^m>..*

\

DEPOSITED IN

THE LIBRARY OE

THE BIOLOGICAL LABORATORIES

A MANUAL

SUB-KINGDOM

CGELENTEEATA.

GALBRAITH AND HAUGHTON'S MANUALS OF SCIENTIFIC AND
POPULAR NATURAL HISTORY.

In fcp. 8vo. with 66 Woodcuts, price 7s. 6d. cloth,

MANUAL of GEOLOGY. By Samuel Haughton,
M.I). F.R S. Fellow of Triu. Coll. and Prof, of Geology in
the University of Dublin. Second Edition, tlioroughly
revised and enlarged.

:\IANUAL of the METALLOIDS. By J. Apjohn,
M.I). F.R.S. Edited by J. A. G-albraith, M.A. and
S. Hat;ghton, M.D. Revised Edition, with 38 Woodcuts.
Fcp. 8vo. 76-. Qd.

MANUAL of the COKALS and SEA JELLIES

(Coclenterafa). By J. Reay Greene, B.A. M.R.I.A. Edited
by J. A. GALnuAiTH, M.A. and S. Haughton, M.D. Re-
vised Edition, with 39 Woodcuts. Fcp. 8vo. 5s,

:\IANUAL of the SPONGES and ANIMALCULES

(Profocoa) : with a General Introduction on the Principles
of Zoology. By the same Author and Editors. Revised
Edition, with 16 Woodcuts. Fcp. Svo. 25.

London : LONGMANS and CO. Paternoster Row.

GALBRAITH & HAUGHTON'S SCIENTIFIC MANUALS.

Experinie)ifal and Natural Science Series.

MANUAL

CCELENTERATA.

JOSEPH EEAY GREENE, B.A.

PROFESSOn OF NATUBuVL HICTORY IN THE QUEEN'S COIJ.EGE, CORK

&C. &C.

NEW IMPRESSION.

LONDON :
LONaMANS, GKEEN, AND CO.

1869.

LONDON: PBINTKD BY

SPOTTISWOODB AND CO., NEW-STREET SQUAHB

AND PAELIAMBNT STEIEI

MANUAL

S U B - K I N G O O M

CGELENTEKATA.

JOSEPH REAY GREENE, B.A.

PROFESSOR OF NATURAL HISTORY I.V THE QUEEN'S COLLEGE, COKK

&C. &C'.

NEW IMPRESSION.

LONDON :
LONGMANS, GREEN, AND CO.

1869.

Ll^l

BIOLOGICAL LABORATORIES LIBRARY

HARVARD UNiVERSITY

PREFACE.

' The house that is a-building looks not as the
house that is built.' The present Manual, though
now issued as complete, is, in truth, but the
abridgment of part of a larger work which the
Author trusts may yet one day see the light.

The general arrangement of the subject-matter
here devised does not seem to require any expla-
nation. Had the Author sought to evade those
delays and difficulties with which, in almost every
paragraph, he has found it his duty to contend,
another, and far easier, plan might have been
chosen. A thoroughly scientific method seemed,
however, more likely to prove useful. And, in
the discussion of questions hitherto considered
unsusceptible of general treatment or, perhaps,
insufficiently known to men of science themselves,
he has not endeavoured, by the invention of diffi-
culties which in nature have no existence, to hide
truth beneath the patchwork veil of a meagre
quasi-originality. Rather has it been his wish to

viii PREFACE.

unfold, in the clearest and simplest language at
his command, phenomena which, as a student,
lie has himself earnestly striven to comprehend.
Keenly, indeed, does he regret the deficiencies of
style and want of artistic combination which, but
too frequently, it is feared, will be found to mar
his pages ; believing, as he does, that for the in-
terpreter of nature there is a standard of literary
excellence not less high than that of the poet or
historian.

In the select bibliographical list appended to
the end of the Manual will be found the names
of those writers from whose published works has
been derived that assistance which the Author
would now, gratefully, acknowledge. In particular
to Professor Huxley are his best thanks due, for,
without access to the original memoirs of that
naturalist, the second chapter, on the Class Hy-
drozoa, could never have been rightly completed.
But the Author must confess himself under deeper
and less formal obligations to the same philosophic
investigator, whose rich and suggestive seeds of
thought could not, from their nature, fail to fall
fruitfull on the soil of any patient mind.

From Professor Allman, also, who has done so
much to promote a right knowledge of the Coelen-
terat'a, the Author has not been denied kind aid.

PREFACE. IX

And of foreign naturalists, personally unknown to
him, he would especially single out, for courteous
thanks, Professors Gregenbaur, E. Leuckart, Milne
Edwards, and Agassiz.

To Mr. Grosse the Author is indebted for the
loan of the beautiful drawings from which two of
the woodcuts have been copied. The wood-en-
graver, Mr. William Oldham of Dublin, has exe-
cuted his share of the following pages in by no
means an unsatisfactory manner.

Mr. Busk, the Eev. Thomas Hincks, and Dr.
Strethill Wright have also supplied the Author
with some valuable facts touching the structure of
the fixed Hydrozoa.

Queen's College, Cork ;
April, 1 86 1

August, 1 861.

Professor ]Max Schultze has just published a
memoir on Hyalonema in which he confirms the
opinion that the beautiful siliceous fibres of this
organism are, in truth, to be regarded as spicules
of a Sponge, allied, in some respects, to Euplec-
tella.

Very recently. Professor Agassiz (op. cit. (71)
p. 256), from personal examination of the living
animal of Millepora, has concluded that the

X PREFACE.

entire division of Tabidata, and, perhaps also,
the Rugosa, can no longer be associated with the
undoubted Actinoid polypes, but find, rather, their
true place in the neighbourhood of the genus
Hydractinia. The details of these observations
having not yet fully appeared, it seems premature
to adopt the important systematic change thereby
indicated.

CONTENTS,

CHAPTEE I.

THE SUB-KINGDOM CCELENTERATA.

Page
I. General characters.— 2. Classes • . • .3

CHAPTER II.

THE CLASS HYDROZOA.

SECTION I.

MORPHOLOGY AKD PHYSIOLOGY OF HYDROZOA.

I. Typeof the Class: Hydra.— 2. General Morphology. — 3. Or-
gans of Nutrition, — 4. Prehensile apparatus. — 5. Tegu-
mentary Organs. — 6. Muscular System and Organs of
Locomotion. — 7. Nervous System and Organs of Sense. —
8. Reproductive Organs . . . . 2C

SECTION II.

PEVELOP31ENT OF HYDROZOA . . . . . 5I

SECTION in.

CLASSIFICATION OF HYDROZOA.

I. Classification. — 2. Order i: Hydridae. — 3. Order 2: Cory-
nidae. — 4. Order 3 : Sertularidae. — 5. Order 4 : Calycopho-
ridae.— 6. Order 5 : Physophoridse. — 7. Order 6 : Medusidse.
— 8. Order 7 : Lucernaridae . . . . '79

XU CONTENTS.

SECTION IV.

DISTRIBUTION OF HYDUOZOA.

Page
I. Relations to Physical Elements. — 2. Bathymetrical Distri-
bution.—3. Geograpki^ml Di«tributioa . . .126

SECTION V.

RELATIONS OF IIYDROZOA TO TIME . , • . I30

CHAPTER III.

THE CLASS ACTINOZOA.
SECTION T.

MORPHOLOGY AND PHY'SIOLOGY OF ACTINOZOA.

1. T^'pe of the Class : Actinia. — 2. General Morphology. — 3.
Organs of Nutrition. — 4. Prehensile apparatus. — 5. Tegu-
mentary Organs. — 6. Corallum or Skeleton. — 7. Muscular
System and Organs of Locomodon. — 8. Nervous System
and Organs of Sense. — 9, Reproductive Organs 131

SECTION 11.

DEVELOPMKNT OF ACTINOZOA , ... I70

SECTION III.

CLASSIFICATION OF ACTINOZOA.

I. Classification. — 2. Order i : Zoantharia.— 3. Order 2: Alcyo-
naria.— 4. Order 3 : Rugosa. — 5. Order 4: Ctenophora . i96

SECTION lY.

DISTRIBUTION OF ACTINOZOA.

I. Relations to Physical Elements. — 2. Bathymetrical Distri-
bution.—3. Geographical Distribution . . . 231

CONTENTS.

Tn\-i

SECTION V.

RELATIONS OF ACTINOZOA TO TIME.

I, General History of A ctinozoa. — 2. Histor}' of Zoantharia.-
3. History of Rugosa. — 4. History of Alcyonaria. — 5. Silu-
rian Corals. — 6. Devonian Corals. — 7. Carboniferous Corals.
— 8. Permian Corals. — 9. Triassic Corals. — 10. Jurassic
Corals. — II. Cretaceous Corals. — 12. Tertiary Corals. —
13. Recent Actinozoa

Page

Bibliography of the Ccelexteuata
Questions on the Ccelenteuata
List of Illustrations .
Index . . . • .

236

249
257
261
263

THE

SUB-KINGDOM

y CCELENTERATA.

^ y:

COELENTERATA.

CHAPTEll I.

THE SUB -KINGDOM CCELENTERATA.

I. General characters. — 2. Classes.

I. Creiipral characters. — The animal forms
included under the sub-kingdom Coelenterata pre-
sent modifications of a type of structure better
marked than that which is characteristic of the
Protozoa. All are furnished with an alimentary
canal, freely communicating with the general, or
somatic, cavity. The substance of the body consists
essentially of two separate layers, an outer, or
' ectoderm,' and an inner, or ' endoderm.' These
two membranes, but especially the former, are in
general provided with cilia.

Another distinctive characteristic of the Coelen-
ierata is found in the presence of the peculiar
articating organs, or ' thread-cells,' which are met
with so constantly in the integument of these
organisms [Jig, i).

Thread-cells, for which the term ' cnid?e ' has
been proposed, usually occur as colourless, trans-
parent, elastic, double-walled sacs, rounded or

B 2

4 THE SUB-KINGDOM CCELENTEIIATA.

oval in form, and containing a fluid in their
interior. Tiie outer wall of the sac is entire and
very delicate ; the inner one is much stronger, having
its open extremity produced into a stout, rather
fusiform, sheath, Vv^hich terminates in a long thread,

Fig. r.

TJrticating organs of Ccelenteeata: — a, e, and/, thread-eeils of
Caryo'phyllia Smithii ; h, tliread-cell of Corynactis Allmani ; (\
portion of the marginal canal of Willsia stcllata, with peculiar
receptacle, containing thread-cells, arising therefrom ; d, a single
thread-cell of the same ; g, thread-cell of Actinia (or Bunodes)
crassicornis. {AM magnified.)

or 'ecthorseum.' A number of barbs or hooks are
sometimes disposed spirally around the sheath,
the ecthoraeum itself being often delicately ser-
rated. In the ordinary condition of the thread-

THE SUB-KINGDOM C(ELENTERATA. 5

cell the ecthoraeum lies twisted in many irregular
coils round its sheath; the barbs of the latter
being closely ajopressed to its sides, while it com-
pletely fills up the open end of the inner sac, into
whose interior it projects. Under pressure or
irritation, the cnida suddenly breaks, its fluid
escapes, and the delicate thread is projected, still
remaining attached to the sheath. So quickly is
this done that the eye can by no means follow the
process, but, in all probability, a complete ever-
sion of the cell's contents takes place. In some
cnidae the presence of a sheath has not yet been
discovered.

Thread-cells vary much both in form and size.
They are unusually large in the Portuguese Man-
of-war (Physalia), where they are spherical in
ligure and attain a diameter of -003 of an inch.
The relative dimensions of the thread and cell
also vary. Sometimes the ecthorasum is scarcely
longer than the sac; in other cases itt* length is
nearly iifty times as great.

The disagreeable stinging sensations experienced
when the human skin is brought into contact with
the bodies of some Coelenterata is, by most zoolo-
gists, attributed to the influence of the thread-cells.
It is supposed that the irritation is in part me-
chanical, arising from the friction of the fl lament
or its sheath, and in part chemical, from the
assumed poisonous nature of the fluid contained
within the cell. The ease with which many
Coelenterate animals seize and, as it were, para-
lyze their struggling prey, is also ascribed to the
same agency. These stinging propensities were
evidently known to Aristotle, who refers to dif-
ferent forms of the present group imder the name

B 3

6 THE SUB-KINGDOM CCELENTERATa.

of aicaX7]<f)ii, a term understood by some modern
naturalists in a more restricted signification.

A few of the Coelenterata are microscopic, but
l)y far the majority are of appreciable size, and
some attain considerable dimensions. Multiplica-
tion by gemmation is of common occurrence among
the members of this sub-kingdom, and when, as
frequently happens, the growths thus formed re-
main permanently in connection with the organ-
ism from which they originally sprouted, it is
evident that this process, repeated several times,
may give rise to aggregate masses, the limits of
which it is not possible to define. In form the
Ccelenterata vary considerably, presenting, in many
cases, an external resemblance, sufficiently re-
markable, to certain members of the vegetable
kingdom.

The Ccelenterata possess no proper blood-vascu-
lar apparatus, distinct from the somatic cavity or
its processes. The cilia which line the endoderm
promote by their motion the circulation of the
nutritive, or somatic, fluid occupying the general
cavity of the body, and, in like manner, respiration
is effected by the cilia of the ectoderm. Both of
these ciliary movements are assisted by the con-
tractions of the body walls, within which muscular
fibres may, not unfrequently, be observed. In-
dications of a nervous system and organs of sense
have been met with only in a few instances. Other
structures, whose function would seem to be se-
cretive, are not, however, wanting. Most of the
Coelenterata are provided with prehensile append-
ages, or ' tentacula,' and, in many of these animals,
special organs, adapted for locomotion, are super-

THE SUB-KINGDOM CCELENTEEATA. 7

added. Throughout the entire of this department
the elements necessary for discharging the
function of true reproduction would appear to be
present.

The power of emitting a phosphorescent light
is eminently possessed by several Coelenterata.
This is more especially seen among the oceanic
species which, together with Noctilv.ca, and other
floating organisms, serve to produce the luminosity
of the sea.

The Coelenterate organism, therefore, has not
only a plan of structure, or relative position of
parts, peculiar to itself, but, viewed also as a mere
animal machine, is seen to be, physiologically, in
advance of the Protozoon. A comparison of the
ultimate morphology of the two groups may serve
further to elucidate this proposition.

The body of the Protozoon, as elsewhere we have
sho^vn, consists chiefly of the elementary tissue
known as sarcode, or animal protoplasm ; a soft,
often transparent, elastic and extensile substance,
albuminous in composition, and presenting the
faintest traces of organisation.

The sarcode body is also remarkable for the
manifold diversities of outward form wiiich it may
assume, though in many Protozoa there is little
which deserves the name of integument, and an
inner cavity, whether it exists under the form of
contractile vesicle or alimentary track, is rudi-
mentary in the highest degree. Some authors
consider the Sponges as Coelenterate, but the
aquiferous system of these animals, however other-
wise it may appear, is, in truth, lined by the outer
surface of the organism.

B 4

8 THE SUB-JvINaDOM C(ELENTERATA.

Nevertheless, the homogeneity of the primi-
tively simple sarcode is liable to become diver-
sified by the two processes known as 'vacuolation'
and * fibrillation.' By vacuolation, clear spaces
and granules arise in its substance, of which ex-
a,mples are furnished by Actinoplivys and the
Gregarince ; by fibrillation, the same tissue may
dispose itself in definite lines, as in the so-called
stem muscle of Vorticella, and perhaps also the
cortical investment of Tethya. Other structures,
still further differentiated, are also seen to occur,
as the nucleus, pigment-masses, reproductive ele-
ments, and the various kinds of cellseform bodies.
But no true nervous or muscular tissues are pro-
duced, although these creatures manifest, in an
humble manner it is true, some amount of con-
tractility and sensibility.

An attempt miglit even be made to arrange
the several forms of Protozoa in an artificial
ascending series, the successive steps of which
Avould ditfer in the relative degree of distinctness
between the body-substance proper, and the outer
portion, or conventional integument, to which it
may give rise. In the lowest members of the
group, as Dujardin has remarked, this external
investment resembles nothing so much as the film
which forms on the surface of flour paste when
left to cool. Here pseudopodia are readily
emitted from all parts of the body ; but in Paim-
phagus, which, like Amoeba, is naked, they are
protrusible from one extremity only, the general
surface of the body acting, as it were, the part of
a more consistent membrane. PVom Pamphagus
to Difflugia, and thence to the higher Rhizopoda,
in which the outlying portions of the sarcode

THE SUB-KINGDOM CCELENTEBATA. 9

curiously differ, in their greater mobility, want of
colour, and feebler tendency to undergo histolo-
gical change, from the more highly vitalised bod}^-
mass within, it were not difficult to effect a natural
transition. In the Sponges structural relations
akin to those just mentioned are still more easily
to be traced. In the Gregariiice an external en-
velope becomes sufficiently distinct from the
granular or vacuolated protoplasm which it
bounds. And in the Infusoria the more contrac-
tile body-substance not merely serves to enclose
a softer sarcode, but is itself protected by a cuticu-
lar covering, on which the styles and cilia are borne.

But the changes which the sarcode substance
undergoes are not simply structural or mechanical.
Other modifications, of a more purely chemical
nature, may either accompany or replace the pro-
cesses above mentioned. Thus, by * conversion '
into horny matter, the fibrous skeleton of the
Sponges, the manducatory apparatus of Chilodon
and its allies, the carapace of the Arcellina and
Infusoria, and perhaps even their cilia, appear
to be produced; or, by 'deposition' of mineral
particles, withdrawn from the envii'onment, shells
and other hard structures have their origin. Nay
more, the diverse forms of Protozoa have the
power of appropriating certain elementary matters
to the exclusion of others. The Polycystine
sculptures its own siliceous shell ; the Foramini-
fer, living beside it, a calcareous one, not less
complex or beautiful ; while from various parts of
the body of the same Sponge a corresponding
diversity of curiously wrought spicules may be
obtained.

Thus, the naturalist, first struck by the varia-

10 THE SUB-KINGDOM CCELENTEKATA.

tions in external aspect presented by the Protozoa,
does not find his astonishment lessen when he
begins to contemplate the manifold endowments
of which each definable form is, as it were, the
index, and perceives the fundamental sameness of
organisation on which these complexities are based.
All this, however, reflection should have led him
to expect. For the body of every Vertebrate
animal was once of as simple a structure as that
of the Protozoon, and might even be said to cor-
respond with it. But the life-history of the former
plainly shows what it is capable of becoming. Some
would add that the vital nature of the two organisms
is less dissimilar than morphology and develop-
ment would seem to indicate, and that those ener-
gies, which the lower animals spend so rapidly in
acquiring the many outward modifications by
which they are soon distinguished, might, if duly
husbanded, and turned in another direction, give
rise to very different structural products. Such a
speculation is not wholly unworthy of mention.
At present its discussion would lead us too far
into the wide region of conjecture.

Turning now to the sub-kingdom Coelenterata,
the members of this group are at once seen to dif-
fer widely from the Protozoa, in that their body-
substance resolves itself into the two layers already
mentioned under the names of ectoderm and en-
doderm ; the former serving the purpose of an
integument, the latter lining the large internal
cavity constantly present.

These layers are very similar, though not iden-
tical, in structure. Both consist of a number of
vesicular bodies, or * endoplasts,' embedded in a
h omogeneous matrix, or ' periplast.' The endoplasts

THE SUB-KINGDOM C(ELENTERATA. 11

of the inner layer are more closely applied to one
another, their size is somewhat larger, and their
contents more transparent, than are the same parts
of the outer layer. The chief difference between
the layers is, however, in mode of increase, the
ectoderm growing from within outwards, the endo-
derm from without inwards.

Even in Hydra, the lowest of Coelenterate or-
ganisms, these two primitive layers are readily
observable. But this animal exhibits, at certain
seasons of the year, a tendency to break up into
particles of a sarcode aspect, which retain for a
long time an independent vitality. Nor are such
amoeboid masses wanting in the tissues of higher
Coelenterccta. The significance of such facts should
not escape our notice, since, at least, they serve to
indicate the nature of the foundations on which
the house of life has been constructed.

But the body-substance of the Gcelentei^ata by
no means always presents that simple structure of
its layers which the above expressions might seem
to imply. Vacuolation and fibrillation here like-
wise perform their part, and the latter metamor-
phosis is often carried to such an extent as to give
rise to true muscular fibres, though these are not,
as in the higher animals, accompanied, in most
cases, by an evident nervous system. Thread-cells,
already described, the body-layers elaborate, as
also reproductive elements, pigment-masses, and
those other granular structures, which seem adapted
for secretion. By conversion and excretion, outer
growths are formed which serve either for support,
ornament, or protection ; while, by deposition of
calcareous salts, the beautiful internal skeletons
known as ^' corals " become variously produced.

12 THE SUB-KINGDOM C(ELENTEEATA.

Lastly, a wonderful diversity of external pro-
cesses, some from the ectoderm, either alone oi
in great part, others from the ectoderm and en-
doderm combined, are seen to arise ; and these
may subsequently multiply to an almost indefinite
extent, or, even separating from the primal or-
ganism, enjoy a brief but independent existence.

The student who has perused the account now
given of the general structure of the lower animals
is warned to be on his guard against the errors
which still but too widely prevail on this and
other kindred branches of histology. These errors,
for the most part, have their origin in the well-
known cell-theory of Schleiden and Schwann : a
theory which, when first announced by its dis-
tinguished promulgators, possessed, indeed, a high
dignity and utility, but in the hands of inferior
naturalists has tended not a little to check inde-
pendent thought, and to render obscure much
that, but for it, would have been intelligible. In
particular the view that the cell-nuclei, or endo-
plasts, constitute special originating centres of
vital activity, is worthy of all reprobation ; con-
tradicted, as it appears to be, by so many careful
observations on the various modes of development.
Cells indeed exist, but only as the differentiated
products of a primitively homogeneous protoplasm,
not as morphological or physiological entities. To
borrow the fine metaphor of Professor Huxley
(whose views on animal structure we have here
more than once sought to interpret and extend),
" they are no more the producers of the vital
phenomena, than the shells scattered in orderly
lines along the sea-beach are the instruments by
which the gravitative force of the moon acts upon

THE SUB-KINGDOM CCELENTERATA. 13

the ocean. Like these, the cells mark only where
the vital tides have been, and how they have
acted."

A general survey of the development of Coelen-
terate animals is best deferred till definition has
first been given of their classes. Here, as in other
sub-kingdoms, too little attention has been paid
to vital changes succeeding what is called the
" adult " condition, more especially to those which
immediately precede the death of the organism.
Such phenomena, in the present instance, would
possess a peculiar interest ; for individuality, the
distinguishing characteristic of living beings, is
displayed by many Coelenterata in so remarkable
a manner, that the phraseology by which, in the
case of the higher animals, we are wont to de-
signate its manifestations, cannot to the former be
applied without considerable qualification.

Excepting two fresh-water genera, all Coelen-
terata are marine. Few, if any, seas appear to
w^ant these animals, the several forms of which,
both fixed and oceanic, enjoy also a varied bathy-
metrical range. The coral-reefs, so widely spread
throughout the tropics, and the floating banks of
jelly-fishes, amid which ships have been known to
sail for days, testify, in like manner, to the abund-
ance of a group of beings, whose place in the
general economy of nature is not less extensive
than significant.

Equally numerous were the Coelenterata at
former periods of the earth's history, nor does
their wide distribution in space fail to find its
parallel in a long-enduring existence through
time. In the lower Silurian rocks Coelenterate

14 THE SUB-KINGDOM C(ELENTERATA.

remains occur, to reappear in every stratified de-
posit of importance intervening between that
epoch and the present day.

2. Classes. — Two leading modifications of
structure may be traced among Ocelenterate ani-
mals, which admit, therefore, of being arranged
under two principal groups or classes, the Hydro-
zoa and the Actinozoa,

In the Hydrozoa, the wall of the digestive
apparatus is not separated from that of the somatic
cavity, and the reproductive organs are external.

In the Actinozoa, the wall of the digestive sac
is separated from that of the somatic cavity by an
intervening space, sub-divided into chambers by a
series of vertical partitions, on the faces of which
the reproductive organs are situated.

In order to understand these relations aright,
it will be necessary to contemplate, from a genera]
point of view, the development of the Cwlente-
rata.

The scanty knowledge which we possess of the
life-history of the Protozoa has, in previous pages,
been sufficiently pointed out. From their known
simplicity of structure, it may, indeed, be conjec-
tured, that the changes which they undergo during
the course of development must be comparatively
slight. As already shown, the very existence of
reproductive elements has yet to be ascertained in
by far the greater number of the members of this
sub-kingdom.

In the other four departments, true reproduc-
tion, by contact of ova and spermatozoa, univer-
sally occurs. Within the nutrient mass, or ' yolk,

THE SUB-KINGDOM C(ELENTERATA.

15

composing the bulk of the ovum, may be perceived
a small cavity, the * germinal vesicle,' which, in
its turn, contains a stiU smaller particle, the ' ger-
minal dot ' {fig. 2, h). In addition to these parts,
many ova are provided with an outer envelope,
known as the yolk-sac or ' vitelline membrane.'
The spermatozoa vary much in form. More corn-

Development of CcELENTEEATA : — a, spermatozoa of Ccdentc-
rata; 6, section of Cceleuterate OYiim, with germ-vesicle and germ-
spot ; c, ovum after segmentation ; d, section of the same, more
advanced, showing its division into two layers ; e, longitudinal
section of typical Hydrozoon ; / and g, longitudinal sections of
typical Actinozoon, in different stages of development. (These
drawings are diagrammatic.)

monly they appear as delicate filaments, swollen
at one extremity into a somewhat oval body {a).

After fecundation, the ovum exhibits a series of
changes inaugurated by the process of ^ segmenta-
tion ' or yolk-division.

First, either the whole or a portion of the yolk

16 THE SUB-KINGDOM CCELENTERATA.

separates into two halves ; each of tliese, again,
into two others, and so on, until, finally, the seg-
mented yolk is resolved into an aggregation of
minute spherules, forming the so-called * mulberry-
mass ' (c). The interior of this mass, in a large
number of cases, liquefies, while its outer portion
constitutes the * blastoderm,' or ' germinal mem-
brane,' from which the several embryonic struc-
tures are produced.

Next, the blastoderm divides into two layers,
an outer, or * serous,' and an inner, or ' mucous '
(d). The outer layer more especially contributes
to the formation of the organs of animal life,
while from the inner layer is fashioned the first
rudiment of a large portion of the alimentary
canal, and various parts of the body with which
it is connected.

The preceding remarks apply to the ova of most
Coelenterate, Molluscous, Annulose, and Vertebrate
animals. In all of these, with the exception of
the Goelenterata, a further differentiation of the
blastoderm would seem to take place. Each of
its primary layers divides again into two others.
In the outer portion of the serous layer arise
the primitive rudiments of the nervous system,
while the inner surface of the mucous layer forms
the lining of the digestive canal. Between these
two structures lies the " membrana intermedia "
of embryologists, in the composition of which both
serous and mucous layers appear, therefore, to
take part. From this membrana intermedia the
greater mass of the organic systems which make
up the body of the adult animal are subsequently
developed. Very soon after the formation of the
intermediate layer, a number of important changes

THE SUB-KINGDOM C(ELENTERATA. 17

begin to ensue ; but it is sufficient here to state,
that, eventually, the principal nervous and vascular
trunks are found to occupy opposite aspects of
the body, the axis of which is traversed by the
alimentary canal. Thus in every Vertebrate, An-
nulose, and Molluscous animal it is possible to re-
cognise two distinct regions, a nervous, or * neural,'
and a vascular, or ' haemal.'

In the Coelenterata, on the other hand, no dis-
tinction between neural and haemal regions can
be noticed. Furthermore, whatever outward com-
plexity the organism may present, all its parts
are found on examination to be readily resolvable
into the two layers previously referred to as ecto-
derm and endoderm. These correspond, both in
structure and mode of growth, with the primitive
layers of the germ in the higher animals, so that
a general analogy may be traced between the
permanent condition of the Ccelenterata and a
well-marked embryonic stage in the Mollusca,
Anmdosa, and Vertebrata.

This is especially the case with the Hydrozoa,
in which class a body-wall, composed of ectodermal
and endodermal layers, invests the simple undivided
cavity which constitutes the whole interior of the
animal {fig. 2, e). The alimentary and somatic
cavities are, therefore, in these beings identical,
though, among many members of the group,
certain parts of the organism are more imme-
diately concerned in the performance of the di-
gestive function.

But in the Actinozoa an oral fold of the blas-
toderm grows inwards to form a distinct digestive
sac ; thus, as it were, suspended, in the general
cavity of the body, with which it communicates
c

THE SUB-KINGDOM C(ELE^^TEKATA.

by means of a wide inferior aperture {fig, 2, /
and g).

Some Actinozoa exhibit a further sub-division
of each of the two primary blastodermal layers
into other secondary membranes, foreshadowing a
structure so constantly met with among the higher
animals, just in the same manner as, in a few of
the more advanced stomatode Protozoa, an in-
distinct differentiation of the body into layers
indicates a condition w4iich is manifested, without
exception, by the immature forms of the four re-
maining sub-kingdoms.

Development of Animats.

The organism does not ex-
hibit true layers.

Protozoa.

A blastoderm is formed,
which divides into inner and
outer layers.

The alimentary canal freely
communicates with the somatic
cavity. There is no distinction
between neural and haemal
regions.

CCFXEXTERATA.

\

• The hffimal region is first
developed. The mouth opens
on the neural aspect. There
is no segmentation of the blas-
toderm.

MoLLrSCA.

The two layers of the blas-
toderm become further diffe-
rentiated. The alimentary
canal has no direct communi-
cation with the somatic cavity.
Distinct neural and ha:ual
regions appear.

I

The neural region is first^
developed. The blastoderm
may divide into segments.

The mouth opens on the
neural aspect, towards which the
limbs are turned.

AXNX^LOSA.

The mouth opens on the \
hpemal aspect, towards which
the limbs are turned. A pri-
mitive groove, dorsal and vis-
ceral plates, are formed.

Veu'iebrata.

THE SUB-KINGDOM CCELENTERATA.

19

Sub-kingdom CGELENTEKATA.

Animals whose alimentary canal freely communicates
with the somatic cavity.

Substance of the body made up of two foundation
membranes, an outer or ectoderm, and an inner or en-
doderm, which correspond, in mode of growth, with the
primitive layers of the germ.

No distinct neural and haemal regions. A nervous
system absent in most.

PecuHar urticating organs, or thread-cells, usually
present.

Class I. Hydeozoa.

Ccelenterata, in which the
■u-all of the digestive sac is not
separated from that of the
somatic cavity, and the repro-
ductive organs are external.

Class II. ACTINOZOA.

Calenterata, in which the
wall of the digestive sac is
separated fi-om that of the
somatic cavity hy an interven-
ing space, subdivided into
chambers by a series of verti-
cal partitions, on the faces of
which the reproductive organs
are developed.

20 HYDROZOA.

CHAPTER IT.

THE CLASS HYDROZOA.

Section I.

MORPHOLOGY AND PHYSIOLOGY OF HYDROZOA.

I. Type of the Class : Hydra. — 2. General Morphology. — 3. Organs
of Nutrition. — 4. Prehensile apparatus. — 5. Tegumental y Or-
gans.— 6. Muscular System and Organs of Locomotion. — 7. Ner-
vous System and Organs of Sense. — 8. Reproductive Organs.

I. Type of the class : Hydra. — The Hydra,
or fresh-water polype, is the t3^e of the class
Hydrozoa {fig. 2, e, and fig. 3). ^

The Hydra possesses a gelatinous, sub-cylin-
drical body, liable, from its contractility, to undergo
various mutations of form, having one end ex-
panded into an adherent disc, or foot, a mouth
being situated at its opposite extremity. The
mouth leads into a capacious cavity, excavated
throughout the entire length of the animal. From
the margin of the oral aperture, or rather, at a
little distance below it, arises a circlet of prehen-
sile tentacles, varying in number from five to
twelve, or even more. These are exceedingly
contractile, at one moment assuming the appear-
ance of delicate filaments, the next, shrunk up
into little wart-like knobs. Numerous thread-cells
are embedded in their substance, and project
freely from their surface. These are of two kinds,
one being much larger than the other. The

HYDROZOA. 21

larger thread-cells, which are fewer in number,
are further distinguished by the possession of a
sheath, surrounded b}^ three recurved barbs, and
terminating in a long slender thread (c). The
length of the Hydra, exclusive of its tentacles,
seldom exceeds three fourths of an inch.

More minutely examined, the body of the Hydra
is found to be composed of two membranes, an
ectoderm and an endoderm. The former consti-
tutes the outer layer of the animal, and has one
of its sides always exposed to the water wherein
the Hydra lives, the other side being in rather
close contact with the endoderm, whose free surface
forms the lining of the large internal cavity. The
tentacles, which open into this cavity, are tu-
bular prolongations of both the above membranes.
It has been asserted by Trembley that the body of
Hydra may be turned inside out, without thereby
sustaining any injury, or being checked in the
performance of its proper functions ; but this
experiment needs repetition. Both ectoderm and
endoderm are vacuolated, especially the latter, and
hence the well-known granular appearance which
the Hydra presents under the microscope. Some
describe the endodermal lining as produced into
a number of villous elevations, projecting into the
digestive cavity, and placed at right angles to its
axis. The thread-cells are chiefly developed in
connection with the ectoderm, and the numerous
nodules, filled with these bodies, which the ten-
tacles exhibit are merely enlargements of this
outer layer. The ectoderm of the tentacles (and
perhaps, also, of the body-wall) shows slight traces
of the presence of muscular fibres. Around the
margin of the mouth, the ectoderm and the endo-
c 3

22 HYDROZOA.

derm unite together, and " the junction between
the two is distinctly marked by a clear line."

The food of the Hydra consists, for the most
part, of minute fishes, crustaceans, worms, and
such other living creatures as come within the
reach of its tentacles ; and it is curious to observe,
how, by means of these apparently fragile cords,
animals are secured which would be deemed, at
first sight, superior to their captor in strength and
activity. There can, however, be little doubt, that
the tentacles are aided in the performance of
their prehensile function by the action of the
thread-cells, with which they are so well provided.
The elastic filaments of some of these are usually
projected into the body of the captured organism,
over whose motions they would appear to exert a
potent benumbing influence, to the production of
which the fluid contained in the interior of the
cells probably serves to contribute. For it has
even been observed that soft-bodied animals, which
succeed in effecting their escape from the grasp of
the Hydra, do not, in some instances, recover,
but, soon afterwards, die. The entire surface of
the tentacles is not at once brought into contact
with the body of their victim, and, in the use of
these organs, much instinctive caution is shown.
When sufficiently mastered, the prey is thrust into
the internal cavity, though the act of ingestion
does not, in all cases, immediately put an end to
its struggles. Grradually, the nutritive matters
contained in its body are imbibed by the Hydra,
all indigestible portions being finally expelled
through the mouth. But some writers describe a
short narrow canal, leading from the inner cavity
to a small aperture, situate in the centre of the

HTDROZOA. 23

foot, through which particles of an excrementitious
nature occasionally pass.

By means of its adherent disc the Hydra at-
taches itself to submerged stones, plants, and the
surfaces of floating sticks or leaves. It is not,
however, permanently fixed, but has the power of
effecting changes in its position, either by the
slow gliding motion of its base, or the repetition
of certain leech-like movements, in which both
tentacles and disc take part. Occasionally, the
disc is protruded above the surface of the water,
and, thus acting as a float, ena])les its possessor to
remain, for a time, in this suspended condition.

The reproductive organs of the Hydra do not
admit of being observed at all periods of the year,
seldom making their appearance before the ap-
proach of the cold weather of autumn. Their
position, however, is constant, the spermatozoa
being contained in conical processes of the body-
wall which arise close to the bases of the tentacles,
while the ova are enclosed in rounded elevations
of much greater size, and situated nearer the fixed
extremity (^fig. 3, d). Sometimes Hydrse are met
with in which only one set of reproductive ele-
ments can be detected. Not more than one of
the large protuberances, containing but a single
ovum, is usually developed at the same time, and
when two occur, they always arise from opposite
sides of the animal. The ovum, having pushed
itself through the body-wall, is seen to be invested
with a spherical envelope, brownish or rosy in
tint, and studded with a number of rough points,
which some writers describe as bristles. Mr. Han-
cock, however, more properly regards them as

c 4

24

HTDROZOA.

Fig.^,

Morphology of Hydea : — a, Hydra vulgaris, with a young
polypite sprouting from its side, attached to a piece of stick ; b,
portion of a tentacle ; c, one of its larger thread-cells ; d, H. vi-
ridis, on a fragment of an aquatic plant, with elevations contain-
ing spermatozoa arising near the bases of the tentacles, between
which and the attached extremity one side of the body-wall is
seen to be much distended by an ovum ; e, this ovum, ruptured ;
f, spermatozoa escaping from their receptacle, burst under pres-
sure, (a is about twice the uatiiral size ; the others are much
magnified.)

HYDROZOA. 25

minute cells or sacs, " probably composed of some
tenacious mucus with which to glue the egg to
any substance on which it may happen to settle ; "
an inference supported by the fact, that, soon after
the attachment of the egg, they disappear. Each
spermatozoon consists of an oval body, furnished
with a very delicate tail. The act of fecundation
most probably takes place after expulsion of the
ovum.

2. CJeneral Morphology. — Simple in struc-
ture as is the Hydra, it must, nevertheless, be
viewed as the representative of an extensive assem-
blage of animal forms, whose outward aspect is
singularly diversified, whilst, at the same time,
the utmost uniformity prevails throughout all the
more essential features of their organisation. In
every Hydrozoon, the wall of the digestive appa-
ratus coincides, or is continuous, with that of the
somatic cavity, and the entire body, or * hydro-
soma,' howsoever modified, is resolvable into ecto-
dermal and endodermal layers, processes of one
or both of which constitute, in like manner, the
manifold and complicated appendages, frequently
superadded {Jig. 5, h).

One end of the hydrosoma is always found to
increase more quickly than the other, which in
some cases, though not in all, is absolutely fixed.
The term ' distal ' is employed to distinguish this
rapidly growing end from the opposite, or ' proxi-
mal,' extremity.

In Hydra, and a few of its more immediate
allies, the hydrosoma consists, as has been shown,
of an alimentary region, or ' polypite,' together

26

HTDROZOA.

Morphology of Hydbozoa : — a, Hydrid; b, Corynid; c, Sertu-
larid ; d, Calycophorid ; e, Physopliorid ; /, Medusid ; g, Lucer-
narid; — ir, polypite; t, tentacles; k, ccenosarc; tt', hydro theca
or pol}^e-cell; v, nectocalyx or swimming-bell; a, pneumatophore
or float; v', umbrella. (These drawings are diagrammatic.)

with a ' hydrorhiza,' or adherent disc, prehensile
* tentacles,' and organs of reproduction {fig, 4, a).

HYDROZOA. 27

More frequently, however, it is composed, not, as
in Hydra, of a single polypite, but of several
similar structures, connected with one another by
means of a common trunk, or ' coenosarc' This
coenosarc may branch, presenting an erect tree-like
aspect, and in such cases is permanently attached
by means of the hydrorhiza which terminates its
proximal extremity (6, and^^.^r. 5, a). Often too, it
excretes from its outer layer a strong chitinous
investment, from which peculiar cup-shaped pro-
cesses, or ' hydrothecse,' serving as protective
envelopes for the delicate poljrpites, may be deve-
loped {^fi{). 4, c). In other members of the class
this firm layer has no existence, the coenosarc re-
maining soft, flexible, and highly contractile ; a
modification which prevails among the complex
oceanic Hydrozoa, creatures of great beauty and
delicacy of structure, whose graceful movements
through the element wherein they live are further
assisted by the * nectocalyces,' or swimming bells,
with which the hydrosoma may be provided
(d and e). In one group of these, the proximal
end of the coenosarc becomes transformed into a
peculiar organ termed the ^ somatocyst.' In
others, this same extremity expands to form the
' pneumatophore,' or float, which enables its pos-
sessor to remain without effort near the surface of
the water (e). There are, also, simple oceanic
Hydrozoctf whose hydrosoma is represented by a
single nectocalyx, from the under surface of which
a polypite is, as it were, suspended (/). Finally,
in another division of the group, the proximal
extremity of the polypite is modified so as to form
a special organ, the ' umbrella,' which usually
simulates the function of a nectocalyx, though its

28 HTDKOZOA.

structure and mode of development are very dif-
ferent {g).

In accordance with these several modifications,
the class Hydrozoa has been divided into seven
orders: Hydridce; Corynidce; Sertularidce; Caly-
cophoridce ; Physophoridce ; Medusidce ; and Lu-
cernaridce.

The Hydrozoa vary exceedingly in size. When
a coenosarc is present, it indicates a tendency on
the part of the organism to increase by a process
of continuous gemmation, and such forms often
attain considerable dimensions, though the sepa-
rate polypites are often so small as to be almost
microscopic. This is the case with most of the
complex fixed Hydrozoa, furnished with a branched
horny coenosarc. In the oceanic species, the poly-
pites are somewhat larger, yet, when the hydro-
soma consists of only one of these, its size is
usually inconsiderable. But, should the hydrosoma
develop an umbrella, its subsequent increase may
be extremely rapid, and, in this manner, the most
gigantic members of the class appear to be pro-
duced.

The animal fabric of the Hydrozoa is, perhaps,
best described as consisting of —

a. Organs of nutrition,

b. Prehensile apparatus,

c. Tegumentary organs,

d. Muscular system and organs of locomo-

tion,

e. Nervous system and organs of sense, and
/. Eeproductive organs.

3. Organs of IVutrition — In every Hydro-

HYDKOZOA. 29

zoon the entire somatic cavity may be said to
perform the functions of a nutritive apparatus
{Jig. 5, 6). But the true digestive process is
chiefly effected within the bodies of the poly-
pites.

Each polypite exhibits two regions, a distal, and
a proximal. The distal extremity terminates in a
delicate, more or less extensile, lip, which, in
the Calycophoridce and Physophoridce, becomes
everted and trumpet-shaped. Not unfrequently,
the lip is lobed, its lobes, usually four in number,
being, in some cases, very much prolonged (fig.
7. b).

In Hydra, and a few of the simpler forms of
Coin/nidcB, the proximal end of the poljrpite is
closed by the hydrorhiza, but throughout the re-
mainder of the class, it freely opens into the
somatic cavity.

jNIauy CcdycojyJwridce and Fhysophoridce have
the proximal or attached division of the polypite
produced into a more or less elongated peduncle,
beyond which may be recognised two distinct re-
gions ; a median, or gastric, and a distal, or oral.
The gastric cavity is separated from the interior of
the proximal region by a peculiar inward growth,
or * pyloric valve,' which is best seen among the
Calycophoridce. Professor Huxley, its discoverer,
describes it as "a strong circular fold of 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

30

Morphology of Cordylophora : — a, hydrosoma of Cordylo-
pkora lacustris, growing on a dead valve of the swan-mussel ; b,
longitudinal section of a portion of its ccenosarc, with a single
polypite ; c, another fragment of the same ccenosarc, to which
are attached, a polypite, and three reproductive bodies, or gono-
pliorcs, in various stages of development, the largest and most
advanced containing ova; — o, mouth; 5', cavity behind the
mouth ; 5, gastric cavity ; «', interior of the ccenosarc ; fv, endo-
derm ; /it, muscular layer; (k, ectoderm; €«', outer hard layer,
or polypary, excreted by ectoderm ; ex", processes of ectoderm,
connecting it with the inner surface of this hard layer. (All, ex-
cept a, magnified.)

HTDROZOA. 3 1

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 walls of the peduncle are thin and muscular.
Those of the gastric chamber are comparatively
thick, while its cavity is wider than any other
part of the interior of the polypite. In addition
to the cilia, which clothe its endoderm, the surface
of this la3'er is often elevated into a number of
villi, or conical processes, which in Physalia attain
a length of '01 of an inch. Such villi, or ridge-
like enlargements which arise in their stead, have
been observed within the polypites of many Hy^
drozoa {Jig. 5, 6). The coloured contents, occa-
sionally noticed in these, and similar elevations,
are regarded by some anatomists as the most
rudimentary form of hepatic apparatus. In Ve-
lella and Porpita, more certain indications of a
liver are presented by a dark brownish mass, which
arises in connection with the digestive cavity of
the large central polypite (jig. 21, a).

The nutrient matters elaborated within the
bodies of the polypites are finally transmitted to
the somatic cavity {fig. 5, h). Here they undergo
an imperfect sort of circulation, a phenomenon
the occurrence of which has been more especially
observed within the long coenosarc of Tuhularia
incUvisa. Currents have been seen to course up
and down the long stem of this Hydrozoon which
occasionally appear to flow through distinct tubes,
but these are nothing more than irregular cavities
produced by vacuolation of the endoderm {fig.
16, c, d, and e). A circulation of albuminous
particles also takes place mthin the peculiar

32 HYDEOZOA.

nutrient system of the Medusidce and Lucernaridce

{fig- 23).

The cavity of the coenosarc, or of the peduncles
of the polypites in connection therewith, may be-
come, as in Tuhularia, partially obliterated by
vacuolation, a process, however, which does not
seem to impair its vital efficiency.

Some have conjectured that the short canal
which penetrates the attached extremity of the
Hydra represents the coenosarcal cavity of the
higher Hydvozoa.

4. Preh ensile apparatus. — The tentacles, or
prehensile organs, which present so striking a
feature in the physiognomy of the Hydrozoa, vary
exceedingly, both in position and structure.

In the Hydridw, Sertularidce, and some genera
of Corynidce, they arise, in one or more circles,
either immediately around the mouth or at a short
distance below it {fig» 3, d). But, in other Coiy-
nidce, they spring from various parts of the walls
of the polypite {figs. 16 and 17), and one genus
of this group, Hydractinia, possesses, in addition,
long isolated tentacles, having an independent
oriofin of their own from the coenosarc. Sinsfle
tentacles of this kind are the only ones which
occur in the genera Physcdia, Velella, and For-
pita, among the Physophoridce (fig. 11, c). In
other Physophoridce, and all Calycophoridce, the
tentacles are inserted on the sides of the polypites,
about the junction of the gastric and proximal
divisions {figs. 20 and 22). In the Medusidce
and Lucernaridce y the tentacles usually surround
the open border of the bell-shaped swimming
organ {figs. 7, 23-25).

HTDROZOA. 33

Transverse section of a tentacle shows it to con-
sist of ectodermal and endodermal layers, enclo-
sing a process of the somatic cavity, which, in
very many cases, is wholly obliterated. The
ectoderm is often found to exhibit muscular fibres,
and is always provided with numerous thread-
cells, which may accumulate near its surface in
minute rounded papilla?, imparting to the tentacle
a roughened apjDearance (fig. 19, h).

Except in the Calycophoridcti and PhysoijJioridce,
the tentacles are very seldom branched. In some
genera of these orders the tentacles are as simple in
structure as those already described, but, in others,
they attain a much greater complexity. Each
tentacle of Fhysalia has a sac-like expansion at its
base, while numerous reniform enlargements, each
well packed with thread-cells, are disposed trans-
versely along its sides {fig. 11, d). Both these
and the sac communicate with a canal which runs
through the entire length of the tentacle. The
side opposite the reniform enlargements is bor-
dered by a wide muscular band, which connects
itself, above, with the basal dilatation. The reni-
form enlargements are, in other genera, replaced
by lateral branches, some of which present three
well-defined regions : a ' pedicle,' or proximal
slender portion; a ^ sacculus,' or "median divi-
sion, vrith one wall much thicker than the other,
containing numerous elongated thread-cells, ar-
ranged in transverse rows perpendicularly to the
wall, and flanked on each side by a longitudinal
series of larger oval thread-cells ; " and, finally,
a ' filament,' or '• terminal cylindrical thread, full
of large rounded thread-cells." Such, according

D

34 HYDROZOA.

to Huxley and Kolliker, is the structure of the
tentacular branches in Forskalia, and which,
under slight modifications, repeats itself in many-
other forms of Physophoridce and Calycophoiidw,
Within the sacculus of the last-mentioned order,
occurs a peculiar zig-zag muscular cord, known as
the ^' angel-band," the nature of which is but im-
perfectly known.

Where the pedicle and sacculus unite, a solid
process of the ectoderm has been observed to
originate in some Physophoridce. From its in-
vesting the sacculus in the form of a hood, this
orofan has received the name of ^involucrum.'

The mode of action of the tentacles, as appen-
dages for prehension, has been sufficiently explained
in our account of the Hydra.

The name of ' nematophores ' has been given by
Mr. Busk to peculiar csecal processes, distinct from
the oral tentacles, which are found on the coenosarc
of some Sertularidw. Like the rest of the coenosarc
these processes are invested with a stiff, horny,
layer, open at the distal end of the nematophore,
beneath which are embedded many large thread-
cells. The nematophores probably serve as organs
of offence. They are most numerous in the genus
Plumularia.

5. Te$cunieiitary Organs. — The tegumentary
system in the Hydrozoa is composed wholly of
the, in general, ciliated, ectoderm, and the rich
supply of thread-cells to which this layer gives
rise. Usually it appears more or less vacuolated,
or it may even become changed into a gelatinous
mass. The thickened disc of the Medusida' and
TAicernaridcv, in some structureless or but faintly

HYDROZOA. 35

cellular, in others has its homogeneous periplast
traversed in all directions by a complex mesh-
work of threads, which remain quite distinct
from the endoplasts about which they diverge, and
with whose processes they appear never to become
continuous. The threads themselves seem elastic,
transparent, of different diameters, frequently
dividing, soon to unite again, and, occasionally,
disposing themselves side by side in such a manner
as to form extended plates. On the convex aspect
of the soft mass, which this thread system streng-
thens, the surface of the periplast is broken up
into a number of polygonal cells, each furnished
with an endoplast ; and in the delicate epithelial
layer thus produced thread-cells may, without
difficulty, be observed.

In the Coryiiidce and SertulaHdce the ec-
toderm excretes a firm, structureless, cuticular
lamina, to which the name of * polypary ' has been
restricted by Professor Allman. This may so far
separate itself from the outer surface of the ec-
toderm as to present, at first sight, the aspect of
a distinct layer (^Jig. 5, 6,). In such cases its
connection with the ectoderm is maintained by
means of transverse processes arising from the
latter, and these may present {fig. i(),h) consider-
able regularity of arrangement. The cup-shaped
chambers, or hydrothecae, commonly known as
polype-cells, which are so characteristic of the
order Seriidaridce, are merely prolongations of
this excreted layer.

The polypites of the Ccdycophoridw and Physo-
phoridce are, in some genera, protected by over-
lapping appendages, termed bracts, or *hydro-

D 2

36 HTDROZOA.

phyllia.' These derive their origin from both
ectoderm and endoderm, though chiefly from the
former, and always enclose a ' phyllocyst/ or csecal
process of the somatic cavity.

The pigment-granules of the Hydrozoa are, in
general, of irregular form, and, though usually
found in the ectoderm, are by no means, as we
have seen, peculiar to it,

6. Muscular System and Orgasms of l^oco-
motion.— In connection with the more or less con-
ti^actile body-substance itself, separate muscular
fibres, whose arrangement is most frequentl}^ longi-
tudinal, present themselves among many Hydrozoa^
and especially in the highly contractile coenosarc of
the Calycojphoridce and Physophoridce. Similar
fibres may also be traced in the walls of the polypites
and tentacula, or on the concave surface of the
swimming organs, with which several of the oceanic
species are provided. They occur most abundantly
in the ectodermal layer (fig. 5, b). In the Medusidce
and Lucernaridce both radiating and circular fibres,
not without distinct indications of transverse stria-
tion, have, by different observers, been detected.

Special swimming organs, or nectocalyces, are
found in the Ccdycophoridw, Medusidce, and many
of the Physophoridce (figs, 20, 22, and 23). Each
nectocalyx is a cup, or bell-shaped body, the open
cavity of which is provided with a muscular lining,
and has received the name of * nectosac' Around
the margin of the nectosac, the wall of the
nectocalyx is produced inwards, forming a shelf-
like membrane, or ' veil.' By means of this mem-
brane, which is highly contractile, the aperture of
the bell may be more or less narrowed. Within

HYDROZOA. 37

the thickness of the roof of the nectocalyx there
occurs, also, a second cavity, from which issue four
or more canals, having no direct communication
with the uectosac, beneath the walls of which they
are prolonged until they reach a circular vessel
surrounding its margin (/i^. 23). The substance
of the nectocalyx consists chiefly of ectoderm, but
a continuation of the endoderm lines the ' necto-
calycine canals ' and the cavity from which they
arise.

The function of the nectocalyx is sufficiently
simple. By the rhythmic contractions of its mus-
cular lining, the water within the nectosac is ex-
pelled, and the organism moves in a contrary
direction.

The umbrella of the Lucernaridce bears some
resemblance to a nectocalyx, and, like it, may
perform the function of a natatorial organ. It
is easily distinguished by the absence of a veil :
its size, also, in the great majority of cases, is
much more considerable (figs, 7 and 25).

7. IVervous System and Or$;ans of Sense.

— The swimming organ, whether it take the form
of a nectocalyx or umbrella, is usually furnished
around its margin with a number of supposed
organs of sense, known as the marginal bodies.
For the best account yet given of their structure
and relations, we are indebted to the researches of
Will and Gregenbaur.

Two kinds of these bodies are found in the
Medusidce, — 'vesicles,' and 'pigment-spots.'

The vesicles are thin-walled, rounded or ovate,
sacs, lined internally with an epithelial layer, and
D 3

38 HTDROZOA.

containing one or more solid, motionless concre-
tions, immersed in a transparent fluid. These
concretions are oval or spherical in form, and
appear to be composed of carbonate of lime.
From the inner wall of the comparatively large
vesicle in Gei^onia arises a short stalk, which
expands to form a delicate membrane around the
solitary concretion. In some forms, a much
thicker covering invests, along with the concre-
tion and a number of minute molecules, a round
or oval body, not unlike an endoplast. Many
other modifications of the vesicles have been de-
scribed.

The pigment-spots, otherwise termed " ocelli "
and "eye-specks," consist of aggregations of colour-
ing matter, enclosed in distinct cavities. The
tint of these bodies is often extremely brilliant,
shades of yellow, red, and black being most pre-
dominant.

Oceania turrita is the only known Medusid in
which vesicles and pigment-spots co-exist (^i^. 23).

In the umbrella of the Lucernaridce, both
vesicles and pigment-spots seem to become united
into a single organ, termed the ' lithocyst.' These
marginal bodies are protected, externally, by a
sort of hood, and present often a very complex
arrangement. Most frequently they occur as
ovate vesicles, mounted on short, hollow stalks,
each of which communicates by means of a canal
with one of the radiating vessels of the umbrella.
Within, the vesicle is delicately ciliated, and en-
closes at its free extremity a broad, thin-walled,
oval sacculus, packed with a number of six-sided
crystalline prisms, obliquely truncated at each
end.

HYDROZOA. 39

Many zoologists describe the vesicles as "au-
ditory organs," the crystals or other bodies which
they contain being designated '''otolites." But
this view of their nature is altogether hypothetical.
Gegenbaur hints that they, perhaps, constitute an
apparatus of excretion. The pigment-spots may
be regarded with some shade of probability as the
earliest indications of organs of sight, which appear
amongr the lower forms of the animal kino-dom.
In some ocelli, a spherical, highly refractive cor-
puscle has been detected by Gregenbaur within the
mass of pigment.

It is by no means certain that a nervous system
exists in any of the Hydvozoa. Professor Agassiz
describes what he considers as such a system in
the nectocalyces of some of the free swimming
forms. " In Medusae (he writes) the nervous
system consists of a simple cord, of a string of
ovate cells, forming a ring around the lower
margin of the animal, extending from one eye-
speck to the other, following the circular chymi-
ferous tube, and also its vertical branches, round
the upper portion of Avhich they form another
circle. The substance of this nervous system,
however, is throughout cellular, and strictly so,
and the cells are ovate. There is no appearance
in any of its parts of true fibres." But the struc-
ture of the tissues here described as nervous is
very susceptible of a different interpretation.
M'Crady and Fritz Miiller have also speculated
on the presence of a nervous system in the Medii-
sidce. The former naturalist states that, among
several species, he has observed a distinct ganglion
in the neighbourhood of each marginal body.

D 4

40 HYDROZOA.

Among the varied appendages attached to the
coeno.'^arc of the PhysojjJwvkke occur certain pro-
cesses of doubtful nature, which Kolliker and some
other observers appear disposed to regard as organs
of touch (fig. 22). The * hydrocysts,' or feelers,
for so have these bodies been termed, bear some
resemblance to polypites in their structure and
mode of attachment, but differ from them in pos-
sessing caecal extremities, beneath which large
thread-cells are embedded. In some genera they
are with difficulty distinguished from polypites in
a young state of development.

8. Reproductive Organs. — Processes of -the
body-wall, within which are developed true gene-
rative organs, the ^ spermaria ' and ' ovaria,' con-
stitute the reproductive apparatus of the Hydrozoa.
These processes are always external, and are re-
markable for the interesting series of modifications
which they present among the several members of
the class {figs. 6 and 7).

In Hydra, as already shown, they are of the
simplest possible structure, differing from other
parts of the body-wall in their contents alone.
Such, also, is their aspect in Lucernaria, Pelagian
and, probably, all the MedusidcB.

In the Corynidw, Sertularidce, Ccdycophoiddce,
and PhysophoridWf the reproductive bodies appear
externally as distinct buds, or sacs, for which Pro-
fessor Allman has proposed the name of * gono-
phores.'

The simplest kind of gonophore consists of a
well-defined protuberance from the body-wall, the
* sporosac,' containing within its substance ovaria

HTDROZOA. 4 1

or spermaria, and euclosing a diverticulum of the
somatic cavity {fig. 6, a). The proximal ex-
tremity of this and other kinds of gonophore
usually becomes naxrowed into a short stalk of
attachment. Such A form of the reproductive
bud is of comparatively rare occurrence. Ex-
amples may, however, be found in some species of
Hydractinia, Coryne, and Clava.

To understand the structural modifications of
the gonophores in other Hydrozoa, it is necessary
to trace briefly the principal phenomena which
the more complex of these bodies present in the
course of their development.

All gonophores first appear as simple processes
of some portion of the body-wall, with its two
layers, the ectoderm and endoderm. Next, the
process becomes better defined, and exhibits a
peduncle or stalk. " When this process has at-
tained a certain size, its distal wall becomes
thickened, and projects as a sort of rounded boss
into the cavity, which, in consequence, becomes
cup-shaped. As the process enlarges, the upper
part of the cup-shaped cavity extends between the
rounded central boss and the outer wall, under the
form of four canals, which run up parallel with
the axis of the process, but stop short of its ex-
tremity. Their csecal ends then send out lateral
processes, so as to become T shaped, and ulti-
mately the lateral processes unite together, so as
to give rise to a circular canal uniting the ends of
the four longitudinal canals. Contemporaneously
with these changes, the axis of the boss becomes
hollowed out by a canal continuous with the
original cavity of the process, and like it lined by
the endoderm. A separation now takes place be-

42

HID BOZO A.

Q^. G

Fig. 6.

'-U^.f"

Eeproductive processes of H ydbozoa: — a — d, simple processes of
body-wall, variously modified, containing generative organs; e — m,
more complex series of reproductive bodies, each giving rise to a
gonocalyx : — ;u, portion of body-wall ; /w', wall of gonocalyx ; ,u",
inward projecture of m', or veil of gonocalyx ; k', simple diverti-
culum of the somatic cavity; >?, lateral prolongation of the same*
forming one of the gonocalycine canals ; k", cavity of manubrium ;
p, spermaria or ovaria. (These drawings are diagrammatic.)

HTDROZOA. 43

tween the central and peripheral portions of the
thickened boss, commencing at the distal ex-
tremity, and extending down very nearly to the
proximal end of the boss, so as to leave the thick
central portion, enclosing the central cavity, at-
tached to the thin peripheral portion (which re-
mains as the wall of the cavity of the bell) only at
its proximal extremity. In the perfect condition
of the zooid thus produced, the endoderm lines
the cavity of the peduncle by which it is attached,
the canals, and the central cavity of the suspended
axial body ; while the ectoderm forms the whole
of the outer walls of both natatorial organ and
central sac."

Still further changes are liable to ensue. The
central sac, or * manubrium,' may acquire a mouth
at its distal extremity, thus, as it were, transform-
ing itself into a polypite, while the natatorial organ,
or ^ gonocalyx,' enlarging, loosens its attachment,
and swims freely in the sea as a veritable Medusid
( fig. 6, in). Indeed, there is every reason to be-
lieve that a great majority of the organisms de-
scribed as Medusidce are, in reality, the detached
reproductive bodies of other Hydrozoa {figs. 13
and 14).

Such bodies, however, are more than mere
organs. Many of them, when first liberated, pre-
sent no distinct traces of generative elements,
pending the formation of which essential products,
their independent existence is of necessity pro-
longed. At this period they lead a very active
life, incre se rapidly in size, and eagerly devour
such minute marine animals as they are able to
secure. During the calmer seasons of the year
they abound in our seas, but before the approach

44 ETDROZOA.

of rougli weather usually disappear, their func-
tion having been, in all probability, previously
discharged. Yet nothing can be more perfect
than the series of transitional forms which estab-
lish the connection between these highly differen-
tiated organisms and the simple reproductive ap-
paratus occurring in Hydra. A few gradations
may be indicated. Thus, the closed gonophore of
Coixlylophora sends off from its manubrial cavity
a system of prolongations, evidently homologous
with true gonocalycine canals {fig. S, g). In
Tuhularia indivisa, fully developed canals are
exhibited by a distinct gonocalyx, but this never
becomes detached {fig. 9). Neither does it pre-
sent marginal tentacles, though even these sur-
round the fixed gonocalyces of Campaiiularia Ld-
veni {fig. 1 0). And so we at once pass to the free
swdmming generative cups of other Hydrozoa.
But it would be easy to dwell on further modifica-
tions. Plumularia pinnata, for example, has its
manubrium irregularly lobed, the lobes being, in
all probability, as Professor Allman suggests, in-
cipient gonocalycine canals, while in Campanu-
laria caliculata canals exist, though the manu-
brium itself is suppressed. And in some gonophores,
the canal system, at first easily recognisable, be-
comes obliterated by age.

A gonophore, therefore, may exhibit in its de-
velopment four distinguishable stages, which cor-
respond, respectively, to the permanent forms of
the reproductive body in particular members of
the group. These conditions are : —

I. That of a simple expansion of the body- wall,
as in Hydra.

HYDBOZOA. 45

2. That of a well-defined process, or sporosac, as
in Hydr actinia,

3. That of a manubrium with closed gonocaly-
cine investment, in which case the medusoid
structure is said to be " disguised," as in the
gonophores of Cordylophora and numerous other
forms.

4. That of a manubrium, with open gonocalyx
and well-developed canal system. Such " medusi-
form gonophores" may either remain attached,
as in Hippopodius and Vogtia, or become free,
as in Velella, Porpita, and many of the fixed
Hydrozoa,

The same gonophore does not contain more
than one kind of generative elements, and these
are situated either between the ectodermal and
endodermal layers of the manubrium, or in the
walls of the gonocalycine canals. When male and
female gonophores differ externally in form, the
special terms ' androphore ' and ' gynophore ' are
employed to distinguish them. But, apart from
such sexual distinctions, two kinds of gonophores
appear occasionally to be produced by the same
Hydrozoon, while, on the other hand, similar
gonophores may arise from the bodies of appa-
rently difi'erent species.

So much, then, for the structure of the gono-
phores ; next, as to their position. They may be
seated either —

1. on the polypites ; or,

2. on special processes termed *gonoblas-

tidia;' or,

3. directly on the ccenosarc.

The first of these methods is characteristic of

46 HTDROZOA.

the Ccdycophoridw, the second of the Physopho-
ridce and Sertidaridai, while all three find acces-
sible representatives in the order Corynidve.

In certain species of this last order the gono-
phores, even on the same individual, obey different
modes of attachment. Thus in Clava multicorniSf
some are inserted on the polypites, others on gono-
blastidia ; while in Hydractinia, besides the gono-
phores borne on the gonoblastidia, a few are found
to arise, without intervening support, from the
sides of the coenosarc.

The gonoblastidia are either simple or branched.
Often they present a curious resemblance to true
polypites, from which, however, they differ in
wanting a mouth, and having usually shorter ten-
tacula. Such polypoid gonoblastidia may be ex-
amined with ease in Hydractinia^ where they are
less than the polypites in size. In this genus the
free extremity of each is seen to end in a pear-
shaped, tapering enlargement, w^hose surface is
studded with minute conical swellings containing
thread-cells, which increase in size so as to re-
semble rude tentacles, ten or twelve in number,
around the largest portion of the pyriform pro-
cess. Beneath these the gonophores are borne.
At the base of the process is inserted a proble-
matical body, presenting the appearance of a short
stalk, which terminates distally in a rounded ex-
pansion, filled with very small, dark orange, masses
of pigment.

In general, gonoblastidia arise from the sides of
the coenosarc, though, in some cases, they are
attached to the bodies of the polypites.

A curious structural modification distinguishes
the oronoblastidia of the Sertidaridce. In Cam-

HYDROZOA. 47

panularia, for example, columnar gonoblastidia
arise in the angles between the stem and branches
of the coenosarc, or from the sides of the branches
themselves (Ags. lo and 19). The lower portion
of each gonoblastidium forms a sort of peduncle,
above which the cuticular investment of its ec-
toderm becomes separated as an urn-shaped cap-
sule, the 'gonotheca.' Such capsules, or "ovigerous
vesicles," are very variable and beautiful in form.
True gonophores, protected by the gonotheca, are
borne along the sides of its axial column.

In some Calycophoridce and Physojphovidce,
particular regions of the hydrosoma may devote
themselves to the performance of the reproductive
function, and, becoming separated from the rest
of the fabric, subsequently undergo a surprising
amount of modification.

Finally, in the Liicernaridce, with the exception
of Lucemaria and a few closely-allied genera, the
reproductive bodies are produced by fission from
polypites of almost microscopic minuteness ; and,
in their detached condition, grow with such ra-
pidity as ultimately to attain a weight of many
pounds, or even hundreds. A corresponding ad-
vance in structure attends this vast increase of
size. Each, at the outset of its free existence.
includes a complete transverse segment of the
polypite from which it has separated. This soon
forms a lobed swimming organ, or umbrella, with
the hooded lithocysts before mentioned. Fromi
the centre of the umbrella hangs a large polypite,
whose lips, in such genera as Aurelia, Cyanea,
and Chrysaora, form lobes of considerable length,
the folds of which serve as temporary receptacles

48

HYDROZOA.

for the ova during the earlier stages of their
development (Jig. 7, h). The interior of the poly-
pite leads to a large central cavity, situate in the
substance of the thick gelatinous umbrella, and

Fig. 7.

Oceanic forms of LrcEKNAEiDJE : — a, Ehizostoma pulmo ; b,
Chrysaora hysoscella ; c, its lithocyst. (All reduced.)

lined by a layer of endoderm which sends pro-
longations into the system of anastomosing canals,
communicating with a marginal vessel, fringed, in
its turn, by a series of tentacular diverticula. The
generative products are lodged in saccular pro-

HYDROZOA. 49

cesses of the lower portion of the central cavity,
immediately above the bases of the radiating
canals, and, being usually of some bright colour,
form a conspicuous cross shining through the
thickness of the disc.

But in RJiizostoma, Cephea, and Cassiopeia, sl
different arrangement prevails, which is best de-
scribed in the words of Professor Huxley.

"In the Rkizostomiike, a complex, tree-like
mass, whose branches, the ' stomatodendra,' end
in, and are covered with, minute polypites inter-
spersed 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 um-
brella, 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 re-
productive elements are developed within radiat-
ing, folded diverticula of the roof of this genital
cavity " {fig. 7, a).

This is, without doubt, the most complicated
structural product presented bj the class, and its
description forms a not inappropriate conclusion
to the preceding general survey of their organisa-
tion.

The majority of Hydrozoa are dioecious, the
same hydrosoma not bearing both male and female

E

50 HYDEOZOA.

reproductive bodies. Exceptions, however, occur
in Hydra itself, in Cordylojphora, in Plumularia
pinnata, in many Physophoridos and Calyco^
p)horidw, Diphyes being an exception. The re-
productive zooids of the Lucernaridw, except
in the case of Chrysaora, appear to be unisexual,
but it is not yet ascertained whether generative
bodies of dissimilar sexes can be produced by the
fission of one primitive hydrosoma.

As in other animals, fecundation is effected by
the contact of ova and spermatozoa : the sper-
maria and ovaria, when fully developed, becoming
wholly resolved into these essential elements.
The spermatozoa present the form of ovate cor-
puscles, from the broad end of which a filament
projects. The ova are, in most cases, spherical,
destitute of vitelline membrane, with distinct
germ-vesicle and germ-spot. Diffusion of the
spermatozoa in the surrounding water seems, in
the present class, the usual prelude to the act of
fertilisation. But, in Cordylophora, it has been
supposed that the male elements can alone obtain
access to the ova by reaching them from within
along the general cavity of the body.

HTDEOZOA. 51

SECTION II.

DEVELOPMENT OF HYDKOZOA.

The fertilised ovum, in all the HydrozocCy under-
goes yolk-division. This process would seem to
be determined by the previous division of the
germ-vesicle, which, according to Gregenbaur, in
some of these animals at least, does not disappear
immediately after fecundation.

The embryo which results may be developed
from the whole, or only a portion, of the vitellus.
It usually appears as a minute, free-swimming
ciliated body, but, in some instances, presents a
different aspect. The ectoderm and endoderm of
the adult Hydrozoon correspond with the inner
and outer layers into which the blastoderm of the
embryo soon separates, the cavity which is at the
same time formed representing the somatic cavity
of the future animal.

HydriDjE. — The modification of one end of
the body into a hydrorhiza, the formation of a
mouth, and of tentacular processes, are the only
changes, save those of growth, which seem needed
to bring such an embryo into the condition of a
perfect Hydra. Eut observations are yet w^anting
on the development of this organism. The re-
searches of Laurent point to the conclusion that,
in the production of the young Hydra, a part
only of the ovum is directly concerned.

The polypite thus resulting from a true genera-
tive act may subsequently, by gemmation, give

52 HYDROZOA.

rise to several others, in all respects similar to the
organism from which they were produced. These
for a time may remain in connection with each
other, but, more usually, they separate, each in its
turn, under favourable conditions, to repeat the
same budding process. The number of inde-
pendent beings into which a single Hydra, when
well supplied with food, and stimulated by a warm
temperature, may resolve itself, is certainly as-
tonishing. Not less so are its reparative powers,
which seem almost to defy the knife of the ana-
tomist. Full details on this subject are given by
Trembley, whose researches on the Hydra, pub-
lished in 1 744, are still well worthy of perusal.

Some years ago, Ecker compared the periplastic
tissue of the Hydra to aggregate masses of the
sarcode, or " formless contractile substance," com-
posing the body of Amoeha. Mr. G-. H. Lewes
has also recognised distinct " contractile masses,"
which he says were so very like Amoebae, as to
make him at first believe that the Hydra had
swallowed them. Such amoeboid particles occa-
sionally become detached by the method denomi-
nated *' diffluence," each usually including one or
more endoplasts ; but there is good reason to infer
that their apparently contractile movements are,
for the most part, the result of a process of en-
dosmose. Jager, however, has shown that two
budding Hydrae, each kept by him in a small
vessel of water, broke up into several isolated
particles, which, after the lapse of a month, were
still living, performed amoeba-like movements,
and, in some instances, passed into a peculiar
stage, resembling the encysted condition of Infit
soria. In this state, Jager supposes, they may

HYDEOZOA.

remain throughout the winter, and again, on the
return of spring, once more assume the aspect of
the primitive Hydra.

Fig.Z,

Development of Cordylophora : — «, gonophore of Cordylo-
phora lacustris, showing embrj'oes in its interior ; b, the same.
more advanced, ^^ith embryoes escaping from its ruptured ex-
tremity ; c, an embryo, in its free swimming condition ; d, the
same embryo, having assumed a pyriform figure; c, the embryo
in its attached condition ; /, primitive polypite, developed there-
from ; g, androphore of the same Cordylophora, its contents es-
caping under pressure ; h, caudate cells liberated therefrom ; ^,
spermatozoa. (All magnified.)

54 HTDROZOA.

CoRYNiD^. In CordylopJiora, the free swim-
ming ciliated embryo, on emerging from the
ruptured gonophore {fig. 8, b), is usually of an
elongated oval form, but very contractile, so that
often it assumes a pyriform figure {fig. 8, c and d).
Eventually, the embryo loses its cilia, and, fixing
itself, developes a hydrorhiza at one extremity
and a mouth at the other, thread-cells being at
the same time formed in the ectoderm {fig. 8, e
and /). Next, a series of about four tentacles
make their appearance ; these are soon succeeded
by others ; the somatic cavity becomes fully formed,
and the young Cordylophora, increasing in size,
is invested with a delicate cuticular layer. The
rudimentary ccenosarc, with its single polypite,
formed in this manner, soon commences to send
forth prolongations, and these, by gemmation,
develop the polypites and other appendages of the
adult organism.

A somewhat different series of changes occurs
in TvMdaria {fig. 9). The embryo of this genus
is not ciliated, but first makes its appearance as
a discoid body, from the circumference of which
short thick processes, the rudiments of tentacula,
are produced {fig. 9, /). The disc then becomes
more gibbous at the side turned away from the
axis of the gonophore ; a mouth, leading into a
newly-formed digestive cavity, soon occupying the
centre of the opposite side. The mouth then
elevates itself on a conical prominence, around
which a second series of tentacles arise. In this
state the embryo issues from the gonophore {fig. 9,
d). Eemaining free for a short time, it finally
becomes fixed, and developes a coenosarc with its
cuticular layer {fig. 9, e and g).

Development of Tubulaeia indivisa : — a, a polypite, attached
to the extremity of the ccenosarc, bearing clusters of reproductive
bodies between the two series of tentacles ; b, one of these clus-
ters consisting of several gonophores, borne on a long branching
stalk ; c, a single gonophore, containing two young polypites, one
of which has commenced to extricate itself; d, the same gono-
phore, in a more advanced condition ; e, a young polypite, thirty-
six hours after having escaped from the gonophore ; /, the
younger polypite, shown within the gonophores c and d ; g, the
polypite e, six weeks after extrication, (a is about thrice the
natural size : the others are much magnified.)

Tuhidaria, like Hydra, and, in all probability,
many other Hydrozoa, possesses well-marked re-
parative powers. When living specimens have

£ 4

56

HTDROZOA.

been kept for some days in vessels of sea-water,
it often happens that the polypites drop from their
stalks. Soon, however, new polypites are budded
forth, each having usually a smaller number of
tentacles than its predecessor. Similar are the
results produced by artificial fission. In this
manner, by section of a single stalk. Sir J. Gr.
Dalyell obtained, in the course of 550 days, twenty-
two successive polypites.

Sertularidje. In most Corynidce, the course
of development closely corresponds with that above

Fiff. lo.

Development of Campanulaeia : — a, Campanularia Lovem; h,
oneof itsgonoblastidia from whose summit project two medusiform
gonophores, one of which is giving exit to a ciliated embryo ; c,
the same embryo, in its free swimming condition ; d and f, suc-
cessive stages of the young ccenosarc developed therefrom ; /,
medusiform zooid of Campamdaria dichotoma ; — ir, polypite; tt',
hydrotheca ; p', gonoblastidium. (a and / are about twice the
natm-al size; the others are much magnified.)

described as taking place in Cordylophora. Of a
similar character, in its more general features at

HYDROZOA. 57

least, is the life history of the Sertularidce. The
young Campanularia or Antennularia^ at first
free, soon loses its cilia, fixes itself, and contracts
into a circular disc, which exhibits a division into
four lobes {fig. lo, c and cV). In the centre of
the disc an opaque spot makes its appearance, and
over this the surface becomes gradually elevated,
until, finally, a young coenosarc is the result (^fig.
10, e). From this, by gemmation, the branching
hydrosoma of the complete organism, with its
crowded assemblage of polypites, is subsequently
produced.

Thus the young condition of a Sertularid would
appear to differ from that of a Corynid in having
a portion of its coenosarc more or less completely
developed before distinct traces of a polypite can
be observed. Such a conclusion accords Avell with
the composite structure always assumed by the
adult hydrosoma. And in this respect the Sertu-
laridce, while departing from the Gorynidce, seem
to agree with the oceanic orders, Calycophoridce
and Physophoridce.

Calycophoridje, Of the earlier embryonic
changes in the Calycophoridce little is knoA\Ti.
In Diphyes, according to Gegenbaur, the blasto-
derm at first appears as an elevated protuberance,
occupying only a portion of the segmented vitellus.
Soon, this blastoderm forms a rudimentary necto-
calyx, from which a short canal leads to the ciliated
cavity of the yolk below. The nectocalyx then
rapidly enlarges, while polypites are seen to arise
between it and the appended yolk-mass.

PhysophoriDjE, Our knowledge of the Qxa-
hryologjoi the Physophoridce is confessedly scanty.

58 HTDROZOA.

The youngest examples of the group seen as yet,

Fig. II.

Development of Physalia: — a, young Ph^salia, with a single
polypite and tentacle ; b, the same, more advanced ; c, adult Phy-
salia; d, a tentacle, with its basal sac; — o, pneumatophore ; a',
pneumatocyst ; tt, polypite ; t, tentacle, (a and b are magnified ;
c is reduced; d is about the natural size.)

exhibit a well-defined pneumatophore, with a single
polypite and tentacle (^fig. ii, a).

HTDKOZOA. 59

In a Velelhif less than -1 of an inch long, ob-
served by Mr. Huxley, " the horizontal disc of the
adult was represented by a bell-shaped, mem-
branous 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 extremit}^, and around its base
were a few short, csecal 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 continuously over the pneumatocyst into
the crest, near whose free edge they terminated
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 [ciliated] canals, united
at their ends by two marginal canals at right
angles with one another, one in the disc, the other
in the crest. The pneumatocyst shone through
the disc, and did not extend into the crest at all."
It appeared "as an almost hemispherical body,
convex above and flat below. On two of its sides,
in a plane perpendicular to that of the crest, there
was a double crescentic mark, caused by a depres-
sion. 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 '05 of

60

HYDROZOA.

an inch in diameter, with strong and thin walls,
w^hich, when it was burst and the air expelled, fell
into sharp folds."

Medusidm. — The development of the true Me-

dusidce has yet to be effectively studied. From
the observations of J. Miiller on jEginopsis, of
Gregenbaur on Trachynema and Cunina, and of
Fritz Miiller on Liriope, it seems highly probable
that these genera proceed at once from the con-
dition of the embryo to assume the aspect of the
organism which gave them birth. Still more con-
clusive on this point are the results of some recent
researches of Claparede on a Medusid closely
allied (if not belonging) to the genus Lizzia.
Within the substance of the body- wall of the de-

Fiff.v.

Development of Lizzia: — a, adult Lizzia, the walls of whose
polypite are seen to bear numerous ova ; b, supposed free-su-im-
ming young of a, viewed from below ; c, the same, seen in pro-
file, (a is slightly, h and c are very much, magnified.)

pendent polypite were observed numbers of what
seemed to be true ova, some furnished with germ-
vesicle and germ-spot, others in a more advanced
stage of development. These last resembled in

HYDROZOA. 61

form and structure certain free floating bodies
(fig. 12), each of which, though still enclosed in
an outer covering, presented, on a reduced scale,
distinct indications of various parts observed in the
adult Medusid; four radiating canals, and eight
tentacular enlargements, being especially notice-
able. The full-grown Lizzia possessed twelve
tentacles, four single, and eight others arranged
in foiu alternating pairs. In the young form four
of the rudimei^tary tentacles were much longer
than the others, and it seems not improbable that
each of them represented one of the four pairs of
tentacles in the perfect Lizzia, No males of this
species have been observed. It is worth adding
that another form placed by zoologists in the same
genus appears to be only the detached bud of one
of the Corynidce. Yet, as Professor Huxley has
said, "it is within the limits of logical possi-
bilit}^ that the adult forms anatomically similar,
should be genetically different ; that they should
have arrived at a similar point by different
roads."

The observations of M^Crady on the direct
development of an other Medusid, Cunina octo-
naria, also deserve attention. This creature oc-
curs as a parasite within the nectosacof a distantly
allied form, Turritopsis nutricfula, from whose
mouth, by means of a long proboscidiform poly-
pite, the young Cunina obtains its food. At an
early stage the Cunina appears as a clavate
body, presenting a short, rudely globular proxi-
mal, and a more attenuate, somewhat cylindrical,
distal, region. From the sides of the posterior
margin of the former, two long flexible tentacles
soon sprout, while, at the same time, a nutrient

62 HTDROZOA.

cavity is formed throughout the central portion of
the mass. Even at this period the larva produces
free buds from its proximal extremity, not more
than two appearing to arise at the same time,
though the process of gemmation may frequently
be repeated. Next, the distal region elongates ;
the nutrient cavity opens at its free extremity,
forming a mouth ; and thus a young polypite is
produced, while from the proximal margin two
new tentacula soon make their appearance. From
this region a rudimentary nectocalyx now arises,
a fold, in which are developed marginal bodies,
appearing, distally, in front of the tentacles,
between which four other tubercular lobes are
now seen to bud forth. The growth of the nec-
tocalyx slowly proceeds; eight marginal bodies
distinctly come into view ; the polypite diminishes
in size, finally becoming inconspicuous ; and the
animal attains the adult form characteristic of its
family, save only that reproductive organs have
not yet been observed.

That instances of the above kind should be mul-
tiplied and re-observed seems, for many reasons,
very desirable, since, as already remarked, not a
few of the forms known as Medusidm are but the
free-swimming gonophores of various other Hy-
drozoa. Thus, from the ovum of Ttirris, one of
the so-called genera referred to this order, a poly-
pite is produced, which sends forth a creeping
coenosarc, giving rise to a hydrosoma, clearly seem
to belong to the Corynidce {fig. 13). Dr. T.
Strethill Wright has further proved that Bour/ain-
tillea Britannica, a common form of Medusoid,
is, in truth, the reproductive body of Atractylis
ramosa, one of the Corynidce,

Development of Turris : — a, oviim of the medusiform zooid {h)
known as Turris ncghcta ; c, polypite, with creeping hydrorhiza,
developed therefrom. (All magnified.)

Certain medusiform gonophores, and, it may-
be also, some true Medusids, possess the power of
producing, by gemmation, free-swimming forms
which directly resemble themselves. Such buds
have been observed to start from the sides of the
polypite in Sarsicc gemonifera and Lizzicc octo-
pundata, from the reproductive region of the
calycine canals in two species of Thaumantias,
from the bases of the tentacles in Steenstrupia
Oiuenii and Sarsia prolifera, and from the de-
pendent portion of the tentacles themselves in
Diplonema {fig. 14). Hence these medusoids
ought, perhaps, to be regarded as free gonoblas-
tidia. Here, also, it may be added, that multi-
plication by fission has been observed by Kolliker
in a species of Stomobrachium.

Gemmation of MEDUsoros : — a, Diplonema Islandica, showing
young Medusoids budding from the tentacles; b, Sarsiagemmifera,
with Medusoids arising from the sides of the polypite; c, Sarsia
proUfira, in which Medusoids are seen to sprout from the junc-
tion of the tentacles with the marginal canal. (All magnified.)

LucERNARiD^. Still more singular phenomena
appear in the life-history of Lueernaridce, In
Aurelia, Cyanea, and Chrysaora, the ova originate
within the generative cavities of the gigantic repro-
ductive bodies previously described. Thence they
are transferred, in some unknown manner, to the
peculiar pouches formed along the margins of the
dependent lips of the polypite, and on their way
to these pouches are, in all probability, fertilised by
contact with the diffused spermatozoa. Segmenta-
tion of the vitellus, and other primordial changes,
are undergone by the young ovum while yet
within the pouch, from which, about the close of
the third day, it comes forth, to enjoy, for a brief
period, an active, free-swimming existence. At
first it appears as an oblong, flattened, ciliated
body, or ' planula,' of very minute size, composed

HTDROZOA. 65

of outer and inner layers, enclosing- a central
cavity {fig- 15, c). Soon it assumes a somewhat
pyriform fignre, enlarging at one extremity, in the
centre of which a depression is observable. Next,
the narrower end attaches itself to some sub-
marine object, while the depression at the oppo-
site extremity, becoming deeper and deeper, at
length communicates with the interior cavity.
Thus a mouth is formed, around which may be
seen four small j^rotuberances, the rudiments of
tentacula (ti). In the interspaces of these four
new tentacles arise ; others, in quick succession,
make their appearance, until a circlet of numerous
filiform appendages, containing thread-cells, sur-
rounds tlie distal margin of the " Hydra tuba,"
as the young organism, at this stage of its career,
has been termed by Sir J. G-. Dalyell (e and/).
The mouth, m the meantime, from being a mere
quadrilateral orifice, grows and lengthens it?elf so
as to constitute a true polypite, occupying the axis
of the inverted umbrella, or disc, which supports
the marginal tentacles. A continuous, wide, open
space occupies the whole interior of the umbrella
and polypite, whose relations to the rest of the
organism, and, indeed, the whole structure of
Hydra-tuba, closely resemble what may be seen
in Lucernavla. Externally, it presents a delicate,
translucent aspect, and in height averages some 3
of an inch. But though dissimilar to Hydra in
organisation and want of locomotive capacity, the
Hydra-tuba recalls to mind its fresh-water con-
gener, first, in its remarkable reparative powers ;
and, secondly, in the extent to which it multiplies
by gemmation. Not merely do buds arise from
the sides of the body, but, in addition, creeping
r

66 HTDROZOA.

tubes, or stolons, are sent forth, from which fresh
gemmae spring up, it may be, to detach themselves,
and so one or several large colonies become
formed, all the produce of a single fertilised ovum.
For years the hydrosoma may continue in this
stage, undergoing no further development. But
under certain conditions, similar, perhaps, to those
which determine the formation of reproductive
organs in the Hydra, a new and striking series
of changes is inaugurated. First, each Hydra-
tuba elongates, increasing somewhat in size.
Then, from just below the tentacles to within a
short distance of the proximal extremity, a succes-
sion of transverse markings begin to appear,
which quickly take on the aspect of circular con-
strictions {g). When the organism was first dis-
covered in this condition by Sars, he, thinking it
a new animal, called it " Scyphistoma." The same
naturalist, observing the Scyphistoma at a still
later stage, with the constrictions more strongly
marked, and the several segments included between
them cleft and lobed around their margins, gave
it, from its resemblance to an artichoke, the name
of Strobila (/i). Still further do the constrictions
deepen until the Strobila becomes not unlike a
pile of cups or saucers. The marginal tentacles
then disappear, but a new row arises in their stead
from the summit of the short, undivided, proximal
extremity (^). The disc-like segments above the
tentacles gradually fall off, and, swimming freely
by the contractions of the lobed margin which
each presents, have been described by Eschscholtz
as true Medusidcc, under the generic title of
Ephyra {k). But each Ephyra soon acquires a
nutritive system, lithocysts, tentacles, and genera-

HYDROZOA.

67

tive organs ; thus eventually becoming similar to
the huge reproductive body, from whose fertilised
ovum the primitive Hydra- tuba was produced.
This, and the stock which it developed, does not,
however, perish, but ma}^ again, by growth and
fission, give rise to fresh successions of generative
bodies.

Fig. 15.

Development of Chrysaoea : — a, ova, with gelatinous invest-
ment, from Chrysaora hysoscella ; h and c, free ova ; d, young
Hydra-tuba, -with four marginal tentacles, developed therefrom ;
e, the same, with eight tentacles ; /, Hydra tuba, in its ordinary
condition ; g, another Hydra-tuba, marked with constrictions ;
h, a more advanced form, with deeper constrictions ; i, a speci-
men undergoing fission, in which the tentacles are seen to arise
from below the constricted portion, while its upper segments
separate, and become free-swimming zooids (k).

Similar to the above appears the life -history of
Cephea and Cassiopeia, notwithstanding the very
different structure of the detached reproductive
zooids which these genera present. On the de-
velopment of Rhizostoma itself accurate observa-
tions are wanting.

In the Lucernariadce proper, no free zooids are
produced, but the generative elements are formed

bS HYDEOZOA.

in longitudinal folds, which arise, opposite to each
other, along the four inner angles of the polypite's
digestive cavity.

In Pelagia, a permanently free form of tho
same order, the ova are developed directly int
the likeness of the organism ^\dthin which they
are evolved. The young Pelarjia, according to
Krohn, presents a minute, semi-transparent, some-
what cylindrical body, invested in a thin, whitish,
ciliated, covering. By means of its cilia, the em-
bryo swims rapidly, often turning round on its
longitudinal axis. At one extremity, which is
truncate, a very small mouth appears, leading
into a distinct nutrient cavity, or stomach. This
cavity quickly enlarges ; the mouth, also, becomes
protruded, whilst at the same time, the hinder
end of the body is developed into an umbrella.
On the third day, traces of eight lobes indent the
margin of the umbrella, an equal number of sacs
arising from the sides of the stomach. The mar-
ginal lobes lengthen, each becomes further in-
dented, and soon rudimentary lithocysts can be
distinguished. By this time the oral region is
changed into a perceptible polypite, but. the or-
ganism still moves chiefly by the aid of its cilia,
the contractions of the umbrella being at first only
occasionally repeated. Afterwards, the ciliated
coat disappears ; thread-ceils are produced ; the
lips of the polypite enlarge ; the umbrella shortens
and assumes its proper function ; the crystalline
contents of the lithocysts make their appearance.
The stomachal sacs increase in number and take
on the aspect which they present in the adult
animal. Finally, marginal tentacles are acquired,
four of these being equal in length to the diameter

HTDROZOA. 69

of the swimming organ, while the four other ten-
tacles, and the lips of the polypite, are as yet
slightly developed.

The development of the various appendages
which arise along the coenosarc of the composite
Hydrozoa now requires to be noticed. The same
hydrosoma often exhibits these in different stages
of evolution, so that their formation admits of
being studied, with more or less hope of success,
in specimens which may seem to have reached the
adult condition.

All the lateral appendages, except the hydro-
theca?, appear first as simple processes of the two
layers of the body, and in outAvard form are won-
derfully similar, the characteristic aspect of each
manifesting itself as groAvth advances. In the
hydrophyllia and nectocalyces the ectoderm en-
larges to a much greater extent than the endoderm ;
in most other appendages, the relations of these
layers are not so disproportionate.

As above remarked, there is but slight differ-
ence between hydrocysts and polypites at certain
stages of their development. But while the
hydrocysts remain closed, an opening is formed at
the distal extremity of each polypite, villi and
distinct regions soon becoming visible in its endo-
dermal lining.

The tentacles, as already shown, differ as to the
degree of vacuolation undergone by their endo-
derm. The lateral threads of branched tentacles
"appear successively as close-set buds on one side
of the proximal end of the tentacle, the younger
buds being ahvaj^s developed on the proximal side

F 3

70 HTDROZOA.

branch is complex, consisting of two or three
distinct portions, these are gradually produced
as each of the bud-like processes elongates. "The
involucrum is formed as a process of the ectoderm
of the distal end of the peduncle. In Physophora,
the distal end of the peduncle itself undergoes a
singular dilatation, and helps to form the envelope
for the sacculus."

The development of the gonophores has, in the
account given of their structure, been sufficiently
described. That of the nectocalyces is, at first,
precisely similar, but the central mass does not, in
these, give rise to a manubrium, while the cen-
tral cavity, into which the longitudinal canals open,
remains very much larger.

The hydrothecae of the Sertularidce are formed
by the gradual separation from the body of each
pol3rpite of the outer layer, or polypary, excreted
by its ectoderm, which, opening distally, displays
the cup-shaped cavity, characteristic of different
species.

The relative succession of the appendages also
demands attention. In the fixed Hydrozoa the
distal polypites are first developed, whereas the
proximal appendages are the youngest in the
Physojphor'idoe and Galycojphoridce. But this rule
does not appear to govern the nectocalyces in the
last-mentioned group, their precise order of deve-
lopment still remaining involved in some com-
plexity.

The phenomena indicated in the preceding
brief detail of the life-history of the Hydrozoa
have, in addition to their special value, a wider
interest for the philosophic student of zoology,

HTDROZOA. 7 1

from their bearing on the subject of animal deve-
lopment in general. A few words of explanation
may therefore, in this place, not appear unneces-
sary.

The life of every animal species may, from a
certain point of view, be regarded as consisting in
the alternate performance of two distinct series
of acts; the one of reproduction, the other of
development.

Each act of reproduction consists essentially in
this, that two dissimilar bodies, an ovum and a
spermatozoon, are brought into mutual contact. In
some cases the spermatozoon penetrates the coats
of the ovum, or even enters it by a proper aper-
ture, known as the * micropyle.'

Thus defined, the process of reproduction is the
same in all animals, though in some its simplicity
is masked by the occurrence of a variety of other
phenomena, all, however, of secondary importance.

It must also be borne in mind that the evolu-
tion of ova and spermatozoa, obviously necessary
as a prelude to the reproductive function, cannot
be considered as forming a part of it. An ovum
or spermatozoon is, in truth, nothing more than a
highly differentiated portion of the parent or-
ganism, the result of a process of development.

But no sooner has the act of reproduction been
duly effected, than that of development forthwith
begins. The fertilised ovum gives rise to an
embryo, which tends to evolve itself into the like-
ness of its parent. This embryo, together with
all the structures subsequently developed there-
from, is said to constitute, in the zoological sense
of the term, an animal individual.

Should the resulting organism develop an

F 4

< *Z IIYDROZOA.

ovum in its turn, then, that ovum, if fertilised,
forms the hiisis of a new individual ; and so on
for every additional ovinn concerned in a genera-
tive act. So that each performance of the repro-
ductive process is, as it were, the natural boundary
between two successive individuals, or, in other
words, between two distinct cycles of development.
Thus, while the individual perishes, the species,
by reproduction, is constantly renewed. Much
also, might be said on the analogy which exists
between the different individuals of the same
species, on the one hand, and the constituent parts
of each individual, on the other.

Again, fission and gemmation are not, as many
writers incorrectly state, modifications of the re-
productive process, but rather, acts of development.
For, as already shown, every ovum is at first a
bud, which at length, by fission, becomes separated
from the body of the parent. All this takes place
quite independently of its fecundation. So that
an unfertilised ovum is no more entitled to be
considered an individual, than a wart or any other
excrescence.

The antagonism between development and re-
production, or even between development in
general and those particular stages of the vital
process by which reproduction is preceded, is
sometimes shown by the fact that certain external
conditions which seem to favour the one exert an
oj^posite influence on the other. Thus, on the
approach of cold weather, the Hycha is prone to
develop organs of true reproduction, while, if
kept in a warm room, it still, as in summer time,
continues the formation of ordinary buds.

It is, therefore, the object of the reproductive

HYDIIOZOA. 73

fimction to confer individuality upon that which
previously was but a detached part of the parent
organism. Howsoever complex the body of an
adult animal may seem, it was once an ovum,
whose extreme simplicity of structure might al-
most be said to verge upon homogeneity. What
inaugurated the wonderful series of changes by
which the ovum fashioned itself into the likeness
of its parent ? Contact with spermatozoa, or, in
one word, reproduction. To say, then, that sper-
matozoa possess a peculiar individualising in-
fluence can scarcely be viewed as a metaphorical
form of expression. Hoio they are capable of
exerting this influence is, however, a problem to
which, as yet, science has furnished no definite
solution. Bischofif has compared their action to
that of a ferment, such as the yeast of beer ; but
this hypothesis, as Claparede truly observes, only
removes the present difficulty a single step back-
wards.

The zoological individual being, therefore, de-
fined as the entire product of the developmental
changes of a single fertilised ovum, we have now
to consider the principal modifications which the
cycle of development presents.

If all the parts of an individual remain mutu-
ally connected, its development is said to be
' continuous ' ; if any of them separate as inde-
pendent beings, it is ^discontinuous'.

Continuous development may manifest itself
imder the three principal modes of * growth,'
'metamorphosis,' and 'gemmation without fis-
sion.' In metamorphosis, growth alternates with
certain well-marked changes of form. In gem-
mation without fission, a tendency to vegetative

74 HTDROZOA.

repetition is more or less distinctly marked. An
example of the first of these methods is presented
by Felagia; of the second, by jEginopsis ; of the
third, by Cordylophora or Sertularia.

In discontinuous development the detached
portions of the individual are termed 'zooids,'
that which is first formed being distinguished as
the * producing,' that which separates from it, the
'produced' zooid. If there be more than two
successive series of zooids, the terms ' protozooid,'
' deuterozooid,' and ' tritozooid,' may then be re-
spectively applied to them. Thus, the medusoids
budded by Sarsia are, probably, tritozooids. The
term zooid is also extended to the several parts of
a connected structure which increases by vegeta-
tive repetition ; for example, to the polypites, and
other appendages of the composite Hydrozoa.

The producing zooid may either possess or want
generative organs. In the latter case the pro-
duced zooid may take on the performance of the
reproductive function, as in so many orders of
Hydrozoa.

In this class we have seen that the produced
zooid may resemble the producing zooid, as in
Hydra, or be dissimilar to it, as shown by the
free-swimming gonophores of the Corynidce and
SertuJaridce. The first case affords an illustration
of simple ' gemmation with fission ' ; the latter, of
the process known as * metagenesis.' If the pro-
ducing zooid possess sexual organs, and the pro-
duced zooid present the morphological, but not
the physiological, characters of an ovum, then the
process is one of 'parthenogenesis.' All these
varieties of discontinuous development are col-
lectively denominated ' agamogenesis,' as distin-

HTDROZOA. 75

guished from ' gamogenesis,' in -which the ovum,
to be developed, must first be brought into con-
tact with spermatozoa.

But such modifications are, in nature, less
distinct from one another than the systematic
definitions just given might appear to imply.
Furthermore, recent investigations on the de-
velopment of Insects and Crustaceans have tended
alike to confuse our old-established notions of
animal individuality and of the true nature of
the generative process. For certain Insect ova
have been observed to undergo development in
the ordinary manner, though no previous contact
with spermatozoa had taken place. And in un-
impregnated female Vertehrata ovarian tumours
are said sometimes to occur, which contain traces
of hair, teeth, bone, nerves, and other tissues pro-
per to the adult organism. If, therefore, cases
exist in which the influence of a male element
seems rather accessor)^ than essential to the normal
evolution of the germ ; nay, can even be dispensed
with, there are others in which, without such
influence, no proper individuality is manifested,
though development, to a certain extent, must as-
suredly be considered to have taken place. For
the present we have preferred to advocate the
views entertained on these disputed points by
Professors Huxley and Carpenter, while, to avoid
needless ambiguity, we have thought it better to
employ the precise terminology which the former
naturalist has suggested.

But other attempts have been made to explain
the phenomena in question. Steenstrup, followed
in Britain by the late Professor E. Forbes, and a
host of minor investigators, proposed to consider

76 HYDROZOA.

both the free zooids of the Hydrozoa and the
organisms from which they sprung as alike en-
titled to the rank of individual beings, belonging
to allied groups, and mutually reproducing each
other by a process of '* alternate generation." But,
in addition to the more general objection whicli
may be raised against this hypothesis, confounding,
as it does, true generation with gemmation or
fission, the one, an act of reproduction, the other,
of development, it is sufficient to show that, in
the present instance, its application is based on
a very superficial examination of the facts to be
explained. The gradual series of transitional
homologous forms, so surely connecting the com-
plex free gonophores of certain Hydrozoa with
the simple reproductive processes of Hydra or
Hydr actinia, could not have been very familiar
to the minds of those who would have hesitated,
if called upon, in accordance with Steenstrup's
theory, to impute individuality to the latter.
Professor K. Leuckart, however, consistently does
this, and would regard as true individuals the
independent polymorphic buds of the same com-
posite Hydrozoon. And Mr. Lubbock has justly
remarked that, "whether we retain the old nomen-
clature, or dropping the idea of unit}^ in the term
' individual,^ adopt the system proposed by Pro-
fessor Huxley, we shall be met by great difficulties
and inconsistencies." It behoves us, therefore, to
follow that explanation which embodies in the
simplest manner all the observed phenomena, and
which is, at the same time, least characterised by
inconsistency.

Recently, Professor Agassiz has proposed a
modification of Leuckart's theory, and suggests

HYDROZOA. 77

the distinction of four kinds of individuality in
the anim.il kingdom. First, hereditary indi-
viduality, when from a single e>gg a single inde-
pendent being is produced. Secondly, derivative
or consecutive individuality, or " that kind of
independence resulting from an individualisation
of j^arts of the product of a single egg ; " as in
many Lucernaridce, Corynidce, and Gamjjanu-
lariadcc. Thirdly, secondai^ individuality, where
the product of one egg multiplies by continuous
gemmation, giving rise to an immoveable com-
munity ; as in the Sertulariadcc. Lastly, there is
coraiolex individuality, w^here a similar but move-
able community is formed; as seen in the Cahj-
coj)horidce and Physophoridcc. In this case, he
adds, " the individuals of the community are not
only connected together, but, under given cir-
cnmstances, they act together as if they were one
individual, while at the same time each individual
may perform acts of its own."

Others were for regarding the gonophores of the
fixed Hydrozoa as the perfect or adult stages of
the forms by which they were produced, the whole
process being viewed as one of ordinary meta-
morphosis. The particular objection just stated
applies also to the opinion under consideration,
which has, nevertheless, found its advocates in a
few writers of distinction. There is, no doubt,
some degree of plausibility in a view which consi-
ders the fixed Cor}Tiid or Campanularid as the
young condition of the more complex Medusoid to
which, by gemmation, it gives rise. It is now,
however, certain not only that the Ccdyeophoridcv
and Physojohoridce agree closely in structure with
the Hydrozoa just named, but likewise, that they

78 HYDROZOA.

bear the same morphological relation to their
reproductive bodies. Extend the case to these;
let Veldla, for example, be henceforth the larva
of its free medusiform gonophores, and the doc-
trine which we have contested is at once seen to
become untenable.

Another explanation emanated from Professur
Owen. His ingenious theory of " parthenogenesis "
supposed that the primitive result of each genera-
tive act retains within its body unchanged a cer-
tain portion of the germ-mass from which it was
first evolved, together "with so much of the
spermatic force inherited by the retained germ-
cells from the parent-cell or germ-vesicle as
suffices to set on foot and maintain the same
series of formative actions as those which consti-
tuted the individual containing them." So that
*' every successive generation, or series of sponta-
neous fissions, of the primary impregnated germ-
cell, must weaken the spermatic force transmitted
to such successive generations of cells." Or, to
confine ourselves to the class under consideration,
that a Corynid produced, as the resultants of the
germ-cells and spermatic force stored up within it,
successions of free or fixed gonophores, until the
generative force became exhausted. But here, at
least, it can be proved, that the unchanged germ-
masses alluded to have no objective existence,
while the more subjective spermatic force, in
these, as in all other, animals, has hitherto suc-
ceeded in escaping the ken of the anatomist.

HTDROZOA. 79

Section HI.

CLASSIFICATION OF HYDROZOA.

I. Classification. — 2. Order i: Hydridae. — 3. Order 2: Corynidae.
— 4. Order 3 : Sertularidae. — 5. Order 4 : Calytophoridae. — 6.
Order 5 : Physophoridae. — 7. Order 6 : Medusidae. — 8. Order 7 :
Lucernaridae.

I. ClassiHcatioiB. — The seven orders into which
the class Hydrozoa is divided may be defined as
follows:

1. Hydridce. — Hydrozoa, whose hydrosoma con-

sists of a single locomotive polypite, wdth
tentacles, hydrorhiza, and reproductive organs
which appear as simple processes of the body-
wall.

2. Corynidce, — Hydrozoa, whose hydrosoma is

fixed by an hydrorhiza, and consists either of
one polypite, or of several connected by a
coenosarc, which usually developes a firm
outer layer. Eeproductive organs in the
form of gonophores, which vary much in
structure, and arise from the sides of the
polypites, from the coenosarc, or from gono-
blastidia.

3. Sertularidce. — Hydrozoa, whose hydrosoma

is fixed by an hydrorhiza, and consists of
several polypites, protected by hydrothecse,
and connected by a coenosarc, which is usually
branched, and invested with a very firm outer
layer. Eeproductive organs as gonophores,
arising from the coenosarc, or from gonoblas-
tidia.

80 HTDROZOA.

4. Calycophoridce, — Hydrozoa, whose liydrosoma

is free and oceanic, consisting- of several
polypites connected by a flexible, contractile,
unbranched coenosarc, the proximal extremity
of which is furnished with nectocalyces, and
dilated to form a somatocyst. Eeproductive
organs as medusiform gonophores, budded
from the peduncles of the polypites.

5. Physophoridtv. — Hydrozoa, whose hydrosoma

is free and oceanic, consisting of several
polypites connected by a flexible, contractile,
seldom slightly branched, coenosarc, the
proximal extremity of which expands into a
pneumatophore, and is sometimes provided
with nectocalyces. Eeproductive organs,
more or less complex in structure, developed
upon gonoblastidia.

6. Medusidce. — Hydrozoa, whose hydrosoma is

free and oceanic, consisting of a single poly-
pite suspended from the roof of a nectocalyx,
furnished with a system of canals. Eepro-
ductive organs as processes either of the
sides of the pol3^ite, or of the nectocalycine
canals.

7. Liwernaridcc, — Hydrozoa, whose hydrosoma

has its base developed into an umbrella in
the walls of which the reproductive organs
are produced.
The characteristics of these orders are indicated
more briefly in the subjoined analytical table.

2. Order i : Hydridso The order Hydridce

contains but a single genus, Hydra, distinguished
from the few marine Hydrozoa which it approaches
in physiognomy by its peculiar habit and locomo-

HYDROZOA.

81

Class
HYDROZOA.

I

/

No umbrella.

An umbrella.

r

An hydrorliiza.

No hydrorbiza.
Oceanic.

Locomotive.
Inhabiting fresh-water.

Order I.
Hydride.

Fixed.

No hydrothecse.
Order II.

CORTNTD^.

Hydrothecse.

Order III.
Sebtularid^.

r

Polypites free from nectocalyces.

Polypite sus-
pended from
the roof of a
nectocalj'x.

No pneumatophore,
Nectocalyces.
A Somatocyst.

Order IV.

CALYCOPHORIDiE.

A pneumatophore.

Nectocalyces
present or absent.

Order V.
Physophobidje.

Order VI.

MEDUSmiB.

a

Order VII.

LUCERNAEIDJI.

82 HTDROZOA.

live powers. Several species of Hydra have been
described under such names as H. viridis, H.
rubra, IT. vulgaris and H. fusca. These differ
in size, colour, the form of the body, or in the
relative proportions of the polypite and tentacles.
The polypite of H. vulgaris is cylindrical, its
colour variable, but usually orange-brown, and its
tentacles of moderate length. H. viridis has a
polypite of a grass-green tint, furnished with
comparatively short tentacula. H. fusca is larger
than either of these, its colour is deep brown, and
its tentacles very long and extensile ; the proximal
extremity of the polypite becoming suddenly
attenuated for about a third of its length. When
living Plydrse are removed from the water, they
appear to the eye as minute specks of jelly, which
quickly, however, recover their true form on re-
immersion. In confinement they readily thrive,
seeking the light and feeding voraciously. Spe-
cimens of the Hydra may be kept in glass vessels,
and their singular habits observed by the student,
with little difficulty.

3. Wrder 2 : Coryniilje. — In Corymorjjha,
Vorticlava, and Myriothela, the hydrosoma, like
that of Hydra, presents only a single polypite,
but, in the greater number of Corynidw, it is
composite, exhibiting numerous polypites con-
nected by a coenosarc, which may be either erect
and branching, as in Cordylophora, or reduced
to a delicate creeping tube, as in Clava and Tri-
cliydra. A hydrosoma of this kind may be com-
pared to a Hydra in the act of budding, while as
yet the young zooids remain in connection with
the primitive polypite, by the hydrorhiza of which

HYDROZOA. 83

the entire fabric continues to attach itself And,
since gemmation may take place in many different
ways, so, in like manner, result the great variety
of forms due to modifications of what is essentially
the same process of growth. In the flower-like
Tubular i a indlvisa, the coenosarc consists of
several simple tubes, intertwined one with another
near their attached extremities, and sometimes
rising to a height of ten or twelve inches {fig. i6).
From the distal ends of these tubes, which are of
a vStraw-yellow colour, the polypites, tinged with a
bright scarlet, conspicuously project. The hydro-
soma of Euclendrium rameum, though seldom
more than six inches in height, bears a singularly
close resemblance to a forest-tree in miniature, its
surface being studded with minute reddish poly-
pites, not less than a thousand of which may
crowd the branches of a single specimen. Such
arborescent structures strikingly contrast with the
slender mossy threads which compose the con-
necting stem of smaller species. In HyclTactinia,
the meshes of the very intricate creeping coenosarc
are aggregated so as to form a compact lamina or
crust, investing the surfaces of univalve shells,
which, by a coincidence hitherto unexplained,
usually atibrd shelter to the Hermit-Crab. The
living coating of Hydractinia presents to the
naked eye the appearance of a rather coarse fioc-
calent nap, pale grey or milk-white in tint, the
polypites, w^hen fully expanded, attaining a height
of nearly half an inch, and waving to and fro
wdth every agitation of the shell ( fig. 1 7, c.) In
other genera the colour of the coenosarc is usually
yellowish-brown. All the preceding forms have
the hydrosoma rooted, or attached to other objects,

G 2

84

HTDROZOA.

and in no Corynid hitherto observed does it appear
to be altogether free, unless, indeed, an exception

Fig. 1 6.

Morpjiology of Tubula^iud-^; : — a, Corymorpha nutans; b,
tuft of gqpophores from Corymbogoniumcapillarc; o, Tuhidaria in-
divisa ; d, distal extremity of its ccenosarc ; e, transverse section
of the same, {a fii^d c are of tjae natural size ; b, d, and e, are
magnified.)

be made in favour of the doubtful genus Ne-
mojpsis.

HYDROZOA. 85

The firm horny layer, or polypary, which the
coenosarc excretes in Tubular ia and its allies,
remains in a comparatively rudimentary condition
among most other Corynidce. In few, however,
is it absent altogether. In Hydractinia, it be-
comes elevated at intervals to form numerous
rough processes or spines, while over the general
surface of the ectoderm its presence is almost im-
perceptible. A very different modification is pre-
sented by the genus Bimeria. Here the polypary
is not, as in other members of the order, restricted
to the coenosarc, but extends itself so as to clothe
the entire body of each polypite, leaving bare
only the mouth and tips of the tentacles.

The chief differences which prevail among the
polypites of the Corynidce have reference either to
size or the disposition of their tentacula. The
comparatively gigantic polypite of CoryTnorpha
nutans, which attains a leng-th of 4*5 inches, is
described by Forbes and Goodsir as presenting
the appearance of a beautiful flower, nodding
gracefully upon its stem (Jig. 16, a). Another
species of the same genus, C. nana, does not
exceed '5 of an inch in length, though this, in its
turn, is double the size of the tiny Vorticlava
humilis (fig. 17, a). Still more minute are the
delicate polypites of some species oi Eudendrium.
In most members of the present group the general
form of the polypite is more or less clavate.

The tentacles exhibit several distinct modes of
arrangement. In Tubularia and Gorymorpha a
fringe of short appendages immediately surrounds
the mouth of the polypite, from the base of which,
close to the distal extremity of the coenosarc, arises
a second circlet of much longer filiform tentacula,
G 3

86

HYDROZOA.

fewer in number than those of the upper row {fig.
9, a). Vorticlava, also, possesses a twofold series
of tentacles, but here, those of the lower circlet are
twice as numerous as the upper, which are five in
number, short, stout, and capitate {fig. 1 7, 6). In
Clava, Cordylophora {fig. 5,c), and GoTyne{fig. 1 7,

Fig. 17.

Various forms of Coeyniad^: — a and i, Vorticlava humilis ;
c, four polypites of Hydractinia echinata, growing on a piece of
shell; d, portion of 8yncoryne Sarsii, with medusiform zooids
(p) budding from between the tentacles (t) of the polypite (0).
(All, except a, magnified.)

d\ the tentacles appear irregularly scattered along
the sides of each polypite, though most abundantly
towards its distal extremity. In Coryne the mouth
is highly flexible, possessing the power of bending
towards that tentacle which has seized the prey,
and of converting itself, upon occasion, into a kind

HTDROZOA. 87

of sucking disc. The polypites of the allied genus
Stauridia are distinguished by the possession of
two or more cycles of dissimilar tentacles, separated
from one another by a considerable interval, each
cycle including four tentacles ; the lower row fili-
form, the upper whorl, or whorls, capitate, and
placed at right angles to one another and the
polypite. In Peiuiaria, there is a basal circlet
like that of Tubularia, between which and the
mouth of the polypite lie scattered numbers of
shorter tentacles, resembling those of VoHiclava,
In Myriothela, multitudes of wart-like tentacles
crowd the whole surface of the- club-shaped, soli-
tary, polypite. Similar to these are the tentacles
of Acaulis, which exhibits, in addition, a basal
series of long prehensile appendages. These, how-
ever, disappear as the organism approaches ma-
turity, so that this form may possibly be but a
young condition of Myvioihela, A single series
of rather long tentacles, inserted as in the fresh-
water Hydra, arises at a short distance below the
mouth in Trichydra, Clavatella, Peiigonimics,
Bhneria, and Eudendrium. Some species of
these genera seem to foreshadow an arrangement
of the tentacles which in Hydradinia becomes
sufficiently conspicuous. Around the mouths of
the digestive zooids in this genus two rows of al-
ternating tentacles are placed, so close to each
other that they appear, at first sight, to constitute
a single series ; the lower tentacula, which are
shorter, projecting at right angles to the body of
the polypite, from the axis of which the upper
tentacles very slightly diverge. These, however,
have no direct connection with the lonof tentacular
appendages, arising directly from the coenosarc, to

G 4

88 HYDROZOA.

which allusion has already been made. Lastly,
in Lav, each polypite supports but two tentacles,
above which the mouth is furnished with a pair of
wide projecting lobes, capable of being approxi-
mated closely to each other, and serving, doubt-
less, as efficient organs of prehension.

The gonophores of the Corynidce vary not a
little both in structure and mode of attachment.
In Cordylophora, Perigonimus, Garveia, Bimeria,
and some forms of Eudendrium and Atracfylis,
they spring directly from the stem or branches of
the coenosarc. In other species of the two last-
mentioned genera they are seated either beneath
the tentacles of the polypites, or on the summits
of special branches, arising from the proximal
region of the hydrosoma (fig. i6, b). In Myrio-
thela, Acaulis, and Clavatella, the gonophores
have their origin on the polypite, not far from its
attached extremity: in Coryne and Stauridia, they
are produced between the tentacles {fig. 1 7, d).
In Co'iymorpha and some species of Tuhularia,
they are supported on long branching gonoblas-
tidia, inserted immediately within the basal circlet
of tentacula {fig. 9, h) : in other Tuhidarice, T.
calaiiiaris and T. Dumortierii, as also in the genus
Fennaria, these long stalks appear to be absent.
The arrangement of the reproductive bodies in
Clava and Hydractinia has already been pointed
out. In the closely allied genera, Dicoryne and
Fodocoryne, they originate, in a somewhat simi-
lar manner, on proper gonoblastidia, never on the
ordinary polypites. But the proliferous stalks of
Fodocoryne are furnished, each, with a mouth,
and differ little from true polypites save in their
smaller size and the possession of fewer tentacula.

HTDROZOA. 89

Transitional forms of this kind should not, however,
surprise us, when we consider the common bond,
community of descent, which connects the two
kinds of appendages in question.

In the genera Ferigoninius, Atractylis, Pen'
naria, Corymoiyha,Acaulis, Stauridiaf and some
forms of Coryne, Tubular ia, and Eudendrium,
the gonophores assume the aspect of free-swimming
medusoids. In most other Corynidce they are
fixed, exhibiting many remarkable gradations of
structure. An intermediate condition is presented
by the curious reproductive zooids of Clavatella,
which, though locomotive, scarcely merit the ap-
pellation of medusiform. They are described by
ilr. Hincks as free polypoid buds, furnished with
six forked processes, set round the margin of a
central hemispherical disc, one limb of each fork
being capitate, like the tentacles of the polypite
itself, the other terminating in a peculiar sucker-
like enlargement. By means of these organs the
zooid, when detached, moves freely about, until
finally it proceeds to mature its generative pro-
ducts.

The gonophores of Trichydra, Vorticlava, and
Lav have hitherto remained unknown.

Two families of Corynidce have been distin-
guished, though the character employed to sepa-
rate them appears to be somewhat artificial.

Order COEYNID^.

Family i. Coryniad^.

Polypary absent, or rudimentary.

Family 2. Tubulariad^.

Polyjpainj well developed.

90 HTDROZOA.

The entire order is sometimes denominated
Tubularidcc, and agrees with the group Tubu-
larina of Ehrenberg.

4. Order 3 : §ertularidap. — Like the mem-
bers of the preceding order, all the Sertularidce,
after the expiration of their embryonic condition,
become permanently fixed by means of the hydro-
rhiza which forms the proximal extremity of the
coenosarc {fg. 4, c). In this group the tendency
to increase by gemmation is even greater than
among the Gorynidce, for no example of a Sertu-
larid has yet been recorded in which the hydro-
soma exhibits but a single polypite. The coenosarc
is plant-like and, frequently, much branched, the
main stem either losing itself in its own ramifica-
tions or remaining distinct throughout the entire
length of the arborescent mass. A good example
of the latter mode of growth is afforded by the
Sea-Fir, Sertidaria cupresslna, the hydrosoma
of which may attain a height of two, or even three,
feet, and bear on its branches so many as 100,000
distinct polypites. In contrast with this, the
largest of our native species, may be mentioned
the delicate Sertularia tenella, the length of whose
slender creeping hydrosoma scarcely reaches one
inch. The waving fronds of Oar-weed on various
parts of the coast afford a suitable habitat to the
anastomosing thread-like coenosarc of another char-
acteristic species, Gampanularia geniculata fVfhich
sends up at intervals its peculiar zig-zag branches,
from the angles of which the polypite stalks
arise. Other Sertularidce attach themselves to
stones or shells, and not a few of the smaller forms
occur parasitically on the stems of more conspic-

HTDROZOA. 91

uous Species. Examples of this habit are afforded
Fig. 1 8.

i

Morphology of Sertltlaeiad^ : — a, dried hydrosoma of Sertu-
laria tricuspidata ; b, portion of the same ; c, fragment of the
ccenosarc from a dead specimen of Halecium kalecinum ; d, gono-
blastidium of H. Beanii ; e, three gonoblastidia of Sertularia
argentca ; — ir', hydrotheca; k, ccenosarc; p\ gonoblastidium,
(All, except a, magnified.)

by the genera Coppinia and Reticularia, and by
several of the true Sertularice,

92 HYDROZOA.

The coenosarc, in all cases, excretes a very firm I
chitinous polypary, usually of a pale horny colour, |
which may either remain throughout in close conti- I
guity with the ectoderm, or become separated from
it at regular intervals, so as to impart an elegant
ringed appearance to portions of the tree-like
structure (y?(/. 19,6). This semi-transparent, horny,
sheath persists long after the destruction of the
soft parts of the organism, so that, among the
larger species of Sertularidce, the peculiar form
of the hydrosoma is sufficiently well seen in dried
specimens. Here, therefore, the polypary differs
from that of the Gorynidce in its firmer texture,
but the most important distinctive feature of the
present order is found in the occurrence of the
hydrothecse ; organs which do not exist in any
other group oi Hydrozoa {jig. 18, 6). The nature
of these appendages has already been explained.
Their numerous, often beautiful, diversities of
form and mode of arrangement afford aids to the
definition of the minor types of structure
which occur within the limits of this circumscribed
group. In CaTupanularia fastigiata the distal
end of the hydrotheca forms, according to Mr.
Alder, a sort of " operculum, which, when closed,
slopes down on each side like the roof of a house,
the two opposite angles forming the gables. When
the operculum is fully open, the folds disappear,
and the edges unite into a continuous rim round
the top of the cell."

The polypites of the Sertularidce, more minute
than those of the Gorynidce, differ little from one
another, either in form or the general arrangement
of their tentacles. In Sertulariadce proper the
polypites are sessile, while in the Campanulariadce

IIYDROZOA. 93

each is elevated on a conspicuous stalk. An
intermediate condition is presented by the genus
Haleclum, the polypites of which are 'sub-ses-
sile,' each hydrotheca being jointed to a short
procef!s of the coenosarc {fig. 1 8, c).

The tentacles, though apparently disposed.
Hydra-like, in a single row below the mouth, are
found, on close examination, to exhibit an indis-
tinct alternate arrangement ; shght differences in
length and position distinguishing those of the
two series. The peculiar rough appearance which
each tentacle presents resolves itself under the
microscope, into rows of minute elevations, or
* palpocils,' witliin which numbers of thread-cells
are lodged. The tentacles are filiform, tapering
gradually towards their free extremities. In
Camjpanidina a delicate web-like extension from
the body of the polypite unites these appendages
for about a sixth of their entire length.

Allusion has elsewhere been made to the nema-
tophores, or characteristic organs of offence, noticed
by ^Ir. Eusk in the genus Plumulai^ia, and one
or two of its immediate allies. These singular
appendages are well deserving of minute investi-
gation. Their offensive nature seems proved by
the abundance of thread-cells in their interior,
coupled with the fact that certain species of Plu-
mularia have been observed to sting wdth some
severity. In Plumularia proper one of these
organs arises on either side of each hydrotheca,
while in Halicornaria they are situated, between
the pol3^pites, on the general surface of the
coenosarc.

The reproductive organs vary, perhaps, less
than those of the Corynida\ and are usually sup-

94

HTDROZOA.

ported on the curiously modified gonoblastidia,
whose structure has previously been described.
Among the Campanulariadce they frequently
assume the form of free-swimming medusoids;

Fiff. 19.

Morphology of Canpanxjlaeiadje : —a, Laomedea neglecta ; b,
portion of the same; c, gonoblastidium of Camiyanularia voliobi-
lis ; d, gonoblastidium of C. Jolmstoni ; e, gonoblastidium of
C. syringa ; f, the same in an earlier stage ; g, upper portion of
c, slightly compressed. (All, except a, magnified.)

but in the Sertulariadoi seldom, if ever, become
detached.

In some Plumularioi the gonophores appear to
be naked. In P. cristata the branch bearing
these o-rgans undergoes a" curious metamorphosis

IITDROZOA. 95

by the development from its opposite sides of
alternate leaflets, which eventually arch over, and
unite with one another, forming a basket-like
receptacle, or ' corbula,' within which the repro-
ductive bodies are lodged.

In Sertularia pobjzonias and some other spe-
cies only one gonophore, consisting of a simple
closed sac, arises from the gonoblastidial column,
and, by the protrusion of this sac beyond the
orifice of the urn, an external capsule, or ' acro-
cyst,' is formed, into which the ova are trans-
ferred at a certain period of their development
{Jig. 19, e,/, and^).

In Camioanularia Loveni the ripe gonoblas-
tidium displays at its summit the medusa-like
gonophores already alluded to, whose form is, in
many respects, so peculiar that Professor Allman
has proposed to designate them by a distinct
name, *meconidia' {Jig. 10). The reproductive
elements of this species are developed, as in the
Goi'ynidcv, between the ectoderm and endoderm
of the manubrial wall, while in other Sertularidce,
with medusa-like gonophores, they arise in the
course of the calycine canals.

The order Sertidaridce includes two families.

Order SERTULARID^.

Family i. Sertulariad^.

Hydrothecce, and polypit^s, sessile.

Family 2. Campanulariad^.

Hydrothecoe, and polypites, stalked.

A more extended acquaintance with the posi-
tion of the nematophores may perhaps afford
grounds for modifying this arrangement.

96 HTDEOZOA.

5. Or€ler4: Calyeophoricliip. — The members
of the next order, Calycophoridcc, appear, at first
sight, very dissimilar in aspect to the fixed Hy^
drozoa, which, nevertheless, in all essential cha-
racteristics, they closely resemble, Diphyes, the
type of the group, presents a delicate filiform
ccenosarc, to the proximal extremity of which are
attached two large, firm, mitrate, nectocalyces
[fig. 20, a). To these appendages, which differ
slightly in form, the distinctive terms of * proxi-
mal ' and ' distal ' have been assigned. The former,
as its name imports, precisely terminal in position,
is furnished with a conical cavity running parallel
with, but distinct from, its nectosac. Into this
cavity is fitted the apex of the distal nectocalyx,
along the inner surface of which it prolongs itself
as a lengthened groove, with its sides arched over
in such a manner as to form a more or less per-
fectly closed canal. The cosnosarc, with its nu-
merous appendages, freely glides up and down the
peculiar chamber, or ' hydroecium,' thus produced,
into which it can, upon occasion, be completely
retracted. The ccenosarc itself dilates slightly
towards its proximal extremity into a small ciliated
chamber, which, narrowing above, becomes con-
tinuous with a sac of larger size, termed the ' soma-
tocyst.' This, too, is ciliated, its cavity appearing
in most cases almost obliterated through excessive
vacuolation of the endoderm. The somatocyst
is firmly embedded in that portion of the prox-
imal nectocalyx which forms the upper boundary
of the hydroecium, while from the smaller ciliated
chamber two ducts are given off, one to the distal,
the other to the proximal nectocalyx, where
each communicates with the small cavity common

HTDROZOA.

97

to the nectocalycine canals. Along the sides of the
coenosarc are placed the several appendages, con-

Fig. 20.

w

Morphology of Caxycophobid* : — a, Diphyes appendiculata ;
b, Vogtia pentacantha ; — v, proximal nectocalyx of DipAyes ; e',
its posterior contour ; 6, its nectosac ; v", distal nectocalyx ; c",
its nectosac ; |, somatocyst ; i, proximal portion of hydrcecium ;
f, proximal extremity of coenosarc ; k, its distal extremity ; ir,
polypite, with its tentacle ; v, v, nectocalyces of Vogtia ; ir, it,
its polypites ; t, t, their tentacles ; 6, androphore ; w, gynophore.
(Natural size.)

sisting chiefly of polypites, tentacles, hydrophyllia,

and organs of reproduction. Large specimens of

II

98 HYDROZOA.

Diphyes attain, when fully extended, a length of
several inches, their coenosarc giving support to at
least fifty distinct polypites. Of the great beauty of
these, and other oceanic Hydrozoa, no description
can adequately treat. So transparent, in many
cases, is the delicate coenosarc, that its course upon
distant inspection is revealed only by the bright
tints of some of its appendages. A touch is often
sufficient to separate it from the nectocalyces,
which, from their size and firm consistence, con-
stitute the most conspicuous portions of the or-
ganism. Hence the origin of the generic name,
Diphyes, devised by Cuvier, who regarded the two
swimming organs as distinct animals, imperfectly
united with one another.

An unbranched, filiform, coenosarc occurs in all
Calycophoridcc. In Hippopodias its proximal
extremity lolds inwards to form a loop, so that the
true position of the nectocalyces is thereby some-
what confused.

Of the many appendages to the coenosarc by far
the most remarkable are those just mentioned. In
accordance with the relative number, structure,
and arrangement of these organs, the few genera
of the order hitherto carefully examined may
readily be identified and separated from one ano-
ther ; as shown in the accompanying table.

Artificial Arrangemext of CALYCOPHORIDiE.

f Nectocalyces two in number ..... i
Nectocalyces numerous, similar 4
A single, proximal, spheroidal, nectocalyx . Sphozronectcs.
r Nectocalyces unlike in size and form .... 3
|_ Nectocalyces similar ..... Praya.
i Proximal nectocalyx equal to, or larger than,
3 < the distal one ..... Diphyes.
[ Proximal ncetocalj'x shorter than the distal Ahyla.

HYBROZOA. 99

Nectocalyces horse-shoe shaped . . . Hipj^opodin.s.

Nectocalyces concave externally, " and pro-
duced into five points of which the three
upper are much longer and stronger than
the two lower." Vogtia.

Pray a, Ilippopodkis, and Vogtia have * in-
complete' hydroecia, the nectocalycine groove along
which the coenosarc glides not forming, in these
genera, a closed canal. In Praya, however, the
two, nearly symmetrical, terminal nectocalyces have
their open grooves so applied to each other as to
form, by their apposition, a short tube (^fig. 4-, d).

The polypifes and tentacles of the several genera
of Calycoph.rldce present no very striking differ-
ences of structure.

Xot so, however, the hydrophyllia. Ahyla, the
genus most closely allied to Biphyes, is distin-
guished from tliat form not merely by its necto-
calyces, but also in having thick, facetted, hydro-
phyllia, the edges of which do not overlap one
another. In Diphyes the hydrophyllia are folia-
ceous, smooth externally, slightly convex, and folded
so that their edges freely overlap.

In Praya, " 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 ca3cal processes." But in Vogtia, Hippopjo-
dius, and, perhaps also, Sphceroiiedes, these or-
gans are absent altogether {fig. 20, h).

The reproductive bodies of the Calycophovidce
are always medusiform, and attached to the pe-
duncles of their respective polypites. In Vogtia
and Hippopodms the manubrium attains a large
H 2

100 HYDROZOA.

size, extending far beyond the margin of the short
gonocalyx. In other genera the reverse is usually
the case, the manubrium being shorter than the
swimming cup within which it is suspended. Each
gynophore, when fully developed, appears to con-
tain several ova. In most Calycojphoridce, except
Diphyes itself, both male and female reproductive
appendages appear on the same hydrosoma.

Four families of Ccdycophoridce have been de-
fined by Professor Huxley. Their characters we
subjoin.

Order CALYCOPHOKID^,

Family i. Diphyd^.

Galycophoridce with not more than two,
polygonal, nectocalyces. Proximal hydroeoium
complete. Hydrophyllia.
Family 2. Spu^ronectid^. I

Galycophoridce with probably not more
than two nectocalyces; the proximal one
being spheroidal, with a complete hydroecium.
No hydrophyllia?
Family 3. Prayidjs.

.Galycophoridce with only two nectocalyces,
whose hydrcBcia are both incomplete. Hy-
drophyllia, >
Family 4. Hippopodiid^. ■
Galycophoridce with many nectocalyces,
)vhose hydroecia are incomplete. No hydro-
phyllia,,

The same naturalist has proposed the distinctive
term of ^Diphyozooids' for those singular detached
reproductive portions of adult Galycophoridce
which received the name of '* monogastric Diphy-

HTDROZOA. 101

cZoe" from earlier observers. Their true nature
was first demonstrated by R. Leuckart, who several
times witnessed the separation of these bodies
from a well-known species of Ahyla. Groups
of organs became detached from the cceno-
sarc, each group consisting of a hydrophyllium,
polypites, tentacles, and gonophores, with a small
portion of the coenosarc itself. More frequently,
however, the actual detachment of the Diphyozooid
has not yet been observed, so that the precise origin
of many still presents a subject for inquiry. Pend-
ing further investigation, it seems right to designate
such forms by provisional generic and specific
names, of which not a few have already been con-
ferred.

6. Order 5: Physophoridae. — The Physo-
phoridce differ much more among themselves
than do the members of the order just mentioned.
All, however, agree in having the proximal end of
the coenosarc modified to form the pneumatophore,
or float, which presents so characteristic a feature
in the physiognomy of these animals. The cavity of
this pneumatophore is a simple enlargement of that
of the coenosarc, the walls of both being directly
continuous. To the apex of the cavity is attached
a firm, elastic, apparently chitinous sac, known as
the ' pneumatocyst,' containing a greater or less
proportion of air. A layer of endoderm, reflected
from the pneumatophore, invests the whole outer
surface of its contained pneumatocyst, which is thus
completely cut off from the somatic cavity below.
The lower extremity of the pneumatocyst is usually,
if not always, entire. Its apex, though most fre-
quently closed, is open in Pkysalia and Rhizophysa,
H 3

102 HYDBOZOA.

and, the free extremity of the pneumatophore
being likewise perforate, a communication exists,
in these genera, between the cavity of the pneuma-
tocyst and the surrounding medium. In Rhizo-
jjliysa, moreover, peculiar long branched processes
freely depend from the distal surface of the pneu-
matocyst. Each process consists of a layer of the
investing endoderm containing in its axis clear cel-
l^form bodies, -02 of an inch long, each of which
includes an opaque oval endoplast, about -i-th of
these dimensions, and this, in its turn, a more mi-
nute particle or nucleolus, oval or circular in form,
and 'OOoSth of an inch in diameter. In Agahna
and Forshalia radiating membranous partitions
connect the walls of the pneumatophore with those
of the pneumatocyst, below which each terminates
in a free arcuated edge. In Velelia and Forpita
the pneumatocyst is furnished with several open-
ings, or stigmata, communicating with the exterior,
while to its distal surfa^ce are attached a number
of long slender processes enclosing air, and hence
termed the ^ pneumatic filaments.'

Excepting the presence of the pneumatophore
and the absence of a somatocyst, the general plan
of structure in these Hydrozoa differs little from
that of the Calycophoridw. In Apolemia, as in
Dipliyes, the numerous groups of appendages are
supported at intervals along a slender, unbranched,
connecting stem. Physophora, the type of the
order, has a filiform, but comparatively short, coe-
nosarc, terminated proximally by a pneumato-
phore of moderate size, below which the greater
portion of its length is occupied by a double
series of nectocalyces, each alternating with its
successor on the opposite side, and deeply grooved

HTDROZOA. 103

on its inner face for attachment to the coenosarc,
[jir). 22, h). The distal extremity of the latter
forms an expanded bulb, above which are disposed,
in a spiral or circular manner, the various appen-
dages; consisting of polypites, tentacles, hydrocysts,
and organs of reproduction. Of these the hydro-
C3'sts are uppermost, or external; next come the
polypites, with a tentacle at the base of each,
between, or above, which the gonophores, of both
sexes, are arranged. The usual length of Fhyso-
phora is about two inches.

The typical genus just described may advan-
tageously be contrasted, on the one hand, with
such forms as Apolemia or Halistemma, on the
other, with the mdely different, though equally
aberrant, genera, Porpita and Velella.

In Halistemma ruhrum the appendages are
attached to a thread-like stem, nearly forty
inches in length, having a float of only three or
four lines in its longest diameter, close beneath
which the swimming-bells, about sixty in number,
extend in two parallel rows for a distance of six
or seven inches. The remainder of the coenosarc
is occupied by the polypites, tentacles, hydrocysts,
bracts, and reproductive buds, all associated in one
continuous series. Especially conspicuous, from
their bright vermilion hue, appear the complex ten-
tacular sacculi, while fainter longitudinal bands of
the same colour mark the hepatic striae of the
polypites, whose size, in this genus, is considerable.
The general aspect of this most i)eautiful, yet
withal, extraordinary being, has been compared
by Vogt, its discoverer, to that of a delicate, trans-
parent garland of flowers, endowed, in a marvel-
lous manner, with life and activity.

104

HYDROZOA.

Far different is the physiognomy of Velella,
whose coenosarc appears almost wholly lost in the
horizontal, or slightly convex, rhomboidal pneu-
matophore, which distinguishes this singular genus.
The proximal surface of the pneumatophore is
traversed diagonally, from one of its angles to the
other, by an upright, triangular crest, which, in
common with the horizontal disc, consists of a soft

Fig, 21.

<^m

Morphology of Velella : — a, Vclclla spirans, somewhat en-
larged ; h, one of its smaller polypites, much magnified ; — u,
crest ; A, liver ; o, mouth of polypite ; 8, its digestive cavity ;
tp, rounded elevations, containing tliread-eells ; p, medusiform

zoiiids.

marginal membrane, or " limb," bounding the
"firm part," or central portion {fig. 2i, a). To the
distal surface of the firm part of the disc are
attached the several appendages; including (i) a

IITDIIOZOA. 105

single large polypite, nearly central in position ;
(2), numerous small gonoblastidia, which resemble
polypites, and are termed ' phyogemmaria ' ; and,
(3), the reproductive bodies to which these last
give rise (6). The tentacles are attached, quite
independently of the polypites, in a single series
along the line where the firm part and limb of
the disc unite. There are no hydrocysts, necto-
calyces, or hydrophyllia. The average length of
Velella may be estimated at two inches, its
heicfht at one inch and a half. The entire or-
ganism is semi-transparent and tinged with an
ultramarine blue, which changes to a deeper shade
in the tentacles and limb of the disc.

On closer examination the firm part is seen to
enclose a hard, shell-like, pneumatocyst, con-
sisting of a horizontal division, included within
the disc, and continuous with the simple solid
vertical plate, which gives support to the sail
or crest. The upper surface of the pneumatocyst
is crossed at right angles to the direction of the
crest by a linear diagonal groove, indicated on its
under surface by a slightly elevated ridge, " 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 consists of two thin laminse,
passing into one another at their free edges, and
united by a number of concentric vertical septa,
between which are corresponding chambers filled
w^ith air. All these chambers communicate toge-
ther 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. Kolliker made

1 06 HTDROZOA,

the interesting discovery that many of the cham-
bers 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 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. Conse-
quently, as there are more than six concentric
chambers, some of them must communicate with
these stigmata only indirectly." To the under
surface of the five or six innermost chambers are
attached from ten to fifteen elongated hollow pro-
cesses containing air, the pneumatic filaments
already mentioned.

But complicated as the pneumatocyst of Velella
may seem, not less so is its curiously modified so-
matic cavity. On all sides the limb is traversed
by an anastomosing system of canals, which are
ciliated, and communicate with the cavities of the
phyogemmaria and large central polypite. Within
the roof of the latter, close beneath the pneuma-
tocyst, is lodged a peculiar brownish mass, the so-
called liver. This, also, is furnished with a canal
system of its own, which eventually becomes con-
tinuous with the sinuses of the limb.

In addition to the preceding organs Velella pos-
sesses certain large " glandular sacs," for the dis-
covery of which we are indebted to Vogt. He
describes them as presenting a very curious minute
structure, and as arranged in a single series around

HTDROZOA. 107

the margin of the limb, to open on its dorsal sur-
face, where they secrete a clear, viscid, mucus. The
true nature of this mucus, whether excretory or
lubricative, is still very imperfectly known.

Thus the most striking modifications of the
common plan of the Physo'phorldw depend on
differences in the relative size and shape of the
coenosarc and pneumatophore. Athorybia and
PhysaUa have, like Velella and Porpita, a dispro-
portionately large pneumatophore; but, in these ge-
nera, it is globular or pear-shaped, not, as in those,
discoidal. In PhysaUa, the true Portuguese Man-
of-war of sailors, often wrongly regarded as the
type of the present group, the float sometimes
attains a long: diameter of eio:ht or nine inches;
tentacles, several feet in length, being attached
directly to the coenosarc along its under surface
{jig. II, c). But, more frequently, the coenosarc
is filiform, with a small pneumatophore ; and, ex-
cept in the case of Rhizophysa, swimming-bells
are also present. Xectocalyces and hydrophyllia
are alike absent in Porpita, Velella^ Physalia, and
Rhizopjliysa. Athorybia has hydrophyllia, without
nectocalyces ; Physophora nectocalyces, but no
hydrophyllia. All other genera possess these two
kinds of appendages.

The swimming-bells of the Physophoridce, when
present, are more numerous than in the Galycopho-
ridce, and, among the different genera, vary much
in size, shape, and mode of attachment, as also in
the relative proportions of the nectosac. Each fre-
quently has its surface marked with grooves and
ridges, and may send forth processes which serve
to embrace the coenosarc, and connect it with its
fellow of the opposite side. In some genera, more

108 HTDEOZOA.

particularly Physophora itself, two of the necto-
calycine canals, which coincide with what may be
termed the medial plane of the coenosarc, remain,
as usual, straight, while the two other, or lateral,
vessels become convoluted in a most complicated
manner before reaching the circular canal. As in
the Calycophoridce, the common cavity of each
nectocalyx is connected with that of the coenosarc
by means of a tubular pedicle.

The hydrophyllia present variations both in
their structure, mode of attachment, and relations
to the other appendages. They may be either
foliaceous (Athorybia), or clavate {Apolemia), or
thick and wedge-like, or even pyramidal (Agal-
TYia) ; while their surface is liable to be diversi-
fied with excavations and ridges, having smooth or
serrate lateral margins. Their arrangement is,
in general, more or less whorled. In Apolemia
they are proximal to all the other appendages, in
each separate group, and here, as in Halistemma
and Stephanomia, they become connected with
the coenosarc by more or less distinct peduLcles.
In Agalma, the attached apex of each is pierced
by a duct which terminates in a caecal phyllocyst,
about the middle of the hydrophyllium, while its
opposite end opens into the somatic cavity. In
Forskalia the hydrophyllia are attached directly
to the peduncles of the polypites. The graceful
Athoryhia rosacea possesses from twenty to forty
of these organs, inserted, in two or three circlets,
immediately below the pneumatocyst, and above a
much smaller number of polypites (fig. 22, a).
In all other PhysophoiHdce with hydrophyllia,
nectocalyces also are present ; but Athoryhia,
though destitute of the latter appendages, has the

HTDROZOA.

109

power of alternately raising and depressing its
hydrophyllia, so as to render them agents of pro-
pulsion.

Fig. 22.

Morphology of Physophobid^: — a, AthoryUa rosacea; b,
Physophora Pkilippii; o, pneumatophore ; v, nectocalyx; v\ rudi-
mentary nectooalyx ; <p, hydrocysts ; tt, polypite ; t, tentacles ;
3, hydrophyllia. (About the natural size.)

110 HTDROZOA.

The polypites of the several genera differ chiefly
in size and mode of attachment. They may be
inserted in continuous series, along one side of
the coenosarc only, as in Stephanomia, or indif-
ferently on either side, as exemplified by Rkizo-
physa. In some cases they are attached directly
to the coenosarc ; in others, supported, with their
tentacles and hydrophyllia, upon special stalks.
In Apolemia they are arranged in groups of two
or three, along with the other appendages, at
intervals, as above mentioned; in PJtysophora
and Athoryhia they form a spiral or circlet
around its distal extremity, while in Pkyscclia,
Velella and Porpita, they are restricted to the
inferior surface of the much modified hydrosoma.

Two kinds of polypites, a larger and a smaller,
appear on the same coenosarc in certain genera.
Much doubt exists as to the true nature of the
latter, which in some cases appear to have been
mistaken for hydrocysts, in others, for gonoblas-
tidia. In Agalma the smaller polypites often
equal in length the true digestive zooids, but are
always much narrower, the rudimentar}^ tentacle
at the base of each presenting a striking contrast
to the highly complex prehensile organs attached
to the pedicles of the polypites properly so called.

The hydrocysts, though present in many other
genera, are especially conspicuous in Physopliora
and Athoryhia. In the former they are disposed,
in a circular series, external to the polypites,
around the expanded distal extremity of the
coenosarc, and, from their bright pink colour
and larger size, are more noticeable, on a first
inspection, than the true digestive appendages.
{fig. 22, h.)

J

IIYDrvOZOA. Ill

Attention has, in a previous section, been
directed to some of the modifications which the
tentacles of the Physojphoridce present. They
appear in Apolemia as simple tubular processes,
with numerous large thread-cells on one side : in
Velella, they are equally simple, but much shorter,
and slightly enlarged at their free ends. In
Porpita there is, in addition, a series of longer
prehensile appendages, having their distal ex-
tremities clavate and beset with stalked knobs, or
capitula, containing urticating organs, which are
wantinor in the smaller marorinal " cirrhi." In
Phi/sal i a, as above described, each tentacle con-
sists of a broad conical basal sac, and a long
simple ribbon-like process, having transverse reni-
form enlargements {fig. ii, d). In all other
Physophovidce the tentacula are furnished with
lateral branches, which in Rkizophysa alone
appear to be destitute of sacculi. These may
want involucra, as in HaUstemma or Forskalia,
or possess these structures, and become further
modified, as in all the remaining genera.

The structure of the gonophores, though always
medusoidal, is, in other respects, liable to much
variation. Among many Physophovidce well-
marked differences of size, aspect, or relative
number, distinguish the male and female reproduc-
tive bodies in the same species. Thus in Agcdma
and Physophora the gynophores are only half the
length of the androphores, than which, however,
they are more numerous. In Velella and Porpita
both androphores and gynophores become com-
pletely detached, and the same is probably true of
the female organs in Physalia. But the andro-
phores of this genus, and the two kinds of gene-

112 HTDROZOA.

rative appendages in many other forms, discharge,
in all probability, their proper functions, without
previous separation from the parent hydrosoma.
The gonophore either presents a well-developed
swimming-cup, with open margin and conspicuous
canals : or the calyx, with its canal system, may
remain in a very rudimentary condition, so as
scarcely to be distinguishable from its contained
manubrium ; while in the male organs of Stepha-
nomia and Athorybia the apex of the latter slightly
projects beyond the margin of the bell. In these
and many other genera each gynophore gives rise
to only a single ovum ; but there is reason to infer
that it may be otherwise with the free-swimming
gonophores of Physalia or Velella.

Most of the Physophoridce hitherto examined
appear to be monoecious, the androphores and
gynophores being borne on the same gonoblastidia,
as in Physalia, Agalma, and Athorybia, or, more
rarely, on separate stalks, as in Stephanomia.
Besides the reproductive bodies, the gonoblastidia
may give support to hydrocysts and other appen-
dages, as in Physalia or Athorybia. In Halis-
temma they are absent or so extremely short, that
the gonophores seem attached directly to the
coenosarc. The phyogemmaria of Velella and Por-
pita are obviously homologous with the gonoblas-
tidia of Hydractinia or the small polypites of
Podocoryne ; and just as the former genera differ
from other Physophoridce, Physalia indeed ex-
cepted, in their laterally expanded coenosarc and
independent tentacles, so, likewise, may Hydrac-
tinia be distinguished from all the more typical
forms of Corynidcc,

Of Physophoridoi Mr. Huxley has established

HTDROZOA. 113

seven definable families, whose characters may be
Rtated thus : —

Order PHYSOPHORID^.

Family i. ApolemiadyE.

Pneumatocyst small. Coenosarc filiform.

Nedocalyces and hydrophyllia present ; the
latter united with the other appendages into
groups, arranged at distant intervals along the
coenosarc.

Tentacula without lateral branches.

Family 2. STEPHA^'OMIADyE.

Pneumatocyst small. Coenosarc filiform.

Nedocalyces and hydrophyllicc present ; the
latter arranged with the other appendages in
continuous series.

Tentacula with lateral branches, terminated
by sacculi.

Family 3. Physophoriad.e.

Pneumatocyst small. Coenosarc filiform,
but short, and dilated at its distal end.

Nedocalyces present, occupying most of
the coenosarc below the pneumatophore. No
hyclrophyllia.

Tentacula wdth branches and involucrate
sacculi.

Family 4. ATnoRTBiAT).E.

Pneumatocyst occupying almost the whole
of the globular coenosarc.

Nedocalyces absent. Hydrophyllia.

Tentacula w4th branches and involucrate
sacculi, each having two filaments and a
median lobe.

114 HYDHOZOA.

Family 5. Rhizophtsiad.e.

Pneumatocyst small. Coenosarc filiform.
Nectocalyces and hydrophyllia absent.
Tentacida having branches, without sac-
culi.

Family 6. Physaliad^.

Pneumatocyst occupying almost the whole
of the thick and irregularly fusiform coenosarc,

Nectocalyces and hydrophyllia absent.

Tentacula with basal sacs, but no lateral
branches.

Family 7. Velellid^.

Pneumatocyst flattened, divided into cham-
bers by numerous concentric partitions, and
occupying almost the whole of the discoidal
coenosarc.

Nectocalyces and hydrophyllia absent.

Tentacula short, clavate, simple or branched,
submarginate.

A single central, principal polypite.

y. Order 6: Ifledusidae. — If the phyogem-
maria of Velella and Porpita be regarded as
gonoblastidia, the hydrosoma of these genera may
then be said to present not more than one true
polypite ; a character in which they differ from
all other Physophoridce, but agree with the mem-
bers of the next order, Medusidce, Cuvier, in-
deed, associated the VelelUdce, Medusidce, and
free zooids of the Lucernaridce in a single group,
under the name of " Acalephes Simples " ; the re-
maining Physophoridce, together with the Caly-
cophoridce, being distinguished as " Acalephes
Hydrostatiques ". But the Velellidce, like all other

IIYDROZOA. 115

PJil/sophoridce, differ, as has been shown, from
CalycoplLoridnif in possessing a float, and from
Mediisidce in the characteristic mode by which
the poh'pite is connected with the rest of the
hydrosoma. In both the Calycophoridce and Pliy-
sophoridiVy the nectocalyces (when present) and
polypites are separately attached to different parts
of the coenosarc. In the Meduslda', on the other
liand, the hydrosoma presents but one nectocalyx,
from the roof of which a single polypite is sus-
pended {fig. 4,/). The endodermal lining of the
polypite passes into the central cavity of the
swimming-organ, from which, as in other necto-
calyces, canals radiate, to join a circular vessel
surroundino: the margin of the bell. From this
margin depend tentacles, which may be either
hollow processes of both layers, in immediate con-
nection with the canal system, or, more rarely,
prolongations of the gelatinous ectoderm itself.
Around the outer margin of the nectocalyx, be-
tween the endoderm of the circular vessel and its
ectodermal investment, are embedded the mar-
ginal bodies, vesicles or pigment-spots, whose
peculiar structure has already been described

ifif/' 23).

The outward form of the poljrpite varies greatly.
It may be long and highly contractile, or stoutly
cylindrical, or so short and broad as to be with
difficulty discernible on the under surface of the
bell {fig, 24). Very often it is curiously con-
stricted. In internal structure it is not known to
present any peculiar features. The oral margin may
be either simple, everted, or produced into lobes,
which, most frequently, are four in number, though
in some forms it is much divided. In Liriope

116

HYDEOZOA.

Catharinensis, it is surrounded by a series of little
sacs, each well packed with thread-cells.

The size and shape of the nectocalyx in relation
to the polypite with which it ia connected may also
vary considerably. The veil which surrounds the
open margin of the nectosac in no case appears to

Fig. 23.

Morphology of Medusid^ : — a, Mednsid, seen in profile ; h,
the same, viewed from below ; c, its pol}-pite ; d, part of its mar-
ginal canal, and other structures in connection therewith; - v,
nectocalyx ; ir, polypite ; x\ marginal canal ; ^, veil ; t, tentacle ;
X, radiating canal ; w, reproductive organ ; o', coloured spot ; o",
marginal vesicle.

be absent. More than four longitudinal canals
sometimes occur. In Willsia these canals are seen
to bifurcate, each branch again dividing into two
others, so that, in this form, the six canals open by
twenty-four ducts into the circular vessel, (Ji.g. 24,
c). In Ounina, JErjina, and ^ginopsis, both cir-

HTDROZOA.

117

cular and radiating canals disappear, their place
being supplied by peculiar pouch- shaped processes,
communicating with the digestive cavity of the
polypite.

The tentacles scarcely require any special mention.
Usually the distal extremities of their cavities

Fig. 24.

Various forms of Medusid^ : — a, Mquorea formusa, seen in
profile ; b, the same, viewed from above ; c, upper view of fVillsia
stcUafa ; d, Slabberia conica ; e, portion of the marginal canal of
Tiaropsis Pattersonii ; f, polypite of Bougainvillea dinema ; g,
part of its marginal canal ; h, Stccnstrupia Owenii. (a, b and d
are about the natural size ; the others are magnified.)

become more or less obliterated through vacuolation
of the endodermal lining. In Trachynema and its
allies the tentacles are stiff; not contractile, as in
other Medusidce,

1 3

118 HYDROZOA.

The reproductive organs have been stated to be
of the simplest kind, consisting of mere expansions,
either of the polypite wall, or radiating canals,
within which the generative elements are produced.

x\t a time when the free gonophores of the
Hyclrozoa had been as yet imperfectly studied, it
was the custom of naturalists to regard these bodies
as independent individuals, worthy of being ar-
ranged under definable genera and species. The
singular resemblance of such gonophores to the
Medusidce began, at length, to attract attention.
Then it was suspected that many of the Medusidce
were not individual organisms properly so called,
but merely the free reproductive buds of various
Hydrozoa. Eventually it was proposed to abolish
the whole group of Medusidce, and distribute their
several forms among the different orders of the
class.

On the other hand, certain observations of J.
INIiiller, Fritz Miiller, Gegenbaur, and Claparede,
to which we have already referred, indicate the
probable existence of a group of Medusid forms
which appear to be the immediate products of
true generative acts, not of gemmation or fission,
(fig. 12).

In the present state of our knowledge, it seems
better to sum up the several aspects of this doubt-
ful question in the following series of conclusions,

1. That several of the organisms formerly de-
scribed as Medusidce are the free gonophores of
other orders of Hydrozoa.

2. That the homology of these free gonophores
with those simple expansions of the body-wall
which in Hydra and some other genera are known
to be reproductive organs by their contents alone

HTDROZOA. 1 1 9

is proved alike by the existence of numerous tran-
sitional forms, and an appeal to the phenomena of
their development.

3. That many other so-called Medusidce may,
from analogy, be regarded as, in like manner, me-
dusiform gonophores.

4. But that there may exist, nevertheless, a
group of Medusid forms, which may give rise, by
true reproduction, to organisms directly resembling
their parents, and, therefore, worthy of being
placed in a separate order under the name of Me-
dusidce.

All the Trachynemidce and ^ginidce belong,
according to Gregenbaur, to the order in question.
And to the same group may be referred, provision-
ally, that large assemblage of forms anatomically
similar to true Medusidce, but whose development
is unknown ; just in the same manner as genera
and species are established for those Diphyozooids
which, there is every reason to believe, are but
the detached fragments of other Calycophoridce.
Pending the study of the life-history of these am-
biguous Medusoids, their true nature must, also,
remain undetermined.

Such are the forms brought together by Gregen-
baur in the systematic table here annexed, which,
at the same time, concisely displays their most
striking anatomical peculiarities.

Order MEDUSID^

With radiating canals.

Eeproductive organs in the poly-
pite-wall. Ocelli at bases of the

tentacles . . . Family 1. Oceanid^b.

I 4

120

HTDRQZOA.

Eeproductire organs in the course
of the radiating canals.
Kadiating canals arising from
the base of the polypite.

Ocelli

Eadiating canals arising from
the outer margin of the poly-
pite. Vesicles. .
Reproductive organs as rounded
protuberances of the radiating
canals. Vesicles.
Tentacles contractile. .
Tentacles stiff.
Reproductive organs as flattened
expansions of the radiating
canals. Vesicles. .
With pouch-shaped prolongations
of the polypite, in which the re-
productive products are formed,
Vesicles

Family 2. Thaumantiada

Family 3. JEquoridjjj.

Family 4.
Family 5.

EucopiD^.

TuACHYNEMIDiE.

Family 6. Gebyonid^.

Family 7. JEaimDM,.

The Medusidce were termed by Eschscholtz,
Cryptocarpge ; by E. Forbes, G^ymnophthalmata ;
and by Gegenbaur, Craspedota. These words
contrast, respectively, with the names Phanero-
carpae, Steganophthalmata, and Acraspeda ap-
plied by the same naturalists to a large section of
the Lucernai'idce, In this group the family
Lucernariadce is not usually included, many na-
turalists, from a mistaken view of its organisation,
referring it to the class Actinozoa, of which
Professor Milne Edwards has recently made it a
distinct sub-class, under the title of Podactinaria.

8. Order J: l<iieernaridsp. — In these Lucer-
nariadce the body is more or less cup-shaped, and
frequently about an inch in height, terminating
proximally in a stalk of variable length, and
furnished with a hydrorhiza, which, like that of

IIYDROZUA. 121

Ilydixif is not permanently attached. Round the
distal margin of the cup arise a number of short
tentacles which, in Lucernaria itself, are disposed
in eight or nine tufts, but in Carduella form one
continuous series. Their free extremities appear
Bucker-like or capitate ; in Depastrum, however,
they are simply clavate. The whole organism is
semi-transparent, variously coloured, and of a
gelatinous consistence {fig. 25).

The cup, in Carduella, presents at its centre
a four-lobed mouth, which is easily seen to form
the free extremity of a distinct polypite, occu-
pying the axis of the entire hydrosoma. The
oral margin of this polypite is simple and slightly
everted. Its gastric region exhibits a number of
tubular filaments, arranged in vertical rows, and
projecting freely into the digestive cavity. In
transverse section the polypite may be described as
somewhat quadrilateral, with a sinuous outline,
which expands at its four angles to form as many
deep longitudinal folds, within which the simple
generative bands are lodged. The space between
the polypite-wall and the inner surface of the cup
is divided in the following manner. From each pro-
jectiug angle of the gastric region run a pair of ver-
tical septa, which diverge widely from one another
so as to reach the wall of the cup at points precisely
opposite the two sinuosities on either side of the
generative band. Thus four of the equidistant
lines along the inner surface of the cup receive
two converging septa, each, however, belonging to a
different pair. These last septa, with the polypite
wall, serve to enclose four wide longitudinal
canals, outside of which are four other spaces,
bounded, within, by two septa of the same pair.

122

HYDROZOA.

and, externally, by the cup itself. The outer
canals are closed superiorly by a roof, consisting
of four inflected lobes from the summit of the
cup ; the inner spaces remaining open. There are
also four very narrow canals coinciding with the
lines where the vertical septa and inner surface of
the cup meet ; and formed, it would seem, partly
by these septa and partly by folds in the cup's
substance. Lastly a circular sinus has its course
immediately beneath the insertion of the ten-
tacles.

Fig. 25.

LucEBNABiA : — Two specimens of Lucernaria auricula, attaclied
to a piece of sea-weed. That figured to the right is somewhat
abnormal, haying a ninth tuft of marginal tentacles. (Natural
size.)

By means of a band of muscular tissue which
traverses its margin, and another set of fibres
which radiate towards the polypite, the distal ex-
tremity of the cup can fold inwards and contract
itself at the pleasure of the animal. Some Lucer-
nariadce have been observed to detach themselves,
and swim in an inverted position by the slowly
repeated movements of their cup-like umbrella.
In this respect they agree with Pelagia, a much
more active and permanently free member of the
same order.

HTDROZOA. 123

The swimming organ of Pelagia is sub-globose,
about three inches in diameter, divided at its
margin into sixteen lobes. Under eight of these
lobes are seen notches, each lodging a hooded
lithocyst, while from the remaining lobes depend
an equal number of long, contractile tentacula.
A polypite, short and broad, is attached, proxi-
mal] y, to the concave centre of the umbrella ;
distally, it terminates in four furbelowed lips,
which extend to a length of nearly four inches.
A number of csecal sacs, corresponding with the
lobes of the umbrella, are prolonged from the
digestive cavity. In other characters Pelagia
resembles the free zooids of Aurelia and its allies

ifif/' 7. b)-

The Lucernaridai admit of being arranged

under two principal sections, in one of which the
development is continuous, in the other, discon-
tinuous. The first section includes Pelagia and
the Lucernariadce, in which reproductive elements
are produced by the organism immediately re-
sulting from a generative act. In other members
of the order, the primitive result of this act is a
fixed and sexless * Lucernaroid,' which gives rise
by fission to free zooids of disproportionate size,
in which the reproductive organs are developed.
The first section, again, includes two minor di-
visions, in one of which the umbrella is perma-
nently free, in the other, furnished with an organ
of attachment. But the developmental cycle of
each Lucernarid belonging to the second section
presents these two principal forms.

It is worthy of remark, that the Lucernaroids
of very different genera — such as Cephea and
Chnjsaora — are often wonderfully alike in struc-

124 HTDEOZOA.

ture ; so that the relation between the producing
and produced zooid is here by no means the same
as in the other orders of Hydrozoa, The true
import of this fact should not escape attention.

All the Lucernaridce may be at once distin-
guished by their umbrella. The cup or disc in
the Lacernariadce and Hydra-tubae, the swimming
organ of Pelagia and of the free zooids, are alike
included under this designation. A free umbrella
differs from a nectocalyx, with which it is often
confounded, (i), in the absence of a veil ; (2), in
its mode of development; and (3), in the nature of
its canal system and marginal bodies. The ra-
diating canals, never less than eight in number,
send off numerous anastomosing branches, which
form a very intricate net-work. The peculiar
structure of the lithocysts has been previously
explained. Each is supported on the end of a
short double-walled stalk, the cavity of which
runs into one of the radiating canals. Protection
is given to this apparatus by a hood-like, cres-
centic fold of the ectoderm, at the base of which,
and on the convex surface of the umbrella, a funnel-
shaped orifice has been observed, whereby the ra-
diating canal communicates with the exterior.
Apertures similar in function, but not in position,
have been met with by Mr. Huxley in certain of
the Medusidoe, There are no lithocysts in the
Lucernariadce, unless the simple tubercles, placed
between the tentacular tufts, on the margin of
Lucernaria auricula, be regarded as these organs
in a rudimentary condition.

The Lucernaridw manifest another characteristic
feature in their gastric filaments, the presence of
which appears to be universal throughout the

HTDROZOA. 125

order. They may, without difficulty, be observed
in the common species of Lucernaria. They are
usually solid, perhaps through vacuolation ; contain
thread-cells, and, even when detached, execute a
peculiar writhing kind of movement. In Chrysaora,
according to Fritz Miiller, they attain a length of
several inches.

Three families of Luceimaridce may be defined
as follows : —

Order LUCERNAEID^.

Family i. Luceenaeiad^.

Reproductive elements developed in the
primitive hydrosoma, without intervention of
free zooids.

Umbrella with short marginal tentacles
and a proximal hydrorhiza.

Polypite single.

Family 2. Pelagid^.

Reproductive elements developed in a free
umbrella, which either constitutes the primi-
tive hydrosoma or is produced b}'' fission from
an attached Lucernaroid.

Umbrella with marginal tentacles.

Polypite single.

Family 3. Rhizostomid^.

Reproductive elements developed in free
zooids produced by fission from attached
Lucernaroid s.

Umbrella without marginal tentacles.

Poly piles numerous, modified, forming with
the genitalia a dendriform mass depending
from the umbrella.

126 HYDROZOA.

Not far from Pdagia, but in a family by itself,
Gegenbaur has placed the genus Chm^bdea,
Fritz jNIiiller, however, shows, that in the closely
allied Tamoya, a distinct veil is certainly present,
while Gharyhdea itself is furnished with marginal
processes, which seem to represent the same ap-
paratus.

Section IV.

DISTRIBUTION OF HYDROZOA.

I. Relations to Physical Elements. — 2. Bathymetrical Distribution.
— 3. Geographical Distribution.

1. Relations to Physical Elements. — The

Ilydrozoa, as a class, are almost exclusively marine;
Hydra and Cordylophora being the only fresh-
water genera hitherto described.

2. Bathymetrical Distribution. — The ma-
rine Hydrozoa, with reference to their distribution,
may conveniently be divided into two groups, the
fixed and the oceanic. The fixed Hydrozoa, Cory-
nidce and Sevtidaridoe, are less abundant between
tide marks than at depths of a few fathoms, some
forms extending their range to very deep water.
The Coinjnidce are, perhaps, on the whole, more
partial to shallow waters than the Sertularidce,
certain species of the latter order, especially of the
genus Campaiudariay being found at considerable
depths. But the vertical distribution of several
forms is more limited than that of others. Thus
Clava and Coryne appear usually not to wander

nTDROZOA. 127

far from low-water mark, while Tuhularia occurs
at depths varyiniy from less than one to more than
fifty fathoms.

The oceanic Hydrozoa in fine weather swim
near the surface of the water, the approach of rain
or wind compelling them to retire for safety to the
more tranquil depths below. The large "jelly-
fishes " which, during summer and autumn, occur
so abundantly in our seas, are, with few exceptions,
the reproductive zooids of Aurelia, Cyanea, and
Chrysaora. Equally numerous with these, but less
conspicuous from their extreme transparency,
appear hosts of minute medusoids, while Diphyo-
zooids, Velella, Physalia, and one or two other
Physophoridw may, at rarer intervals, be de-
tected.

3. Oeo^raphical Distribution. — The genera
of Hydrozoa are very widely distributed, renewed
investigations tending rapidly to diminish the
number of those supposed to be peculiar to certain
regions of the globe.

The limits of the area inhabited by Hydra have
not yet been definitely ascertained. The other
fresh-water genus, Gordylophora, has been met
with only in Denmark, Great Britain, Ireland and
North America.

Not much is known accurately of the geographi-
cal range of the Corynidoi ; the Sertularidce,
from the ready preservability of their polypary,
having been far more extensively studied. Sertu-
lariaf Plwinularia, Antennularia, and Campa-
nularia are truly cosmopolitan, and the same may,
likewise, be said of some species of these genera,
for example, S. opercidata. Many South African

128 EYDEOZOA.

Sertularidce are identical with European forms,
both being, in a large number of cases, sufficiently-
distinct from the Australian, Phillipine, and New
Zealand species. Several British species cannot,
however, be distinguished from those of the Atlan-
tic coasts of America, while on the other hand,
greater differences prevail between these last and
the North Pacific forms. Among purely exotic
genera may be mentioned CryiDtolaria, one species
of which has been found in Madeira and another
in New Zealand, and Lineolaria, a remarkable
Australian Sertularid, having gonophores mth two
longitudinal rows of stroug spines elevated m ridges,
between which a few smaller spines are scattered
over a flattened, transversely furrowed area.

The Calycophoridw and Physophovidce have
hitherto been most successfully studied in the
Mediterranean and Southern seas ; some genera,
such as Diphyes and Agalma, having been obtained
by Sars at a latitude of 6iJ° N. off the shores of
Norway. It were premature to describe any of
these forms as peculiar to certain regions, many of
the species and genera ranging over areas of consi-
derable magnitude.

Precise information is much wanting on the dis-
tribution of the Medusidce and Lucernaridce,
The free zooids of some species are very extensively
diffused, and are occasionally met with by sailors
in numbers so immense as almost to impede navi-
gation. Our common Aurelia aurita has been
obtained in the Red Sea, off the east coast of North
America, and in various parts of the southern
hemisphere.

A few words on the phosphorescence of the

HYDROZOA. 129

Hydrozoa may here be inserted. This property-
has been observed in most orders of the class,
though, among the Fhysophoindce, 8tephaii07nia,
and, of the LucernaTidcv, Pelagia are most re-
markable for its manifestation. Some, however,
of our more common jelly-fishes are also luminous.
It does not appear that, in any of these, special
light-giving organs exist. In the Medusidce, the
phosphorescence chiefly arises from around the
marginal bodies, but, in some instances, it is
emitted by the reproductive swellings, and, occa-
sionally, by the walls of the central polypite.
Our own Thaumantias lucifera, a species by no
means rare, displays this phenomenon in a very
beautiful manner. The little creature, when ir-
ritated b}'- contact of fresh-w^ater, marks its position
by a vivid circlet of tiny stars, each shining from
the base of a tentacle.

Such small Medusidce are, doubtless, more
efficient in promoting the luminosity of the ocean
than their larger and, at times, more brilliantly
conspicuous congeners. But the fixed Hydrozoa,
which, obviously, can take no share in this display,
are, also, eminently phosphorescent. A remark-
able greenish light, like that of burning silver,
may be seen to glow from many of our native
Sertidaridce, becoming much brighter under
various modes of excitation. It is an error to
suppose, however, that thus alone do these cold,
oily, flames emanate. "If (writes Professor E.
Forbes) a bunch of one of the bushy corallines,
such as Sertularia abietina, be plunged when
active and alive imto fresh-water or spirits, a
gorgeous display of living stars is instantaneously
produced."

130 HTDROZOA.

SECTION V.

RELATIONS OF HYDROZOA TO TIME.

Well-preserved remains of extinct Hydrozoa
are wanting. Obscure indications of fossil Sertu-
laridce and, perhaps, also of Lucernaridce, have
on a few occasions been met with, but of too
fragmentary a character to permit of definition.
Professor Agassiz, indeed, states that, many years
ago his " attention was attracted by two slabs of
limestone slate from Solenhofen, the counterparts
of one another, upon which a perfect impression
of a Discophorous Acaleph was distinctly visible."

The G-raptolites and ()ldhamiae have, by some
naturalists, been referred to the present group.
Both, however, may, with more propriety, find a
place in the Molluscan class of Polyzoa,

131

CHAPTER TIL

THE CLASS ACTINOZOA.
Section I.

MORPHOLOGY AND PHYSIOLOGY OF ACTINOZOA.

I. Type of the Class : Actinia. — 2. General Morpholoo:y. — 3. Or-
igans of Nutrition. — 4. Prehensile apparatus. — 5. Tegumentary
Organs. — 6. Coralkun or Skeleton. — 7. Muscular System and
Organs of Locomotion. — 8. Kervous System and Organs of Sense.
— 9. Reproductive Organs.

I. Type of the Class: Actinia The ^C-

tinia, or Sea-anemone, is the type of the class
Actinozoa {fig. 26, c).

The body of Actinia presents a soft, fleshy, ot
leathery consistence, and varies much both in
form and size, according as it assumes its con-
tracted or expanded condition. Average speci-
mens attain, when expanded, a diameter of from
one to three inches, their height being rather
less; but these dimensioos are often exceeded.
The expandedJ.c^i}iia is somewhat cylindrical in
figure, attaching itself by one of its flattened ends,
known as the * base,' a mouth being placed in the
centre of the ' disc,' or opposite extremity. Nu-
merous tentacles, disposed in alternate series,
surround the disc's outer margin, between Avhich
and the mouth a region destitute of any append-

132 ACTINOZOA.

ages, the * peristomial space,' is usually observ-
able.

The mouth is of a slightly elliptical form, a
pair of tubercles, including between them a
groove, being situated at each of two opposite
points of its circumference. The stomach, or
digestive sac, into which the mouth directly leads,
is a short, distensible tube, open at both ends,
and extending about half-way towards the base of
the animal ; in diameter scarcely exceeding the
mouth itself, with which its form, when viewed
from above, is seen to correspond. The folds, or
grooves, between the oral tubercles are continued,
in the form of semi-canals, along the inner surface
of this stomach, until, finally, they reach the wide
aperture by which it communicates with the so-
matic cavity.

A transverse section of the body of Actinia
exhibits two concentric tubes, the outer being
constituted by the body-wall, the inner by the
digestive sac. The wide space which intervenes
between these tubes is divided by a number of
radiating partitions, or ^mesenteries,' arising at
definite intervals from the inner surface of the
body-wall. The 'primary,' or first-formed and
widest, mesenteries serve to fix the stomach in its
place, their inner edges being inserted throughout
the entire length of its outer surface. From the
base of the stomach, the inner edge of each me-
sentery, becoming free, arches, at first, abruptly
outwards, and then, more gradually, downwards
and inwards, until at length it reaches the centre
of the base, from which all the primary mesen-
teries appear to radiate. Other partitions, de-
veloped in successive cycles between those just

ACTINOZOA.

133

mentioned, and having no connection with the
stomach wall, are distinguished, in accordance
with their relative na^^o^vness, as ^ secondary me-
senteries,' * tertiary mesenteries,' and so on.

Fig. 26.

Morphology of Actinozoa: — a, polype of Alcyonium; b, ideal
transverse section of the same ; c, longitudinal section of Actinia ;
— k', somatic cavity; ff, mesentery; 5', digestive cavity; 5, wall of
digestive cavity ; 0, mouth ; t, tentacles ; c/c, ectoderm ; ev, eudo-
derm ; fi, muscular layer ; )3', base ; p, reproductive organs ; p',
convoluted filaments, containing thread-cells, (a and b are en-
larged ; c is of the natural size.)

The mesenteries of each cycle are arranged in
alternate pairs, while those belonging to opposite
sides of the body correspond and are similar to
one another. Externally, the mesenteries are
often indicated by lines or ridges which traverse
the whole length of the column, and are continued,
K .3

134 ACTINOZOA.

in radii, along the base and disc. Their arrange-
ment is best seen in living, semi-transparent
species, without any recourse to dissection.

Thus, a number of imperfect chambers are
formed, all opening into one another below, or
beyond the free edges of the mesenteries ; and, in
some cases, apertures occur in the sides of the
mesenteries themselves by which a further com-
munication is kept up. These apertures usually
appear in the midst of the wide upper portion of
the mesentery, not far from the under surface of
the disc. They are most constant in the primary
partitions ; the secondary mesenteries being fre-
quently imperforate. The tentacles, which are
hollow, and, in many Actinise, perforate at their
free extremities, open directly into the somatic
chambers.

To the faces of the mesenteries are attached
the reproductive organs, which occur as thickened
bands of a reddish tint, containing ova or sper-
matozoa. The male and female organs appear
perfectly similar, previous to examination of their
contents. Most Actinise are dioecious, but, by no
external character can the individuals of both
sexes, which seem to be about equally numerous,
be distinguished from each other. Accurate obser-
vations are yet wanting on the reproduction of
Actinia, It is probable that the spermatozoa,
first diffused in sea-water, find their way through
the mouth to the ova contained in the general
cavity of the body.

A long convoluted cord, or ' craspedum,' arises
in front of the reproductive apparatus, along the
free edge of each mesentery. In addition to the
craspeda, other organs of similar structure, termed

ACTINOZOA. 135

'acontia,' are occasionally met with. Both cras-
peda and acontia are richly furnished with thread-
cells, for the emission of which special apertures
along- the wall of the somatic cavity have, in some
species, been observed. Mr. Grosse, who gives the
name of ^cinclides' to these apertures, describes
them as varying considerably in size and opening
directly into the somatic chambers. " Each is an
oval depression, with a transverse slit across the
middle." The sides of the cinclis can be opened
or closed at the animal's pleasure, yet, when sepa-
rated to their utmost extent, the front of the
orifice is seen to be protected by a very thin
superficial film.

In the common Sea-anemone the margin of the
disc is furnished with a series of whice or bright
blue specks, which some writers describe as a
rudimentary apparatus of vision. The structure
of these organs is not yet fully understood. Like
the body-warts, mentioned elsewhere, they are
probably to be regarded as sac-shaped prolonga-
tions of the outer layer.

Grood evidence has not yet been brought forward
of the existence of a nervous system in Actinia,
A muscular apparatus is, however, well developed,
and has been described in detail by M. Hollard.
In the inner layer of the body-wall are two sets
of flattened muscular fibres ; a superficial circular,
and a deeper longitudinal. Each mesentery has
four muscles, two for each of its faces. The
stomach wall is also provided with its own mus-
cular fibres, these being so arranged in the vicinity
of the inferior aperture as to permit the latter to
be closed at pleasure. The existence of this
sphincter is denied by some observers. A similar

K 4

136 ACTINOZOA.

arrangement has been noticed in the delicate
muscles which surround the tips of the tenta-
cula.

Though, histologically, the several structures of
Actinia admit of being resolved into two founda-
tion membranes, an ectoderm and an endoderm,
yet each of these, more especially the former,
manifests a tendency to differentiate into other
secondary layers, so that several apparently dis-
tinct tissues are recognizable in the body of the
adult animal. This is well seen in the column
wall, the principal thickness of which is composed
of the two sets of muscular fibres mentioned above.
That portion of the ectoderm which serves as an
external investment to this muscular wall appears
to consist, in some Actinise at least, of two separ-
able, transparent membranes ; an outer, or epithe-
lial, forming the general surface of the body, and
an inner or dermal layer in immediate contact
with the muscular substance. The dermal mem-
brane is almost wholly made up of a structureless
periplast containing very few endoplasts ; in the
epithelium, however, endoplasts are more abundant.
Between these two membranes thread-cells are
sometimes found embedded in such numbers as
almost to form a true layer, while close beneath
the epithelium occur masses of the pigment gran-
ules, to which the varied, and often gorgeous,
colours of these animals would seem to be due.
The endodermal lining of the muscular wall is,
in like manner, composed of two membranes, the
one superficial, the other in direct contact with
the deeper longitudinal fibres. Both ectoderm
and endoderm have their free surfaces more or
less abundantly ciliated. The structure of the

ACTINOZOA. 137

digeptive sac does not differ, in any essential respect,
from that of the column, than which it is much
thinner and more delicate, its endoderm being
richly furnished with cilia. Ordinary pigment
granules are here absent, but in their stead occurs,
within the upper portion of the stomach wall, a
thin layer of a red or yellowish brown tint, to
which some writers have ascribed the function of
a liver. The mesenteries are to be regarded as
processes of the column wall. The thin layers of
endoderm which invest the two sides of each
mesentery are produced beyond its free edge to
form the sac-like covering, within which the repro-
ductive elements are lodged. Having enclosed
these, the two layers are brought into mutual con-
tact, a narrow band being thus produced, to which
the cord-like craspedum is attached.

The flower-like appearance of the fully expanded
Actinia is sufficiently familiar to every sea-side ob-
server. While the animal is in this condition any
passing object likely to serve as food is firmly
grasped by one or more of the tentacula, which,
aided by the muscular contractions of the body
wall, soon force it into the interior of the digestive
sac. The morsel thus swallowed is usually, after
a time, rejected by the mouth ; while the nutritive
matters withdrawn from its substance by the action
of the stomach secretions are transferred to the
somatic cavity, within which, as in the Hydrozoa,
the process of nutrition is completed. Of the
voracity of the Actinia many amusing accounts
have been made known. It may, nevertheless, be
kept in captivity for several months, if supplied
with water containing minute particles of organic
matter.

138 ACTINOZOA.

Contact of food^ mechanical irritation, or the
withdrawal of the sea-water within which it dwells,
causes the Actinia to assume its contracted con-
dition. In this state it appears as a somewhat
conical, inert mass, often much flattened, the
mouth and tentacles being more or less completely
concealed by the folded margin of the disc. In the
act of expanding, this is gradually rolled backwards,
displaying the tentacles, which, as the margin con-
tinues to unfold itself, are soon distended to their
full extent by the pressure of the fluid contained
in the somatic cavity.

The Actinia has the power of effecting con-
siderable alterations in the general form of its
body by the alternate contraction and expansion
of the muscular fibres mentioned above. It can
also, like the Hydra, shift its position at pleasure,
though some species, under ordinary circumstances,
attach themselves so firmly as not to be removed
without laceration of the base.

2. Oeneral Morphology. — In no essential
respect does any Actinozoon depart from the typi-
cal structure above described, nor do the members
of the present group present such varied modifi-
cations of a common plan as have been shown to
appear in the Hydrozoa. In all Actinozoa the
digestive apparatus, though communicating freely
with the somatic cavity, is furnished w^ith a wall
of its own, between which and the outer boundary
of the body the generative elements are produced.
By these characters they may readily be distin-
guished from the Hydrozoa, with which, in the
more minute details of their structure, they closely
agree; the body of an Actinozoon, like that of

ACTINOZOA. 139

a Hydrozoon, wholly consisting of ectoderm and
endoderm.

The entire class is divided into four orders. In
the first of these, Zoantharia, represented by
Actinia and its immediate allies, the number of
mesenteries, tentacles, and other parts in connec-
tion therewith is, in general, some multiple of five
or six. In the three remaining orders some mul-
tiple of the number four appears to prevail. Thus
in the Alcyonaria there are eight somatic cham-
bers, eight mesenteries, and eight tentacles, not
simple, as in Actinia, but furnished with pinnate
margins {Jig. 26, a and h). The members of the
third order, Rugosa, known only through its fossil
representatives, seem to have possessed a structure
in some respects intermediate between that of the
two preceding sections. Lastly, the Ctenophora
are free-swimming, gelatinous animals, in physiog-
nomy widely different from the other Actinozoa,
though evidently akin to these in their leading
anatomical features. They are the most highly or-
ganized of Coelenterate animals, a common repre-
sentative of the order, Pleurohrachia, possessing
a complex nutritive apparatus, together with
well-defined organs for prehension, reproduction,
locomotion, and, in addition, unmistakeable indi-
cations of a nervous system {Jigs. 27 and 39).

Like the members of the preceding class, many
of the Actinozoa multiply freely by gemmation,
complex plant-like individuals being thus formed
which consist of numerous zooids united by a
coenosarc {figs. 34, d and 35). In such instances,
each nutritive zooid, or that portion of the organism
which answers to the polj^Dite of a Hydrozoon, is
distinguished by the name of ^polype.' When

140 ACTINOZOA.

gemmation does not occur, as in several species
of Actinia, the name polype is often employed to
denote the entire animal.

Though the soft parts of the Actinozoa have
only of late years received proper attention from
zoologists, yet the hard structures to which these
animals give rise have, under the general name of
** Corals," been objects of interest from a very
remote period. The outward aspect of Corals, as
preserved in our museums, is familiar to most
persons. Their true nature, in relation to the
living organisms by which they are produced, is
knowTi only to the student of the Actinozoa.

The limits in size presented by the several
forms of Actinozoa are not very readily defined.
The polypes of this group are usually much larger
than the polypi tes of the Hydrozoa, and, in a few
cases, attain a diameter of even eighteen inches.
The gigantic dimensions of some of the coral
structures, produced by a combined process of
growth and gemmation, are well known. Though
the separate polypes of such a mass may, in certain
instances, be little larger than pin's heads, yet,
very often, they are half-an-inch in diameter, and
not unfrequently, their size is much more consider-
able. It can scarcely be said that any Actinozoa
are of microscopic dimensions. All the Ctenophora
are conspicuous animals ; PleurobracJiia, already
alluded to, one of the smaller members of the
group, being often about the size of an ordinary
marble.

The various structures of the Actinozoa may be
described under the general heads of

a. Organs of Nutrition,
6. Prehensile apparatus,

ACTINOZOA. 141

c. Tegumentary organs,

cL Corallum or Skeleton,

c. JMuscular system and organs of Locomotion,

/. Nervous system and organs of Sense,

g. Keproductive organs.

3. Organs of Nutrition. — The whole interior
of the polypes and, in the budding species, that
of the ccenosarc by which they are connected, con-
stitutes the nutritive apparatus of the Actinozoa.

In most Zoantharia the structure and functions
of the polypes are best illustrated by reference to
the account of Actinia, above given. But in some
members of this order the digestive sac is rela-
tively much shorter than the somatic cavity, being,
according to Dana, little more than '2 of its length
in the genus Palythoa.

So, likewise, among the Alcyonaria, the somatic
cavity of each pol 3^2)6 usually appears as a long,
somewhat slender, tube, in the upper portion of
which the comparatively short stomach is, as it
were, suspended ; the proximal, or post -stomachal,
region of the body cavity becoming gradually
much narrowed. In the Gorgonidce, however,
the somatic cavity is shorter and slightly dilated
towards its basal extremity.

There are two apertures to the digestive cavity of
every Actinozoon ; first, the mouth, and secondly,
the proximal or inferior outlet, which opens freely
into the somatic cavity.

In many, though not all, Alcyonaria, the somatic
cavities of the separate polypes which make up
the compound mass are prolonged into canals,
freely communicating with one another, inoscu-
lating, and forming a sort of aquiferous system.

142 ACTINOZOA.

within which the nutritive products circulate. In
Meandrina and certain other Zoantharia, the
general cavities of the polypes open by wide aper-
tures into each other ; but in very many forms of
coralligenous Actinozoa it is erroneous to suppose
that any connection, available for nutrient pur-
poses, is maintained between the different polypes
of the same compound stock.

Although, in a large number of Actinozoa, the
somatic cavity has no communication with the
exterior, save through the digestive sac or the
free ends of the tentacles, when these are perforate,
yet, among other members of the class, the exist-
ence of apertures in the body-wall seems to have
been satisfactorily ascertained. Mention has al-
ready been made of the cinclides of Actinia, nor
are these orifices wanting in several allied genera.
In Philomedusa, according to Fritz Miiller, twelve
rows of such openings, which appear to the naked
eye as minute pale dots, and capable of inde-
pendent contraction, radiate from the posterior
extremity of the animal. In the centre of this
extremity, as also in Peachia and Cerianthus,
exists a much larger aperture, or, rather, short
canal, which the animal has the power of closing
effectually whilst its somatic cavity remains dis-
tended with fluid. And Milne Edwards has
shown that in Corallium the ccenosarcal canals
communicate directly with the surrounding me-
dium by means of numerous perforations in their
walls. Lastly, to the category of structures now
under consideration must be referred the * apical
pores ' of the Gtenophora, whose nutrient system
presents peculiar features which render it necessary
that some account of Pleurobrachia ( = Cydippe),

ACTINOZOA. 143

as a typical example of the group, should in this
place be given {fig. 39, e).

The body, or * actinosome,' of Pleurobrachia is
sub-spherical or melon-shaped, colourless, gela-
tinous, and perfectly transparent, but displaying,
in sunlight, tints of a beautiful iridescence.

Two poles, an oral and an apical, mark the
opposite extremities of the axis of the animal.
The slightly protuberant mouth appears, when
closed, as an elliptical fissure, presenting two
flattened sides and two opposing edges.

Eight meridional bands, or ' ctenophores,' bear-
ing the comb-like fringes, or characteristic organs
of locomotion, traverse, at definite intervals, the
interpolar region, which they divide into an equal
number of lune-like lobes, termed the 'actino-
meres.' But this division of the body does not
extend into the immediate vicinity of the poles,
before reaching which the ctenophores gradually
diminish in diameter, each terminating in a point.
Around the apical pole, in particular, may be
noticed a somewhat oblong, depressed, area, dis-
tinctly circumscribed by the adjacent converging
actinomeres.

The eight actinomeres are by no means
equal in size, and, to understand their relations
aright, it seems desirable to distinguish three
principal kinds of these parts as the antero-
posterior, the lateral, and the accessory actino-
meres.

The two antero-posterior actinomeres, wider than
their fellows, are opposite to each other and the
edges of the elliptical mouth. At right angles to
these, but in like manner opposite one another,
lie the two lateral actinomeres, which, therefore.

144 ACTINOZOA.

face the flattened sides of the mouth. The acces-
sory actinomeres, slightly narrower than either of
the preceding, serve to occupy the four inter-
spaces which occur between the lateral and an-
tero-posterior pairs.

The lateral actinomeres are further distinguished
by the presence in each of a large sac, which opens
obliquely, outwards and downwards, about mid-
way between the equatorial region and the apical
pole of the body. From this sac the animal has
the power of protruding at pleasure a long, highly
contractile, beautifully fringed, tentacle.

Immediately within the apical pole is situate a
peculiar body, supposed to be an organ of sense,
which is best termed the ' ctenocyst.' Upon this
rests a nervous mass from which issue small fila-
ments. The structure of these parts, as also of
the prehensile, locomotive, and reproductive appa-
ratuses are described in their appropriate para-
graphs. At present let us chiefly notice, in con-
nection with the form of the body, the arrange-
ment of its somewhat complex nutrient system.

{fig- 27-)

This system may be said to commence in the
stomach, or digestive sac, a cavity having the
general form of an elliptic cylinder, and extending
from the mouth through the longitudinal axis
of the body, for about -6 of its entire length.
Slightly contracting below, the digestive sac is
seen to open into a much wider and shorter cavity,
also axial in its direction, known as the ' funnel,'
which gradually diminishes in diameter as it ap-
proaches the apical pole of the body, to terminate
just above the ctenocyst and nervous mass. From
the funnel three pairs of canals are given off.

ACTINOZOA.

145

Two of these, the ' apical canals,' very short and
narrow, run directly downwards and outwards on
either side of the ctenocyst, and soon open exter-
nally as the ' apical pores,' situate immediately
beyond the margin of the ' apical area.' Some
writers describe the apical canals as lateral, others
as antero-posterior in their direction. Both
opinions are partially correct, the apical pores

Fig. 27.

Morphology of Pleueobeachia : — a, diagrammatic longitu-
dinal section of Pleurobrachia ; b, the same in transrerse section;
0, mouth ; 5', digestive cayity ; k', funnel ; 0', ganglion and
ctenocyst ; t', sac of the tentacle ; r, tentacle, with one of its
branches ; x. radial canal ; x\ one of the ctenophoral canals ; x">
apical canal ; x, paragastric canal. The numerals i and 2 indicate
the first and second bifurcations of the radial canals.

being obliquely opposite one another, though
placed on different sides of the body. Two other
canals, the ' paragastric canals,' assume an upward
course, parallel to and not far from the flattened
sides of the digestive sac, but terminate csecally
before quite reaching its oral extremity. A third
pair of canals, much wider and shorter than those
just mentioned, radiate from the funnel in a

146 A.CTINOZOA.

horizontal or sligMly oblique direction, proceeding
towards the bases of the pits in which the tentacles
are lodged. Before gaining these, however, each
' radial canal ' divides into two branches, the
secondary radial canals; each of these again into
two others, and, thus, eight tertiary radial canals
are formed, which run towards the equatorial
region of the body, where they open at right
angles into an equal number of longitudinal
vessels, the ' ctenophoral canals,' whose course
coincides with that of the eight locomotive bands.
These canals end csecally both at their oral and
apical extremities.

If, now, a comparison be made between this
nutrient system and that of Actinia, the digestive
sacs of the two organisms are clearly seen to cor-
respond ; in -form, in relative size, and mode of
communication with the somatic cavity. The
funnel and apical canals of Pleurobrackia, though
more distinctly marked out, are the homologues
of those parts of the general cavity which in
Actinia are central in position and underlie the
free end of the digestive sac. So, also, the para-
gastric and radial canals may be likened to those
lateral portions of the somatic cavity of Actinia
w^hich are not included between the mesenteries.
Lastly, the ctenophoral canals of Pleurobrackia ?ind
the somatic chambers of Actinia appear to be truly
homologous, the chief difference between the two
forms being that while in the latter the body
chambers are wide and separated by very thin
partitions, they are in Pleurobrackia reduced to
the condition of tubes ; the mesenteries which
intervene becoming very thick and gelatinous, so
as to constitute, indeed, the principal bulk of the

ACTINOZOA. 147

body. lu both, the nutrient system is lined by
a ciliated endoderm, the vibratory action of which
serves to maintain a circulatory motion of the
included fluid. The contractile tissues of the
ectoderm may further assist such movements.
And in Pleurobrachia, whose bilateral symmetry
is more strongly marked than that oi Actinia, the
nutrient fluid, as Agassiz has shown, is at times
alternately impelled between the right and left
sides of the somatic cavity.

Very many curious modifications are presented
by the canal system among the different genera of
CfenopJiora, to some of W'hich reference will be
made in the more particular account to be given
of that order.

Xo manducatory apparatus exists in the Acti-
nozoa. The oral margin may, however, be some-
what thicker and firmer than the surroundinof
parts, or otherwise become altered in a23pearance ;
and the cilia of the digestive sac may also differ
from those which occur on other regions of the
body.

As in Actinia, one part of the digestive cavity
may undergo some amount of modification, coloured
granular masses appearing in its walls which have
been supposed to indicate a liver. Such coloured
cells in the Ctenoj^hora usually arrange themselves
as vertical ridges which surround the innermost,
or stomachal, division of the otherwise transparent
digestive sac.

^lilne Edwards has also shown the existence in
Cestum of another structure whose function is
probably secretive. Between certain of the cili-
ated bands and their corresponding ctenophoral
canals; parallel to, and in close connection with

L 2

148 ACTINOZOA.

the latter, runs a tube occupied by a number of
granular bodies, and giving off, at right angles to
itself, a series of short vertical branches, which
open along the line of the locomotive fringes to
communicate with the surrounding medium.

4. Prehensile apparatus. — The tentacles of
the Adinozoaf like those of the Hydvozoa, appear
usually, if not always, as hollow appendages, in
immediate connection with the somatic cavity,
their walls being richly provided with thread-cells
and consisting throughout of two layers, an ecto-
derm and an endoderm.

Among the Zoantharia, the tentacles vary ex-
ceedingly in size and external form. Viewed from
without, they are seen to arise, save in Eumenides,
from the distal extremities of the polypes,
between the mouth and the outer margin of the
disc {figs. 33 — -35). Dissection shows them to be
hollow processes in free communication with the
somatic chambers, each of which is furnished with
one or more of these appejidages. Their most usual
form is that of a slightly curved, more or less
tapering, cone, as in many species of the genus
Actinia itself. But from this typical aspect there
are very many aberrant modifications,

Among the Alcyonaria, the tentacles are com-
paratively short, closely arranged in a single cycle
of eight around the mouth of each polype, their
margins being produced into a number of lateral
pinnas {fig. 26, a). These last, according to Dana,
are perforate at their free ends, the extremity of
the tentacle itself being csecal ; but this statement
is denied by Milne Edwards and others, who more
correctly view the pinnae, in some genera at least,

ACTINOZOA. 149

as destitute of distal orifices. The pinnae are very
contractile, so as to vary in form from mere lobes
or tubercles to long filiform fringes. But little
diversity is exhibited by the tentacles of this order.
Except in the distinctive characters just mentioned,
they agree essentially with those of Actinia.

The tentacles of the Ctenojphova are best des-
cribed in connection with the general survey of
the characters of that order.

5. Tesiinientary Or«;ans. — In but few Acti-
nozoa do the tentacles appear to be processes of
the ectoderm only. This layer, as we have seen
in J. c^/?iia, exhibits a tendency to differentiate into
two diverse planes of growth, which, with Professor
Huxley, we may designate the ' ecderon ' and the
' enderon', respectively. Sometimes, however,
this distinction is not observable. The ectoderm is
usually ciliated, and in the Ctenophora becomes
very thick and gelatinous, presenting a structure
somewhat similar to that which occurs in the
oceanic Hydrozoa. Gregenbaur describes the re-
ticulating threads which traverse the periplastic
mass as tubular in young Ctenophora, but, as
groT\i:h advances, tending to become solid. Other
minor histological modifications have been ob-
served.

The general surface of the body, smooth in most
Ctenophora, is in Chiajea and a few other genera
diversified at intervals by the elevation of numer-
ous simple papilloe. And, in some Sea- anemones, it
exhibits a number of clear warts or vesicles, each of
which, according to M. Hollard, possesses a muscu-
lar arrangement of its own, in connection with a
sort of two-lipped mouth; so that a needle, or
L 3

150 ACTINOZOA.

other small foreign body, introduced into the
vesicle, is quickly and tenaciously secured. In
their natural situations these creatures are often
completely covered by fragments of shell, gravel,
or sand, attached to their bodies by a peculiar
viscid secretion, in the production of which these
warts are, perhaps, concerned.

Or, the epidermic secretion may give rise to a
distinct membranoid coat, protecting the integu-
ment of the animal, from which it is at times cast
off by what may be termed a process of sloughing.
Such a membrane in Cerianthiis Mr. G-osse states
to be " wholly composed " of altered cnidae, which
intertwine one with another to form a wide tube,
investing the entire surface of the column. Here
the connection of the tube is so loose that it can at
any time be removed without much inconvenience
to the animal, but, in other genera, a more
adherent covering may be found. In Adamsia
the base excretes a delicate, somewhat chitinous
membrane, which, upon occasion, may continue
its growth beyond the attached outline of its
possessor, and even form an artificial extension of
the peculiar surface which this genus is wont to
choose for its abode.

The thread-cells of the Zoantharia have been
studied with great care by Mr. Gosse, who dis-
tinguishes four principal kinds of these bodies
by the titles of ' chambered,' * spiral,' ^ tangled '
and ' globate cnidae.' The chambered cnidae
(which are the most common) are of a long oval
form, the ecthoroeum, which varies greatly in
length, presenting in all cases, thecomplex armature
characteristic of these minute weapons ; a number
of delicate barbs, or ' pterygia ', being attached to

I

I

ACTINOZOA. 151

a thickened band, the 'strebia', twisted in a
screw-like manner around the basal portion of the
thread. The tangled cnidse are relatively broader
then the preceding, having a very long ecthorseum,
loosely rolled up into a confused bundle. The
spiral cnida3 present a much elongated, fusiform
chamber, within which the thread lies coiled in a
close regular spiral. Lastly, the so-called globate
cnidse have been seen to push out at each end a cy-
lindrical protuberance, sometimes equal in length
to the cnida itself, which does not contain any
thread.

On the urticating organs of the Alcyonaria
less attention has been bestowed. In general, they
are of minute size and seem to resemble the
tangled cnidse of the Zoantharia. In Sarco-
dictyon they are aggregated on the tentacular
pinnae in minute rounded swellings, homologous
with the palpocils of the fixed Hydvozoa.

The thread-cells of the Ctenophora present a
peculiar structure. Each, in Pleurobrachia, ac-
cording to Professor H. J. Clark, appears of a
rounded or slightly napiform figure, and is covered
externally by a single, dense, layer of very minute
granules. From the summit of a broad conical
projection on the inner surface of its otherwise
uniformly thick, but rather delicate, wall, arises,
in a very oblique direction, the simple thread,
which, after making not more than seven or eight,
equi-distant, spiral turns, set very far apart, ter-
minates suddenly in what seems to be a free
ending, precisely opposite its point of attachment.
The thread is cylindrical, smooth, apparently solid,
of firm consistence, and about eight or nine times
the length of its envelope, from which it is set

L 4

152 ACTINOZOA.

free by the gapiug of the cell itself, around the
thread's distal extremity.

On the whole it seems safe to say that among the
Actinozoa the thread-cells exhibit a greater ten-
dency to become collected in particular organs than
has been shown to be the case with the Hydrozoa ;
though we by no means wish to forget the tentacles
or nematophores of the latter. The mesenteric
cords of the Sea-anemones strikingly illustrate
this, and, in the CtenopJiora, the urticating organs
form a well marked layer on the outer surface of
the tentacles and their lateral fringes. Parallel to,
and agreeing in position with, these last, the twoi
tentacles in Hoimiijpliora are furnished, as Gegen-
baur has proved, with a number of very peculiar,
bright yellow, appendages, one between from about
every ten to fifteen of the ordinary side filaments.
Each of these bodies, which serve as special recep-
tacles for the thread-cells, is hollow, of a flattened
fusiform, or lancet-shaped, form, with a short stalk
of attachment, above which it is prolonged later-
ally into several pairs of tubular processes, which
gradually diminish in length, and finally vanish
altogether, before reaching its free, simply taper-
ing, extremity.

Pigment-masses, irregularly scattered in some
Actinozoa, are in others combined so as to form
more or less definite layers, which may readily
be examined in the commoner species of Sea-
anemones. In the substance of the body-wall
and tentacles, outside the muscles of the mesen-
teries, or even in the digestive tube itself, such
interrupted layers of colouring-matter have been
observed.

The exquisite roseate tint of some Ctenophora

ACTINOZOA. 153

is due to the presence of pigment-streaks or less
regular stellate masses, in various parts of the
ectoderm.

6. Coralliim or S»keleton. — Intimately con-
nected with the tegumentary organs of these
animals, under which head, indeed, it might with-
out impropriety be described, is the so-called
skeleton, or ' corallum ', with which so many of
them are furnished.

The term coral, or corallum, is properly re-
stricted, in zoology, to the hard structures deposited
in the tissues, or by the tissues, of the Actinozoa.
Any form of this class which possesses such a
framework is called a ' Coral '.

All Actinozoa are not coralligenous. The Cteno-
phora and several species of Zoantharia deposit
no corallum. On the other hand, the order
Rugosa is kno^vn only from the remains of extinct
Corals.

Of coral structures there are two principal
kinds, which must be carefully distinguished from
one another. First, the ' sclerobasic ' corallum,
a true tegumentary excretion, formed by the
conversion of successive growths from the outer
surface of the ecderon. Secondly, the * sclero-
dermic ' corallum, which better merits the name
of skeleton, deposited, as it is, within the tissues
of the animal, and, in all probability, by the en-
deron.

The sclerobasic corallum is by Mr. Dana termed
" foot secretion " ; the sclerodermic, " tissue se-
cretion ".

Let us first notice the sclerobasic corallum,
which is found only in certain budding composite

154 ACTINOZOA.

Actinozoa, Most frequently its texture is simply
corneous, but in Corallium proper and a few
other forms, it becomes calcareous by deposition ;
and in Hyalonema and Hyalopathes, if these be
true Actinozoa, it is siliceous. In Isis and
Mojpsea it consists of alternately disposed calca-
reous and horny segments, thus, as it were, com-
bining strength with a yielding pliancy. In Isis
branches are developed from the calcareous, in
Mopsea from the horny segments of the sclero-
basis. Melitcea presents a like structure, save
that, in it, the corneous segments are replaced by
others which assume a porous and suberous as-
pect.

Section of a sclerobasis shows it to be, in some
cases, solid or nearly so; in others, distinctly
resolvable into concentric layers, which serve,
also, to illustrate the manner in which it has been
produced; while, more rarely, it is composed of an
aggregation of separate fibres.

Two principal modifications of form distinguish
the sclerobasis. In some Actinozoa it constitutes
a free axis, virgate or pinnately divided, and
varying much in composition and thickness. In
others it is attached, simple or branched, and
often singularly plant-like in physiognomy, as in
those Gorgonidce to which the name of Sea-shrubs
has been applied.

The relations of such structures to the soft
parts of the animal are, with little difficulty,
discerned. The sclerobasic corallum is, in fact,
outside the bases of the polypes and their con-
necting coenosarc, which, at the same time, receive
support from the hard axis which they serve to
conceal. Thus the coenosarc of these corals ap-

ACTINOZOA. 155

pears as a soft, fleshy covering, from which the
several polypes arise, their somatic cavities freely
communicating one with another.

Far different in its nature is the sclerodermic
corallum, deposited, as above stated, within the
bodies of polypes, which, in some cases, remain
separate, but, in others, multiply by continuous
gemmation. And, just as the whole body of an
Actinozoon is made up either of one polype or of
several united by a coenosarc, so, too, may the
fully developed sclerodermic corallum consist of
a single ' corallite ' or of several connected by a
* coenenchyma '.

The parts of a typical corallite are these {jig.
28). First, an outer wall, or 'theca', somewhat
cylindrical in form, terminating distally in a
cup-like excavation, or * calice ', and having its
central axis traversed by a ' columella '. The
space between this and the theca is divided into
' loculi ', or chambers, by a number of radiating
vertical partitions, the ' septa '. These do not, in
certain instances, quite reach the columella, but
are broken up into upright pillars, or * pali ',
arranged in one, two, or three circular rows,
termed ' coronets '. All the preceding parts are
best brought into view by transverse section.
Longitudinal division of a corallite shows, fre-
quently, the existence of imperfect transverse
partitions, or 'dissepiments', which, growing from
the sides of the septa, interfere, to a greater or
less extent, with the perfect continuity of the
loculi. Sometimes the septa have their "sides
covered with styliform or echinulate processes,
which, in general, meet so as to constitute nume-
rous ' synapticulge ', or transverse props, extending

156

ACTINOZOA.

across the loculi like the bars of a grate." In
other cases, the dissepiments are replaced by the
development of successive horizontal floors, or
* tabulae ', which do not grow from the septa^, but

Fig, 28.

I

Morphology of Zganthajiia sclerodeemata : — A diagram-
matic section of a living corallite. A, digestive cavity ; B, its com-
munication with the somatic cavity ; r, intermesenteric chamber ;
r', mesentery ; A, tentacle ; E, mouth ; Z, ectoderm ; H, endo-
derm ; 0, epitheca ; I, sclerobase ; K, theca ; A, septa ; M, palu-
lus ; N, columella ; O, dissepiments ; n, tabulae ; P, ccEnenchyma.
(The septa should be seen between the dissepiments, but are left
out for distinctness' sake.)

extend, without interruption, across the entire
space bounded by the theca. On the outer
surface of the latter may occur 'costse', or vertical
lines, corresponding in position to the septa within;
' exothecae ', which arise from the sides of the

ACTINOZOA. 157

costoe, thus representing the dissepiments ; and a
continuous layer, or ' epitheca ', consisting of the
coalesced, external, indications of tabulae.

It needs scarcely to be stated that an organism
producing such a structure as the foregoing must
closely have resembled, in every essential respect,
the Actinia^ or typical polype, previously de-
scribed. The relations of the septa and pali to the
mesenteries, of the theca to the column wall, of
the columella to that part of the enderon which
forms the floor of the somatic cavity below the
digestive sac, are, indeed, sufficiently obvious.
The septa, too, like the mesenteries, are primary,
secondary, and tertiary, according to their degree
of approximation to the columella ; the primary
septa alone being in direct contact therewith. All
these parts are, in the living animal, completely
concealed by the soft integuments : the digestive
sac, and much of the somatic cavity, especially its
upper portion, performing, as in the soft-bodied
species, their proper nutrient and reproductive
functions (Jifj. 33).

In a similar manner is the coenench3^ma de-
posited within the coenosarc {fig. 28). It may be
united with the corallites at their bases only, thus
forming a creeping expansion or stalk, or become
connected with them throughout the greater por-
tion of their height. There are even cases in
which the corallites appear sunk amid a very
abundant coenenchyma, while, in others, the same
structure is but sparingly developed. The relative
distance of the corallites from one another is also
subject to much variation.

But the typical structure of the coral lite above
described does not admit of being studied in any

158

ACTINOZOA.

single species. Its, nearest approach, as Milne
Edwards has stated, is found in the genus Acervu-
laria, which wants, however, synapticulse and
columella, the pali, also, being rudimentary. This
genus is a member of the extinct order Ru-gosa,
in which the sclerodermic corallum may, per-
haps, be said to attain its most remarkable deve-
lopment. Both septa and tabulae here occur in
the same corallite, the former being always ar-
ranged in multiples of four.

Fig. 29.

COLTJMNAEIA FrANKLINTI.

Portion of corallum, of the natural size.

Among the sclerodermic Zoanthavia tabulae and
septa are scarcely known to co-exist, a special
section of this group, Tabulata, being distin-
guished by the nearly exclusive possession of the
former (^fig. 29). In two other large divisions,
the Aporosa and Perforata, including several
families, septa, in sets of five or six, normally occur,
and in some are associated with dissepiments,
more rarely with synapticulse. In a fourth section,

ACTINOZOA. 159

Tubulosa, the septa are indicated by mere streaks
( fig. 36, c). And in the Tubiporidw, a family
of Alci/oimria, sei^tii are absent; each corallite be-
ing a simple tube, connected with the thecse around
it "by horizontal plates, which represent the inner
transverse floors of the Tabulata {fig.30).

Fig. 30.

TUBIPORA MUSICA*

Fragment of corallum, of the natural size.

From the Tubijooridcc to other Alcyonaria in
which the corallum, though sclerodermic, soon
ceases to present traces of thecae, a transition, not
very abrupt, may be effected. Such intermediate
stages, though not of much value to the sys-
tematic zoologist, are of great interest in a mor-
phological point of view, since they show well the
manner in which the complete sclerodermic co-
rallum has been formed ; thus at once illustrating
its minute structure and the several stages of its
development. In Telestho, the corallum is made

160 ACTINOZOA.

up of a number of branching tubes, which are not,
as in all the preceding forms, perfectly calcareous.
In Cornularia and its allies a corallum, never
wholly tubular or of a firm calcareous consistence,
has yet been detected ; and in Sarcodictyon masses
of spicules only can be observed. In some species
of Alcyonidce proper, the spicules attain a com-
paratively large size, and become aggregated into
definite nodular masses. These ' dermosclerites ',
as Milne Edwards has shown, are of two principal
kinds, the fusiform, and the irregular. The for-
mer are somewhat cartilaginous in consistence,
and have their surface studded with slight asperi-
ties. The irregular nodules are stronger and more
decidedly calcareous, presenting six faces, each, in
general, furnished with a tubercular enlargement,
which sometimes prolongs itself into a number of
spines, bearing on their sides other secondary
tubercles. By the coalescence of such masses and
the deposition of more minute particles among
their interstices, a thecal corallum, in other Acti-
nozoa, at length comes to be formed. In Alcy-
onium itself the spicules, though numerous, are
not of large size, and are most conspicuous in the
column wall below the margin of the disc. Re-
turning to the Zoantharia we find, in the genus
Zoanthus, a spicular corallum still more feebly
developed than that of Alcyonium. In many of
the Sea-anemones no spicules have been observed,
though traces of a corallum are not, even in these,
absolutely lost. Finally among the Ctenophora
we in vain search for the faintest indications of its
existence.

From what has been said it were easy to infer
that but little minute structure would be presented

ACTINOZOA. 161

by the perfect scl-erodermic corallum. Its decalcifi-
cation, liowever, reveals delicate shreds of the
periplastic substance by which it had been de-
posited, usually exhibiting an irregular reticulating
arrangement. The * sclerenchyma/ or coral tissue,
presents every gradation between this nearly solid
condition and the spicular stage permanently ex-
emplified in Ah'i/onium. Thus, in the Aporosa,
it is iirm and compact; in the Perforata, porous
and granular, or even spongy and reticulate.

In the accompanpng table the chief modifica-
tions of the corallum, from an artificial point of
■y;iew, are systematically exhibited.

It must not, however, be supposed that the
presence of a sclerobasis renders the deposition of
tissue secretions wholly impossible, for, among the
GorcjonidcB it is certain that, in addition to the
basal corallum, true sclerodermic spicules appear,
within the substance of the investing mass. When
such a Gorgonia is dried, and the soft parts washed
away, a thin layer of calcareous sjDicules will
be found gently adhering to the brown, horny
sclerobasis below. i\i. Valenciennes has proposed
to distinguish five kinds of these spicules, or
*sclerites,' by the names of capitate, fusiform,
massive, stellate, and squamous, respectively.

KEY TO MODIFICATIONS OF COEALLUM.

Corallum wholly sclerodermic.
Corallum thecal, calcareous.
Tabulae present.

Septa in X of 4 Eugosa.

Septa in x of 5 or 6, rudimentary or

absent Tabulata.

Tabulae absent.

Septa weU marked, in x of 5 or 6.

SclerenchjTna porous. . . . Perforata.
M

162 ACTINOZOA.

Sclerench}Tna imperforate. . Apobosa.

Septa indicated by mere streaks. The-
cse pear-shaped, in some connected
by a basal, creeping coenenchyma. . Tubulosa..
Septa absent, Thecse crowded, cylin-
drical, united at various heights by
distinct, horizontal epithecae. . TuBiPORiDa^

Corallum spicular or, if thecal, corneous or
sub-calcareous.

Spicules numerous, in some replaced,
either wholly or in part, by an imper-
fect, tubular corallum. . . . Axcyonidje.
Spicules scanty, or replaced by particles

of sand Zoanthid^.

Corallum sclerobasic.

Sclerobasis spinulous or smooth. . . Z. Sclerobasica.
Sclerobasis sidcate.

Sclerobasis attached proximally. . Gorgonid^.
Sclerobasis free Pennatuledje.

7. IVInseular ISysteni and Or@;ans of liOCO-
mction. — Reference has already been made to
the muscular system of Actinia.

A like apparatus, presenting, however, some
differences of detail, appears to become differen-
tiated from the general periplastic substance in
most other Zoantharia and Alcyonaria. But the
power of altering the position of the body by the
slow alternate contractions of a normally attached
base is possessed only by those Zoantharia to
which the name of Sea-anemones is usually applied.
Their non-adherent allies, such as Ediuardsia and
Cerianthus, have a highly contractile column-
wall, capable of greatly varying its length, and
of executing movements, for the most part, of a
feeble worm-like character. The Alcyonidas and

ACTINOZOA. 163

Gorgon'idce are permanently fixed, as are also
many of the higher coralligenous Actinozoa, es-
pecially those which multiply by continuous gem-
mntiou. Others, however, and these chiefly the
simpler forms, are free, but, like the unattached
Pennatuliihc, not truly locomotive. Yet in the
greater number of the Actinozoa each polype,
though fixed, is contractile to some extent, shrink-
ing down under irritation, and again unfolding
itself at pleasure, while, among the Alcyonaria^
with a few exceptions, it is also retractile into the
fleshy substance of the coenosarc. Even this, too,
has its own share of contractility, most evident in
those sj^ecies which possess an elastic sclerobasis.
Thus, on the South American coast, Mr. Darwin
observed a Sea-pen which, on being touched,
forcibly drew back into the sand some inches of
its compound, polj'pe-covered, mass.

All the Ctenophora are free-swimming animals,
but doubt yet hangs over the nature of certain
exceptional Zoantharla, reputed to be of similar
habit. The apical extremity of the genus Minyas
and its allies is represented by Lesueur and Lesson
as dilated into a large air-sac, excavated beneath
the floor of the somatic cavity, and furnished be-
low with an opening into the surrounding medium.
By means of this sac the creature is said to float
without effort, its oral disc being turned down-
wards; but further observations on its structure
are much wanting. Again, the Arachnactis albida
of Sars, possesses, according to Professor E. Forbes,
not merely the power of smmming like a Medusid,
but " it can convert its posterior extremity into a
suctorial disc, and fix itself to bodies in the man-
ner of an Acilicia.''' But the aspect of the tentacles

164 ACTINOZOA.

in this organism strongly suggests the possibility
of its being an immature form, nor is the suspicion
weakened by the discovery of Haime, that the
young of Ceriaiithus, while resembling AracJt-
nadis in physiognomy, enjoys a similar oceanic
mode of life.

The muscular fibres of the Actinozoa are in-
teresting to the histologist, as wanting, among
many forms, those distinct transverse striie, which,
elsewhere, they so frequently present. Such strisB
are not, however, always absent. In the body-
substance of this class we have, in truth, obvious
transitions from a simple contractile periplast to
muscular fibres, Vv^hich in no essential respect
differ from those of various invertebrate animals.
In the typical Ctenophora, the contractile tissues
appear to be disposed in two principal sets; a
transverse or circular, and a longitudinal.

Some Zoantharia employ their tentacles as
aids to locomotion, though neither in these nor in
the Alcyonaria can it rightly be said that special
motile organs exist.

Of this nature, however, are the ' ctenophores,'
or ciliated bands, which constitute so obvious a
feature in the physiognomy of the Ctenophora.
The normal number of these bands would seem to
V>e eight, though in Gestmn, and one or two other
forms, their typical structure and arrangement is
somewhat modified. Each ctenophore is of a
much elongated ovate form, widest at the equa-
torial region of the body, and tapering gradually
to end in a point at some distance from the oral
and apical poles ; slight differences in degree of
approximation to these parts, and such-like minor
characters, distinguishing the ctenophores of the

ACTINOZOA. 165

several genera. The surface of the ctenophore is
transversely elevated at intervals throughout the
greater portion of its length into a number of
successive ridges, to each of which a row of strong
cilia is attached in such a manner as to form a
paddle-like plate, or comb, the free extremities of
the cilia remaining separate. The cilia are not all
of equal length, those of the middle portion of
the comb usually having the advantage in this re-
spect, while the cilia on either side symmetrically
correspond ; their degree of elongation varying
so as to impart to the edge of the entire comb a
gently curved outline, when seen at rest. This is,
indeed, seldom the case during the life of the
animal, throughout which the combs manifest an
astonishing amount both of simultaneous and
successive activity. Nay, even after death, de-
tached portions of these creatures, bearing frag-
ments of the ctenophores, exhibit for many hours
no apparent diminution of their ordinary vibratile
efficiency.

8. Nervous System and Or$irans of lienf^e.

— In no Actinozoa, save the Gtenophora, has
good evidence of the presence of a nervous system
or organs of sense yet been obtained. Nor should
this appear surprising, for the sensitivity which,
in more highly differentiated organisms, has its
course restricted to definite tracks, is here diffused,
in a less appreciable manner, through the more
general and comparatively ill-developed tissues
of the body. The white or blue marginal sacs of
some Actiniae, and the body-warts in allied species,
have, it is true, been regarded as sensitive in
function, and the former have even been dignified

M 3

166 ACTINOZOA.

by the title of rudimentary eyes. The radiating
system of ganglia and nerve-fibres which Spix
described as existing within the base of Actinia
has not come under the notice of other observers.

But in the Ctenophora occurs a well-marked
sense-organ, the 'ctenocyst,' upon whose precise
function, whether oculiform or auditory, naturalists
are far from being agreed. Such differences of
opinion are in truth based on the prejudices which
most anatomists acquire from a too exclusive
attention to the structural peculiarities which the
higher animals present. The ctenocyst, in all
probability, neither sees nor hears, but would seem
to be the localised recipient of those obscure
general impressions to which its lowly-organised
possessor is capable of responding.

The ctenocyst occupies a central position amid
the soft substance of the ectoderm, immediately
within the apical pole of the body. In form it is
ovate or spherical, smooth externally, but, in some
cases, invested with an adventitious layer of pig-
ment granules. Its wall appears to be very firm
and elastic, so as quickly to recover its proper
figure, should this be changed in accordance with
the ordinary contractions of the body. Within, the
ctenocyst is hollow, and apparently distended with
a fluid. In the midst of this fluid lie a number of
rounded or polyhedral concretions, semitransparent,
colourless or somewhat tinted, occasionally coa-
lesced into a single mass, and composed, probably,
of carbonate of lime. Each granule is little more
than '0003 of an inch in diameter. The concre-
tions appear subject to a peculiar vibratory move-
ment, but some observers have disputed the fact of
its occurrence.

ACTINOZOA. 167

The nervous system of the Ctenophora consists
either of a single ganglion or of a pair of ganglia
closely approximated, giving origin to a number
of delicate nerve-like cords. The ganglion lies
deeply seated within the pyramidal mass of ecto-
derm included between the apical canals, towards
the narrow extremity of which its apex is directed,
while its base rests upon the surface of the cteno-
cyst. In form it is sub-pyriform or bluntly conical :
anatomically, it seems resolvable into a thin tran-
sparent w^all, enclosing granular contents ; in
colour, it is most frequently pale 3^ellow. From
this central mass issue two principal series of
nervous cords, one of which arches inwards towards
the walls of the digestive cavity, and, in some
cases, separates into four sets or bundles to supply
the principal regions of the body. The nerves of
the second series, usually eight in number, are
distributed along the rows of swimming combs so
as to lie between the latter and their corresponding
canals. These cords appear dilated at intervals
into numerous minute ganglionic enlargements,
giving off secondary filaments, one for each of the
ciliated plates ; an arrangement, which, if corrobo-
rated by subsequent investigations, would go far
to throw some light upon the singular and quasi-
independent movements which these combs per-
form in the living animal. There is still, however,
much diversity of opinion as to the true interpre-
tation of the parts just described. Kolliker, while
recognising in Chiajea the presence of delicate
cords extending from comb to comb, expresses
himself, nevertheless, as very doubtful of the ex-
istence of a nervous system in any of the Cteno-
phora which he had himself investigated. Agassiz

M 4

168 ACTINOZOA.

is equally sceptical. On the other hand, the careful
observations of Will, Milne Edwards and, more
recently, of Gegenbaur, point to an opposite con-
clusion. Somewhat similar are the views of Frey
and Leuckart. All the preceding writers are un-
animous in rejecting the prior account of the
nervous system of Fleurobrachia given by Grant,
who describes a double nervous ring surrounding
the mouth, in the course of which he thought he
could detect eight ganglia, each giving off on
either side two fibres and a fifth larger filament,
traceable onwards beyond the middle of the body.
Yet this description, when carefully considered, is
less irreconcileable with the views expressed on
the same subject by other observers than seems to
be usually supposed.

Certain curious appendages, which possess, per-
haps, a tactile function, have been observed by
E. Wagoner in two genera of Ctenophova, Beroe
and Pleurohrachia. These organs appear as long
hair-like threads, which arise from either side of
the ctenophores along their whole length, and
form around each pole of the body a sort of
wreath, composed of several concentric rings. The
threads have not been seen to exhibit any inde-
pendent movements. Each swells once or twice
in the course of its length into a smooth or angular,
rounded or flattened, expansion, the entire surface
of which is abundantly beset with minute stalked
knobs. In some threads smaller swellings, want-
ing the capitate stalks, take the place of those
just noticed. Occasionally the threads branch
and, instead of ending in points, terminate in
dilatations of a like nature to those which inter-
rupt their filiform axes.

ACTINOZOA. 169

9. ]fepro«1uetive Organs. — The reproduc-
tive organs, in most Alcyonaria and Zoantharia^
agree, both as to position and striictiire, with the
same parts in Actinia; each spermarium or ova-
rium consisting of a prolongation of the peritoneal
membrane which clothes the sides of the mesen-
teries, and forms along their free edges a double
band, within which true generative elements are
produced {Ji<j. 26, c). Each band usually con-
tains only ova or spermatozoa, but ovaria and
spermaria may occur either in the same or in
different pol3T)es. Cerkinthus and some other
forms are monoecious, but more frequently, as in
the majority of Sea-anemones, the sexes appear
to be distinct. Not so, however, the Ctenophora,
in which group bisexuality may certainly be said
to prevail. An ovarium and spermarium occur as
thickened folds along the opposite sides of each
ctenophoral canal, beneath its endodermal lining.
But the same mesentery gives rise on both of its
free sides to only one kind of generative element.
Here, as in other Actinozoa, the male and female
organs differ in their contents alone.

The ova of the Actinozoa are, in general, of a
rounded form, smooth or dilated, and often bril-
liantly coloured. Their structure is typical, pre-
senting the parts common to ova in general. The
spermatozoa are caudate, with a broadly conical
or even heart-shaped body, to the apex of which
the tail is usually attached.

Eeproduction does not, as in so many Hydrozoa,
devolve upon specially ^modified zooids. Some
Actinozoa have been known to become everted
and die shortly after the maturation of their genital
products. But in others, no such exhaustion

170 ACTINOZOA.

seems to occur. Fertilisation is probably effected
while yet the ova remain within the somatic cavity
of the parent. Here, in many cases, the early
stages of development also take place.

Section II.

DEVELOPMENT OF ACTINOZOA.

The life-history of the Actinozoa presents a
series of phenomena by no means so diversified as
those which have been shown to characterise the
developmental cycle in most of the Hydvozoa.
For here the embryo, by an easy and gradual suc-
cession of changes, tends finally to assume the
condition of an organism similar to that which
brought it forth.

In the evolution of the embryo the whole or a
greater part of the fertilised ovum seems to be
concerned. The product of the reproductive act
usually soon appears as a ciliated body, while yolk-
cleavage, division into layers, and formation, by
liquefaction, of an internal nutrient cavity, take
place in the ordinary manner.

ZoANTHARiA and Alcyonaria. — Among the
Zoantharia and Alcyonaria the further develop-
ment of the primitive polype into which the embryo
is resolved would seem to be, in most cases, as
follows. The young animal, still retaining its
cilia, assumes a somewhat oblong ovate form. A
central depression then makes its appearance at
one extremity, indicating the rudimentary mouth.
The internal cavity enlarges. Meanwhile, the

\

ACTINOZOA. 171

tegumentary system is gradually becoming dis-
tinct, thread-cells accumulating to form a super-
ficial layer, beneath which pigment globules may
also be observed. Soon the muscular substance
begins to be differentiated, its longitudinal fibres
and the first rudiments of the mesenteries being,
at an early period, discoverable. Next, small pro-
tuberances arise round the mouth, each of which
gradually elongates to form a tentacle. The initial
number of tentacles in the embryonic polype
always bears some relation to that observable
among the adult forms of the group of which it
is a member. Thus, in most Zoantharia either
five or six tentacles first sprout forth, but this
number is rapidly doubled, an increase in size of
the older tentacles being simultaneously effected.
But, in very young Alcyonaria eight tentacles
appear, as in the mature polype.

Within the body-substance transverse contractile
tissues may now, at length, be detected. Minor
changes of external form also take place; the
cilia disappear, or are replaced by others of smaller
size; and the proximal extremity modifies itself
in accordance with the habits of the adult animal.

But before the formation of its tentacles, the
young polype undergoes that important structural
change which distinguishes it from the rudimen-
tary polypite of the Hydrozoa. A circular fold of
the body-substance surrounds the oral extremity
and grows inwards in such a manner as to produce
the wide digestive sac, open above and below, and
freely communicating with the somatic cavity,
from which it, nevertheless, remains distinct. The
histological composition of the wall of this sac

172 ACTINOZOA.

differs, in no essential respect, from that of the
outer boundary of the body.

The polype, while yet immature, presents a well-
marked bilateral symmetry. The oral fissure is
produced more in one direction than in another,
its form being by no means, as some have wrongly
stated, circular. At its opposite angles, gonidial
grooves, in certain cases one only, arise. Two of
the mesenteries in Actinia, as Haime has pointed
out, are developed opposite to each other before
the rest make their appearance, and these in direc-
tion correspond with the two mouth-angles. The
mesenteries grow from above downwards and, in
some long-bodied polypes, do not extend much
farther than the level of the free end of the
digestive sac, or, becoming narrowed and much
convoluted, are finally lost in the proximal portion
of the wall of the somatic cavity. In Cerianthus
two of the mesenteries descend, far below the
others, almost to the orifice at the base of the
general cavity. The remaining mesenteries, much
shorter than the preceding, gradually diminish in
length till they reach two points at either side of
the larger mesenteries and, like them, opposite to
one another.

The rudimentary tentacles, also, afford proofs of
the symmetry just noticed. In those young Zoan-
tharia which possess five of these appendages,
four, as Agassiz has stated, are arranged in pairs
on either side of the mouth, while the fifth lies
opposite one of the oral angles.

The subsequent development of the tentacles
has been well illustrated by Haime from the case
of the common Sea-anemone. The succession of
these organs is effected from within outwards in a

ACTINOZOA. 1 73

series of concentric circlets, each of wliich, save
the second, includes twice the number of tentacles
proper to its predecessor. Thus, the first circlet
contains 6 tentacles, the second 6, the third I2,
the fourth 24, the fifth 48, and the sixth 96. In
Antijjctthes and some other polypes never more
than the first six tentacles arise. Among certain
other Zoantharia, one or more tentacles are occa-
sionally aborted, and hence the somewhat puzzling
numerical proportion of these organs noticed in
slightly abnormal forms of this group.

Should the young polype give rise to a calcareous
corallum, the early stages of its deposition consist
chiefly in the formation of spicules which, at first
small and detached, gradually increase in size and
coalesce in a greater or less degree to form the
various structures whose nature has elsewhere been
explained.

Where a septal apparatus occurs, the develop-
ment of its several elements follows the same
definite law by which the number of the mesen-
teries and tentacles is determined.

The young Zoantharian has at first six septa,
the Eugose Coral four, equidistant from one
another, and separating similar loculi. In a few
genera only can a smaller number of initial septa
be detected. But among the great majority of
Zoantharia, the typical grouping of the septa is
hexameral ; among the Rugosa, tetrameral.

In some Zoantharia the number of the septa
never exceeds six, but, more frequently, new septa
appear midway between those first formed, at the
lower portion of the theca. These gradually grow
inwards, at the same time increasinof in heiofht.
Thus the primary loculi become divided into

174 ACTINOZOA.

secondary chambers ; these, by the formation of
other intermediate septa, into tertiary chambers,
and so on till the development of the corallite has
been accomplished. The primary loculi alone
are complete, for the septa limiting the other
chambers neither extend so high, nor so closely
approach the columella, as do those which are
-afterwards formed ; the size of all succeeding
series of septa being in direct ratio to the order
of their development. So that the latter may be
almost as clearly pronounced in an adult corallite
as in a collection of specimens of different ages
belonging to the same species. In like manner
those stages of the septal apparatus which are
transitional among the more complex coral lites
are well illustrated by an appeal to the various
permanent conditions of the same system among
less developed representatives of the group. Thus
in some species of StylojDhora we can count but
six septa, and an equal number of primary cham-
bers. In a much larger number of Zoantharia
twelve septa may be observed, of which six are
primary and six secondary. In others, there are
twenty-four septa; six primary, six secondary, and
twelve tertiary. The septal formula in all these
types admits, therefore, of being, respectively,
stated thus :

I.

. . 6 =S. 6.

II.

. . 6 + 6 =S. 12.

III.

. 6 + 6 + i2 = S. 24.

in S. 12 one, and in S. 24 three additional septa
being developed between each of the primary
pairs. And, since the normal number of primary
septa is 6 among the Zoantharia, it might be

ACTINOZOA. 175

expected that for oil corallites of this order having
a higher formula than S. 24, the number of new
septa produced within each of the first formed
chambers would be represented by the successive
terms of the series 7, 15, 31, 63, &c., in which
each number is double plus unity of its antecedent.
Practically, however, we find that, as soon as the
formula S. 24 has been reached, only two septa
arise at the same time in each primary chamber,
or, in other words, the corallite developes simul-
taneously not more than twelve additional septa.
So that the simple expression n x 6 + 6 at once
determines the normal septal formula for all
Zoantharian corallites having twenty-four or more
septa. Here n is = the number of septa between
each primary pair, and corresponds to the succes-
sive terms of the arithmetical progression 5, 7, 9,
II, 13, &c.

Since, therefore, the number of septa in process
of formation is often less than the number of
loculi, it becomes necessary to determine those
chambers in which the new septa first appear, and
the precise order of succession which they observe.
But first let us explain the few technical terms by
which the facts to be announced have been ren-
dered susceptible of definition.

All septa which commence their growth simul-
taneously are said to be of the same order, while
those which divide chambers of equal size belong
to the same cycle. It is evident that septa of the
first three cycles must correspond to those of the
first, second, and third orders, respectively. But
the fourth cycle includes septa of the fourth and
fifth orders ; the fifth, of the sixth, seventh, eighth,
and ninth; and the sixth of the lOth, nth, 12th,

176

ACTINOZOA.

13th, I4tli, 15th, 1 6th, and 17th orders. In but
few corallites does a seventh cycle occur.

The lines in the subjoined diagram are supposed
to represent the septa of part of a corallite having
the formula S. 48, each septum being indicated
by the numeral proper to its order.

The same numerals also enable us readily to
point out any one of the loculi included between
the primary septa. In the present case the
chambers, from left to right, would be denoted
as 1+4, 4 + 3, and so on, respectively. Again,
chambers of equal size are said to be similar, and
those represented by corresponding numerals are
of the same expression. Thus, in the diagram,
the chambers 1+4 and 4+1 have the same ex-
pression. So, also, just before the development
of septa of the fifth order, the chambers then
separated from one another by the septa marked 3
could not have been denominated similar.

ACTINOZOA. 177

We are now in a position to understand the five
rules of septal development laid down by Milne
Edwards and Haime, to whom science is indebted
for most of what we yet know on the subject
under consideration.

Kule I. The formation of new septa takes place
simidtaneously in all loculi having the
same expression.

Eule 2. The formation of septa takes place sue-
cessively in loculi having a different ex-
pression.

Eule 3. The order of succession of the septa is
determined in the first place by the age of
the cycle to which they belong, and those
of a new cycle do not commence to be
formed till the development of the pre-
ceding cycle is complete.

Eule 4. Among loculi of the same cycle having
different expressions precedence in the for-
mation of new septa is determined by the
inferiority of the sum of the two terms of
this expression.

Thus, if septa of the 6th order were to be de-
veloped in the corallite indicated by the above
diagram, we should expect them to appear in the
chambers 1+4 and 4+1.

Eule 5. Among loculi of the same cycle having
different expressions, but which yield the
same sum by the addition of the two terms
of each expression, the order of appearance
of the septa is determined by the relations
which exist between the lowest terms of
these expressions, the new septa being
formed first where the lowest term occurs.

178 ACTINOZOA.

Or, recurring to our diagram, we might look
for the appearance of new septa in the chambers
5 + 2 and 2 + 5 sooner than in 4 -f 3 and 3 + 4.

As might be expected, from abortion and other
causes, variations from the above arrangements now
and then occur, the careful investigation of which
is far from exciting that attention which it so well
deserves.

It is to be remarked that some coralligenous
polypes, when in confinement, appear to attain
a considerable bulk before any traces of their
skeleton can be observed.

Except in its minute size, and the comparative
paucity of its tentacula, the young polype, when
first excluded, closely resembles its parent, through
whose mouth it usually makes its entrance into
the surrounding world. The student will ex-
perience little difficulty in obtaining Actiniae
containing young in several stages of development
for detailed anatomical examination.

It is to be wished that the embryology of the
composite Alcyonaria and Zocmtharia were more
efficiently worked out, since it is just possible
that, apart from other modifications of the de-
velopmental process, a rudimentary coenosarc may,
under certain circumstances, be produced before
the formation of a distinct polype.

CTEywFHORA. The development of the Cteno-
phora, while presenting some peculiar features,
resembles that of other Actinozoa in the compara-
tive rapidity with which its early stages progress,
so that the product of the reproductive act, while
yet of small si^e, attains a form and structure
similar to the parent.

ACTINOZOA. 179

The young Bolina, according to Mc Crady, is
not very unlike an adult rieurobrachia in shape,
but the oral extremity appears somewhat truncate,
and smaller than the more rounded apical region.
From around the latter radiate eight short cteno-
phores, each containing from five to seven combs.
The mouth leads into a very large digestive sac,
tapering rapidly as it proceeds inwards to meet
the narrow funnel, from which are given off two
broad lateral sinuses, opening by eight " short
pointed projections" into the embryonic indica-
tions of the ctenophoral canals. The extremity
of each horizontal sinus also connects itself with
a wide excavation, or pit, at the side of the body,
from which freely issues a short tentacle, furnished
with three or four fringing threads. Apical canals,
also, may be observed to the right and left of the
large rudiment of the ctenocyst. At a later stage
are developed the paragastric canals, while as yet
the ctenophoral tubes remain short and incomplete.
These last are seen to approach the apical region,
before lenofthening so as to come near the oral
pole of the body. Their development would seem to
be somewhat in advance of that of the ctenophores
themselves. From the wide horizontal sinuses
branching radial canals soon shape themselves
out. The mouth gradually becomes depressed,
while the large antero-posterior lobes of the adult
animal are making their appearance. Meanwhile,
the tentacular pits shift, as it were, rather closer
to the oral extremity, until at length they arrange
themselves in their final position on either side
of the mouth. The tentacles alter very much
in appearance as their development proceeds,
becoming more numerous, but simple and less

N 2

180 ACTINOZOA.

contractile, so as no longer to resemble those of
Pleurohraclila, Short branches are given off from
the extremities of the paragastric canals, and from
these the two canals which run by the sides of the
mouth are probably produced. Not until the large
lobes have become very distinctly recognizable do
the earlets, four small appendages in connection
■with the lateral ctenophores, render themselves
visible.

The observations of Semper on Chiajea onulti-
cori'iis, another lobed representative of the same
order, indicate a still more rapid evolution of the
embryo, which before quitting its egg-covering
has the outward form of the adult animal, the
canal system and ctenophores being as yet rudi-
mentary. As in Bolina, the digestive tube appears
the first formed part of that system, and is, indeed,
developed earlier than any other internal organ.
At no period of its career is the young Chiajea
provided with a uniform covering of cilia.

In Beroe, a genus destitute of tentacles, the
funnel of the embryo is comparatively large, before
the ctenophoral canals are fully developed. The
circular oral vessel is formed from two lateral
tubes, whose e^^tremities anastomose with four of
the ctenophoral canals, while as yet the four others
have not approached more than half way the oral
pole of the body, towards which, as in the pre-
ceding, they are gradually developed. We have
here an interesting proof of the bilateral symmetry
of the Ctenophora, for in the adult Beroe, as from
its want of tentacles might have been expected,
this symmetry is, on a hasty inspection, less obvious
thai), in most other members of the order.

Like Chiajea, Pleurobrachia is developed within

ACTINOZOA.

181

an egg-covering, and with an equal degree of
rapidit}^ After yolk-cleavage the embryo appears
rudely cylindrical in form, a belt of cilia passing
round the middle of its body {fig, 31). This soon

®

Development of Pleubobrachia : — a, newly extruded ovum,
showing the yolk with its external covering ; b, the same, after
segmentation ; c, d, and e, successive stages of the embryo. (All
magnified.)

breaks up into two lateral groups which eventually
disappear altogether ; the ctenophores, at first very
broad and few in number, at an early period taking
on the performance of their special function. The
tentacles are at first destitute of lateral fringes,
traces of which appear while the rest of the canal
system is but imperfectly indicated. The form of
the adult animal is, according to Agassiz, fully
assumed before the young organism attains a
length of '04 of an inch.

N 3

182 ACTINOZOA.

Of the duration of life among the Actinozoa,
as among Coelenterata in general, we are still very-
ignorant. Some Ctenophora appear to last but a
single season, yet this statement can by no means
be regarded as true of the entire group. The
Alcyonaria and Zoantharia seem long-lived and
hardy animals, a specimen of the common Sea-
anemone, kept in confinement for forty years,
showing no visible signs of decrepitude or old
age.

In reparative power the members of this class,
notwithstanding their' increased differentiation of
tissue, appear fully to rival the Hydrozoa. Many
experiments show how complete may be the healing
of wounds and regeneration of lost parts among a
large number of Actinozoa, Occasionally the pro-
cess of reparation displays itself in a curiously
abnormal manner. Thus, a section having been
made below the disc of a Sea-anemone, tentacles
were developed from both of the fresh surfaces
thus exposed.

A common British Zoantharian, Anthea cereus,
adapts itself well to the conditions of such experi-
ments. Artificial fission of this species, if per-
formed with care, does not always result in death
of the parts so divided.

The same animal may also illustrate the mode
in which spontaneous fission occurs amoug many
other forms of Actinozoa. A longitudinal cleavage
of the polype commences, in most cases, across
the region of the disc, and thence proceeds down-
wards towards its proximal extremity.

Less frequently is fission effected by the separa-
tion of small portions from the attached base of
the primitive organism, whose form and structure

ACTINOZOA. 183

they subsequently, by grcadiial development, tend
to assume.

Further observations are wanting on the occur-
rence of fission and oremmation amongr the Cteno-
jjhoiXL In none of these animals do we find
colonies of zooids resulting, as in the Hydrozoa,
from a process of continuous budding.

But in other Actlnozoa continuous gemmation
abundantly takes place, and in this manner are
formed those composite structures, consisting of
numerous pol}^es, met with among so many
genera of Zoantharia, Alcyonaria and Rugosa.
In a few of these organisms, discontinuous gem-
mation may also be noticed.

Young polypes may be budded either (i) from
the base of the primitive structure, or (2) from
the sides of the polypes, or (3) from their oral
discs. Since these surfaces are but parts of a
common integument it might be anticipated that
intermediate positions of buds would now and
then occur. Nevertheless, it has been found con-
venient to distinguish three principal modes in
which gemmation of polypes may be effected, as
the basal, the parietal, and the calicular, respec-
tively.

In basal gemmation the polype sends forth a
rudimentary coenosarc, from which, after a time,
the young polype-bud is produced, and so on for
all the zooid forms subsequently evolved.

The extent to which a coenosarc may be deve-
loped varies, however, considerably. It must not
be inferred that in every composite Actinozoon
such a structure is present ; for the mass may
exhibit nought else save a congeries of polypes
in immediate mutual connection. In this case

184 ACTINOZOA.

multiplication by fission or by parietal gemma-
tion has probably occurred {fig, 32).

If parietal gemmation be repeated several times
during the growth of the budding organism, it is
said to be indefinite ; if once only, definite.

Indefinite gemmation is termed regular when
the young polypes arise at points which are deter-
minate for the same species ; irregular, when
buds are produced indifferently from different parts.

The results of definite gemmation vary accord-
ing as the producing and produced zooids are
turned towards the same side, or in opposite direc-
tions.

Calicular gemmation is only known among cer-
tain extinct coralligenous Actinozoa. In Cyatho-
phyllum and its allies, members of the order
Rugosa, the primitive pol3rpe sends up from its
oral disc two or more similar buds, these, in their
turn, produce other young polypes, and thus, the
process is repeated, until an inverted pyramidal
mass of considerable size is formed, all the parts
of which rest upon the narrow base of the fii'st
budding polype.

Milne Edwards has carefully insisted on the
necessity of distinguishing between fission and
gemmation among the Actinozoa. The oral disc
of the budding polype always remains entire, and
the bud, when it first appears, wants not only a
mouth, but most of the other structures which
subsequently it acquires ; whereas the polypes
produced by fission resemble each other in organ-
isation and, not unfrequently, in size, as soon as
they become distinct. Here, it need hardly be
said, oral fission is referred to ; basal fission, a
mere variety of discontinuous gemmation, being,*

ACTINOZOA. 185

as above stated, of comparatively rare occur-
rence.

Every polype-bud is, therefore, at first, no more
than a protuberance from the two parent layers,
enclosing a caecal diverticulum of the somatic
cavity. Thus, the nutrition of the young zooid is
provided for, till it developes for itself a mouth ;
after which it may either still continue its primi-
tive connection with the common mass or, as
already stated, by deposition of tissue secretions,
become, physiologically, a separate organism,
though morphologically associated with other
zooids of the same composite fabric.

The growth, gemmation, and fission of the
composite coralligenous Actinozoa require, in
addition to what has been said, the following more
particular explanations.

The whole compound structure in the sclero-
basic species may be regarded as the result of a
peculiar modification of the process of basal
gemmation ; the rudimentary coenosarc developed
around the base of the primitive polype, instead
of spreading only at its circumference, shaping
itself into a slender and, at first, slightly elevated
stem which, gradually increasing in height, con-
tinues, at the same time, to excrete a succession
of epidermic layers. Thus the young coenosarc
enlarges in diameter, and soon a number of buds
are seen to spring from its surface. Some sclero-
basic Corals remain thus throughout life, the
coenosarc, with its axial excretion, merely becom-
ing taller. But, more frequently, branches are
sent forth, which resemble, in every essential
respect, the stem from which they originated.

1 86 ACTINOZOA.

These branches may be apparently quite irregular
in their arrangement, or they may arise in the
same vertical plane from opposite sides of the
stem, sometimes uniting one with another by the
formation of an intricate network, as in the well-
known and beautiful Fan-Corals.

Such is the growth of the coralhim in the fixed
Gorgonidce and Antipathidce. Among the Penna-
tulldce, which are free forms, the proximal end of
the coenosarc usually becomes produced into a
gently swelling or tapering mass, supported by a
comparatively slender, elongated sclerobasis.

More varied are the modifications of the com-
posite sclerodermic corallum, the increase of which
may be the result of fission alone, or of gemmation
alone, or of gemmation and fission combined.

Among coralligenous polypes fission is usually
effected by oral cleavage, as indicated above ; the
proximal extremity of each, for a greater or less
extent, remaining undivided. The two zooids thus
produced either supply by reparation those parts
which were wanting to render them complete, or,
it may be, continue through life in a more or less
imperfect condition.

The coral structures which result from a repe-
tition of the fissiparous process are of two principal
forms, according as they tend most to increase in
a vertical or horizontal direction. In the first of
these cases the corallum is coespitose, or tufted,
convex on its distal aspect, and resolvable into a
succession of short diverging pairs of branches,
each resulting from the division of a single coral-
lite. In some parts of the mass the walls of the
coral lites blend, so that a few of them may even
become compacted with one another.

ACTINOZOA. 187

When growth takes place chiefly in a horizontal
direction, the so-called lamellar form of corallum
results. Here, the secondary corallites are united
throughout their whole height, and disposed in a
linear series, the entire mass presenting one con-
tinuous theca. Sometimes it is possible to count
the number of corallites, but often their several
calices merge, as it were, into a single groove,
traversed, perhaps, by a columella running parallel
to its sides, towards which two opposite rows of
septa are seen to converge.

Both the lamellar and ccespitose forms of coral-
lum are liable to become massive by the union of
several rows or tufts of corallites throughout the
whole or a portion of their height. An illustration
of this is afforded by the large gyrate corallum of
Mceandrina, over the surface of whose spheroidal
mass the calicine region of the combined corallites
winds in so complex a manner as at once to suggest
that resemblance to the convolutions of the brain
which its popular name of Brain-stone Coral has
been devised to indicate.

Basal gemmation, among sclerodermic Corals,
affords very different products, according as the
coenosarc remains soft, or deposits a coenenchyma;
appears under the form of stolons, or of stouter
connecting stems ; or even spreads out in several
directions as a continuous horizontal expansion.
In this case it is evident that the latest formed
parts of the mass are those which are situated
nearest to its circumference.

The corallites may either appear widely sepa-
rated from one another, or closely aggregated
and, perhaps, confused by reason of the scanty
development of an intervening coenenchyma.

188 ACTINOZOA.

In some cases the basal deposit encroaches but
little on the sides of the corallites ; in others it
rises upwards till on a level with their calicos, or
even above them. Again, instead of remaining
horizontal, it may become folded in such a manner

Fig. 32.

TuEBiNAEiA PALiTEBA. Corallum, of the natural size.

that one part of its distal surface is brought into
more or less close contact with the other; the
corallites, which previously seemed to arise from
the same plane, having now their calicos turned in
opposite directions. The resulting corallum will be
dense or foliaceous, according to the degree of de-
velopment of the included coenenchyma, and the
greater or less elongation of its projecting corallites.

Budding from the sides of the corallites takes
place, however, more commonly than basal gem-
mation, not that the latter must be considered of
infrequent occurrence.

The aspect of the corallum to which parietal
gemmation gives rise varies in accordance with

ACTINOZOA. 189

1. The number of buds which each corallite pro-

duces :

2. The frequency with which the budding process

is repeated; whether once only or several
times during the life of the corallite, but at
different stages of its growth :

Fig. 33.

Dendeophtllia nigrescens: — Part of a branch, in outline,
of the natural size. A single polype is shown separately, mag-
fied.

3. The height which each corallite attains :

4. The angle at which the growing buds diverge

from the parent corallite :

5. The degree of rapidity with which such diver-

gence takes place :

6. The position of the buds produced ; from one

side only of the budding corallite, or indif-
ferently from any part of its circumference ;
at a short distance from its base, or relatively
nearer to the edge of the calice :

190 ACTINOZOA.

7. The degree and nature of the union between
the several corallites; whether this be pro-
duced by the close contact and blending of
their walls, or by the development, both in
a horizontal and vertical direction, of an
epitheca, coenenchyma, and other similar
structures {fig. 33):
and several less essential modifications of the same
process, manifested either separately or in com-
bination.

It is, therefore, not surprising to find among
budding Corals forms analogous in every respect
to those produced by fission and, in addition, many
others whose physiognomy, copious as is the list of
descriptive terms, it is scarcely possible to define.
Nor does each Coral always restrict itself to a
single mode of growth, but, on the contrary,
several kinds of gemmation, or of gemmation
and fission in unison, have been observed to take
place in the same species. So that, even with the
aid of figures, and extensive suites of museum
specimens, the development of the composite
Corals can receive illustration in but a limited
and imperfect degree. For these organisms, like
all others, rightly to be understood, must be
studied amid ^' the glorious variety of Nature '*
itself, "living and multiplying in their destined
homes and habitats."

Mr. Dana, who devoted much time to the exami-
nation of the Corals of the Pacific, thus endeavours
to describe some of their general diversities of
form :

" Trees of coral are well known ; and although
not emulating in size the oaks of our forests, —
for they do not exceed six or eight feet in height.

ACTINOZOA. 191

— they are gracefully branched, and the whole
surface blooms with coral polyps in place of leaves
and flowers. Shrubbery, tufts of rushes, beds of
pinks, and feathery mosses, are most exactly
imitated. JNIany species spread out in broad
leaves or folia, and resemble some large-leaved
plant just unfolding; when alive, the surface of
each leaf is covered with polyp flowers. The
cactus, the lichen clinging to the rock, and the
fungus in all its varieties, have their numerous
representatives. Besides these forms imitating
vegetation, there are gracefully modelled vases,
some of which are three or four feet in diameter,
made up of a net-work of branches and branchlets
and sprigs of flowers. There are also solid coral
hemispheres like domes among the vases and
shrubbery, occasionally ten or even twenty feet
in diameter, whose symmetrical surface is gor-
geously decked with polyp-stars of purple and
emerald green."

Under such aspects appear the living organisms
whose combined efforts have mainly constructed
those reefs and islands of Coral orio-in which now
lie scattered far and wide over the surface of the
tropical ocean. Three principal forms under which
such reefs occur have been distinguished by the
names of Fringing-reefs, Barrier-reefs, and Atolls.

Fringing-reefs skirt the shores of favourably
situated lands, towards which, at a gentle slope,
they incline, ending abruptly seawards, where
soundings reveal a depth of from 20 to 30 fathoms.
The surface of the reef is covered at high water,
and forms a nearly level platform, from a few feet
to more than a mile in breadth, according to the
degree of inclination which the land presents. Its

192 ACTINOZOA.

outer margin may be rendered sinuous by bays,
or the continuity of the whole reef completely
interrupted by one or more irregular inlets.

Barrier-reefs differ from those just mentioned
in occurring at a greater distance from shore, a
wide channel of relatively smooth and shallow
water flowing between, mthin which terraces of
fringing corals sometimes find their proper en-
vironment. Outside the reef depths almost un-
fathomable have been obtained, as close as it is
possible to venture amid the rolling surf which
breaks in unceasing billows against its surface.

Besides the larger Barrier-reefs which, in cer-
tain cases, attain a length of many hundred miles,
there are others, in all respects similar, but more
variable in size, which are often found surround-
ing the smaller islands of the Pacific.

From such reefs to Atolls we may trace every
possible transition. An Atoll differs from an
encircling Barrier-reef in enclosing, instead of a
central island with its intervening channel, an
uninterrupted surface of calm water, or lagoon.
The low-lying strip of land separating this lagoon
from the line of white breakers which marks the
outer boundary of the Atoll is seldom more than
half-a-mile in breadth, and presents an imper-
fectly circular or prolonged crescentic form,
though occasionally, as in Whitsunday Island, the
circle is complete.

Eeefs, in general, grow only along their outer
edge and, when upraised above the sea-level,
always appear highest to the windward side.

Clear sea-water, well aerated, at a certain tem-
perature, and a depth of not more than 25 or 30
fathoms, are among the most important of those

ACTINOZOA. 193

external conditions which seem favourable, if
not essential, to the growth of reef-building
Corals.

But to what other influences do the above three
classes of reefs, which present so much in common,
owe their occurrence? It has been said that they
are of Coral origin ; yet how is it that some of
them rise from depths so considerable, seeing that
those living Corals, by which they have been
constructed, build only in seas comparatively
shallow ?

It is not our business here to discuss the various
speculative views which have from time to time
been put forward on the subject of Coral-forma-
tions. I..et it suffice to say that Mr. Darwin's
theory of elevation and subsidence offers the only
consistent explanation of most of their known
phenomena which science is prepared to receive.

If we suppose a Fringing reef, together with
the area which it surrounds, to sink at a rate not
more rapid than the upward growth of its consti-
tuent Corals, the reef itself will undergo little
apparent alteration, while a channel of water,
gradually increasing in width, will appear between
it and the more elevated regions of the slowly
submerging land. Thus a Barrier-reef is formed.
Depression still going on, the land encircled by
the reef is reduced to one or more projecting
peaks, as in those islands of the Pacific to which
allusion has been made. Further subsidence
causes these peaks to disappear beneath the
sea-level, and the Barrier-reef changes into an
Atoll.

Fringing-reefs, therefore, show that the shores
which they skirt are stationary or rising, while

0

194 ACTINOZOA.

Atolls and Barrier-reefs attest that subsidence has
taken place.

That such slow subsidence does occur over
many parts of the extensive area occupied by
Coral-reefs is proved by a number of considera-
tions ; some of which were forcibly suggested by
Sir Charles Lyell, several years before Mr. Darwin's
theory was promulgated.

Observations and experiments made with a
view to ascertain the rate of growth of the reef-
building Corals have not hitherto yielded a suffi-
cient number of accurately recorded results.
That, in certain instances, their growth is rapid,
varying, however, mth the species, may be re-
garded as proven ; but it is also far from impro-
bable that there are many species which have the
same average rate of increase. It may likewise
be conceded that the growth of the same species
varies at different periods and under difterent
external conditions. In future investigations on
this subject the particular form of the corallum,
and its mode of fission or budding, should in each
case receive attention. For it is obvious that the
same amount of calcareous deposit, if appropriated
respectively by a massive and by a dendroid
species, would give rise to apparently dissimilar
quantitative products.

The Coral-polypes are, however, powerless to
raise their structures higher than the line of low-
water. To effect this, various agencies, but chiefly
the forces of wind and ocean, acting upon masses
detached from the reef and other marine debris,
are brought into play. But the mode of operation
of these agencies, the general theory of elevation
and subsidence, the conversion of the irregular

ACTINOZOA. 195

surface of the reef into one continuous level, and
the alterations to which its dead and deeply-sub-
merged portions become exposed in the lapse of
time ; these, and other kindred subjects of inquiry,
fall rightly within the province of the geologist.
As monuments of past change. Coral-reefs form
the basis of some of the most " splendid general-
isations " which his science has deduced. For
him an island occupied each region where now
without interruption flow the quiet waters of a
lagoon. And seas must have once rolled over
those existing continents amid whose mountain-
chains remains of ancient Coral-reefs abound.

Yet the zoologist, in taking leave of his own
department of this subject, cannot, without satis-
faction, contemplate how large a portion of the
earth's substance must, during the long lapse of
geological time, have formed part of the organised
structures of a group of beings who still continue
to fulfil, with no impairment of efficiency, the
great task for ages allotted them in the scheme of
universal nature. Not matter only, but force, he
sees made subject to their sway. The physical
agencies, which seemed at first to threaten destruc-
tion to the growing Coral, are soon successfully
overcome, and then pressed into its service. Nay,
without their aid, so much of the reef as rises
above the ocean level, forming the abode of plants
and animals, and finally of man, could not even
have existed. But had Coral-pol3rpes not pre-
viously laboured, the same forces would have been
potent only to destroy.

02

196 ACTINOZOA,

Section III.

CLASSIFICATION OF ACTINOZOA.

I. Classification. — 2. Order i; Zoantharia, — 3. Order 2: Alcyon-
aria. — 4. Order 3 : Kugosa. -— 5. Order 4 : Ctenophora.

I. Classification. — The class Actinozoa may
be divided into the four orders here defined :

1. Zoantharia.^ Actinozoa,mwhich the tentacles

are simple or variously modified, in general
numerous and, together with the mesenteries,
disposed in multiples of five or six. Corallum
absent or sclerobasic, in most sclerodermic,
the septa of each corallite following the
numerical law of the soft parts.

2. Alcyonaria. — Actinozoa, in which each polype

is furnished with eight pinnately fringed
tentacles. Mesenteries and somatic chambers
in number some multiple of four. Corallum
sclerobasic or spicular, rarely thecal, and
never presenting traces of septa.

3. Hugosa. — Actinozoa, presentinga sclerodermic

corallum, with tabulae and well developed
septa arranged in multiples of four. Soft
parts unknown.

4. Otenophora, — Transparent, oceanic, delicate

gelatinous Actinozoa, swimming by means
of ctenophores, or parallel rows of cilia dis-
posed in comb-like plates. No corallum.

Z. Order i : Zoantharia. — The chief modi-
fications of the plan of ^^^nicture proper to the

ACTI.NOZOA.

197

Zoantharia have reference to the form of the
polype, with its tentacles and corallum, and that
of the compound mass resulting therefrom by
a repetition of the process of gemmation. The
varieties which these animals present in size,
colour, and the superficial aspect of the body are
too multifarious to admit of any precise general
enunciation.

%. 34.

Zoantharia maxacodermata : — a, Actinia (or Sagartia)
rosea; b, Uyanthus Scoticus ; c, Arachnactis albida ; d, Zoanthus
CoucMi. (Natural size.)

The form and structure of the polype has been
best studied in the soft-bodied Zoantharians, more
o 3

198 ACTINOZOA.

familiarly known as Sea-anemones. A typical
Sea-anemone, when contracted, may be compared
to a more or less depressed cone, rounded above,
with a gently spreading base; when expanded,
to a column, for the most part cylindrical, but
widening somewhat towards its extremities {fig.
34, a). The base is scarcely broader than the
column in some species ; in Adamsia, on the
other hand, it is ovate in form, sending forth two
lateral lobes, which extend so far as to surround
the aperture of the univalve shell on which this
curious animal-flower is found, the lobes at length
uniting by a zigzag suture along the outer lip of
the shell. As in the case of Hydractinia, the
shells which Adarasia selects appear always to be
tenanted by a species of Hermit-crab.

The column may have its surface marked by a
variety of epidermic growths, or pierced by sundry
apertures. In Actinoloba its summit rises into a
conspicuous ridge, separated from the outer series
of tentacles by a deep depression or furrow. Oc-
casionally, as in the same genus, the margin of
the disc is waved or thrown into sinuous folds, so
that the arrangement of the tentacles appears
thereby somewhat confused. The peristomial space
may also vary in size, and relative depression or
elevation. The lips of the mouth undergo their
own modifications, and, in some genera, a groove
with mouth-tubercles occurs at but one of the oral
angles. In Actinopsis these tubercles are produced
upwards to form a pair of long, rigid, semi-
cylinders, the lateral margins of which again bend
downwards to terminate in cleft extremities.

In the non-adherent Sea-anemones, such as
Hyanihus and its allies, the column is propor-

ACTINOZOA. 199

tionally larger than in most of the Adinidce
proper. The base is either rounded or bluntly
tapering, and, in certain genera, becomes at times
much distended, as in Saccanthus or Edwardsia.

In some the base is furnished with a central
perforation : in others this appears to be wanting.
In Peackia the oral region is singularly modified ;
the tubercles of its single groove uniting to form
a tube, the expanded summit of which, * conchula '
of Mr. Grosse, presents a more or less thickened,
everted edge, cleft into a variable number of lobes.

The polypes of the Zoantliidce and true coral-
ligenous Zoantharia, save in characters merely
generic, resemble those of the Actinidce. Their
average size is, perhaps, smaller, though the Ac-
tinia Pawnotensis of the Pacific, whose expanded
disc measures a full foot in diameter, is, in this
respect, certainly exceeded by large specimens of
Fungia. This genus presents a widely extended,
circular or elliptic disc, destitute of the usual
folding margin, and blending, by insensible de-
grees, with the shallow, ill-defined column, over
the radiating septa of which the tensely stretched
soft parts converge towards the prominent, central
mouth. Such a simple form contrasts strikingly
with Meandrina and those allied genera in which
the several polypes produced by fission fuse to-
gether into a convoluted linear track, with tentacles
arising from its opposite sides.

Of these appendages and their variations, a
brief notice seems here required. In Anthea
{=Anemonia) they are long and slender; in
Fungia and Discosoma, reduced to mere warts
or papillae; in Capnea, very short, resembling
oblong tubercles ; while in Arachnactis, the outer

o 4

200 ACTINOZOA.

tentacles are capable of being extended to three
or four times the length of the body {jic). 34, c).
In Eumenides the tentacles are fusiform, iu Hete-
ractls moniliform, in Corynactis and Caryophyl-
lia they terminate in globose heads. Their free
extremities are often perforate, the animal having
the power of opening or closing the orifice at
pleasure. Dana describes the inner tentacles of
his Actinia flagelUfera as terminating in a re-
tractile pencil of hairs, but it is possible that these
hairs may have been, in reality, large everted
thread-cells. Although in most Zoantharia the
tentacles are simple, yet in Thalassianthtis and
its allies they are branched, and have their surface
studded with tubercles or papillae. In a few
genera, two kinds of tentacles appear on the same
polype; the one simple, the other lobed or
branched, as in Phyllactis.

The number of the tentacles, though in general
some multiple of five or six, is, in other respects,
liable to considerable variation. Their arrange-
ment, also, is correspondingly diversified. They
may dispose themselves in one, two, or more con-
centric series, and in some species they appear
irregularly scattered. Antipathes exhibits six
tentacles in a single circlet ; Peachia twelve, simi-
larly disposed ; while in the common Sea-anemone
there are nearly two hundred of these organs,
arranged in the manner noted above. In this
species, as in most other Zoantharia, the tentacles
of contiguous rows alternate. Cerianthus and
Saccanthus, however, possess two distinct circlets
of tentacles, the one oral, arising close round the
mouth, the other marginal, not far from the edge
of the disc, the tentacles of the inner row being

ACTINOZOA. ' 201

equal in number and opposite to those of the
outer.

The tentacles of most Zoantharia.siTe retractile,
but in CerianthuSf Anthea, and a few other
forms, this power is either absent, or imperfectly
exercised.

The numerous families of the present order
have been conveniently arranged by Milne Ed-
wards under three sub-orders : Malacodei^iata ;
Sclerobasica ; and Sderodermata.

In the Z. Malacodeimata, the corallum is
either absent or represented by scattered spicules.
The actinosoma, in most of these animals, presents
but a single polype. Exceptions to this rule oc-
cur, however, among the ZoantJiidce, the budded
polypes of which remain permanently united by a
coenosarc, in some linear, in others carpet-like or
incrusting. In certain Actinidce also, for example,
the Corynadis mediterranea of Sars, a similar
connection is maintained {fig. 34, d).

Within the soft parts of the Z. sderobasica
spicular tissue secretions seem wanting (fig, 35).
All the members of this sub-order are composite
structures. Antipatkes, the type of the group,
presents a stem-like, simple or branched coenosarc,
which in one species tapers to a length of more
than nine feet, with a basal diameter of scarcely
•3 of an inch. In this genus the sclerobasis is
horny, and each polype, according to Dana, has
but six tentacles ; but in the allied family of
Hyalodioetidce, the tentacles are twenty in num-
ber, while the basal excretion resolves itself into
numerous siliceous threads, transparent, twisted
into an erect axis. Doubts, however, are yet
entertained of the true nature of these so-called

202 ACTINOZOA.

" Glass-plants," whose siliceous stem may be the
product, not of the polype-mass, but of the sponge
on which it is parasitic.

Fig. 35.

ZoANTHAKiA SCLEROBASICA : — Part of a living stem of Anti-
•patlies avguina, of the natural size. Two polypes are shown
separately, magnified.

Of the Madrepores, or Z, Sclerodermata, want
of space forbids us to say much. A short sketch
of the several families into which the sub-order is
divided appears to be the best form into which
this part of our subject may be condensed.

Among the Turbinolidoe the coral lum is usually
simple, never presenting a coenenchyma. In
Coenocyathus continuous lateral gemmation takes
place, the corallites so formed remaining con-
nected in a close irregular tuft. In Blastotrochus
gemmation also occurs, but here it is discon-
tinuous. The septa are very perfectly developed,
not giving rise to either dissepiments or synap-
ticulse {fig. 36 a).

In the Oculinidce there is a very abundant
coenenchyma, which blends gradually with the

ACTINOZOA. 20B

thecae through their entire height. Each corallite
has its chambers slightly interrupted by a few
dissepiments.

The members of another family, Astrceidce, are
either simple or composite, but there is no proper
coenenchyma, as in the Ocalinidce. The epithecse
or costse form the chief substance of the mass by
which, sometimes, the corallites are separated.
Many J. s^imc^cc present that rapid fissive develop-
ment of the corallum, whose curious results have
been already indicated in the case of Meandrina.
Dissepiments, in most, are numerous. In number
of genera and species, perhaps, also, of indi-
viduals, and, apparently, in general importance,
the Astrceidce seem to surpass all other families of
the class Actinozoa.

The Fungidce are at once distinguished from
the three families just noticed by the possession of
synapticulse. Dissepiments are absent, nor can
it be said in many cases that a proper theca exists ;
the septa passing, without interruption, into the
costae, save at the base of each corallite. Both
simple and composite Fungidce occur, the latter
multiplying by lateral gemmation.

In the somewhat porous condition of their ill-
developed theca the Fungidce may be said to
differ from other Aporose Zoantharia, and ap-
proach the two next families, collectively distin-
guished by the title of Perforata. These, like
the Aporosa, have a well-marked septal appa-
ratus, and present no traces of tabulae.

Among the Madreporidce the sclerenchyma is
simply porous, the septa are distinct, and but
very slightly perforate. Save in Fupsamnia and
some of its allies, the corallum is composite, but

204

ACTINOZOA.

the thecae of the separate corallites do not become
lost in the surrounding coenenchyma. In the
Poritidce, on the contrary, such fusion always
takes place, and the septal system, instead of
forming distinct plates, consists wholly of more or
less definite series of trabiculae. The entire
corallum, in like manner, appears to be made up
of a spongy, reticulate sclerenchyma.

The next division, Tuhulosa, contains only a
single family, Auloporidce ; and this but two

Fig. 36.

ZoANTHAEiA scLEEODERMATA : — a, corallum of TurbinoUa
costata; b, the same, in transverse section, showing the columella,
septa, theea, and costae ; c, part of corallum of Aulo^ora tubsfor-
mis, (All, except c, magnified),

genera, Fyrgia and Aulopora, in both of which
the corallite, while destitute of tabulae, has its
septal system indicated by faint markings along
the inner surface of a comparatively smooth tube.
Fyrgia is simple, though having a distinct epi-
theca. In Aulopora the somewhat remote coral-
lites are connected by means of a basal creeping
ccenenchjTna. (Jig. 36, c.)

The four last families of Zoantharia constitute

ACTINOZOA. 205

the great division ©f Tahulata, in which the rudi-
mentary condition of the septa is made amends
for by the extensive development of the transverse
floors, referred to in the name this group bears.

{fiO- 29.)

The corallum of the Tahulata is mostly, if not
always, composite. Among the Mille'poridce an
abundant coenenchyma occurs, and the resulting
compound structure assumes a massive or folia-
ceous aspect. The substance of the corallum is
traversed by interspaces which give to its section
a somewhat tubular or cellular appearance. In
the Seriatoporidai it is more compact, but here,
like^vise, the coenenchyma is abundant, present-
ing, externally, a tufted or arborescent form. In
the Favositidce the corallites have their lamel-
lar walls brought into very close apposition, little
or no true coenenchyma being observable; while
in the Thecidce the septa form by their lateral
union the greater portion of the dense spurious
coenench3rma, of which their massive corallum is
composed.

The following may be given as definitions of
the families of Zoantharia. Those of the Madre-
poric forms, founded wholly on characters derived
from the corallum, admit readily of being ex-
hibited under the guise of an analytical table.

Order ZOANTHARIA.

Sub-order i. Z, Malacodermata.

Family i. AcTiNiDyE.

Corallum not evident. Polypes rarely con-
nected by a coenosarc; in general, locomo-

206 ACTINOZOA.

tive, the flattened base of each adhering at
pleasure.
Family 2. Ilyanthid^.

Corallum not evddent. Polypes unat-
tached, with rounded or tapering base ; no
connecting coenosarc.
Family 3. Zoanthid^.

Polypes attached, united by a coenosarc,
and furnished with a spicular corallum.

Sub-order 2. Z. Sclerobasica,

Family 4. Antipathid^.

Corallum sclerobasic, horny, smooth
or spinulous. Polypes with six tentacles.
Family 5. Hyalochcetid^.

Corallum, sclerobasic, composed of twisted
siliceous fibres. Polypes with twenty ten-
tacles.

Sub-order 3. Z. Sclerodermata.

{Tabulse present. Septa rudi-
mentary (Tabulata) 2
Tabulse absent 5

{Coenenchyma wanting, or ill-
developed 3
Coenenchyma abundant. 4

f Septa forming a spurious, mas-
^ < sive coenenchyma. . . Fami'y 6. Thecidje.
[_ Septa and coraUites distinct. . Family 7. Favositld^.

{Sclerenchyma compact. Coral-
lum arborescent. . . Family 8. Seeiatopobid-ze.
Sclerenchyma tubular or cel-
lular. .... Family 9. MilleporidjE.
{Septa indicated by faint striae
(Tubulosa). . , . Family 10. Auloporidje.
Septa well developed 6

{Sclerenchyma porous (Perfo-
^^JJ^) • 7
Sclerenchyma imperforate
(Aporosa) 8

ACTINOZOA.

207

9 1

Sclerenchyma reticulate. The-
coe not distinct from the sur-
rounding coenenchyma.
Sclerenchyma simply porous
Thec?e distinct. .
' Syniipticulse present. No dis'
sepiments. .
Synapticulae absent.
Dissepiments in general numer
ous. Ccenenehyma absent
or formed only by the deve
lopment of the costse or epl
theca.
Dissepiments few or absent,
'Ccenenehyma compact, abun

dant.
^No ccenenehyma. .

Family ii. PoaiTiD.B.
Family 12. Madeepobid.e.
Family 13. Fungid.e.

Family 14. Astb;eid.s:.

Family 15. Oculinid^.
Family i6. TufiBiNOLrD.s:.

In addition to those here defined, Milne Edwards
has distinguished four other families of Apoi^osa,
which inosculate, so to speak, between the primary
groups just mentioned. The first family. Das-
midai, includes but a single genus, closely related
to the Turhinoliclce, from which it differs in the
peculiar modifications of its septa. Each of these
is represented by three vertical laminae, united
only along their external margin. A second
family, Stylophoridce, appears as a transitional
group between the Oculinidce and Astrceidoe, As
in the former, there is a well-developed ccenen-
ehyma and few dissepiments ; but, on the other
hand, the surface of the ccenenehyma is echinu-
late, while it is smooth in the Oculinidce, and the
thecal of the coral lites do not, as in that family,
increase endogenously, so as almost to obliterate
the loculi. Another osculant family, Echino-
poridce, still more closelj resemhles the Astrceidce,
differing therefrom chiefly in the possession of a
foliaceous, basal ccenenehyma. But one genus

208 ACTINOZOA.

presenting this combination of characters has been
observed. The family Meridinidoe has, lastly,
been constituted for the reception of an equally
aberrant Astra^oid genus, Merulina, which clearly
points in the direction of the Fungidce, resembling
these corals in the perforate condition of its coral-
lum, though, as in the txMQ Astroeidoi^Tio synap-
ticulse occur.

3. Order 2 : Alcyonaria. — The Alcyonaria,
with the exception of one genus, Hairrieia, which
may, however, yet prove to be an immature form,
are composite in structure; their polypes being
mutually connected by a coenosarc, through which
permeate prolongations of the somatic cavity of
each, forming a sort of canal system, whose several
parts freely communicate and are, therefore, rea-
dily distensible.

Throughout the whole order the polypes exhibit
a very close agreement in structure, howsoever
much the coenosarc may vary. Each, when ex-
panded, displays a cylindrical, or somewhat octa-
gonal, tube, with delicate transparent walls, and
eight pinnate tentacula, whose form offers slight
though characteristic variations among the several
genera of the group. In some the polypes are
i-etractile into excavations which occur in the sub-
stance of the coenosarc, while in others such
excavations seem to be wanting.

Alcyonium, the typical genus, presents, when
first dredged up, a sufficiently repulsive aspect,
suggestive of the vulgar names, " Cow's paps " and
" Dead-man's hand," sometimes conferred on it.
But, when placed in sea-water, the lobate fleshy
mass, distending its aquiferous system, is gradu-

ACTINOZOA. 209

ally seen to become exquisitely pellucid, while
from all parts of its surface numbers of tiny
polypes, emerging, expand to the utmost their
star-shaped crowns of delicately fringed tentacula.
Within the somatic chambers circulating currents
may now be observed. These find their way up
one side of the tentacles, following the course of
the several fringes, and, having gained their sum-
mits, again revert, proceeding in a contrary direc-
tion. So that here, as in many Zoantharia, it
would not, perhaps, be too much to say that the
tentacles, by reason of the delicacy of their ciliated
walls, fulfil the proper function of a respiratory
system.

In Sarcodidyoiii as in Alcyonium, a spicular
corallum occurs, but the coenosarc is scanty and
creeping, resembling that of the Zoanthidce. So
also in Cornularia, the corallum of which is,
however, more consolidated. But of all the Al-
cyonidce proper Telestho, with its tufted, sub-cal-
careous, tubular corallum, makes the nearest
approach to the allied family of Tuhijpovidce,

The beautiful Organ-pipe Corals, forming the
several species of the genus Tubipora, appear to
be the sole representatives of this group. Allusion
has already been made to the exceptional structure
of their corallum, the colour of which, in all cases,
is of a bright crimson-red. The polypes are
either violet or grass-green in tint, and, according
to the dissections of Dana, present this anatomical
peculiarity, that two only of the mesenteric edges
are furnished with ova, the remaining six sup-
porting spermaria. The oral extremity of each
polype can be inverted for protection into the
summit of its calcareous tube, but it is wrong to

210 ACTINOZOA.

suppose that the latter completely invests the soft
parts of the animal, the corallum of Tabipora
being a true tissue secretion. Its horizontal outer
plates are suggestive of a distinct analogy to the
Tabulata, nor are traces of internal tabulae wholly
wanting. The characteristic form of Tubipora
seems due to the periodic budding of zooids from
the distal surface of the plates, while at the same
time certain of the older corallites continue to
increase in height. But neither the minute struc-
ture nor development of this interesting genus have
yet received proper attention.

The Gorgonidce differ from all other Alcyonaria
in having an erect branching coenosarc, firmly
rooted by its expanded proximal extremity (^fig.
37, c). Those which possess a horny sclerobasis
have been by many writers confounded with the
AntipatMdce ; but, apart from the anatomical
features of their polypes, they may at once be
known from the latter by the more or less sulcate
aspect presented by the surface of the sclerobasis.
The modifications w^hich this structure displays in
Corallium,lsis,Mopsea,Sind Melitcea have already
received a brief notice. It may suffice to add that
very exaggerated concei^tions seem to prevail as to
the height which the horny Gorgonidce are capable
of attaining. It is doubtful whether their largest
trees ever rise to more than five or six feet,
yet some have been reputed to rival oaks in
size, an assumption which, however incredible, is,
nevertheless, not inconsistent with theoretical con-
siderations.

The Alcyonaiia, as a group, seem destitute of
locomotive power, though one family of this order,
the Pennatulidce, have been often regarded as

I

ACTINUZOA.

211

oceanic, and described by such names as " Pol}"pe8
nageurs." It is more probable, however, that,
under ordinary circumstances, these creatures live
with their proximal extremity plunged firmly into
the sand or mud of the sea-bottom; the distal

Fig. 37.

PENNAxrxiDJE and Gorgoxtd^ : — a, dried stem of Virgidaria
mirabilis ; b, portion of another stem, in the living condition ; c,
corallura of Mopsca costata ; d, fragment of the same, {a is
reduced one-third ; h and c are of the natural size ; d is magni-
fied.)

end of the ccenosarc, which bears the numerous
polypes, freely exposing itself to the influence of
the clearer water above.

The ccenosarc of the Pennatulidce may be
slender and simply elongate, with very short
p 2

212 ACTIXOZOA.

pinnules, or lateral lobes, bearing the polypes, as
in Virrjularia, the sclerobasis of which is rigid,
tapering towards its extremities, and densely cal-
careous {fig. 37, a). In the true Sea-pens, forming
the genus Pennatula and its near allies, the pin-
nules are very conspicuous, and so modified as to
arrangement and comparative size that the whole
mass presents a striking resemblance to a bird's
feather. The proximal end of the coenosarc, often
for nearly half its length, is bare of pinnules or
polypes, appearing swollen and fleshy. In other
Pennatulidce the entire coenosarc is club-shaped,
without any pinnules, the polypes being irregularly
scattered, as in VeretilluiUf or arranged in longi-
tudinal rows on part only of the surface, as in
Kophobelemnon. In Benilla a comparatively
short coenosarc expands distally to support a
pmooth, s}Tiimetrical, kidney-shaped disc, from
the free surface and edge of which the scattered
polypes arise. This aberrant genus appears to
want a sclerobasis, the interior of the stalk and
disc being hollow, and in free communication with
the cavities of the polypes, so that the animal
possesses the power of largely increasing its
dimensions by allowing itself to become expanded
by the ingress of the surrounding sea-water.

Like the members of the preceding families,
many Pennatulidce are liable to have their soft
structures strengthened by the deposition of spicu-
lar concretions.

The Sea-pens still further interest us by reason
of their beautiful phosphorescence. What Agassiz
has observed in Benilla is probably true of the
entire group, " It shines at night with a golden
green light of a most wonderful softness. When

ACTINOZOA. 213

excited, it flashes up more intensely, and when
suddenly immersed into alcohol, throws out the
most brilliant Ught." Experiments performed
on our own British Pennatula phosphorea led
Professor E. Forbes to the conclusion that this
species is " phosphorescent only when irritated by
touch " ; but it seems safer to infer that the light,
itself the index of an energetic display of vital
power, is only evoked in answer to proper stimuli,
which may very well be expected to occur more
appropriately in nature, though, doubtless, under
a form less clumsy, than in the exaggerated con-
ditions of an experiment. Forbes also showed
that the phosphorescence, when thus excited by
shock, sparkles onward from the portion struck in
an upward or distal direction, still, however, con-
tinuing to be emitted from the point of prime
contact. The vividness of the luminosity appears
to bear a direct ratio to the living energy of the
animal. Such are the chief conditions of its
manifestation, but the true cause of this pheno-
menon, as of vital phosphorescence in general,
still remains almost wholly unknown.

Four families of Alcyonaria may be defined:

Order ALCYONARIA.

Family i. ALCYONiDiE.

Corallmn sclerodermic, in general spicular,
without true calcareous thecse. (7ce?iosarc fixed.

Family 2. Tubiporid^.

Corallum consisting of a number of distinct
corallites, destitute of septa, their thecse united
externally by horizontal plates, arranged at
distant intervals. Coenosarc fixed.

P 3

214 ACTINOZOi.

Family 3. Pennatulid^.

Corallum sclerobasic, tissue secretions also
being sometimes present. Coenosarc free.

Family 4. Gtokgonid^.

Corallum sclerobasic, sulcate, with or with-
out additional tissue secretions. Coenosarc
shrub-like, attached by its expanded proximal
extremity.

4. Order 3 : Rugaisa. — Among the Rugosa
a highly developed sclerodermic skeleton occurs,
each corallite beiug very distinct, and presenting,
in many cases, both septa and tabulae. Some
Rugosa are simple, the corallite often attaining a
considerable size ; others are composite, increasing
either by lateral or calicular gemmation, these
two processes, but especially the latter, checking
to a greater or less extent the growth of the
primitive corallite. A true coenenchyma is ab-
sent.

In Stauria, Holocystis, and Polycoelia four of
the septa admit readily of being distinguished,
by their greater development, from the others, but
in Cyathaxonia only one primary septum or
chamber remains conspicuous. So, also, of Za-
phrentis and its more immediate allies {jig. 38).
In other genera the septa radiate, in about an
equal manner, from the inner surface of the theca.
In many Rugosa they are incomplete, that is,
" do not extend from the bottom to the top of the
corallite, in the form of uninterrupted laminae."

The tabulse exhibit various grades of develop-
ment, and, in some species, are wanting altogether.

In a few Rugosa the columella is cylindrical,
and of large size ; in others, styliform or lamellar :

ACTINOZOA.

215

often, it is wanting. In Cyathophyllum and
certain allied forms there is a spurious columella
formed by the " twisting together " of the inner
edges of the septa.

In Cystiphyllum neither columella, septa, nor
tabulae can be distinguished. The whole interior
of the corallite, save its shallow calice, is here, as

Fig. 38.

Zaphbentis cyxindrica.
(Part of a corallite, of the natural size.)

it were, broken up b}^ the numerous laminae of a
sclerenchymatous deposit resembling the vesicular
substance of its epitheca and wall.

Most Rugosa belong to the large group of
Cyathophyllidce, The remaining families include
but a few generic forms.

Order EUGOSA.

Family i. Staurid^.

Gorallum simple or composite. Septa
complete, united by lamellar dissepiments.
p 4

216 ACTINOZOA.

Family 2. CyATHAxoNiDiE.

Corallnm simple. Se]pta complete. No
dissepiments or tabula?.
Family 3. Ctathophyllid^.

Corallum simple or composite. Septa in-
complete. Tabulae, in general, present.
Family 4. Cystiphyllid^.

Corallum simple ; composed chiefly of a
vesicular mass, witb but slight traces of
septa.

5. Order 4 ; Ctcnopliora. — The leading
characters of the Ctenophora, or oceanic Acti-
nozoa, have been already, to some extent, sketched
out in the account given of Pleiirobrachia, se-
lected as the type of the group, and in the com-
parative survey which has been taken of the prin-
cipal organic systems of the present class.

The more striking modifications which appear
within the limits of the order have reference
chiefly to size, the general form of the body, the
several parts of the canal system, and the struc-
ture and arrangement of the tentacles.

The ordinary dimensions of Pleurobrachia are
by many other Ctenophora frequently exceeded, a
common British species of Bolina often attaining
a long diameter of two or three inches, while
Beroe sometimes reaches the size of a large lemon.

In Pleurobrachia alone does the form of the
body approach the spheroidal {fig. 39, e). Its
axis is somewhat lengthened in the Beroidce, so
that the animal, when seen in profile, appears
more or less ovate (c). In Cestum, on the other
hand, elongation takes place to an extraordinary
extent, at right angles to the direction of the

ACTINOZOA.

217

digestive track, a flat ribbon-shaped body, three
or four feet in length, being the result (a), Calli-

Fig. 39.

Various forms of Ctenophora : — a, Cesium Veneris ; b, Eit'
rhamphcBa vexilligera ; c, Beroe rufescens ; d, Callianira triplo-
pfcra ; e, Pleurobrachia pileics. (a is considerably reduced ; b,
slightly so ; c and e are about the natural size : the size of d is
uncertain.)

anira, a genus of which very little is known, is
remarkable for having its ctenophores elevated on

218 ACTINOZOA.

prominent wing-like appendages, and may possibly
yet prove to be but an imperfectly developed
member of the group (d).

In one large division of the order two wide
diverging lobes project from the antero-posterior
regions of the body, far beyond the level of the
mouth, which, by their approximation, they some-
times wholly conceal. Between these lobes, in
Bolina and its allies, occur on either side two
small lateral appendages, or earlets. Four of the
ctenophores, much shorter than their fellows, run
to the bases of these, along which they are con-
tinued as simple ciliated fringes. The earlets,
therefore, may be regarded as specially modified
prolongations of the lateral ctenophores.

Eurhawphoeay one of these lobed Ctenophora, is
remarkable for the general elongation of the axis
of the body, which is, moreover, much compressed
from side to side (6). The two lateral actinomeres
terminate in long tapering appendages, which
project for some distance beyond the apical extre-
mity of the animal, and then curve gently upwards
and outwards. Of a similar nature, but much
shorter and wider, are the apical appendages, or
lappets, of some Beroidcc.

The mouth of the Ctenophora varies as to size,
degree of prominence, and the relative develop-
ment of distinct lips. Even in the same indivi-
dual its form, at different periods, presents many
diversities of aspect. It attains a very large size
in Beroe and its allies, extending right across
almost the entire oral extremity, in its usual
antero-posterior direction. Hence Leuckart has
proposed to divide the Ctenophora into two sepa-
rate groups; the Eurystomata, corresponding to

ACTINOZOA. 219

the Beroidce, and the Stenostomata^ including all
remaining members of the order.

The apical area, also, presents its own peculiar
variations. In general it appears as a somewhat
oblong flattened space, bounded by a distinct
ridge, its long diameter coinciding with that of
the mouth. The ridge is usually smooth, and the
general surface of the included region covered
with very fine cilia. But in the Beroidce a num-
ber of conspicuous, though short, arborescent fila-
ments fringe the margin of the area, which, in this
family, appears divided into a pair of shallow, ovate
lobes, converging to a point at the apical pole of
the body.

The modifications of the nutrient system now
require our attention. The whole of this appa-
ratus may, for purposes of description, be resolved
into the following minor systems or groups of
parts : —

1. An axial system; including the digestive tube,

the funnel, and the apical canals.

2. A paraxial system. To this belong the para-

gastric canals arising from the funnel, and an
oral vessel, or vessels, into which, in some
genera, their distal extremities open.

3. A ctenophoral system ; the eight canals of

which may be variously prolonged to supply
the lobes or other appendages of the body ;
and —

4. A radial system ; or those channels whereby

the several elements of the preceding system
become connected with the funnel.

I. The parts of the axial system vary but little
throughout the order. In Pleurohrachia the di-

220 ACTINOZOA.

gestive sac is relatively shorter than in other
Ctenophora. In the Beroidce the funnel is very
short and wide. Not more than one pair of apical
canals ever appears to have been noticed.

2. Of paragastric canals, one or two pairs may
be present. The single pair of these canals are
csecal in Pleurobrachia ; in Beroe they open into
an oral tube. In most of the lobed Ctenophora
two pairs of paragastric canals occur, of which
one lies close to the sides of the digestive sac, the
position of the second pair being more superficial.
The inner pair may be csecal, while the outer pair
open into the oral tube ; or the latter may anas-
tomose with both pairs of paragastric canals, as
described by Milne Edwards in Ckiajea Paler-
mitana.

The oral vessel may be circular, as in the
Beroidce, or consist of two straight canals run-
ning by the sides of the mouth, as in many of
the lobed Ctenophora; though in one of these,
Eurhamphoea, this vessel is complete.

3. The eight ctenophoral canals may conve-
niently be divided into four lateral, and four
antero-posterior.

The oral extremities of all these canals may be
caecal, as in Pleurobrachia, or open into the cir-
cular tube surrounding the mouth, as in the
Beroidce. Among the lobed Ctenophora, the
same tube, or the two vessels its representatives,
receives only the extremities of the four lateral
canals. Thus, in Bolina, each of the latter, after
supplying the course of its comparatively short
ctenophore, winds round the edge of one of the
smaller lobes or earlets, and having again reached

ACTINOZOA. 221

the general surface of the body, soon joins the
extremity of the oral tube of that side, before
gaining wliich it sends off a branch, destined,
after having pursued a long and devious track, to
anastomose with its fellow from the opposite side
of the body, and the complex system formed by
the much convoluted and prolonged extremities of
the intervening pair of antero-posterior canals.
These last, after quitting their ctenophores, which
are relatively much longer than those of the sides
of the body, are produced to supply the two large
lobes, in front of and behind the mouth, within
the substance of which, after many sinuous turns,
they finally blend and become lost.

The course of the ctenophoral canals is usually
simple. In Chiajea Palermitana each gives off
on either side a number of very short straight
branches, which in the Beroidce are replaced by
somew^hat larger, arborescent tufts.

4. Lastly, the apical extremities of the cteno-
phoral canals, and the manner of their communi-
cation with the axial system, or, in other words,
the disposition of the radial vessels, remain to be
noticed.

In PleuTobrachia, as has been shown, the
primary, secondary, and tertiary radial canals
are very well marked, meeting the cteuophoral
vessels at right angles to, and about midway in,
their course. The apical extremities of the latter
are distinctly prolonged, to end csecally around the
oblong, flattened area.

In the Beroidce a radial system can scarcely
be said to occur, the apical ends of the cteno-
phoral canals curving gently round, to open into

222 ACTINOZOA.

the funnel, which is placed very close to its own
extremity of the body.

Arrangements intermediate between the two
extreme conditions just noticed may be traced
among the several genera of the lobed Cteno-
phora.

Thus, in Le Sueuria, the apical extremities of
the eight canals curve inwards to empty them-
selves into the four short radiating vessels which
issue from the funnel. Somewhat similar is the
radial system in Bolina and, perhaps, in some
species of Chiajea. But in (7. Palermitana,
while the four longer or antero-posterior canals
comport themselves much as in Le Sueuria, each
lateral canal, near the middle of its course, is
connected by a short transverse branch with the
extremity of one of the four short radial vessels,
just where it receives the curved inward prolon-
gation of the adjacent antero-posterior vessel. In
this form the apical extremities of the shorter
ctenophoral canals are csecal, but in Eurhamphwa,
w^hose radial system resembles that just noticed,
the lateral canals of each side open at an acute
angle into one another, at the bases of the two
curved appendages of the apical lobes peculiar to
this genus.

Comparing the preceding account with the little
yet known of the development of the nutrient
apparatus in the Ctenophora, the following con-
clusions seem deducible. The first formed part
of this apparatus seems to be that large rudiment
of its axial system from which, at an early period,
the digestive sac and funnel become differentiated.
From the funnel, or central portion of the whole
nutrient cavity, the apical canals soon branch off".

ACTINOZOA, 223

Thus the axial system is completed. Next, radial
and paragastric canals appear, the former quickly
reaching the surface of the body, where they
branch and give origin to the ctenophoral canals,
the apical ends of which are well developed, while
their .opposite extremities are still on their way
towards the oral region, in striving to gain which
they are outstripped by the paragastric vessels.
The extremities of these either remain caecal
{Pleurohrachia)f or, by branching, give origin to
the two lateral rudiments of the oral canal.
These may still continue distinct {Bolina, Chia-
jea, Le Sueuria), becoming connected with the
extremities of the lateral ctenophoral canals ; or
unite to form a complete circular tube, which
receives, as before, only the four lateral canals
(Eurhamphoea), or both these and the vessels of
the antero-posterior ctenophores {Beroe). Thus,
the oral tube bears to the paragastric canals a
relation comparable, perhaps, with that between
the ctenophoral vessels and the radial system : and
the gradual development of the entire canal sys-
tem may be described as tending in a peripheral
direction, while its several elements bifurcate;
the branches so formed, to a greater or less ex-
tent, again uniting, and prolonging their course
beneath the surface of the body.

In the above survey the canal systems of Cestum
and Callianiva have not been included. Of the
latter nothing whatever is known. In Cestum the
axial system resembles that of Bolina and the
Ctenophora in general. Each half of the cteno-
phoral system is represented by four very long
canals, two of which run side by side along one of
the fringed margins of the ribbon-shaped body,

224 ACTINOZOA.

the other pair, parallel to these, being situate
midway between them and the opposite, or oral,
margin. For, in this strangely aberrant form, the
course of the axial system corresponds with the
short mid-axis of the ribbon, the apparent width
of which represents, therefore, the true height of
the animal, whose breadth answers to the length
of the ribbon, and antero-posterior diameter to its
thickness. On this account, it will be convenient
to speak of two of the ctenophoral canals as mar-
ginal, the two others being medial. There are
four radial vessels, two on either side, each of
which divides into a single pair of branches, com-
municating with the ctenophoral canals. The two
branches of each radial canal are very unequal in
length, and run in opposite directions, the shorter
branch soon becoming continuous with a marginal
canal, while the longer branch trends parallel to
the sides of the digestive cavity, turning round
abruptly to open at right angles into one of -the
medial canals, as soon as it has reached its level.
At each extremity of the ribbon the marginal and
medial canals anastomose with one another and a
long vessel running parallel to their course along
the oral margin of the body. At its opposite
extremity, near the mouth, this canal unites with
its fellow of the other side, where both are joined
by a pair of paragastric canals, which Milne
Edwards has figured as running in an antero-
posterior direction, parallel to the digestive sac.
A second, or inner, pair of these canals has not
yet been observed.

There can be little doubt that the paired medial
and marginal canals of Cestum represent the eight
ctenophoral vessels of other genera of the order.

ACTINOZOA. 225

and that each half of the long unpaired marginal
canal is homologous mtli one of the lateral oral
vessels in such genera as Bolina or LeSueiiria.

Two principal kinds of tentacles occur in the
Ctenophora: long, highly contractile cords, capable
of being retracted into special pits ; and shorter,
isolated threads, which may, in some species,
become ago-reo-ated to form tufts or bunches.
Among the Beroidce, tentacles are absent. In
Pleurobrachia a large tentacular pit excavates
obliquely upwards the substance of the two lateral
actinomeres. The base of this pit is brought into
close connection with the distal extremity of the
short primary radial canal, which opens directly
into a wide heart-shaped sac, from between the two
dee^Dly-cleft lobes of which, at the upper portion of
the fissure formed by their junction, the tentacle
itself makes its appearance. Proximall}^, it is some-
what compressed, but for the greater part of its
length becomes truly cylindrical, giving off on that
side which is turned away from the body a number
of secondary lateral filaments. Both these and
the tentacle itself are hollow, communicating
with the canal system through the medium of
the basal sac, their Avails also, like those of the
body canals, being lined by an investment of
endoderm. On the secondary branches them-
selves still more minute threads are said to have
been observed. Of the grace and beauty which
the entire apparatus presents in the living animal,
or the marvellous ease and rapidity with which it
can be alternately contracted, extended, and bent
at an infinite variety of angles, no verbal descrip-
tion can sufficiently treat. These movements seem
partly caused by the action of the contractile fibres
Q

226 ACTINOZOA.

whicli occur in the tentacular walls, and partly by
the distensive pressure of the fluid forced into
the interior of the tentacle, by means of the
elastic basal sac.

An account somewhat different to the above has
recently been given by Professor Agassiz, both of
the precise structure of the tentacle itself, and the
mode of its connection with the canal-system.
" Nearly two-thirds of the length and breadth of
the proximate side of the actinal or closed end of
the tentacular socket is occupied by an oblong
disc, from the mid-length of which the tentacle
arises. The distal side of the disc, or that which
faces towards the periphery of the body, is convex,
with a shallow furrow extending from the base of
the tentacle to the actinal end of the disc ; and
the proximate side, or that which faces towards
the axis of the body, is a plane, immediately
beneath whose surface and next to the edge, two
chymiferous tubes run parallel ; leaving between
them, along the median line, a thick ridge, which
is nearly as broad as the diameter of the tubes."
— "Only imagine the socket to be removed or
reverted, as oftentimes does happen in a great
measure, and the whole apparatus will appear like
a peripheric ridge, which, at one point, is drawn
out into a slender thread, the tentacle. The base
of the tentacle has the form of a high, narrow
ridge or keel, more or less plicated or distorted,
according to whether the apparatus is extended or
retracted ; but we have never seen it projecting
beyond the aperture of the socket. At the basal
end of the keel it is as broad as the disc from
which it arises, but it suddenly narrows to a uniform
thickness, which it retains to the other end, where

ACTINOZOA. 227

it merges into the cylindrical part of the tentacle."
Professor Agassiz describes the latter not as hollow
but solid, though he recognizes the two layers of
which it is composed.

In Calllanira two long tentacles, relatively
situate as in Pleurobrachia, but destitute of
lateral threads, divide, at their free extremities,
into two or three lengthened branches (fig. 39, d).

In Centum, also, a pair of symmetrical tentacles
appear to be usually present, but these do not, as
in the preceding forms, issue from the equatorial
region, thence turning away from the mouth; but,
rather, take their position in front of and behind
the latter, towards which they are seen to incline.
Milne Edwards figures the tentacles of C. Veneris
as simple ; by other writers, and perhaps in other
species, they are described as variously branched.

Among the lobed Cteiwphora the particular
homologies of the tentacular apparatus have
hitherto been by no means sufficiently studied.
In Chiajea occur two tentacles, one on each side
of the body, but similar in other respects to those
of Cestum, Tentacles of the second type are,
however, more frequently to be met with in this
section of the order, and these may be either
lateral, and arranged in groups, as in most genera,
or disposed in a ring round the mouth, as in
E arhamphoea.

In Bolina a tuft or brush of very short tentacles
is seen to arise on either side of the mouth, where
the oral vessels appear to meet, about midway in
their course, the superficial paragastric canals.
The extremity of each of these canals ends in a
simple socket, within which the tuft of tentacles
may be withdrawn. But there is no proper sac,

228 ACTINOZOA.

as in Pleurohrachia. Agassiz states that, in tliis
genus, only the innermost pair of paragastrio
canals open into the oral vessels, the outer pair,
notwithstanding their close approximation to the
sides of the mouth, being destined solely for the
supply of the tentacular bulbs.

In LeSueuria two tufts of tentacles, similar in
position to those of Bolina, but prolonged to form
a pair of lateral fringes, were first observed by
Milne Edwards, and in addition four simple conical
appendages, of moderate length, arise, one between
each of the four pairs of smaller lobes character-
istic of this genus. The outer paragastrio canals
are seen distinctly to open into the canals repre-
senting the oral vessels, but Milne Edwards does
not notice the existence of any communication
between them and the lateral tentacular fringes,
which are, perhaps, nothing more than filamentous
extensions of the ectoderm.

LeSueuria, according to the same writer, is
frrnished, however, with a pair ofcurious appen-
dages by means of which the oral extremities of
the paraxial canal system communicate directly
with the exterior. Each appendage is cylindrical,
short, and tubular, arising from the midst of one
of the principal tentacular tufts, and terminating
distally in four small lobes, surrounding the orifice
of the canal, which seems to perforate its axis.
Agassiz denies the existence of this opening, and
considers the two appendages homologous with
the tentacular bulbs of Bolina,

In Leucofhea a very complicated tentacular
apparatus occurs ; short threads like those of
Eurhamphoea and LeSueuria being here present,
in addition to three long tentacular organs, arising

ACTINOZOA. 229

on eacli side of the mouth. Of these, one is
simple, while the others display a number or
lateral ramifications. But all these structures
require to be investigated anew.

In habit the Gtenophora resemble the oceanic
Hydrozoa, like them swimming near the surface
in calm weather, and again descending on the
approach of showers. Like them, also, their deli-
cate structures rapidly disappear when removed
from the sea-water and exposed to the rays of the
sun, an almost imperceptible film remaining the
only trace of what was erewhile an active and
beautiful organism. Yet are the Cteiiophora very
voracious, feeding on a number of floating marine
animals, among which their own kindred seem
especially to be preferred. The prey, once swal-
lowed, is assimilated with a rapidity which to
some may seem strange, when the simple structure
of the digestive apparatus is considered. All the
Gtenophora are not equally fragile. Pleuro-
brachia, in spite of its tender gelatinous aspect,
may be preserved in captivity for weeks, or even
nionthB, if properly supplied svith. food.

The Gtenophora swim in various positions, and
some may often be noticed with their apical ex-
tremity turned downwards or forwards. Hence
many writers term this the dorsal aspect ; the
digestive sac, by a strange perversion of language,
being described as situate below the funnel ; and
so with the relative positions of the remaining
organs. This practice is not only objectionable in
itself, but has tended much to confuse almost
every published account of the structure of a
group of beings, than which few anatomical sub-
jects are at once so easy and so accessible. Not
Q 3

230 Acnsxaok,

in leajniiig. bat in imleaniii^ is t i t = - r. "
the Cf^nopkora compelkd to waste -_ii :_r ;^^
iiig|€iiiilLv.

Some ClfMop^byni sre pho^ihoresceiit. In a
ipedes q£ Boiima eanuiMiii around oar dioies this
beaiilifal ptopeity may Teiy i^eadibr be observed.
^peoally dBsdngm^ied for its luminodty is the
much honest GesiHm, Vemms of the Mediterraneaii.
"^i::!: i=^ ■=^-."^ *" '^leam. at nis^t liir- ? "^'"^9.1'

L : 1 _ _;Tii^benc^:i :ir : i^

Mlowsi—

-. r'^rryri'DTTr.'P ^^

f annhr i. Cj.i~

Faziiir 3. r^

1

I

AcnyozoA- 231

lobes, Tentades two in number, lateral,

turned from the mouth.

Sub-order 2. Eurystamata,

Family 5. Besoid^

J?'>iv ovaL elongated, without oral lobes.
Tentadts al:»>ent.

Here ^e Lave slightly modifei :if "t-_- 15
:f G-egenbaur, at the same ri^r ii- : _ 1 :

appears to be the most iianir.il s-r .7_ t : :_t
several femilies. The gronp OiUi :_::--

f :r the present be c>jusiderai as merelj j : — , _ _ .
Tne four other divisions of C'.erv'r^-:'^? 'L:.~r ':rr-
:r?ently elevated by Agassiz t. " - . : - -

: rdersj and the entire numc«er 0: :_: _ . . t v .

to ten. This arrangement, however. rT-r-ir _
advantaore over the more sini^lf :.:: " -^ - : 1 ; ..r
a^i opted by G-egenbanr. wLiL. i_ . . . i_.:^-

be regarded as an iniprc"r ; :.._-...:.:_ .: -Jif
prior elassidcation of Eschs:^ .::.

Si-m-K IT.

I Relations^ to Phy^ieal Elements. — All

the A:::-.::- :ire niAriLf,

2. Bathyxuetrical Disitribotion — I'r.n-he
whole it Diay be sci:l -:. ;: -le .^ ^rr ^vr-

abimdant between tide-iiiirks. -u:.i :-j-ciir mdee^ier

232 ACTINOZOA.

waters, than the Zoantharia. Alcyonium has been
met with at seventy fathoms, but, like Pennatula,
is common in much shallower seas. From so
great a depth as 240 fathoms a species of Virgu-
laria, V. Fhimarchica, was dredged at Oxfjord
by M. Sars, who also obtained, in the same locality,
the widely different Briavemn grandiflorum, a
low creeping Alcyonid, allied to Sarcodictyon.
The Gorgonidce, in like manner, seem to prefer
deep seas, Corallmm having been found at 120,
and Gorgonia itself at nearly 200 fathoms.

Though depths equal to or even exceeding those
just mentioned have yielded many species of
Zoantharia, Ulocyathus, for example, frequenting
water of 200 fathoms, yet, in general, the members
of this order are most abundant in seas of not
more than 50 to 100 fathoms deep. The Actinidai
and Madreporidw include those species which are
most prone to descend below this limit. j\Iany
of the Actinidw, it is well known, are numerous
between tide-marks, the common Sea-anemone
tending to encroach upon the line of high water.

The shallow vertical rans^e of the reef-buildino:
Actmozoa has already been sufficiently explained.
Certain species are chiefly restricted to particular
parts of the reef; Astvoeidce and Seriatoporidce
choosing its more submerged portions, below the
outer exposed edge, upon which Porites and its
allies flourish. On the surface of the reef both
Astroeidw and Fungidw may readily be distin-
guished, the lab3ri'inthic form of Moeandrina,
among other genera, being here especially con-
spicuous.

The soft-bodied non-adherent Zoantharia
usually occur on muddy or sandy banks, at or

ACTINOZOA. 233

near the level of low water. A few appear to be
oceanic. Phllomedusa, a minute form, from the
Brazilian seas, habitually seeks shelter beneath
the swimmiuQ-oro^an of various Medusidce and
Lucemarlda'.

The bathymetrical distribution of the Cteno-
phora, by reason of their oceanic habit, is scarcely
amenable to observation. Some species, during
the storms of winter, appear to seek considerable
depths, on the return of spring again approaching
to the surface.

3. Crcograiilascal Uistrilmtiois. — The Cfe-
nopliova, Alcyonaria, and soft-bodied Zoantharia
appear to be about equally abundant in tropical
and temperate seas, many forms extending their
rano^e to bio-h latitudes. Of corallio-enous Zoan-
tharla two families, Tiivhlnolidce and Madrejjo-
ridce, are not without northern and even arctic
representatives, yet by far the majority of other
sclerodermic species are seldom found to occur
be3"ond the limits of the tropics. The reef-build-
ing Corals, according to Dana, will not flourish in
water wherein the mean winter temjDerature is
lower than 66° F. So that on either side of the
equator a zone of water sufficiently heated for the
gro^\i:h of these Corals extends, the boundary lines
of which have of necessity a somewhat contorted,
irregular course, by reason of the varied com-
binations of circumstances influencing the local
distribution of heat. Even within these limits
other external conditions, not less essential than
a high temperature to the welfare of reef-build-
ing Corals, are often absent. But when once the
nature of these conditions has been carefully

234 ACTINOZOA.

understood, the many apparent anomalies in the
distribution of Coral-reefs, far from being, as
some have stated, unaccountable, become in each
case susceptible of their appropriate physical
explanation.

In the British seas about ten species of sclero-
dermic Zoantliaria occur. The number of Medi-
terranean Corals is much greater, though these,
with few exceptions, are specifically distinct from
those observed by Ehrenberg in the Eed Sea.
The Mediterranean also yields two or three forms
of sclerobasic Zoantliaria, a group apparently
unknown in more northern seas. Corallium
rubrum, the Eed Coral of commerce, would seem
to be restricted to the same region, though other
species of its genus have from time to time been
dredged off Madeira and the Sandwich Isles.

Of Actinozoa, which occur beyond the limits of
the Mediterranean and North Atlantic Seas, our
knowledge still remains very imperfect, save only
in the case of the reef-building Corals and the more
conspicuous forms of Ctenophora. The genera
Cestum, Callianira, Galyminnaf Chlajeaf and Leu-
cothea may be cited as examples of this order cha-
racteristic of the tropical and warmer temperate
zones. Ocyroe, an obscure but interesting Cteno-
phorid, distinguished by the possession of two
antero-posterior lobes, prolonged outwards at right
angles to the true axis of the body, and which,
when better known, may prove to be the imma-
ture condition of some apparently dissimilar form,
has a range not wider than the equatorial regions
of the Atlantic.

In high latitudes several Actinidce, a few Tiir-
binolidcB and Madreporidw, together with various

ACTINOZOA. 235

Alcyonaria and Ctenophora, of which one genus,
Mertensia, is said to be exclusively arctic, chiefly
represent the class. The Pennatulidce appear
more numerous than other Alcyonaria around
the northern colder temperate shores, seven spe-
cies being named in the Norwegian fauna of Sars,
while but three have yet been recorded as British.
Umbel ltd aria, a very aberrant member of this
family, which presents a rod-like coenosarc six feet
in length, crowned with a spreading tuft of polypes
at its summit, is only known from the published
descriptions of two specimens, dredged from a
depth of 236 fathoms, off the coast of Greenland,
about the middle of the last century.

Among genera of Actinozoa which enjoy a
wide distribution. Actinia, Alcyonium, Zoanthus,
and Gorgonia are perhaps best worthy of men-
tion. To this list should be added the names of
several forms of Ctenophora, than which few ma-
rine animals appeared so well adapted to thrive
under every variety of climatal conditions. Two
genera in particular, Beroe and Pleurobrachia,
are remarkable for their unbounded geographical
area.

With less confidence can the names of such
Actinozoa as are restricted in their range be, at
present, insisted on. Eenewed observations show
that the number of extra-tropical genera, once
thought to be peculiar to certain regions, must
undergo considerable diminution. Of specific
forms, however, not a few seem to characterise the
various seas and shores to which they are confined.

The existence of natural barriers, whether of
land or deep water, exercises a marked influence
on the distribution of the Actinozoa. The differ-

236 ACTINOZOA.

ences between the East and "West Indian species
of Corals, or between the several Atlantic and
Pacific forms of the class, often curiously resem-
bling one another under similar conditions of
depth and temperature, but, in a large number of
cases, specifically distinct, may thus be easily
accounted for. Many genera of fixed Actinozoa,
abundant in one hemisphere, are found wholly
wanting in the other. To a less extent is this
observation true of the soft-bodied or free-swim-
ming species.

Section V.

RELATIONS OF ACTINOZOA TO TIME.

1. General History of Actinozoa. — 2. Historj' of Zoantharia. — 3.
History of Kugosa. — 4. History of Alcyonaria. — 5, Silurian
Corals. — 6. Devonian Corals. — 7. Carboniferous Corals. — 8. Per-
mian Corals. — 9. Triassic Corals. — 10. Jurassic Corals. — 11.
Cretaceous Corals. — 12. Tertiary Corals. — 13. Recent Actinozoa.

I, Greneral IS istory of Actinozoa. — Acti-
nozoa appear to have been numerous during each
of the greater artificial geologic epochs. The hard
parts of the coralligenous species only have been
preserved. Hence the expressions "fossil Corals"
and " fossil Actinozoa " may be used as syno-
nymous.

One order, Ctenophora, has no fossil represen-
tatives. The Rtigosa, on the other hand, are
wholly extinct.

The accompanying table exhibits, from a general
point of view, the relations to time of the prin-
cipal groups of Actinozoa. Lists are appended of
those orenera of Corals which rano,e through more
than one geological period.

ACTIXOZOA. 237

Chkonologicax Areangemext of ACTINOZOA.

Names of Groups.

Names of Periods.

c

3
O

u

o

CO

>>

c

«

•S

.2

1

1

0.

•

2

3

!

1

c

i

Actinozoa

Zoautharia

—

—

—

—

—

Malacodermata

Sclerodermata

—

—

Aporosa

1 —

—

—

Turbinolidae

Dasmidffi

Oculinid?e

Stylophoridae

—

—

Astrseidce .

—

Echinnporidge

Merulinidie

Fungidge .

—

__

—

—

Perforata .

Madreporidse

—

—

—

Poritidae .

—

—

—

—

Tubulosa .

. ?

—

Auloporidge

?

—

—

Tabulata .

—

—

—

—

Thecidse .

Seriatoporidse

—

—

—

FaTositidffi

—

—

—

—

—

INIilleporidse

—

—

—

—

_.

—

—

—

Sclerobasica .

Antipathidffi

—

Eugosa

—

Cystiphyllidfe

—

_

Cyathophyllida

' —

—

—

Cyathaxouidai

—

—

Stavu-idfe .

—

Alcyonaria .

?

Alcyonidffi

II

Tubiporidfe

Ponnatulidse .

9

__

Gorgonidse

i ?

Ctenophora

!

—

238

ACTINOZOA.

PALEOZOIC COE.ALS,

^V^HICH OCCUE IN MORE THAJf OKE GeOLOGICAI- PeEIOD.

,

Names of Periods.

Names of Genera arranged in order
of their appearance.

1

J

1

i

3

0.

'S

1

Cystiphyllum

—

Syringophyllura
Eridophylium
Ptycliophyllura
Aulacophyllum
Heliolites .

—

—

Psammopora

Aulopora

Cyathaxonia

?

—

Clisiophyllum
Propora
Cyathophyllum
Zaphrentis .

—

—

—

Syringopora .,
Favosites

z

—

—

Emmonsia ,

—

Alveolites .

__

Ch»tetes

—

Philippsastrea
Lithostrotion

—

—

Campophyllum
Amplexus .
Lophophyllum
Beaumontia .

—

—

Miehelinia .

Fistulipora .

—

—

ACTINOZOA.

239

MESOZOIC, CAINOZOIC, and RECENT CORALS,

"WHICH OCCXJK IN MORE THAN ONE GEOLOGICAL PeRIOD.

1

Names of Periods.

Names of Genera

o

.

S^

arranged in order of their

"s

s

i

cS

c

appearance.

.2

3

►-5

i

I

Hyboeoenia . .

_^

__

G-oniocora .

Rbabdophvllia

—

:

CladophyUia

—

—

—

Isastrea .

—

Montlivaultia

—

—

Euiymeandra

—

—

Thamnastrea .

__

Enallohselia

Coelosmilia

__

Pachygyra

—

Rhipidopfyra

Stylosmilia

Stylina

Cyathopliora

Calamophyllia

—

Haplophyllia

Adelastrsea

Pleiiroccenia

___

Trochocyathus

—

Trochosmilia

—

Astrocoenia

—

Stephanocoenia

—

Thecosmilia

__

Oroseris .

—

Mseandrina

_

__

Favia

^

—

—

Heliastrpea

—

Ulophyllia .

— .

—

—

Pleras'traea .

__

— .

_

Millepora ,

—

—

240

ACTIXOZOA.

MESOZOIC, CAINOZOIC, and EECENT COEALS, ^^hich
OCCUR IN MOEE THAN ont; Geoi-ogical Period — Continued.

1
Names of Periods,

Names of Genera

C ' "^

arranged in order of their

•i

"Z

o

b

*i

appearance.

<«

£

o

.2

c

.2

x.

u

3

H

»-5

O

H

Barysmilia

Stylocsenia

—

Centrocsenia

— .

Phyllocsenia

Brachyphyllia

Ehizangia .

Cyclolithus

Stephanopliyllia

Eatliycyathus

Lopliosmilia

—

_™

Diploria .

.

Leptoria .

Groniastrsea

C}-phastrgea

—

—

Pavonaria .

Caryophyllia

Mycetophyllia

*

Hydnophora

.

Cladocora .

,

Cycloseris .

Corallium .

. ,

Isis .

,

Acanthocyathus

Paracyathus

Sphenotrochus

Desmophylluni

Oculina

Lophoha-lia

Stylophora

___

Euphyllia .

__

Galaxea .

—

Lithophyllia

___

Dasyphyllia

—

—

ACTINOZOA.

241

MESOZOIC, CAINOZOIC, and EECENT CORALS, avhich
OCCUR IN MORE THAN ONE GEOLOGICAL Peeiod — Continued.

Names of Periods.

Names of Genera

arranged in order of their

6

u

3

£•

*:

appearance.

S

S

s

.2

c

1

1

1

Sjmpliyllia

_

_

Plesiastrpea

_

Solenastrsea

__

Astrsea

__

Prionastrgea

__

__

Astrangia .

—

—

Phyllangia

—

—

Eupsamnia

—

Endopachys

—

_

Ealanopliyllia

—

Dendrophyllia

—

—

Maclrepora

—

—

Turbinaria

Astraeopora

—

Porites

—

—

Rhodarsea .

Yirgularia

Mopsea

—

—

Hyalopathes

—

—

Choetetes passes up into the Trias. With this
exception no genus of Corals survives the Paleozoic
epoch except, perhaps, lisis, of which doubtful in-
dications have been met with in rocks of very-
ancient date.

No Triassic genus of Corals has recent represen-
tatives. Of genera which occur in the Jurassic
series seven still survive. Fourteen recent genera

R

242 ACTINOZOA.

first appear in the Chalk, while very many are
common to the Tertiary and Eecent periods. But
few Recent species of Corals occur in a fossil state.
It appears also from the preceding tables that
six genera of Corals range through four periods,
thirty-four through three, and sixty-eight through
two. Some genera, however, arise in one forma-
tion, are apparently absent from the next, but again
present themselves at a subsequent period. Of
this seeming anomaly Mlllepora furnishes an ex-
ample. Such instances must always be received
with suspicion, since they are probably due to
defective observation.

2. History of Zoantfiaria. —

All extinct Zoantharia belong to the group of
Sclei'odeimata, with the exception of a few slight
indications of Antipathidce which appear in the
Tertiary period. The Malacodemicda are wholl}'
recent. On the other hand, the small group of
Tuhulosa does not survive the Paleozoic epoch.
But two families of Zoantharia, Thecidce and
Aidoporidce, have altogether disappeared. On
the whole it may be said that Tabulata prevail in
the Paleozoic deposits, Aporosa and Perforata in
those which succeed. Tabulata are comparatively
scarce in strata anterior to the Carboniferous,
though no geological period is without some
representative of this division, and in modern
seas four genera have been observed. A single
genus, Pakeocijdiis, which occurs in the Silurian
period, is the only known representative of Aporosa
in strata older than the Trias. The Perforata are
represented in the Paleozoic rocks by two genera,
Ivut, excepting these, no other forms of the group

ACTINOZOA, 243

have been met with in deposits of earlier date
than the Jurassic.

3. History of Rugosa. —

All Ru<josa are confined to the Paleozoic epoch,
with the exception of the genus Holocystis, which
is peculiar to the Lower Grreensand, where it is
represented by a single species, H, elegans.

The Rugosa first appear in the Lower Silurian.
They are especially abundant in the Upper Silu-
rian, Devonian, and Carboniferous deposits. In
the Permian rocks they are represented by only
one generic form.

4. History of Alcyonaria

Few genera of Alcyonaria have hitherto been
found in a fossil condition, and scarcely three of
these are wholly extinct. The existence of this
order during the Paleozoic epoch must be regarded
as doubtful, though the genus Protovirgidaria
has been constituted for the reception of a Silurian
fossil, supposed to belong to the family Peniiatu-
lidce. The genus Isis, also, has been stated to occur
in some of the Paleozoic formations. With these
exceptions no Alcyonaria have been found in
rocks more ancient than the Chalk. The family
Alcyoniclce is without an extinct representative.

5. Silurian Corals. —

The Silurian Corals consist chiefly of Rugosa
and Tabulata. The Aporosa are represented by
the geniisPalceocyclus; the Perforata hj Protarcea,
Doubtful indications of Tubulosa and Alcyonaria
also occur. At least nine families of Corals first
make their appearance in this period, and one,

B 2

244

ACTINOZOA.

ThecidcB, does not survive it. The following genera
are peculiar to the Silurian series :

FUNGIDJE.

Gonioiiihyllum.

Falceocyclus.

Strombudcs.

PORITID^.

Favositid^.

FrotarcBa.

C<£nit(S.

Thecidje.

Halysitcs.

Thicia.

Fletscheria.

Column aria.

Danaia.

MlLLEPORIDJE.

Dckaia.

Lydlia.

Labecheia.

Cyathophyllidjb.

Constdlaria.

Streptelasma.

Stauridae.

Omjphyma.

Stauria.

6. Devonian Corals. —

Excepting the genus Aulo'pora, and the am-
biguous form Pleurodictyurrif the Devonian Corals
consist wholly of Rugosa and Tabulata, One
family, Seriatojporidce, first makes its appearance
in this formation, and one, Cystiphyllidce, does
not survive it. The following genera are exclusively
Devonian :

PORITID^.

Pkurodictyum.

AULOPORID^.

Aulopora.

SERIATOPORIDiE.

Dc7idropora.
Trachypora.
Favositidje.
Thccostegites.
Chonostcgitcs.
Roemrria.

MlELEPORIDJE.

Battersbyia.

Cyathophyllidje.

Smitkia.

Spongophyllum.

Acervidaria.

Endophylhon.

PachyphyUum.

Heliophyllinn.

Chonophyllum.

Anisophyllum.

Baryphyllum.

Hadro'phyllum.

Hallia.

CoiubopTiyllum.
Stattrid^.

Metriophyllum.

of

Car]5>OBiiferoiis Corals. —

In addition to the genus Pyrgia, the Coral fauna
the Carboniferous rocks seems to be wholly

ACTINOZOA. 245

made up of Rurjosa and Tabulata. Three families,
Auloporidce, Cijathophyllidce, and Cyathaxonidce,
do not outlive this period. The following genera
are restricted to the Carboniferous deposits :

AUXOPORID-B.

Seriatopoeid^.
Bhabdopora.

CYATHOPHYLLIDiE.

Axophyllum,

Cyathophyllid^ ;
Lonsdaleia.
Stylaxis,
Chonaxis.
Aidophyllum,
Menophyllum.
Trochophyllum.

8. Persnian Corals. —

The few Permian Corals hitherto found belong
to the Rugosa and Tabulata. The genus Poly-
coelia, of the family Stauridce, is peculiar to this
period.

9. Triassic Corals

Fossil remains of Corals are scarce in the Trias.
The family Astrceidce, so abundantly represented
in all subsequent formations, now first makes its
appearance. To this group most of the Triassic
Corals have been referred. The Favositidce are
represented by the old genus Choetetes. It can
scarcely be said that any genera of Corals are
characteristic of this formation.

10. tFurassic Corals. —

There are no Rugosa in Jurassic rocks, and
Millepora, a recent genus, is the sole representa-
tive of the Tabulata, The greater number of
Jurassic Corals belong to the Aporosa, and cer-
tain beds of this series have received the name of
Coral-Eag from the great abundance of Astrceidce
which they contain. The genera Stylina and
B 3

246

ACTINOZOA.

MontlivauUia are especially rich in species.
Two generic forms represent the Perforata.
The families Titrbinolidce and Oculinidce now
appear for the first time. The following genera
are exclusively Jurassic :

TuRBINOLIDvE.

Biscocyatlnis,
Thecocyathus.

OCULINID^.

Euhdia.

ASTR^ID^.

Axos7mIia.

Haplosmilia.

Pkytogyra.

ASTR^ID^ :

PlacosphyUia.
AngeastrcBa.

Protoseris.
Comoseris.

POEITIDJE.

Microsolena.
Anomophyllum.

II. Cretaceous Corals. —

The Corals of the Chalk are very numerous,
belonging chiefly to the Ajporosa and Perforata.
Here also undoubted indications of Alcyonaria
present themselves. The Tahulata are repre-
sented by two genera. For the last time the
order Pugosa makes its appearance, a single
genus, Holocystis, being its representative. The
families MadreporidcB, Pennatulidce (?), and Gor-
f/onidoi (?) now first appear. The following genera
are peculiar to this period :

TURBINOLIDJE.

Brachycyathtts
Cyclocyathus.
Stylocyathus.
Smilotrochus.

OCULINID^.

Synhclia.
Baryhclia.

ASTR/EID^.

Placosmilia.
Dipfoctenium.
ParasDiilia.
Pej)losmilia.

AsruMTDM :
Holocmnia.
Acanthoccenia.
Placocoenia.
Elas7noc(xnia.
Pentacoenia.
Heterocoenia.
Leptophyllia.
Bactylosmilia.
Hymenophyllia.
Aspidiscus.
Stelloria.
Meandrastrcea.

ACTINOZOA

AsTR^rDiE :

Madreporid^.

DimorphastrcBa.

Actinacis.

Fleurocora.

Faa'ositid^.

FUXGID^.

Koninckia.

Micrabacia.

MlLLEPORIDiE.

Anabacia.

Polytremads.

Genabada.

SxAURID.li.

Holocystis.

247

12. Tertiary Corals. —

The Tertiary formations are abundantly sup-
plied with Corals, chiefly belonging to the Ajpo-
rosa and Perforata. The Tabulata are repre-
sented by a single genus. There are distinct
traces of Alcyonaria. The Sclerobasic Zoan-
tharia now first present themselves. Here, too,
appear for the first time the Dasniidce and Stylo-
phoridw, whose claim to the rank of distinct
families is somewhat doubtful. The Dasniidce do
not survive the period. The following genera are
restricted to the Tertiary deposits ;

TUBBINOLID^.

Conocyathiis.
Dcltocyathus.
Leptocyathus.
Ecmesus.
Turbmolia ?
Piatytrochus.
Ceratotrochiis.
Discotrochus.
Dasmid.b.
Dasmia.

OCUHNID^.

Diplohdia.

Astrohelia.

Stylophorid^.

Aroeacis.

AsTRiEID^.

Cydosmilia.
JJendrosmilia,

Haplocoenia.

Circophyllia.

Tichastrcea.

Mctastrma.

Cryptangia.

Cladangia.

FUNGID^.

Trochoseris.
Cyathoseris.
Madreporidje.
Lobopsamnia.
Stcreopsa vmia,
Dendrads.

PORTTID^.

Litharoea.

MlLLEPORID^.

Axopora.
Pennatulid^.
Graphularia.

&4

248 ACTINOZOA.

13. Recent Actinozoa. —

Except the Rugosa, Tubulosa and Thecidce,
all the orders and families of the sub-kingdom
Coelenterata have living representatives. The
names of the recent genera are too numerous to
be here mentioned. Many of them have already
been indicated in those parts of the work devoted
to the study of their classification.

The systematic form under which we have
sought to exhibit the above selection of facts
touching the general relations to time of the
several groups of Corals must not lead the student
to repose too much trust in a record confessedly
so imperfect, or regard it as aught else than, in
the strictest sense, provisional. In particular it
would appear from certain investigations, not yet
fully published, that the supposed line of demar-
cation between the Paleozoic and Neozoic Coral
forms does not really exist. The entire subject,
like many others discussed in the preceding
pages, still offers a wide and richly promising
field for future inquiry.

249

BIBLIOGRAPHY OF THE CIELENTERATA.

(i.) Frey und Leuckart.— * Beitrage zur Kenntniss Wirbelloser

Thiere,' 1847. (pp. 1—40.)
(2.) Leucicvrt. — ' Ueber die Morphologie der Wirbellosen Thiere,'

1848. (pp. 13—31.)
(3.) Leydig. — ' Lehrbuch der Histologic,' 1857. (passim.)
(4.) Carus, J. v.— 'Icones Zootomies,' 1857. (Taf. II.— IV.)
(5.) Gegexbaur. — ' Grundzuge der Vergleichenden Anatomie,'

1859. (Zweiter Abschnitt, pp. 67 — 103.)
(6.) Bronx. — ' Die Klassen und Ordnungen des Thier-reichs,'

1859-60. (Zweiter Band, Lief. 1 — 6.)
Also, the systematic worlo of Cuvier, Regne Animal (ed. with

plates, by his pupils) ; Lamarck, Hist. Nat. des Animaux

s. Vertebres (2nd ed. by Deshayes and Milne Edwards) ;

De Blainville, Manuel d'Actinologie, 1834; Johnston,

History of British Zoophytes, 1847, 2nd ed. ; Cavolini,

Memorie per servire alia storia de' Polipi marini, 1785;

Bosc, Hist. Nat. des Vers, 1802; Esper, Die Pflanzen-

Thiere, 1806 ; Pallas, Elenclius Zoophytorum, 1766,

Charakteristik der Thierpflanzen, 1787; and the following,

among other, miscellaneous treatises :
(7.) BoHADSCH. — 'De quibusdam Animalibus marinis,' 1761.
(8.) BuscH. — ' Beobachtungen iiber Anatomie und Entwickelung

einiger Wirbellosen Seethiere,' 1851.
(9.) Dalyell. — ' Rare and Remarkable Animals of Scotland,'

1847-8.
(10.) Delle Chlwe. — * Descrizione e Notomia degli Animali inver-

tebrati della Sicilia citeriore osservati vivi negli anni

1822-30,' 1841-44.
(11.) Forbes and Goodsir. — 'On some remarkable Marine Inver-

tebratu new to tlie British Seas,' Trans. Roy. Soc. Edin. 1851.

250 BIBLIOGRAPHY

(l2.) FoRSKAL. — ♦ Descriptiones Animalium,' 1775.

(13.) . — ' Icones Keruin naturaliuin,' 1776.

(14.) GossE. — 'A Naturalist's Rambles on the Devonshire Coast,*
1853.

(15.) Laukent. — ' Zoophytologie ' (in Vaillant's Voyage de la
Bonite), 1844.

(16.) Lesson. — 'Centurie Zoologique,' 1830.

(17.) Leuckart. — ' Ueber den Polymorphismus der Individuen,'
1851.

(18.) MiJLLER, 0. F.— ' Zoblogia Danica,' 1788-1806.

(19.) QuoY et Gaimard. — * Zoblogie ' (in Voyage de I'Uranie,
sous Freycinet), 1824.

(20.) . — ' Zoblogie ' (in Voyage de I'Astrolabe,

sous Dumont d'Urville), 1830-4.

(21.) Sars. — ' Beskrivelser og Jagttagelser, &c.' 1835.

(22 ) . — * Bidrag til Kundskaben om Middlehavets Littoral-
Fauna,' 1857,

(23.) , KoREN et Danielssen. — ' Fauna littoralis Norvegise,*

1846 and 1856.

(24.) Steenstrup. — ' On the Alternation of Generations' (Eng.
Trans, by Busk), 1845.

(25.) Stimfson. — ' Synopsis of the Marine Invertebrata of Grand
Manan' (sep. and in Smith. Cont.), 1853.
It is unnecessary any longer to prolong the above list, since a
complete enumeration of the various memoirs on Coelenterata
may be found in the 'Bibliotheca Zoologica' of Carus and
Engelmann (Leipzig, 1861), pp. 320 — ^45. In the ' Natural
History Review ' (London, 1861 et seq.), continuations of this
catalogue will from time to time appear. The Reports fur-
nished each year by Leuckart to Wiegmann's Archiv. fur Na-
turgeschichte may also be consulted with advantage. Of
the more select memoirs which treat of particular groups of
Coelenterata we have here attempted to subjoin the names:

HYDROZOA.

a. Hydrid^e.

(26.) Trembley. — ' Me'moires pour servir k I'histoire d'un genre de

Polypes d'eau douce, a bras en forme de comes,' 1744.
(27.) Hancock. — 'Notes on a species of Hydra found in the North-
umberland Lakes,' A. N. H. 1850.

OF THE CCELENTERATA. 251

(a8.) Thomson', At.len. — *0n the Co-existence of Ovigerous and

Spermatic Capsules on the same individuals of the Hydra

viridis,' Phil. Journ. 1847.
(29.) EcKER. — 'Zui' Lehre von Bau und Leben der contractilenSub-

stanz der niedersten Thiere,' Z. W. Z. 1849 (or Trans, by Busk

in Q.J, M. S. 1854).
(30.) Jager. — ' Ueber das spontane Zerfallen der SUsswasserpolypen

nebst einigen Bemerkungen iiber Generationswechsel,' Vien.

Sitz. i860: and other memoirs cited in Bib. Zool., especially

those of Allman, Cokda, Laurent, Leydig, and Rouget.

The British species of Hydra are described by Johnston (op.

s. cit.), and Lewes, A. N. H. i860.

b. CORYNID.E AND SeRTULARID^.

(31.) Loven. — 'Beitrag zur Kenntniss der Gattungen Campanu-
lariaund Syncoryne,' Wiegm. Arch. 1837. (Abstract in Steen-
strup) (24).

(32.) Beneden, Van. — ' Memoire sur les Campanulaires de la cote
d'Ostende, conside'res sous le rapport physiologique, embryo-
genique et zoologique.' Bruss. Mem. 1843.

(33.) , . — 'Recherches sur I'embryogenie des Tubulaires

et I'histoire naturelle des diffe'rents genres de cette famille qui
habitent la cote d'Ostende,' Bruss. Mem. 1844.

(34.) ScHULTZE, Max.—' Ueber die mannlichen Geschlechtstheile
der Campanularia geniculata,' Arch. Anat. 1850 (or Trans, in
Q. J. M. S. 1855).

(35.) Mummery.— ' On the Development of Tubularia indivisa,'
Q. J. M. S. 1853.

(36.) Allmax.— ' On the Anatomy and Physiology of Cordylophora,'
Phil. Trans. 1853.

(37.) . — ' On the Structure of the Reproductive Organs in

certain Hydroid Polypes,' R. S. E. Proc. 1857-8; and 'Ad-
ditional Observations on the Morphology of the Reproduc-
tive Organs in the Hydroid Polypes,' ibid. 1858,

(38.) — 'Notes on the Hydroid Zoophytes,' A. N. H. 1859

et seq.
Other memoirs on the reproductive organs and development of
these orders are those of Dujardin, Ann. S. N. 1843 ^^^ 45 '»
Desor, Ann. S. N. 1849; Forbes, A. N. H. 1844; Lister,
Phil. Trans. 1834; Koli.ikkr, Z. W. Z. 1853; and Krohn,
Arch. Anat 1843 and 53, Wiegm. Arch. 1851. Also, the

252 BIBLIOGRAPHY

works of Sars (21), (22), (23); Steenstrup (24); and
Gegenbaur (47). Xo complete monograph of the genera of
CorynidjB and Sertularidaj has yet appeared. For descriptions
and figures of the Britisli species see the works of Dalyell
(9); GossE (14), and Linn. Trans. 1857; Ellis, 'Essay
towards a Natural History of Corallines,' 1755, and Johnston
(op. s. cit.) ; together Avith the papers of Allman ; Alder,
A, N. H. 1856 et seq. ; HixcKs, A. N. H. 1851 et seq. ; Stret-
HiLL Wright, Phil. Journ, 1857-8-9; and WY\nLLE Thomp-
son, A. N. H. 1853. American forms have been described by
Ayres, Bost. N. H. S. 1852 and 5; Desor, Bost. N. H. S.
1848-9; M'Crady (49); Murray, A. N. H. i860; and
Sti.mpson (25): Southern species by Busk, B. Ass. Rep.
1850, in Q. J. M. S. passim, and in supplement to Vol. I. of
Voyage of Rattlesnake; ' and Hincks, A. N. H. 1861,

c. Calycophorid,« and Physophorid^.
(39.) Huxley. — 'The Oceanic Hydrozoa — A Description of the
CalycophoridEe and Physophoridse observed during the voyage
ofH. M. S. " Rattlesnake " in the years 1846-50. With a
General Introduction,' 1859: and the writings of Milne
Edwards, Gegenbaur, Kolliker, Leuckart and Vogt,
cited in the bibliography appended to the same work. Other
memoirs by Gegenbaur, Nov. Act. i860; Glaus, Z. W. Z.
i860; and Keferstein und Eulers (Abstract in Wiegm.
Arch, i860), have since appeared.

d. Medusid.e and Lucernarid^.
(40.) Eschscholtz. — ' System der Acalephen,' 1829.
(41.) Ehrenberg. — 'Die Akalephen des rothen Meeres und der

Organismus dei Medusen der Ostsee,' 1836.
(42.) Wagner. — ' Ueber den Bau der Pelagia noctiluca, und die

Organisation der Medusen,' 1841.
(43.) Lesson. — ' Acalephes,' Nouvelles Suites k Buflfon, 1843.
(44.) Forbes.—' A Monograph of the British Naked-eved Medusae,'

1848.
(45.) Agassiz.— 'On the Naked-eyed Medusa of the Shores of

Massachusetts, in their Perfect State of Development,' Trans.

Amer. Acad. 1849.
(46.) Huxley.—' On the Anatomy and Affinities of the Family of the

Medusae,' Phil, Trans. 1849.

OF THE C(ELE>'TERATA. 253

(47.) Gegenbaur. — 'Zur Lehre vom Generationswechsel und der
Fortpflaiizung bei Medusen und Polypen,' 1854.

(48.) . — ' Versuch eines Systemes der Medusen, mit

Beschreibung neuer oder wenig gekannter Formen,' Z. W. Z.
1857.
(49,) M'Crady. — ' Gymnophihalniata of Charleston Harbor,' Ell. Soc.
Proc. 1857: and the syste;natic papers of Peron et Lesueur,
Ann. d. Mus. 1809-10; Brandt, Petersb. Mem. 1833 and 38;
and LuTKEN, Videns. Med. 1850 ; with other memoirs by
Allman (Structure of Lucernariadaj), Q. J. M. S. 1860^
Milne Edwards (Structure of yEquorea) (67), and (Circu-
lation in Lucernaridae), A. S.N. 1845; Derises (Hermaphro-
ditism of Chrysaora), Ann. S. N. 1850; KoLLiKER (Medusidae
of Messina), Z. W. Z. 1853; Leuckart (Medusidae of Nice),
Wiegm. Arch. 1856; Fritz Muller (Gastric filaments of
Lucernaridaj), Z. W. Z. 1858; Eysenharut (Rhizostoma)
Nov. Act. 1821 ; and Tilesius (Cassiopeia), Nov. Act.
1831.

On Structure of Marginal Bodies see especially Gegenbaur,
Arch. Anat. 1856 (or English abstract in Q.J. M. S. 1858).

On Minute Structure of Medusidae and Lucernaridae, vid. Busk,
Mic. Trans. 1852; Schultze, Arch. Anat. 1856; and Hux-
ley (46).

On Structure of Charybdeidae : Milne Edwards (Charybdea),
Ann. S. N. 1833 ; and Fritz Muller (Tamoya), Halle Abh.
1859.

On Development of Medusidae; J. Muller (.^ginopsis). Arch.
Anat. 1851 ; Gegenbaur (Cunina and Trachynema), (47);
Fritz Muller (Liriope), Wiegm. Arch. 1859; M'Crady
(Cunina), Ell. Soc. Proc. 1856; and CLAPARt;DE, Z. W. Z.
i860.

On Development of Lucernaridae: Sars (21), Tsis, 1833, and
Wiegm. Arch. 1837-41 and 1857; Siebold, Beitrage znr
Naturgeschichte der VVirbellosen Thiere, 1839 ; Steenstrup
(24); Dalyell (9); Desor, Ann. S. N. 1849; and Reid,
*Physiolo:,ncal, Pathological, and Anatomical Researches,' or
A. N. H. 1848. These writers treat chiefly of Aurelia, Cyanea,
and Chrysaora. For the development of other genera see
Gegenbaur (Cassiopeia), (47) ; Frantzius (Cephea),
Z.W.Z. 1853; and Krohn (Pelagia), Arch. Anat. 1855 (or
English abstract in A. N. H. 1856).

254 BIBLIOGRAniT

The British species of Medusidae (and Medu^oids) are ilUistrated
by Forbes (44), and Zool. Proc. 1851; FoitBEa and Good
SIR (11); GossE (14); Greene, Nat. Hist. Rev. 1857-8;
CoBBoLD, Q.J. M.S. 1858; Patterson, D. U. Z. B.A.
1859; and Stketiiill Wright, Phil. Journ. 1859. The
Lucernariadse are described by Johnston (op. s. cit.) ; Owen,
B. Ass. Ivep. 1849; GossE, A.N. H. i860; and Allman, op.
s. cit. and A. N. H. i860. Brief descriptions, without figures,
of the pelagic Lucernaridae are given by Forbes (44), but
most of the species are figured in the other works mentioned
above.

ACTINOZOA.

a. ZOANTHARIA.

(50.) DiCQUEMARE. — * Essay towards elucidating the history of the
Sea-Anemones,' Phil. Trans. 1773. *A second essay on the
natural history of the Sea-Anemones,' ibid. 1775, and a third
essay in ditto, 1777.

(51.) Ellis and Solander. — ' The Natural History of many curious
and uncommon Zoophytes,' 1786.

(52.) Rapp. — ' Ueber die Polypen im Allgemeinen und die Actinien
insbesondere,' 1829.

(53.) Ehrenbkrg. — 'Beitrage zur physiologischen Kenntniss der
Corallenthiere im Allgemeinen, und besonders des rothen
Meeres,' and ' Ueber die Natur und Bildung der Coralleninseln
und Corallenbanke im rothen Meere,' Berl. Abh. 1834.

(54.) Quatrefages. — 'Me'moire sur les Edwarddes,' Ann. S. N,
1842.

(55.) Dana. — 'Report on Zoophytes,' and 'Atlas of Zoophytes,*
(U. S. Exploring Expedition), 1849. The introductory part
of this Report (which is now out of print) has been jjublished
under the title of " Structure and Classification of Zoophytes."
A " Synopsis " of the Report itself has since appeared.

(56.) Edwards et H aime.— ' Recherches sur les Polypiers,' Ann.
S. N. 1848—52.

(57.) . — ' Histoire Naturelle des Coralliaires ou Polypes

proprement dits,' 1857 — 60.

(58.) Hollard. — ' IMonograpbie anatomique du genre Actinia de
Linne, conside're comnie type du groupe g^ndral des Polypes
Zoanthaires,' Ann. S. N. 1851.

OF THE C(ELENTERATA. 255

(59.) Haime. — 'Memoire sur le C^rianthe,' Ann. S. N. 1854.
(60.) GossE. — 'Actinologia Britannica : A History of the British
Sea-anemones and Madrepores,' i860.
See also various memoirs by Milne Edwards, Hollard, and
GossE, cited in Bib. Zool., in addition to those of Spix ;
KoLLiKER ; Lewes, ' Sea-side Studies ; ' Lacaze du Thiers ;
Rathke ; Erdl ; and others.

On the structure of Actinia see especially Hollard (58) ;
GossE (60); Haime, C. rend. 1854; Fkey und Leuckart
(i); Teale, Trans. Leeds Soc. 1837, and B. Ass. Rep. 1838;
and CoBBOLD, A. N. H. 1853.

On the Sclerobasic Zoantharia, vid. Brandt,' Symbolse ad
Polypos Hyalochaetides spectantes, 1859 ; ' and Schultze, C.
rend. i860. Dana (55) describes the polypes of Antipathes.
(57) is a complete monograph of the orders Zoantharia, Alcy-
onaria, and Kugosa. For collections of figures the works of
Esper, Ellis, Dana, and the French voyagers must be chiefly
consulted. The British Zoantharia are described and figurtd
by Gosse (60).
On Coral reefs and islands see Darwin, 'The Structure and
Distribution of Coral Keefs,' 1842 (now forming part of the
.same author's " Geological Observations ") ; and Dana, * On
Coral Keefs and Islands,' 1853.

h. Alcyonaria.

(61.) Rapp. — * Untersuchungen uber den Bau einiger Polypen des
Mittelandischen Meeres,' Nov. Act. 1829.

(62.) Edwards, Milne. — 'Me'moire sur un nouveau genre de la
famille des Alcyoniens,' and 'Observations sur les Alcyons
proprement dits,' in Recherches sur les Polypes, 1838, or Ann.
S. N. 1838.

(63.) Agassiz. — ' On the Structure of the Halcyonoid Polypes,' Am.
Ass. Rep. 1850. Also: Milne Edwards, in Regne Ani-
mal, and several of the works which treat of Zoantharia.
Johnston, op. s. cit., describes and figures the British Alcj'o-
naria.

c. Rugosa and Fossil Corals.

C64.) Edwards et Haime. — 'A Monograph of the British Fossil
Corals' (published by the Palajontographical Society), 1850
-55-

256 BIBLIOGRAPHY OF THE CCELENTERATA.

(65.) Edwards et Haime. — ' Monograpliie des Polypiers Fossiles
des Terrains Paleozoiques,' Arch. d. Mus. 1851.
And many other palasontological works, most of which are
quoted by Edwards et Haime (57).

d. Ctenophora.

(66.) Mertens. — *Beobachtungen Uber die Beroeartigen Acale-
phen,' Petersb. Mem. 1833.

(67.) Edwards, Milne. — ' Observations sur divers Acalfepbes,'
Ann. S.N. 1841.

(68.) . — * Note sur I'appareil Gastrovasculaire de

quelques Acalephes Ctenophores, Ann. S. N. 1857.

(69.) Will. — ' Horse Tergestinai, oder Beschreibung und Anatomic
der im Herbste 1843, bei Triest beobachteten Akalephea,'
1844.

(70.) Agassiz. — * On the Beroid Medusae of the Shores of Massa-
chusetts, in their Perfect State of Development,' Amer. Acad.
♦ Trans. 1849.

(71.) . — 'Contributions to the Natural History of the

United States of America,' — Part II. of Second Monograph,
i860.

(72.) Gegenbaur. — ' Studien iiber Organisation und Systematik
der Ctenophoren,' Wiegm. Arch. 1856: and the papers of
Grant, Z. Trans. 1833; Kolliker, Z. W. Z. 1853; Lesson,
Ann. S.N. 1836; Quoy et Gaimard, Ann. S.N. 1825;
Kang, Bourd. Soc. Lin. 1826, and Isis, 1832: and Wagner,
Arch. Anat. 1847. The systematic works of Eschscholtz
(40) ; De Blainville, op. s. cit. ; and Lesson (43) may
also be consulted. On Development of Ctenophora see the
papers of M'Crady, (Beroe and Boiina), Ell. Soc. Proc.
1859; Price (Pleurobrachia), B. Ass. Rep. 1846; Semper
(Eucharis), Z. W. Z. 1858 ; and Strethill Wright
(Pleurobrachia), Phil. Journ. 1856. The British Ctenophora
are described by Forbes and Goodsir, B. Ass. Rep. 1840-1 ;
and Patterson, R. I. A. Trans. 1839-40.

*^* The abbreviations above used to indicate the titles of periodi-
cal journals are explained in the Bibliography of ' Natural History
Review,' 1861.

257

QUESTIONS ON THE CffiLENTERATA,

1. By what structural features are Coelenterata separated from other

primary divisions of the animal kingdom ?

2. Contrast the two sub-kingdoms, Protozoa and Coelenterata.

3. Describe the typical structure of a thread-cell.

4. Compare the classes, HyUrozoa and Actinozoa.

5. Describe the structure of Hydra. How does the ' polypite ' of

this genus diflfer from that of the non- budding forms of the
Corynidge ?

6. Define the terms,

a. ' hydrotheca ' ;
6. * hydrophyllium * ;
c. 'hydrocyst.'

7. In what order of Hydrozoa do ' nematophores ' occur? Describe

the structure and position of these appendages.

8. Describe the structure of a ' nectocalyx.' How does this organ

differ from an ' umbrella ' ?

9. Describe the structure and relations of the nectocalyces in

Diphyes, and state how the same parts are modified in
Praya.

10. In what Physophoridse are nectocalyces absent?

11. Briefly describe the modifications of the coenosarc, and relative

attachment of its appendages, in the following genera of
Physophoridse ;

a. Physophora;

h. Stephanomia;

c. Apolemia;

d. Athorybia;
c. Velella.

S

258 QUESTIONS ON THE CCELENTERATA.

12. Compare the structure of the tentacles in

a. Physalia;
h. Forskalia;
c. Apoleraia.

13. What is the position of the tentacles in the following genera of

Hydrozoa :

a. Tuhularia;

b. Hydractinia ;

c. Diphyes;

d. Porpita;

e. Pelagia?

14. Describe the * pneumatocyst ' of any of the Physophoridse, and

the principal modifications which it presents among other
genera of the same order.

15. Describe, as to structure and position,

a. the marginal ' vesicle ' of Geryonia ;
h. the 'lithocyst' of a free Lucernarid.

16. How are the 'gonophores' situated in the following genera of

Corynidae :

a. Tubularia;
6. Hydi'actinia ;

c. Clavatella;

d. Cordylophora?

17. Compare the structural relations of the reproductive organs m

a. Lucernaria;

b. Aurelia;

c. Rhizostoma.

18. What are 'gonoblastidia'? Explain the modifications which

these structures present among the Sertularidae.

19. In what genera of Lucernaridai do free reproductive zobids

occur? Trace the development of any of these forms.

20. Describe the development of a medusiform gonophore.

21. Give some account of the early stages of development in

a. Cordylophora lacustris ;

b. Campanularia Loveni ;

c. Cunina octonaria.

22. What rule seems to govern the successive development of the

appendages among the Phy sophoridae ?
a3. Define the orders :

a. Sertularidae;

b. Calycophoridae ;

c. Lucernaridae.

QUESTIONS ON THE C(ELENTERATA. 259

14. What genera of CoDlenterata are known to inhabit fresh water?

25. Give some account of the geographical distribution of the iSertu-

1 arid IB.

26. Describe the minute structure of the bodj'-wall in Actinia.

27. What law appears to determine the number of parts among the

several orders of Actinozoa?

28. What is the number and structure of the tentacles in the

Alcyonaria?

29. Describe the structure of a typical ' corallite.'

30. Define the order Kuiijosa.

31. Explain the formation of the gyrate corallum of Moeandrina.

32. How does a ' sclerobasis ' differ from a true corallum?
33 Compare the nutrient system in

o. Actinia and Pleurobrachia.
b. Pleurobrachia and Beroe.

34. Describe the structural relations of the tentacles in

a. Actinia;

b. Cestum ;

c. Pleurobrachia.

35. What characters distinguish the thread-cells of the Cteno-

phora?

36. Describe the structure and position of the ' ctenocyst.'

37. Give some account of the nervous system of the Ctenophora.

38. What peculiarity of position distinguishes the reproductive

organs of Tubipora?

39. Define precisely the position of the male and female organs in

the Ctenophora.

40. What, according to Haime, is the number and succession of the

tentacles in the common Sea-anemone?

41. Give some account of the development of the canal system in

Beroe.

42. What numerical law governs the development of the * septa ' in

a Zoantharian corallite?

43. Distinguish three principal modes of gemmation among the co-

ralligenous Actinozoa.

44. Describe the structure of a Fringing-reef,

45. How has Mr. Darwin explained the true nature of Barrier-reefs

and Atolls ?

S2

260 QUESTIONS ON THE CCELENTERATA.

46. Define the characters of the famil}- Beroidae, with reference to the

subjoined categories :

a. form of body ;
I. mouth ;

c. canal system ;

d. tentacles.

47. Give some account of the distribution, bathy metrical and geo-

graphical, of the reef- building Corals.

48. What families of Zoantharia seem wholly extinct ?

49. In what deposits does the family of Astraeidae first make its ap-

pearance ?

50. Name those groups of Corals which are most abundantly repre-

sented in the Paleozoic series.

261

LIST OF ILLUSTRATIONS.

Page
4
15

1. Urticating organs of Ccelenterata, after Gosse

2. DeTelopment of Ccelenterata

3. JMorphology of Hydra, after Hancock, Johnston, and Allen

Thomson

4. IMorphology of Hydrozoa

5. Morphology of Cordylophora, after Allman

6. Reproductive processes of Hydrozoa, after Gegenbaur

7. Oceanic forms of Lucemaridcs, after Gosse .

8. Development of Cordylophora, after Allman

9. Development of Tubularia indivisa, after Mummery .
ro. Development of Campanularia, after Loven

1 1. Development of Physalia, after Huxley

12. Development of Lizzia, after Claparfede

13. Development of TMrm, after Gosse ....

14. Gemmation of Medusoids, after Forbes and the Author

15. Development of C/trysaora, after Dalyell .

16. Morphology of Ttibulariadce, after Alder, Forbes, and Stret-

hill Wright . . 84

17. Various forms of Coryniadce, after Alder, Forbes, and Sars 86

18. Morphology of Sertulariadce, after Alder, Dalyell, Forbes,

and Johnston

19. Morphology of Campanulariadae, after Alder and Hincks
ao. Morphology of Calycophoridce, after Kolliker

21. ]\Iorphology of Veldla, after Kolliker

22. Morphology of Physophoridoe, after Kolliker

23. ]Morphology of Medusidce ....

24. Various forms of 3Iedusid(B ....

25. Luccr/iarUi, after Johnston ....

s 3

91
94
97
104
109
116
117
122

262 LIST OF ILLUSTRATIONS.

Page

26. Morphology of ^c<mo2oa, after Gosse and Hollard . . 133

27. Morpliology of Pleurobrachla, altered from Agassiz and

Huxley 145

28. INIorphology of Zoantharia Sclerodermata, after Huxley . 156

29. Columnaria FrankUnii . . . . . . . .158

30. Tubipnra musica . ' . . . . . . .159

31. Development of Pleurobrachla, after Strethill Wright . 181

32. Turbinaria pallfera, after Dana 188

33. DendropliyJlia nigrescens, after Dana . . . . .189

34. Zoantharia Malacodermata, after Couch, Forbes, and Gosse 197

35. Zoantharia Sclerobasica, after Dana 202

36. Zoantharia Sclerodermata, after Milne Edwards and Haime 204

37. FennatulidcB and Cor^omWcCjafter Patterson, Milne Edwards

and Haime 211

38. Zaphrentis cylindrica , . 215

39. Various forms of Ctenophora, after De Blainville, Gegen-

baur, Lesson, Patterson, and the Author. . . . 217

263

INDEX.

Abyla, 98, 99, loi.

Acalephes Simples, 114.

Acaleuhes Hydrostatiques, 114.

Acanthoccctiia, 246.

Acanihoci/at/ius, 240.

Acaulis, 87, 88, 89.

Accrvularia, 158, 244.

Acontia, 1^5.

Arraspeda, 120.

Acrocyst, 95.

Actinacis, 247.

Actinia, 131, 146, 166, 172, 199, 235.

Actinidip, 2oy.

Actinoluba, 198.

Actinomeres, 14J.

Aclinopsis, 198.

Actiiiosome, 143.

Actinozoa, type of, iji ; general cha-
racters of, 19, 138 ; development of,
170 ; classification of, 196 ; distri-
bution of, 231; relations of, to time,
236.

Adamsia, ijo, 198.

Adelastrwa, 239.

jEgina, 116.

jEginidce, 120.

Mginopsis, 60, 116.

JEquoridce, 120.

Agalma, 102, 108, no, lir, 112.

Agamogenesis, 74.

Air-vesicle, oi Physophoridce, loi.

Alcyonaria, general characters ol, 139,
196, 208 ; nutritive cavity of, 141 ;
tentacles of, 148; thread-cells of,
151 ; corallum of, 154, 159, 162 ; de-
velopment of, 170 ; families ot, 2ij;
distribution of, 231, 233 ; relations
of, to time, 237, 243.

Alcyonidce, 213 ; corallum of, 160, 162,
209.

Alcyonium, 161, 208, 232, 235.

Alimentary canal, of Ccelenterata, 3,
29, 141.

Alternate generation, 76.

Alveolites, 238.

Amoeboid condition of Hydra, 11, 52.

Amplextis, 238.

Anabacia, Z47.

Androphore, 45.

Ane77ionia, 199.

Angeastraa, 246.

Angel-band, 34.

Anisop/iyllufn, 244.

Anomophyllujn, 246.

Antagonism, between development
and reproduction, 72.

Antenvularia, 57, 127.

Anthea, 182, 199, 201.

Antipathes, 200, 201.

Antipathidcc, 206.

Apertures, of somatic cavity in Acti-
nozoa, 142.

Apical area, of Ctenophora, 143, 145,
219.

Apical canals, 145.

Apical pores, 142, 145.

Apolemia, 102, 108, no, in.

ApolemiadiS, 113.

Aporosa, 158, 161, 162, 203, 206, 237,
242.

Appendages, of Hydro%oa,i'j,&^% of
Hormiphora, 152 ; of Ctenophora^
218.

Arachnactis, 163, 199.

Arteacis, 247.

Aspidiscus, 246.

Astrangia, 241.

Astroccenia, 239.

AstrcBu, 241 .

Astrwida:, 20j, 207, 232, 237, 245—7.

AstrcBopora, 241.

A&irohtelia, 247.

Athoryhia, 107, 108, no, 112.

Athorybiada;, i\i.

Atolls, 191 -4.

Atractylis. 62, 88, 89.

Aulacophyllum, 238.

Aulophyilum, 245.

Aulopora, 204, 244.

AtiloporidcE, 204, 2c6, 237, 244, 245.

Aurelia, 47, 64, 123, 127, 128.

Axophyllum, 245.

Axopora, 247.

Axosjnilia, 246.

S 4

264

INDEX.

Balanophyllia, 241.

B;iirier-reefs, 191 — 4.

Baryhcelin, 246.

Baryphyllum, 244.

Jiari/S7nilia, 240.

Basal gemmation, of Aciinozoa, 182,
i8j.

Bas.l membrane, of Adamsia, 150,
198.

Base, of Actinia and its allies, iji,
198.

Balhycyathus, 240.

Bathymetrical distriS)ut)on, of Hy-
drozoa, 126; of Actinozoa, 231.

Batter sbyia, 244.

B('ai/7no7itia. 2j8.

Beroe, tactile organs of, 168 ; develop-
ment of, 180; size of, 216; moutli of,
218 ; canals of, 220-1.

Bcroidce, form of body in, 216 ; lappets
of, 218 ; apical filaments of, 219 ;
canal system of, 220—2 ; definition
of, 231.

Bimeria, 85, 87, 88.

Blastoderm, 16.

Blastotrochus, 202.

Body-layers, of Coelenterata, 3, 10,
17-

Body-substance, of Protozoa, 7 ; of
Coelenterata, 10.

Body-wall, of Hydrozoa, 25; of Ac-
tinozoa, 136, 149.

Bolina, development of, 179; size of,
216; lobes and earlets of, 218; canals
01, 2ZO— 3 ; tentacles of, 22/ ; phos-
phorescence "f, 230.

Bougainvillea, 62.

Brachycynthus, 246.

Brachyphyllia, 240.

Urain-stone (oral, 187.

Briareum,ziz

Calamophyllia, 239.

Calice, 155.

Calicular gemmation, of Actinozoa,
183, 184.

Callianira, 217, 223, 227, 234.

CallianiridcB , 230.

Calycophoridce, general characters of,
80, 96 ; development of, 57 ; soma-
tocyst of, 96 ; ccenosarc of, 98 ; nec-
tocalyces of, 98 ; polypites of, 29 ;
tentacles of, 32—4; hydrophyllia of,
99 ; goiiophores of, 99 ; families of,
IOC ; distribution of, 128.

Calymma, 234.

Calymmidce , 230.

Campanularia, 44, 47, 57, 90-5, 126.

Campanulariadce, 94, 95.

Campanulina, 93.

Campophyllum, 238.

Canal system, of Medusidcs and Lu-
cernaridcE, 48, 115. 121— 4; of Cte-
nophora, 144—7, 219—24.

Ciipnea, 199.

Carboniferous Corals, 244.

Carduella, 121.

Caiytphyllia, iso,

Cassiopeia, 49, 67.

Cfll-theory, 12.

Cenlrocoenia. 240.

Cephea, 49, 67, 123.

Ceratotrochus, 247.

CeriaiUhun, 142, 150, 162, 172, 200,
201.

Cestidce, 230.

Cestum, form and size of, 216 ; canals
01, 223 ; tentacles of, 227 ; secretive
organ of, 147 ; phosphorescence of,
230; distribution of, 234.

Chambered cnidae, 150.

Cliarybdea, 126.

Chiajea, papilla? of, 149; development
ot, 180; canals of, 2x0—3; tentacles
of, 227.

Chcetetes, 238, 241, 245.

Chonaxis, 245.

Chonophyllum, 244.

Chotiostegites, 244.

Chrysaora, 47, 64, 123, 125, 127.

Cilia, ot Coelenterata, 3 ; of CtenO'
phora, 165.

Cinclides, oi Actinia and its allies,i3S,
142.

Circophyllia, 247.

Circulation, in Coelenterata, d, 31, 142,
147. 2.09.

Cladangia, 247.

Cladocora, 240,

Cladophyllid, 239.

Classification, of Coelenterata, 14, 19-
of Hydrozoa, 79, of Actinozoa, 196 ;
of Corynida:, 89 ; of Sertularidte, 95 ;
of Calycophoridce, 100; of Physo^
phorida.', 113; of Medusidce, 119; of
Lucerniiridce, 125; of Zonntharia,
205; o\' Alcyonaria,ziy, of Bugosa,
215 ; of Ctenophora, 230.

Clara, 41, 46, 82, 86. 88.

Clavatella. 87, 88, 89.

Clisiophiillwn, 238.

Cnidce, '3, 150.

Coelenterata, general characters of, 3,
6, 13, 19; classes of, 14, 19; thread-
ceils of, 3; minute structure of, 10 ;
rievelouinent of, 14—18.

Ccelusmilia, 239.

Ca?nenehyiiia, 155, 157.

Coetiiies, 244.

Cccnocyathiis, 202.

Ccenosarc, 27; of Corynidce, 82 — 3; of
Scrtularidce, 90 ; of ( alycophoridce,
98 ; of Fhysophoridce, loi— 3, 107; of
Aciinozoa, 139, 157, 2QI, 208.

Ccespitose corallum, 186.

Columella, 155.

Column, of Zoantharia, 198, 199.

Columiiaria, 244.

Covihophyllum, 244.

Combs, of Ctenophora, 165.

INDEX.

265

Conioscris, 146.

Conocyathus, Z47.

Conslellaria, 244.

Continuous development, yj.

Conversion, chemical, of tu.iies, 9.

Coppinia, 91.

Coral, isj.

Coral ol commerce, 234.

Coral -reefs, 191 — 5.

Corals, 140, 15}.

Corallite, 155.

CoraUiu7n, pores of, 142; corallum of,

154, 210; distribution of, 232,234.
Corallum, structure of, 153— 62; de-
velopment of, 173 — 8, 185—90; of
Zoanthaha, 201—5 ; of Alci/imaria,
159, 209 ; of liugosa, 214
Corbula, of Plimiularia cristata, 95.
Cordylupliora, gonophores of, 44, 45,
88 ; development of, 54 ; ccenosarc
of, 82; tentacles of, 86; distribution
of, 126, 127.
Cornularia, 160, 209.
Coronets, 155.
Cort/nactis, 200, 201.
Curyiie, gonophores of, 41, 88, 89;
tentacles of, 86 j distribution of,
126.
Coryniadce, 89.

Coryni'JiC, general characters of, 79,
82 ; tentacles of, 32, 85—8 ; cceno-
sarc of, 82 ; polypary of, 35, 85 ;
polypites of, 29, 85 ; gonophort'S of,
40—6, 88. 89 ; development of,
54 ; families ot, 89 ; distribution of,
126, 127.
Costae, 156.
Craspeda, 134.
Oaspedota, 120.
Cretaceous Corals, 246.
Cryptangia, 247.
Cryptoc-arpte, 120.
Cryptolaria, 128.
Ctenocyst, 144^ 166.
Ctenop/iora, type of, 142; general
characters of, 139, 142, 196, 216 ;
form and size of, 140, 216—9;
canal system of, 144, 219 ; tentacles
of, 225 ; tegumentary organs of,
149, 152 ; tliread-cells of, 151 ; cte-
nophiires of, 164; nervous system
of, 167; organs of sense ot, 166,
168; reproductive organs of, 169;
habits of, 229 ; phosphorescence of,
230 ; development of, 178 ; fimilies
of, 230 ; distribution of, 253—5.
Ctenophoral canals, 146, 219.
Ctenophores, 143, 164.
Cunina, development of, 60, 61 ;

pouches of, 116.
Cyanca, 47, 64, 127.
L yathaxonia, Z14. 238.
Cyalhaxunidie, 216, 237.
Cynthophora, 239.
CyatUvpliyllidcE, 216, 237, 244, 245.

CyathophyUnm, 184, 215, 238.

CyiKhiiscris, 247.
Cycle, of septa, 175.
Cycloci/fifhtis, 246.
Cyclolitcs, 240.
Cycloseris, 240.
Cydosmitia, 247.
Cydippe, 142.
Vyphastrcen, 240.
Cystiphyllidce, 216, 237, 244.
Vystiphylluni, 215, 238.

Dactylostnilia, 246.

Dauaia, 244.

Dasima, 247.

Dasnndce, 207, 237, 247.

Dasyphyllia, 240

Definite gemmation, 184.

Dekaia, 244.

Dt'ltocyathus, 247.

Dtndracis, 247.,

Dendrophyllia, 241.

Dendropora, 244.

Deudrosmilia, 247.

Dendrostyies, of Rhizostomidce, 49.

Dcpastrum, 121.

Deposition, clicmical, in tissues, 9

Derrnoscleiites, 160.

Dcsmophyllu))!, 240.

Development, of animals, 14, 18, 70 ;
of Ccelenterata, 14. 17, iS ; of Hy-
drozoa, 51 ; of Aclinuzon. 170; of
Hydridce, 51 ; of Corymdce, 54; of
Seriularidce 56; oi Calycuphoridte,
57 ; of Fhysophoiidce, 57 ; of MC'
dusidce, 60 ; of Luccrnaridce, 64 ;
of Zuanthiiria and Alcyonaria, 170,
185 ; of Ctenopkora, 178.

Devonian Coials, 244.

Deuterozoiiids, 74.

Dicori,nc, 88.

Digestive sac, of Actinozoa, i}2, 141,
144, 220.

Diniorpliaslr^a, 247.

Dipliyd/E, 100

DiphyeSy 50, 96, 99, loO, 128.

Dipliyozodids, 100, 127.

Diploctenium, 246.

Diplohcelia, 247.

Diploiicma, 63.

Diplorid, 240.

Disc, gelatinous, of MediisidcB and
Luceruaridce, 34; of Actinia and
its alles, 131, 198.

Di^cocyathus, 246.

Discontmuous development, 73.

Discosoma, 199.

Discolrochus, 247.

Dissepiments, 155.

Distal extremity, of hydrosoma, 25 ;
of polypite, 29.

Distal nectocalyx, of Diphyes and
Abi/la, 96, 98.

Distribution, of Ccelenterata, 13 ; of
Hydruzoa, 126; oi Actinozoa, 231.

266

INDEX.

Earlets, of Ctenophora, 218.

Ecderon, 149.

Echmoponrtie, zcrj, 437.

Ecmesus, Z47,

clcthorceuin, 4.

Ectoderm, 3. 10, 17.

Edwardsia, 162, 199.

Elasmocoenia, 246.

Embryo, of Hydruzoa, 51 ; of Actino-

xoa, 170.
Emmonsm, 2j8.
Enallohoelia, 239.
Eiuteron, 149.
Endoderm, 3, 10, 17.
Endopachys, 241/
Endophyllum, 244.
Endoplasts, 10.
Ephyra, 66.
Epitheca, 157.
Epithelial layer, of disc in Medusidts

and Lucer'naridtB, 35 -, of body-wall

in Actinozoa, 136, 149.
Eridophyllum, 238.
EucopidiE, 120.
Eudendrhun, 83, 87 — 9.
EuhcElia, 246.
Eumenides, 148, 200.
Euphyllia, 240.
Eupsatnnia. 203, 241.
Eurhamphcea, 218. 220, Z23, 247, 228.
Eurymceandra, 239.
Eurystomata, 218, 231.
Exotltecas, 156.
Eye-specks, 38, 166.

Families, of CorynidcB, 89 ; of Sertu-
laridce, 95 ; of Cali/cophuridce, 100 ;
of Phi/scphoridce, 111; *^^ Meausidce,
119; of Lucernaridce, 125 ; of Zuan-
tkaria, 205 ; of Alcyonaria^iii ; of
Rugusa, 215; of Ctenophora, 230.

Fan-Corals, 186.

Favia, 239.

Favositcs, 238.

Eavositidof, 205, 206, 237, 244, 245.

Fecundation, ui Hydrozua, 50 ; in .4c-
tinozoa, 170.

Feelers, of Physophoridtef 40.

Fibrillation, of tissues. 8.

Filament, of tentacle in Hydrozoa, 33.

Firm part, of Velella, 104.

Fission, 72; oi Hydra, 52; of Medu-
soids, 63; of Stomobrnchmm, 61; of
Hydra-tuba, 66; of Actinozoa, 182,
185.

Fistulipora, 238.

Flctscheria, 244.

Float, of Physophorid<e, 27, loi.

Foot secretion, 153.

Form of body, in Ccelenterata, 6 ; in
Hydrozoa, 25—7; in Actinozoa, 196,
208, 216.

Formless contractile substance, 52.

Forskalia, 54, 102, 108, III.

Fossil Corals, 236.

Fossil forms, of Hydrozoa, 130 ; of

Actinozoa. 236—47.
Fresh-water forms of Ccelenterata, IJ,

126.
Fringing-reefs, 191 — 4.
Fungi a, 199.
Fut.gida:, 203, 207, 232, 237, 244, 246,

247.
Funnel, of Ctenophora, 144, 220.

Galaxea, 240.

Gamogcnesis, 75.

Ganglia, of Actinia, 166; of Cteno-
phira, 167.

Garveia, 88.

Gastric filaments, of Lucernaridce,
124.

Gastric region, of polypite, 29.

Gemm.iti<in, 72; of Ccelenterata, 6; of
Hydrozoa, 51, 54, 57, 63, 65, 69 ; ol
Actinozoa, 182—90.

Genabacia, 247.

General cavity, of Hydrozoa, 17, 29 ;
of //c/mo2o«, 17, 141.

General morphology, of Ccelenterata,
I, 6, 10 ; <if Hydrozoa, 25 ; of Acti-
ncxoa. 138.

Geographical distribution, of Hydro-
zoa, 127; of Actinozoa, 233.

Germinal dot, 15.

Germinal vencle, 15.

Geryonia, marginal vesicle of, 38.

GeryoTiidcE, 120.

Glandular sacs, of Velella, 106.

Globatecnidse, 150.

Goniastrcea, 240.

Goniocor/i. 239.

Goniophyllitnt, 244.

Gunoblasiidia, 45 ; of Corynidee, 46,
88; of Srrruluridce, 46, g^; of Phy-
sophuriiia:, 112.

Gonocalyx, 43.

Goiiophcires, structure of, 40— 5 ; po-
sition of, 45—7; of Corynidce, 41,
44, 88 ; of Srrtvlarid^, 44, 46, 93 ;
01 Cci/ycopfiun'dce, 4s>99'i of thyso-
phoriilce 45, iii.

Gonotheca, 47.

Gorgonia, 161. 232, 235.

Gorgonidce, general characters of, 163,
210, 214 ; somatic cavity of, 141 ;
sclerobasis of, 154, 162 ; spicules of,
161; develoiment of, 185; distribu-
tion of, 232; relations of, to time,
237,246.

Graphularia, 247.

Graptolires, 130

Gymnophthalmata, 120,

Gynophore, 45.

Hadrophyllum, 244.
Haimeia\ 208.
Haleciutn, 93.
Halicornnria. 93.
Halistevinia, 103, 108, ill, IIZ.

I

INDEX.

267

Hallia, 244.

llulysiUs, Z44.

Hnplucania, 247.

Hnplophyllia, 2J9.

Haplosmilia, 240.

Heliastrcea, 259.

Hcliolites, 238.

Heliophyllum, 244.

Hcteractis, zcxi.

Heterocccnin, 246.

Hippopodiidte, 100.

Hippopbdiiis, 45, 98, 99.

History, ol Ui/ilrozoii, ip; of ^c^?-
nozort, 2}6; ol Zcirtw/Art?va, 242 ; of
liugosa, 243; of Alcyonaria, 24 j.

Hoenial region, 17.

Holvcoenia, 246.

Holocystis, 214, 247.

Hortniphora, 152.

Jlyalochcetidce, 201, 206.

Hyalonema, 154.

liyolopathes, 154, 241.

Hybocania, 239.

Hydiiof hura. 2 p.

Hydra, morphology and physiology
of 20— s; development of, 51; amoe-
boid condition ol, 11, 52 ; s.pecies of,
82; distribution of, 126, 127.

H>dra-tiil)a, 65—7, 124.

Hydractinia, tentacles of, 32, 87 ; coe-
nosarc of, 83; polypary of 85; gono-
phores ol, 41, 45 ; gouoblasudia of,
46.

Hydrides, 51, 79, 80.

H>dri)cysts, of Physopkoridce, 40. no.

Hydr2ecium, of Calycuphoridce, 96, 99.

Hydrophyliia, 35 ; of Cali/cophoridte,
99; of P/iysopfiondiP, 108.

Hydrorhiza, 26.

Hydrosoma, 25.

Hydrothecae, of SertularidcE, 27, 70,
92.

Hydroxoa, type of, 20 ; general cha-
racters ot, 19, 20; development of,
51; classification of, 79; distribution
of, 126 ; relations ol, to time, 130.
HymenophyUia, 246.

Ilyanthus, 198.
Indefinite gemmation, 184.
Individuality, of animals, 13, 71.
Inner layer, of blastoderm, 16.
Integument, of Coelenteratn, 10 ; of

Hydroxoa, 34 ; of Actmuxoa, 149.
Intermediate layer, 16.
Involucrum, 34.
Irregular gemmation, 184.
Isastrcea, 239,
Isis, 154, 210, 240, 241.

Jelly-fishes, 13, 127.
Jurassic Corals, 245.

Koninrkia, 247.
K^p/U'dclcmuon, 212

Lahfcheia, 244.

Lagoon, 192.

Lamellar corallum, 187.

Lappets, of Beroida:, 218,

7>ar, 88.

Layers, of Coelenterata, j, 10.

Lrptocyathns, 247.

Leptophyllia, 246.

Leplorid, 240.

LeSueuria, canals of, 222, 223; tenta-
cles of, 228.

I.rucothea, tentacles of, 228.

Lite, duration of, in Actinozoa, i8z.

Life history, of Hydroxoa, 51 ; of AC'
tinoxoa, 170.

Limb, of Fdella, 104.

Lineolaria, gonophores of, 128.

I.iriope, 60, 115.

Litharoea, 247.

Lithocysts, oi Lucernarida, 38, 124.

Lithophyllia, 240.

Lithostrotion, 238.

Liver, of Hydroxoa, 31, ic6; o{ Acti-
niixoa, 137, 147.

Lixxia, development of, 60, 63.

Lobupsamnia, 247.

Locomotive organs, of Hydroxoa, j6;
of Actinoxoa, 164.

Loculi, 155.

Lonsda/eia, 245.

LophohcBlia, 240.

LopUophyUuin, 23O.

Lovhosmilia, 240.

Lncernaria, 40, 121, 124, 125.

Lucernariadii, 67, 120, 122—5.

Lucernaiidcc, general characters of,
80, 120 — 5 J unibi ella of, 27, 37, 124;
lithocysts of, 38, 124; canal system
of, 48, 124; gastric filaments of, 124;
reproduciive zodids of, 47; develop-
ment of, 64, 123 ; families ol, 125 ;
distribution of, 127, 128.

Lucernaroids, 123.

Lyellia, 244.

Madrepora, 241.

Madreporidec, 203, 207, 232—4, 237,

246, 247.
Malacodermata, 201, 205, 237.
Manubrium, 43.
Marginal bodies, of Medusidte, 37 ; of

Luceniartdie, 38, 124 ; of Actinia,

165.
Massive corallum, 187.
Meconidia, of Campanularia Loveni,

95-

Medusidce, general characters of, 80,
114—20; nectocalyx of, 36, 116;
marginal bcdies of 37 ; polypite of,
Ii5i tentacles of, 117; reproductive
organs of, 118 ; development of, 60,
118; families ot, 119; distribution
of, 128; phosphorescrnce of, 129.

Mfdusiform gonophores, 45.

Medusoids, 43, 45, 62, 118.

268

INDEX.

Meli/cea. 154, 210.

Membrana intermedia, 16.

Mennphyllum, 145.

Mertensia, ijj.

Merulina, zo8.

Merulinidts, 208, 237.

Mesenteries, oi Actinoxoa, ill — 4, 137,

146, 172.
Metagenesis, 74.
Metamorpiiosis, 73.
MeCastrcea, 247.
Michelinia, 2j8.
Micrabacia, 247.
Micropyle, of animal ovum, 71.
Micrusolena, 246.
Millepora, 259.

Millepondce, 205, 206, 237, 244, 247.
Mini/as, 16 j.
M^andrastrcea, 246.
Mceandrina, I4Z, 187, 199, 20J,2J2, 239.
Monoecious forms, of Hydrozoa, 50,

112: 01' Aclinoxoa, 169, 209.
Montlivaultia, 259.
Mopsea, 154, 210. 241.
Mouth, of Hydrozoa, 29, 86, 88; of

^c^jwozoa, i}i,i4i; of Ctenophora,

218.
Movements, of Hydrozoa, 2;, 27, j6;

of Actinoxoa, ij8, 162, 210 ; of CYe-

nophora, 229.
Mucous layer, of blastoderm, 16.
Mulberry-mass, 16.
Muscular system, of Hydrozoa, j6 ;

of Actinozoa, 162. .
MycetophyUia, 240.
Myriotheia, 82, 87.

Nectocalyces, 27, j6, 70 ; of Calyco-

p/ioridce, 96—9 ; of PkysophoridtB,

107; of MedusidcB,\\(>.
Nectocalycine canals, 37.
Nectosac, 36.

Nematophores, of Sertularidce, 34, 93.
Ncmopsis, 84.
Nervous system, of Medusidcs, 39 ; of

Actinia, 166 ; of Ctenophora, 167.
Neural region, 17.
Nutritive organs, of Hydrozoa, 28 ; of

Actinozoa, 141.

Oceania turrita, marginal bodies of,

OceanidiP, 119.

Ocelli, 38.

Oculinn, 240.

Oculinid<B, 203, 207, 237, 246, 247.

Ori/roe, 234.

Oldliaunffi, 130.

Onipliyina, 244

Operculum, of Campanularia fasti-
ii'Ota, 92.

o'lal can.ils, of Ctenophora, 220.

Older, of septa, 175.

Oiders, of Hydrozoa, 79 j of Actino-
zoa, 196.

Organ.pipe Corals, 209.

Oroseris, 239.

Otolites, 39.

Outer layer, of blastoderm, 16.

Ova, of animals, 14, 71 ; of Hydrozoa,
50; of Actinozoa, 169.

Ovaria, of Hydrozoa, 40, 50; of Ac-
tinozoa, 169.

Ovigerous vesicles, of Sertularidce, 4,^.

Pachygyra, 239.

Pachyphyllum, 244.

Pali, iss.

PalcEocyclus, 744.

Palpocils, 93.

Palythoa, 141.

Paracyathus, 240.

Paragastric canals, of Ctenophora,
145, 220.

Parasitic forms, of Hydrozoa, 61, 83 ;
of Actinozoa, 198, 233.

Paras?nilia, 246.

Parietal genim.ition, of /ic/moxofl, 183.

Parthenogenesis, 74; theory of, 78.

Pavonaria, 240.

Peachia, 142, 199, 200.

Pedicle, of tentacle in Hydrozoa, 33.

Peduncle, of polypite, 29, 31.

Pelagia, structure of, 40, 123 ; de-
velopment of, 68; phosphorescence
of, 129.

PelagidcB, 125.

Pennaria, 87—9.

Pennatula, 111 — i, !■}!.

Pemiatulidce, 163, 186,210—3,214,235,
237, 246, 247.

Pentacoema, 246.

Peplosmilia, 246-

Perforata, 158, 161, 203, 206, 237, 242.

Perigonimus, 87 - 9.

Periplast, 10.

Peristomial space, 132, 198.

Permian Corals, 245.

Phanerocarpas, 120.

Phillipsastrcea, 238.

Philomedusa, 142, 233.

Phosphorescence, of Coelenterata, 7 ;
of Hydrozoa, 128 ; of Pennatulidce,
212 ; of Ctenophora, 230.

Phyllactis, 200.

Phyllangia, 241.

Phylloccenia, 240.

Phyllocyst, 36.

Fhyogemmaria, of Velella, 105.

Physalia, thread-cells ol, 5 ; villi of,
3 1; tentacles of, 32, 33, III; develop-
ment of, 58 ; pneumatophore of,
loi, 107; gonophores of, ill, 112;
distribution of, 127.

Physaliadce, 114.

Physical elenents, relations to, of
Hydrozoa, 126; of Actinozoa, 231.

Pkysoptiora, loz. 107 — 12.

Physophoriadcs, 113.

Pkysophuridce, general characters of.

INDEX.

269

80, loi ; development of, 57 ; pneu-
niatophore of, loi, IC7 ; coenosarc
of, 107 ; nectocalyces of, 107 ; hydro-
phyllia of, 108 ; polypitt-s of, 29, 1 10 ;
hydrocysts of, 40, no ; tentacles of,
32—4, ill ; gonophores of, 47, in ;
families of, 113 ; distribution of,
127, 128.

Phyto^yta, 246.

Figment masses, of Hydrozoa, 36, j8 ;
of Actinozoa. 152.

P/acoca'nia, 246.

Placosvitlia. 246.

Placosphyllia, 246.

Plan of struiture, of Ccelenterata, 3,
19 ; of Hydrozoa, 17, 19, 25 ; of Ac-
titwzua, 17, 19, 138.

Planula, 64.

Pialylrochus, 247.

Plerastrcra, 259.

Plesiastrcea, 241.

Pleurubrachia, form and structure
of, 143, 216, 221 ; thread-cells of,
151 ; nervous system of, 168 ; tactile
organs of, 168 ; development of,
180 ; tentacles of, 225 ; distribution
of, 235.

Pleurobrachiadce, 230,

Pli'urocoenia, 239.

P/eurucora, 247.

Pleurodictyuni, 244.

Plumularia, 34, 44, 50, 93,94.

Pneumatic filaments, of Velella and
Porpita, 102, 106.

Pneumatdcyst, of Physophorida:, loi j
of Feli/ld, 105.

Pneumatopiiore, of Physophorid(E,
loi, 107.

Podactinaria, 120.

Podocoryne, 88, 112.

Polyccelia, 245.

Polypary, of Corynidce and Sertula-
ridtB, 35, 85, 92.

Polype, 139; of Zoantharia, 197; of
Alcyonaria, 208.

Polype-cells, of SertuJaridce, 35.

Polypite, 25. 29 ; of Hydrida;, 20, 82 ;
of Coiyni(i(E, 85 ; of Sertularidce,
92 ; of Calycup/iorida:, 29, 99 ; of
Physophoridce, no; of Medusidcc,
IIS J of Lucernaridte, 47, 65, 121,
123.

Polytremacis, 247.

Porites, 232. 241.

Poritida:, 204, 207, 237, 244. 246, 247.

Porp»/a, 32, 45, 102, 103, 107,110—112.

Pouches, of .^ginidie, 116, 120.

Praya, <^, 99.

Prayid(E, 100.

Prehensile organs, of Hydrozoa, 32 ;
of Actinozoa, 148.

Prionastraa, 241.

Propora, 238.

Protartea, 244.

Protoplasm, animal, 7.

Protnsrris. 246.

Protovirguliiria, 243.

Protozoa, their minute structure,
compared with that of Cwlenterata,
7 ; development of, 14, 18.

Proiozooids, 74.

Proximal extremity, of hydrosoma,
25 ; of polypite, 29.

PsamvKipora, 238.

Pterygia, 150.

Ptycfiophyllum, 238.

Pyloric valve, of Calycophoridce, 29.

Pyrgia, 204, 245.

Radial canal system, of Ctenophora,
219, 221.

Recent Actinozoa, 248.

Regular gemmation, 184.

Reniform enlargements, of tentacle
in Physalia, 33, iii.

Renilla, 212.

Reparative powers, of Hydrozoa, 5Z,
55 : of Actittozoa, 182.

Reproductive organs, of Hydrozoa,
40 ; of Actinozoa, 134. 169 ; of Hy-
dridce, 23 ; of Corynidce. 88 ; of
Sertularidte, 93 ; of Calycophoridce,
99 ; ot Physophoridce, in ; of A/f-
dusidcc, 40. 118 ; of Lucernaridcs,
40, 47 ; of Zoantharia and Alcyon-
aria, 134, 169, 209 ; of Ctenophora,
169.

Respiration, in Ccelenterata, 6, 209.

Rettcularia, 91.

Rhabdophyllia, 239.

Rhahdopora, 245.

Rhipidogyra, 239.

Rhizangia, 240.

Rhizophysa, loi— 2, 107, no— I.

Rhizophysiadce, 114.

Rhizostoma, 49, 67.

Rkizostomidce, 49, 125.

Rhodarcea, 241.

Rce7neria, 244.

Rugosa, corallum of, 158, 173, 214;
families of, 215 ; relations of, to
time, 237, 243.

Saccanthus, 199, 200.

Sacculus, of tentacle in Hydrozoa, 33.

Sarcode, 7, n.

Sarcodictyon, 151, 160, 209.

Sarsia, gemmation of, 63.

Sclerenchyma, 161.

Sclerites, ot Gorgonidce, 161.

Sclerobasic corallum, 153 ; develop-
ment of, 185.

.Sclerobasic Zoantharia, 154, 162, 201,
234, 237.

Sclerodermic corallum, 155, 173, 186.

Sclerodermic Zoantharia, 158, 161,
17 J, 202, 234- 237-

Scyphistoma, 66.

Sea- Anemones, 131 162, 198.

Sea-Fir, go

270

INDEX.

Sea. Pens, i6}. aiz.

Sea-Shrubs, 154.

Secretive organ, of Cestum, 147.

Segmentation, of ovum, 15

Sense-organs, ol Hydroxoa, 39; of

Actinozoa, 166, 168.
Septa. 155 : development of, 17J.
Septal formulffi, J74.
Seriatoporidce, 205, zo6, Z32, 237, 244.

245.
Serous Liyer, of blastoderm, 16.
Sertularia, 90, 95. 127.
Sertulariadcc, 'j'j. 92,, 95.
Sertularidce, geniral characters of,
79, 90 ; development of, 56 ; cceno-
sarc of, 90 ; polypary of, 35. 92 ;
hydrotliCLae ol, 35, 92 ; polypites of,
92 ; tentacles ol, 9} ; nematophores
of, 95 ; gonoblasiidia of, 46 ; gono-
phores of, 44, 93 ; families of, 95 ;
distribution of, 126—8.
Silurian Corals, 243.
Size, of Cuelenterata,f)\ of Hydroxoa,

28 ; of Actinozoa, 140.
Smilntrochits, 246.
Stnithia, 244.
Solenastraa, 241.

Somatic cavity, of Ccelenterata, 3, 10,
17; of Hydxizoa, 17, 29. 31 ; oi Ac-
tinozoa, 17. 141, 144—7.
Somatic chambers, oi Actinozoa, 134,

146.
Somatic fluid, 6.

Somatocyst, of Cafi/cophoridir, 96.
Spermaria, of Hydruzoa, 40, 50 ; of

Actinozoa, 169.
Spermatozoa, 15; of Hydroxoa, 50;

of Actinozoa, 169.
Spkefiolrochus, 240.
Sphixronectes, 98. 99.
Sphceronectidce, 100.
Spicular coral lum, 160, 162, 209. 212.
Spicules.of v^/ryon/rftj-and Zoanthidce,

160 ; of Gor^onid(S, 161.
Spiral cnidfe. ijo, 151.
Spongophyllum, 244.
Sporosac, 40, 45.
Spurious ccenenchyma, 205.
Stauria, 214, 244.
Stawidce, 215, 257, 244, 245.
Stauridia, tentai-les of, 87 ; gono-

phores of, 88, 89.
Steenstrtipia, 63.
Steganophthalmata, 120.
Stelloria, 246.
Stenostomala, 219, 230.
Stephanocoenia, 239.
Stephanomia, 108. no, 112.
Stephanitminda;, 113.
Stcphanophyllia, 240.
Stereopsam u ia, 247 .
Stinging p(jwers, of Ccclenternta, 5.
iitomAch,oi Actinozoa, 132, 141, 144.
Stomachal filaments, oiLucernaridce,
IZ4.

Stomatodendrn, of Rhizostomid<s, 49

S/onit/brar/iium, fission of, 6j.

Strebla, 151

Streptelas'iia, 244.

Strobila, 66.

Strombodes, 244.

Stylaxis, 245.

Stt/iina, 239, 245.

Slylucdenia, 240.

Stylocyathus, 246.

Stylophora, 174, 240.

Styluphoridce, 207, 237, 247.

Stylosmilia, 239.

Successive development, of appen*
dages in Hydrozoa, 70 ; of mesen-
teries and tentacles in Actinozoa,
132, 172 ; of septa in corallite, 173.

Swimming-bells, of Hydrozoa, 27, 36 ;
of Caiycophoridce, 96—9 ; of Phyno-
phoridcE, 107 ; of Medusidce, 115 — 6.

Symphyllia, 241.

Synapticulae, 155, 203.

Syndendriiim, o{ Rhixostomidce, 49.

Syrihoelia, 246.

Syringophyllum, 238.

Syringopora, 238.

Tabulata. ic8, 161, 205, 237, 242,
Tabulffi, 156.

Tactile organs, of Hydrozoa, 40 ; of
Ctenophora, 168.

Tamoya, 126.

Tangled cnidae, 150, 151.

Tegnmenfary organs, of Hydroxoa,
34 ; of Actinozoa, 149.

Telestho, 159. 209.

Tentacles, of Hydrozoa, 32; of Ac-
tinozoa, T48 ; of Hydridis, 20 ; of
Corynida, 32, 85; of Sertularidce,
32, 93 ; of Caiycophoridce, 32, 33 ;
of Fhysnphori'dcc, 32, 33, III ; of
Medusidce, 52, 117; of Lucerna'
ridce, 32, 121, 125; of Zoantharia,
148, 199 ; of Alcyonaria, 148 j of
Ctenophora, 225.

Tertiary Corals, 247.

Thalassianthus, tentacles of, 200.

Thamnaslraa, 239.

Thaumantiadce, 120.

Tliauiiiantias, 63, 129.

Theca, 155.

Thecia, 244.

ThecidcB, 205, 237. 242, 244.

Thecucyathus. 246.

Thecosvitlin, 239.

Thecostcfiites, 244.

I'hread-cells. ol Ccelenterata, j ; of
Physalia, 5 ; of Hydra, 20 ; of /Ic-
tinozoa, 150

Tichastrcea, 247.

Time, relations to, of Hydrozoa, I30,;
of Actinozoa, 236.

Tissue secretion, 153.

Tracfiyneina, 6q, 117.

Trachynemidce, 119, IZO.

INDEX.

271

Trachypora, 144.

'J'riaSsic Coral*, 145.

Trichydra 82. 87, 89.

'I'ritozooiiU. 74.

Trochucyathiis, 239.

Trochophi/lluni, 245.

Trochusfiis, 247.

Trochosinilia. 239.

fubipora, 209.

Tubiporidce, 159. 162, Z09. 21 j. an-

TubtUarin, ji, ;2, 44, 54, 8j — 9, 127.

Tuhn/tinud^e, 89.

TubHlarilcE, 90,

Tiihtdaiir.a, 90.

Tubulosa, 159, 162, 204, 206, 2J7, 242,

^^^ .
Titrhmaria, 241.

Turbinolia. 247.

Turbniolidce, 202, 207, 2jj, 2J4, 237,

246, 247.
Turns, 62.
Turn'topiis, 61.
Tvpe, of Hydrozoa, 20; of Actinozoa,

iji

L'locyathHS, 232.

Ulophyllia. 239.

Utubellularia, 235.

Umbrella, of I.ucernaridte, 27, 37.47,

124.
Urticating organs, of Coclenterata, 3.

Vacuolation, of tissues, 8, 32.
Vegetative repetition, 73.
Veil, of nectocalyx, 36, 116.
VeleUa. form and structure of, 104;

development of, 59 ; pneumatocyst
of, 102, 105 ; tentacles of, 32, III ,
polypites of, 105, 110, 114; gono-
phores of. 45. iiz — 3; glaiKiularsacs
of, 106 ; distribution of, 127.

Vtlellida, 114,

Veretillum, 212.

Vesicles, of Medusidce, 37.

Villi, of polypite, 31.

Virgularia, 212, 241.

Vital endowments, of lower animals,
10.

Vitelline membrane, 15.

Vogtia, 45, 99.

I'orticlava, 82, 85—7, 89.

Warts, of Sea-anemones, 149.
Willsia, canals of, 116.

Yolk, 14.

Yolk-division, 15, 51, 170.

Yolk-sac, 15.

Zaphrentis, 214, 238

Zoaiitharia, ueneral characters of,
139, 169, 196; polypes of, 131, 141,
197; tentacles of, 148, 199; thread-
cells of, 150; corallum oi, 158—62,
173, 186, 201— 5 ; free-swimming
forms of, 163 ; families of, 201, 205 ;
development of, 170 ; distribution
of, 232—6; relations of, to time, 237,
242.

ZoanthidcE, 162, 199, 201, 206.

Zonnthus, 160, 235.

Zooids, 74.

LONDON: PKINTED BY

6F0TTISW00DE AND CO., NEW-STREET SQUABB

AND PAELIAMENI STEEET

»L375 .G7 1869

, ,M, I nl III, smI. kl

larvard MCZ Library

III

3 2044 062 441 894

Return this book on or before the last
date stamped below

r '^ 2 S 1969
m 1 7 1973

Cat. no. 1174

■1?;^ ■■«