonophores. (See Pl. XII, fig. 13.)
9. A very young Velella (Rataria).
10. Its pneumatocyst.
11. Part of its disc.
12. A portion of its crest.
13. A portion of the disc of a larger specimen.
14. A still more advanced specimen.

PLATE Xi

J Baswire gc

VELEELA.

TBE adtnat del.

hy tie, .
7 * ae .

Ue ie ae i aa

q y . 7
4 ty ., ae ating oy bra “ei

: Sie ov eit” ‘Dana edaiay tl ot sutipyleonk wir
a dyplsity walt doe oe ul vi nog uta a vowimaan |
baiky: wide yA 1 ea Wi hole Ayal é Dit thor) gilt die
ss wpe ts ¥ “ iy ,
indanyp op) Tata) Ayer ne > eadgiled, cone ote a .
‘o Tp owning otf Yo nih 98, ; onlioton ae

sid it | ubionzien ben sir "ieee j My
-* mi is {ri ub in; HAM Lery eS a A e

. 5 a et ity Ww: : ao hnpyh Hyp) A at :
ma ins, 4 Ae ne 4! Wr al mili BS ndba). avec an etarmpag Ty",
7 ss ry log 4 Foopiye avabay we Tie vtheodtes ee Leap rae

ha i iw Be iis J ipo oir whey ifivy wh el ery Dy »!
6 lice estylanidued ome VIW bron au} Ti) nal Yo | Wi iz aurg . '

a * fi 4 7
3 Oni can ." (4 wit in sgl te Phin he a wail *
| Nyt ones, MWiisieleay =
‘on Gylt te wit nx 13 sng
ry xt i Ne) sent m
uy Af 'y 1 ‘wil
rie S Heys Le

ide

ae sit ka wale ral Noha) aie ~

ef ever Ory x" ae A le re) Viol on ie Lael aie gine
“a
Hog: Midi) / i ie mien eahby shed \ eyliinlay

, meee

sista BaPA\ lua HS i aii i a ate ey otny} i es hap

ufo CR iY ym nme iy! wii be YORE lca ts ys Palle TTI) val res yy
lout qa ; ‘ag 1o% saa a iu hus i rf A \ f) tu2)d iw tis arid At pray
; a ba , ' ; iy; R >) ere» | ” be m i mits, 1) phy ie
A : iveye iff) ‘i wil! Ve ul Lada :

ae teal ae led
Mii) eit PMT) 2 ab ivel me
Auabiy le vi) ail } ® dati) 4) ' tr) P) «
v €
al my ees:
a atten pre” oly . jess oF (lang Hu) wis ae

ay A F ” Laeo wiry? ® entity ig a! A
. . ¢
Lae ae es wren Aes

; 7 sd : ‘J % $2, aylit
y, 0 ar Na bl) vi pang * “1% vd aif eta sae
Reins itl ' An \ a ¥ Agate) hats am
7 > ae j
4 Shy Wy OMe i, ¥ Rire fi 0 orn) sunt ‘hy iy
tial? 5 Fayed pid Fo 4ueel lawton Oj dina in
» iytion Tulbiitnlioaeg A one hon Lrrtime Ay inev vias eo 4 T
a ¢- pheiliy otylecto 7 Hine 6). Dh he ig ;
7 his hs
im me,
: at *
; ; - a

2

om’
=», A

PLATE XII.

1. Galeolaria filiformis, magnified (after Gegenbaur). 4, proximal, 8, distal nectocalyx ;
a, somatocyst; 4, nectocalycine canals; c, aperture of the proximal, d, of the
distal nectosac; e, valvular projection at the aperture of the distal nectosac ;
J, coenosare, with the polypites; 2, nectosac of the proximal, 7, of the distal necto-
calyx.

. Praya maaima, slightly enlarged (after Gegenbaur). 4, B, the nectocalyces; a, a, their
nectosacs; 4, origin of the coenosarc, ¢; ©’, c’, nectocalycine canals; e, polypites,
with their hydrophyllia and tentacles.

3. A single zéoid viewed laterally (after Gegenbaur). a, hydrophyllium; 4, gonocalyx and

manubrium ; ¢, polypite; d, tentacle.

4. Hippopodius neapolitanus (after Kolliker). a, perfect, 4, undeveloped, nectocalyces ;
c, nectosacs ; d, coenosarc, with six undeveloped polypites; e, fully developed poly-
pites, near which gonophores, which are omitted, were seated ; /, tentacles.

. The proximal end of the ccenosarc, with some nectocalyces still attached (after Leuckart).

. Vogtia pentacantha (after Kélliker). a@, nectocalyces; 0, polypites; ¢, androphore ;
d, gynophore ; e, tentacles.

i]

oS Or

7. Different views of the nectocalyces. a, aperture of the nectosac.

8. Apolemia waria (after Gegenbaur). The proximal end of the hydrosoma, about three
fifths of the natural size. a, pneumatophore; 4, hydrocysts among the necto-
calyces; ¢, coenosarc; d, hydrophyllia; e, hydrocyst ; f, tentacles ; g, polypites.

9. Halistemma rubrum (after Kolliker); less than the natural size. a@, pneumatophore ;
6, young nectocalyces ; c, fully formed nectocalyces ; d, e, coenosarc; f, hydrophyllia ;
gy, polypites; /, hydrocysts ; 7, tentacles of the polypites ; 4, tentacles of the hydro-
cysts.

9a. The end of a tentacular branch. a, the pedicle; 4, the sacculus ; ¢, the filament.

10. Forskalia Edwardsii (after Kolliker). a, pneumatophore; a’, young nectocalyces ; 4, 4’, ¢,
cylinder formed by the nectocalyces, and traversed by the coenosarc,7; d, poly-
pites ; e, hydrophyllia; g, hydrocysts ; /, tentacles.

11. The end of one of the full-grown tentacular branches of Physophora Philippi, much
magnified (after Kélliker). @, the pedicle ; 4, the outer layer of the mvolucrum ;
c, its middle, and d, its inner layer; e, its cavity, in which the sacculus les, and
which opens at /; /, the free end of the sacculus; gy, a muscular band fastened to
its other end.

12. Physalia (utriculus)? Natural size (original). a, crest ; 4, c, pneumatophore and cceno-
sarc; d, its processes ; e, the polypites; /, velvetty masses; g, tentacles; /, tenta-
cular sacs.

13. Chrysomitra striata; about twice the natural size (after Gegenbaur). a, gelatinous
substance of the nectocalyx.

14. Porpita mediterranea, seen from below (after Kélliker). a, the central polypite ; 6, b, 6,
small or gonoblastidial polypites ; ¢, c, small tentacles ; d,d, large tentacles; e, mar-
ginal limb. ;

15. A perpendicular section through the same (after KOlliker). a@, pneumatocyst, with its
chambers; a, central chamber; 4, proximal layer of the hydrosoma; ¢, limb ;
d, small, e, large, tentacles; f, liver; g, central polypite ; 4, gonoblastidial polypites,

/

with the gonophores, 4’; 2, renal organ, or white plate.

~~
i 4

TATE, SOUT

%
Ba AB ey lM via!
SAD AF nigh nee

Me

get athe of ‘wat

*

4, a ngoagy, . gar teanugy
", SAF e ag cer tetany, < Lf
Hiya shy ag eee tena eo hu

ae
Seat a ee a? thy ay ."
Sg a ay fut fee eet tae Sacre hy

S Basie.

Hae ‘

f
bi
ats
bY
oO]

~

+) : N HANG
A / sy a

oe Nt bare hn